The Refracted Light

Blue Cameras Take the Best Pictures

2012-02-01T14:50:00.000-06:00

Click the link for something funny: Ah, Managers

Composition, Part 1 - the Frame

2012-01-27T00:43:00.000-06:00

I MUST ADMIT that the technical aspects of photography are easiest for me. Color spaces, exposure and lens calculations, f/stops, shutter speeds, and ISO sensitivity are generally objectively certain and quantifiable. On the contrary, artistic considerations such as the use of color itself and composition seem to be subjective, qualitative, and much less certain. However, we must not oppose technique with art: they are not two things, but are different aspects of one thing, and they both must be taken into consideration when making a final image.

When I became serious about photography a number of years ago, I didn't give composition too much consideration, simply due to the fact that I was taking mainly architectural photos:

Saint Francis Xavier Church, at Saint Louis University, in Saint Louis, Missouri, USA.

The hard work of composition was already done for me by the architect. I merely had to discover good camera positions and angles, and the kind of post processing that would express the work of the architect in a pleasing manner. Fortunately, these discoveries came rather quickly to me.

Likewise, I found it easy to take pleasing photos of flowers:

Flower, at the Missouri Botanical Garden (Shaw's Garden), in Saint Louis.

Flowers are intrinsically interesting, and nature suggests composition.

But I always had a problem with landscapes. So many times, I'd take some photo of a scene that I thought was interesting, but my end results were usually dull:

Bottomland forest, at Castlewood State Park, in Saint Louis County.

With my eyes, this scene pleased me enough that I took the time to set up my tripod, take a series of exposures, and then I spent significant time on the computer attempting to make this image look good. Now when I view this, I cannot imagine what I thought I saw when I was in the forest. This appears to be a waste of my time.

A while back, I was hired to photograph a book on the public parks in my city. This caused me no small amount of anxiety; now, my publisher must have thought that I was up to the task, and many of the photos were to be somewhat like this:

Bandstand, at Tower Grove Park, in Saint Louis.

As this is an architectural landscape, much of the composition was already done for me by the building and landscape architects, and with some judicious cropping, this ought to become a pleasing image. But what about a natural landscape?

Looking back at the dull forest photo above, I knew that something interesting must have been in that image, otherwise I would not have spent so much time on it. But why does this image fail to show this interest? I conclude that either the post-processing is wrong, or more likely, the composition of the final image is wrong.

I am not so naïve to believe that all I need to do is to learn rules of composition, which will automatically produce pleasing images. But neither am I satisfied the advice that I ought to simply adjust my image until it looks good to me. What if all my adjustments are unsatisfactory? Why are they not satisfactory? But I was in the forest, and what I saw from my camera position pleased me. How do I discover a pleasing composition in this image?

It is the goal of this blog to provide objective information on photography. In my very first article here, I wrote:

…I intend to write about the more certain, objective factors in photography, things that are less a matter of personal taste and legitimate varying opinion. The laws of physics are objective, the operation of specific cameras is in principle knowable, but even some subjective factors can be known objectively to a good degree, such as the properties of vision that are fairly uniform among persons…

So how can composition be objective? Is this not something that is completely subjective, varying between individuals? Yes and no. One compositional element is absolutely objective, and this element is a basic element of composition in photography.

Please consider that nearly all photographs are rigidly bound by a fixed frame. For nearly all images, we can determine with high precision what precisely is a part of the photograph and what is not a part of it. All of the photographs above are rectangular, being at most 500 pixels wide. The fixed frame is therefore the basic element of composition, or at least is the most objective element. It is the most certain, most rigid, most obvious, most purely geometrical part of a photograph. The frame defines and bounds the photograph. Please note that I use the term ‘frame’ here to denote the outside edges of the photo, but picture frames also serve the important function of making a distinct, yet broad transition between the image and the wall.

Also note that all consumer digital cameras produce rectangular images. This is significant, but I won’t go into it further here.

[On the contrary, irregular cutouts are more ambiguous, as are feathered edges that fade to transparency. These effects are used at times, but largely in designs where a photograph is a part of a larger composition. Occasionally, an image element will break out of the frame, but this is typically used sparingly as a gimmick. But the frame is fixed for the vast majority of casual or serious photographs.]

OK, so the frame of an image has an undeniable influence over the content of an image. You may re-crop an image in a large number of ways, and every crop excludes a different part of the image. What are we to make of this? Here are some possible opinions:

Every crop of an image is equally valid. Every frame is as good as any other.
You merely have to crop an image so that it looks good to you.
If you fill the frame of your camera when shooting, then you don't need to crop.
You should crop your images so that the ratio of the long to short side is equal the golden mean φ (phi), or 1.61803:1
You should crop your images to fit the standard sizes of the media used; for example, 4:3 or 16:9 for television or computer monitors, 5x7 or 8x10 inches for prints, or the ISO 216 series of standard paper sizes.
You should crop your image according to the Rule of Thirds.
If you don't compose your image according to the Canons of Proportion, then you won't have a good image.
Various systems of harmonic proportion have been developed over the millennia. Certain ratios have long been considered pleasing, and others displeasing. An image ought to be cropped to one of these ratios, such as 1:1, 2:1, 3:2 or 4:3.
The subject of the image and its internal composition is more important than the outside edge of the image. The crop ought to serve the image, reinforcing the compositional elements within the image itself.

OK, if rule #1 is true, then you can stop reading now. It doesn't matter what you do. But that is ridiculous. It is total skepticism and completely ignores the order in the cosmos as well as in the human person. ‘Whatever’ has never been understood as a good rule for art.

The subject of Edvard Munch’s The Scream was driven insane by bad theories of art.

Regarding #2, why do you like any given crop? Do you know why, or does a good crop just feel good? According to ancient Greek philosopher Plato, “An unexamined life is not worth living.” How can you expect to grow in your art unless you can intellectually discern why you do one thing instead of another? I am not saying that this kind of discernment is easy, only that it is highly recommended in order to perfect art.

I cropped the photo this way because that’s the way I liked it. Certainly, I wanted to show the leaf margins on the right and bottom, but why did I leave as much room as I did? How did I determine the top and left crops? This photo was taken within the Climatron greenhouse at the Missouri Botanical Garden in Saint Louis.

In rule #2 is the notion that the artist only has to please himself. But what if an important client requests another crop, different from your preferred one? Would you rather starve than compromise your artistic vision? Is this a battle worth fighting for? Have you considered that your client may actually have a better artistic vision than you do? Why don’t you instead produce an image that is robust enough to survive a wide variety of crops?

#3 is often claimed that filling your cameras frame is the most efficient use of pixels, and certainly if your subject takes up more of your camera’s frame, then you will likely have better image quality. However, there are two limitations. First, very tight framing limits your cropping options, particularly if you have to make a standard size print that may not have the same aspect ratio as your camera. For example, my publisher often crops my images severely when my framing is too tight; better compositions may be had if I do looser framing. Second, the aspect ratio of your camera is likely not ideal for every subject: many photographers prefer an image taller than it is wide when doing portraits, and likewise produce an image wider than it is tall when doing landscapes. Square format is considered to be particularly powerful, but it would need to be cropped in these circumstances.

This is a very tight composition, and would not, to me, be satisfactory if recropped as an 8x10 image.

On #4, or the Golden Mean or Golden Ratio. First, it is generally understood that not all frame aspect ratios suit all images. When I’m taking photos of flowers that show some degree of radial symmetry, as seen in the flower image above, I often find that a square aspect ratio is pleasing for the subject. An aspect ratio of approximately 1.6180339887:1 can’t always be optimal.

An approximate golden rectangle; 500x309 pixels in size, giving an aspect ratio of about 1.618:1.

Furthermore, phi is an irrational number, and its decimal digits go on forever, and so that isn’t well suited for digital photography where we deal with discrete numbers of pixels. But this is a petty complaint. It is claimed that a golden rectangle has the most pleasing aspect ratio of all rectangles, but it is hard to find data that substantiates that claim, and anyway human eyes aren’t good at making precise measurements.

A practical consideration is that cropping your images to this aspect ratio makes printing and framing your prints more expensive and time-consuming.

It is claimed the golden mean has remote antiquity, and was used as an aesthetic number by the ancient Greeks. There is no evidence of this. Where the number is definitely found in antiquity, it was of mathematical use only, and only incidentally found in art, since phi does appear in some geometric constructions such as the pentagram and the Platonic solids. We can be pretty certain that phi, before the modern age, was never used directly as an aesthetic proportion.

You can find an extremely critical article against the golden ratio here.

By the way, the Greek philosophers did use the term ‘golden mean’ but only in discussions of virtue. According to the philosophers, a virtuous action is the ‘golden mean’ between two vices; for example, justice is the golden mean between leniency and severity, and courage is the mean between being a coward and being foolhardy.

#5 is practical advice. If you are selling prints, it is prudent to size and crop them to fit within one of the standard print sizes. Because frames and mattes are pre-cut to a series of fixed sizes, you can save much money and time by doing a standard crop.

But herein lies the problem. Who says that that standard print sizes are going to work with your images? Some compositions are very tight, and if you have a pattern of repeating elements in your image, a bad crop can make it look disharmonious. Also, you probably don’t want to crop out the top of the bride’s hair just to make it fit within an 8x10 inch print. Finally, putting a non-standard image size in a standard frame — where there are gaps outside of either the long or the short edges of the print — looks very unprofessional.

Pinkie, by Thomas Lawrence, badly framed.

Its quite frustrating, but many of the standard American and Japanese print sizes have different aspect ratios; so if you want to produce a small, medium, and large standard print, you have to crop them differently for each size. The three most standard print sizes, 4x6, 5x7, and 8x10 inches all have different aspect ratios, and so different crops are required. Supporting all of these print sizes means that you have to shoot loosely and compose your images so that no one of these three crops looks bad, which may be difficult to do well with some subjects.

Mona Lisa, cropped to the 8x10 inch aspect ratio. The original is approximately 3:2. I am pretty sure that Da Vinci neither put a code nor phi in this painting, and I am positive that an 8x10 crop doesn’t help it.

However, I might mention that any given standard print, matte, and frame combination is usually harmonious. These sizes aren’t arbitrary.

Contrary to the Japanese-American standard print sizes, the ISO 216 standard uses a fixed aspect ratio for all sizes. Different cropping isn’t needed for each size. But the aspect ratio is irrational, with the long size being the square root of 2 times the size of the short one (approximately 1.4142135623731:1). As with the suggestion that the golden mean be the aspect ratio (see #4), this may be misguided. This system gives no choice of aspect ratio in order to harmonize with the subject, nor has this ratio ever been widely considered to be aesthetically pleasing. In practice, the numbers for the paper sizes are rounded off to the nearest centimeter, with different manufacturers rounding off the numbers in various ways. By the way, the ISO 216 standard was designed so that the next size is half that of the one that came before it, so an A3 print is half the size of an A2 print. The standard includes an A, B, and C series, each having a different starting point so as to allow for a larger number of standard sizes, closer to each other.

The alternative to using standard sizes is to crop your images to whatever aspect ratio serves your composition. If you are making images for display on the Internet, there isn’t anything else you need to do besides resizing them.

If you are making a print, you can make a custom matte and frame to harmonize with it. Custom-cut frames and mattes are quite expensive in retail stores, and if you aren’t watching the person cutting them, and help them with the measurements, you might very well get the wrong sizes. Or you can do this yourself, if you are good with cutting tools, but still this isn’t cheap. Although some sort of framing is almost always needed for art prints, be aware that some fine art photography in the past was displayed without frames, tacked on cardboard. Whether or not your spouse or your client will buy this, however, is not for me to decide.

All Star game, in Saint Louis, in A.D. 2009. The 4:3 image at top is stretched to 16:9 on the bottom. Never, never, never, never do this.

Ah, the famous Rule of Thirds, rule #6. Is there any other rule of composition? This one gets repeated constantly. This rule states that main compositional elements ought to be placed one-third of the way between the edges of the photo. I wrote about that rule (see the article Rule of Thirds?), and I do sometimes use it, if I think it makes for a stronger composition.

Many folks say that this rule should be followed in all photos, but when someone claims that it doesn’t work, then they say “rules are meant to be broken”. Ugh. No no no no no. You do not tell a police officer, who is arresting you, that “rules are meant to be broken” nor do you tell that to the judge. You do not break the rules of gravity. Either it is a rule that has significant real consequences if it is broken, or it isn’t a rule at all.

Ancient proportional systems have very many “rules” but in such a wide variety that they aren't rules at all, but rather are harmonious design guidelines. Like painters’ color palettes, the musical scales, or poetic forms, you have a broad freedom to select whatever system that works for your composition. If you are really good, you can devise your own systems. Based on the classical harmonies, I could say that this or that composition follows the “rule of unities”, the “rule of halves”, the “rule of fifths”, the “rule of octaves”, the “rule of increments”, “the rule of inverses" and dozens or hundreds of other so-called rules. Adding harmony to your image is usually accepted as a way of improving it. What particular harmony to add to an image can be a problem.

Flooding on Smallpox Island, near Alton, Illinois. I explicitly used the Rule of Thirds here, as well as the Rule of Halves (which is a rule I made up). I got a number of positive compliments about this composition.

Rule #7: if you don’t follow the canons of proportion, your result will be bad. The Age of Enlightenment was also the Age of Science. Prominent scientists and philosophers determined that certain artistic rules were good for the visual arts, and so academia taught artists to follow these rules. This process was accepted in the 18^th century and it is accepted today, although of course the rules are different. What was absurd was the claim that if art did not follow such rules, it couldn’t be good; also absurd would be the claim that any art that followed those rules must be necessarily be good. In juried competitions, excellent paintings were rejected only because they violated some unreasonable rule or another. Please remember that an unjust rule isn’t a rule at all.

While the state-sponsored academies were perhaps wrong in limiting artistic latitude so strongly, the opponents of these academies were wrong in claiming that there were no rules, that freedom must be total. One of the most absurd claims by the Impressionists was that composition of paintings was not necessary, because nature has her own composition. If this is true, then why is my forest photo above so bad? That there is an intrinsic order to nature cannot be denied, but this order usually exceeds human reason.

An example of these kinds of artistic rules can be found in the “Classical unities,” inspired by The Poetics by Aristotle; these rules state that a tragic play ought to have one plot, one location, and take place over the course of not much more than one day. But Aristotle himself, in the same book, didn’t say that his rules were inviolable; he simply said that the best tragic plays of his day conformed to some rules, and that certainly tragedy could be further improved.

As a general principle, never confuse “is” with “ought”. The great Gothic Revival theorist A.W.N. Pugin said that the Gothic is the greatest Christian style of architecture, worthy of imitation; he also said that some other style in the future could be even greater.

I reject this painting, The Oath of the Horatii, by Jacques-Louis David, because it invokes Godwin’s Law.

Unlike the Modernists, I see no problem with working within the old classical harmonies and canons of proportion. For example, Byzantine iconography has seen a great resurgence, and there are artists who are trying to recover the canons of Gothic painting. Photography is certainly an art form of the Modern period, but that does not limit me from appreciating art in the Western classical tradition, nor prevent me from incorporating these principles into my art.

Classical harmony, #8, is a topic near and dear to my heart. It is certain that many of the great architectural achievements of the past were explicitly constructed using these harmonies. David’s painting above depicts three arches made with harmonious proportion.

We have no evidence that the ancient Greeks or the Gothic cathedral builders used the golden mean as a proportional ratio, and the square root of two was used only as a consequence of that number being the diagonal of a unit square. Rather, we do know very well what kind of proportions in art were preferred by them: ratios of small whole integers. Derived from the musical experiments of Pythagorus and commented on by the philosophers Plato and Aristotle, the theory of numbers and their harmonic proportions was preserved and elaborated during the Medieval period, and served European civilization until Modernism. This number theory, coupled with the geometry of Euclid and his use of the compass and straightedge, led to some of the most awe-inspiring buildings in history. Classical harmony is also the foundation of music, and ought to be seriously considered by every artist.

See the article, Harmonious Proportion and Ratio, for an overview on the classical harmonies.

#9, the value and importance of the frame.

Let’s face it. Photography is an inferior medium. We do not have the great control over our subject as do painters or architects, and even if we use Photoshop to its fullest extent, we end up with digital art rather than photography. Now if I had a photo studio, perhaps I could compose my images precisely and harmoniously, but instead if I’m taking photos of landscapes, I have to take what I see and work from there. I simply cannot compose my images as well as I would like. Photography is a powerful medium, but composition is not one of its strengths. Ironically, much contemporary compositional theory was developed by photographers as they attempted to improve their medium.

There are some situations where framing can harmonize precisely with an image, for example, see the following photograph of metal panels, found at an exhibit at a car show:

The proportion of the sides of this image is 3:2, and this harmonizes with the regular metal pattern found in the image. If I didn’t keep the camera level, or if I cropped it contrary to the pattern in the metal, this would likely be a poorer image.

If your subject has a strong symmetries — like the church in the top photo above — then certainly do your best to conform the frame to the symmetry expressed in the subject. You are given a good subject, and so give it a good frame. Make the frame harmonize with the subject.

But many subjects aren’t so tractable. For a blunt — and very crude — list of things that make bad photos, see 7 things you really don’t need to take a photo of. Included in this list of seven bad subjects is scenery.

We are often told that “beauty is in the eye of the beholder,” that there is no artistic method that will lead to beauty. However, many of these theorists also praise works that are “jarring” and “subversive,” and so by implication there are principles of disharmony and ugliness that these artworks must obviously follow. What is certain, it is impossible to produce a work of art that follows no rules or principles at all. See the article “The Tropeless Tale” for a humorous refutation of this nihilism.

A problem with landscape photography is that we usually aren’t provided with regular patterns. While we do find these harmonic patterns in flowers, leaves, and animals, general landscapes have a logic that is likely indiscernible by the human intellect. So, in landscapes, what exactly is the frame harmonizing with? Do not forget, you always have a frame. Does the frame fit the subject? Quite a problem, but some landscape photographers seem to produce good compositions at alarming frequency.

Perhaps the very best landscape photographers select those subjects that they know will produce good photos, by intellectually analyzing the scene before them, and not allowing their instincts or baser passions to influence their decision to take the photo. There is a huge difference between feeling that a scene will make a good photo, and knowing that it will. If you have a prime subject, say, a mountain surround by a featureless plain, then your composition is easier, and multiple kinds of harmonious crops may be adequate. This kind of analysis does not have to be labored; I think that long study and extensive practice can make this kind of scene analysis second nature.

Perhaps the problem is with human perception. What do our eyes concentrate on? Perhaps in my forest photo above, I saw some tiny detail that is now lost among the leaves. Perhaps I was influenced by the buzzing insects, the humidity in the air, the chirping of birds, and the rustling of leaves when I had my camera in hand. Perhaps I can’t make a good crop of that photo.

[I ought to note that beginning landscape photographers often think that they must use the widest angle lens that can find, to “get the whole scene in”; but professionals say that a telephoto lens is often better, to focus on a single interesting detail in the scene.]

Our vision picks out patterns, or even imagines some pattern where none exists, like seeing faces and animals in the shapes of clouds. We look for order and we analogize what we see with what we already know, which is sometimes called the ‘associative imagination.’ This is essential for survival: if something appears dangerous, even marginally so, it is prudent that we avoid this imagined, and potentially actual, danger. We are also drawn to good things by a slight glimpse of a shape or a subtle curve in the distance, or a distinctive color. Photography and all of the representational arts would be impossible without the associative imagination: we can recognize a portrait of someone famous, or a loved one, even if it is highly abstracted. I cannot imagine how vision would be useful if it weren’t for the associative imagination. Perhaps a good image reinforces our imagination, and makes it more concrete, by abstracting away distracting details, leaving us with our primary associations.

I think this associative imagination is likely one of the reasons why abstract painting of the mid-20^th was so highly disliked and mocked by the general public. If an artist intentionally paints something that appears if it may have associations — but does not — then this kind of art is a cruel joke or is dishonest if it is intended to be displayed to the public. Of course, this kind of art was originally intended to be revolutionary — in the art and in the political sense. But this does not mean that all abstract art is destructive: decorative geometric patterned art is also abstract, but has been wildly popular throughout history, and it is this kind of art that harmonic proportion reigns supreme. See the article Abstraction and the “Grammar of Ornament”.

I think that good composition in photography is the intersection between realism and the kind of geometric abstraction found in the decorative arts, and the ubiquitous frame is the most abstract element in images. But investigating this will be left to a future article.

Photoshop Wishlist

2011-11-30T15:35:00.001-06:00

When I apply noise reduction in Photoshop — no matter what technique I use — the apparent saturation decreases. Why is this? How can this be avoided? I don’t want to merely add saturation back in, but avoid it. I’ve noticed the same thing with small-sensor digital cameras, those that do lots of noise reduction in-camera.
Are there any good techniques to get rid of the brightly colored noise found along edges? This is due to demosaicing the Bayer array of the sensor. Eliminating this noise usually leads to desaturation of the good pixels.
Photoshop does not resize images well, and often generates interference patterns. Lots of research has been done on these kinds of algorithms, and it would be good to see these better solutions in Photoshop. Also, fractal resizing has many benefits and would be valuable to have also.
I’d like to be able to fully edit ICC profiles in Photoshop. For example, I have a profile that uses the GRACoL ink colors, but has other parameters I’d like to change. The Custom CMYK interface is rather lacking.
When you use curves in RGB, you can either do it with the Normal blending mode, which typically causes an increase in saturation, or you can do it with Luminosity blending, which decreases saturation. How about a simple method which does neither? I just want the tonality to change, not the basic coloration. Click here to see how it is done.
A solid method of adding local contrast. ‘HDR toning’ does this a bit, and a lot of other things at the same time too, making it less useful. ‘Clarity’ is pretty good, but tends to produce halos also. There are better methods.
A ‘Vibrance’ tool for CMYK. This increases saturation without causing any channel to blow or plug.
I’ve noticed that there is a distinction between chroma, colorfulness, and saturation; not really sure how or what Photoshop does. A solid colorimetric model would be useful.
More color spaces. XYZ and xyY would be useful.
Using Munsell colors in the color picker would be great, as well as better out-of-gamut warnings.
Better statistical color tools. Something that can show all the colors in an image as a three-dimensional solid would be great.

Yeah, I know. All of this development is expensive, and it is easy to criticize!

Using ICC Profiles for Creative Color Control

2011-11-28T23:45:00.000-06:00

COLOR MANAGEMENT has become far easier since the introduction of standard ICC profiles, developed by the International Color Consortium, a group of various computer, digital camera, and imaging companies. ICC profiles describe the color handing characteristics of cameras, scanners, printers, and computer monitors, as well as allowing for uniform methods of converting the colors from one device to another. For example, ICC profiles are used to convert the colors captured by a camera to those colors available on a printer.

ICC profiles define a mapping from the typically limited color spaces of devices into a universal color space (which can describe all colors), such as CIELAB or CIEXYZ; from these universal spaces, the colors can then be mapped to an output device. As such, ICC profiles recognize limits of what is possible with any given device, and allows us to compare the capabilities of one device with another. If an output device, like a computer display or digital printer, cannot show particular colors, then the software will use the ICC profile to change the colors to the closest ones available (this will often cause the image to lose detail or saturation).

ICC profiles are also stored in image files, thereby interpreting the colors of the file, and these profiles are used during image editing. Typically, JPEG image files delivered by cameras are encoded into standard color spaces which are independent of the particular make and model of camera. The most common color spaces are sRGB and AdobeRGB; these provide a wide enough gamut of color for practical use, while being compact enough to produce reasonably small file sizes. ICC calls these "three component color encoding" standards. Using these standards are convenient because we can use JPEG or TIFF images from a wide variety of cameras without worrying about the specific characteristics of each. For example, an sRGB color value of (255, 0, 0) will describe the same bright red color, no matter where the image came from. Some software, including web browsers, actually assume that all image files are encoded in the sRGB standard.

Click here for an overview of RGB color spaces.

I did not create this image. Click here for source and attribution. This shows the relative color gamuts of various RGB color profiles, as well as the color gamut of the Epson 2200 printer.

Each of the standard RGB ICC profiles defines three primary colors: particular shades of red, green, and blue. A narrow standard such as sRGB can describe about 35% of all colors, and so will have primary colors that aren't quite as bright, colorful, or saturated as the Wide Gamut RGB standard, which can describe about 77.6% of all colors. The ProPhoto standard is even bigger and has mathematical primaries that aren't even real colors. However, be aware that most computer monitors and digital printers cannot display or print color much beyond the sRGB standard, and so sRGB is generally recommended for most uses on the Internet and for consumer-grade printing.

So ICC profiles have these uses:

Describing the color capability of a device.
A standard method of converting the colors from one device to another.
Describing the colors in an image file such as a digital photograph.
A widely accepted color standard for cross-platform image interchange.

May I also propose another use of ICC profiles?

Creative use. Artists can use ICC profiles to precisely manipulate and control the gamut of color in images.

In his book Professional Photoshop (5th Edition), Dan Margulis described the use of 'false profiles': using ICC profiles as an alternative method of image manipulation. He used this method to greatly brighten shadow detail without modifying the color: this is difficult to do with the standard Photoshop tools.

Estimating the colors of Autochrome

Now, while I made note of the false profile method, I rarely used it nor did I fully appreciate the wider potential of this method. But some months ago, I became interested in the old Autochrome Lumière method of color photography, released in the early years of the 20^th century. These color slide plates produced somewhat muted colors, but the color palette produced was beautiful and lacked the often garish colors found in later slide films:

Couple, by Mrs. Benjamin F. Russell, ca. 1910; George Eastman House Collection. Source.

Woman in floral silk robe, by Charles Spaeth, ca. 1915; George Eastman House Collection. Source.

Dancer wearing Egyptian-look costume with wings reaching to the floor, unknown photographer, ca. 1915; George Eastman House Collection. Source.

I wanted to reproduce the colors produced by Autochrome, and thought that if I could determine the primary colors used by the Autochrome process, then I could create an ICC profile that could produce digital images with the approximate color gamut of Autochrome.

By collecting together a number of Autochrome images taken from the Internet, I was able to get a very rough idea of the range of colors found in the these images.

Now there are a lot of problems with my method:

Did Autochrome fall within the sRGB color gamut? Is the full gamut of color captured by these reproductions?
Are these direct photographs of Autochrome slides, or are these scans of printed reproductions?
Are these photos a representative sample of Autochrome?
What color of light was used when these slides were exposed? What color of light was used when these slides were digitally photographed? What white balance was used by the camera? How accurate is the camera color rendering?
How much have the color dyes faded with time? How much variation was there in the Autochrome process?
Autochrome is made up of small, visible colored grains; how do I reproduce the pointillist-like appearance of the method?
So forth and so on.

Now I can make some observations and educated guesses. Many Autochrome images on the Internet do appear to be faded, so I generally used the most saturated and brightest examples I could find. No image had colors that came close to any of the sRGB primaries, and so I thought that this method had some validity. Clearly, blues and cyans must be highly muted, while greens are somewhat better. Red appears to be impossible to capture, while pink, peach, and some unsaturated yellows and yellow-greens are quite prominent. Orange often appears to be blown. Black and white are well-represented, although the white balance often seems to be off.

According to some authorities, the Autochrome process used orange, green, and violet primary colors; based on my inspection of these derivative images, I am convinced that orange and green are correct, but I'm not so sure about violet; it is only marginally violet at best, and unsaturated. Based on these primary colors, we ought to get pink, dull yellow, and dull cyan as secondary colors, and we do see these as prominent colors. Orange is often blown in these images (that is, it lacks detail and texture), something that I don't regularly see with any other saturated Autochrome color; this leads me to believe that the orange primary color is outside of the sRGB gamut. I also looked up the chemical dyes used in the Autochrome process, and those used can produce these primary colors.

Using the ImageJ software package, I was able to produce statistical plots of the colors found in the example Autochrome images.

Palace of Horticulture, Pan American Exposition, unknown photographer, 1915; George Eastman House Collection. Source.

Here is a CIE xy chromaticity diagram of the colors in the image above:

The chromaticity diagram is a standard method of showing color gamut. All of the lightest pixels of any given hue and saturation are plotted. The bright colorful pixels that form a triangle around this diagram are the primary, secondary, and tertiary colors of the sRGB color space: the primary colors are blue at the bottom left; green at top, and red on the right. See the chromaticity diagram, above on this page, which shows various standard color spaces.

Whenever we are mixing together three primary colors of light — additive color mixing — the chromaticity diagram will always be contained within a triangle, with the primaries being the corners of the triangle. Notice that the Autochrome colors also roughly form a triangle. We ought to expect that the corners of this Autochrome triangle are the primary colors of the Autochrome process.

Note that the blue corner of the Autochrome is well within sRGB, but the line between orange and green is right up against the sRGB gamut line. Now perhaps the Lumière brothers were precient way back in 1904, anticipating the ITU-R BT.709 standard of 1990 and choosing primary colors compatible with High Definition Television, or perhaps (more likely) the Autochrome gamut exceeds the sRGB standard. As it happens, I've found some Autochrome images stored in wider gamuts than sRGB, and indeed the orange color is beyond what sRGB can represent.

Now, if all my example Autochrome images had chromaticity diagrams bound by the same triangle, then my work would be done, and I could confidently establish the Autochrome primary colors. Rather, the chromaticity diagrams are variable:

Be aware that since the example Autochromes were digitally captured, the camera itself, plus any processing done on them could alter the final colors. The colors would be different if these were scans of printed Autochromes, or if the dyes faded, or if variable illumination was used to capture these images, or if these images were over or underexposed.

But in many cases, orange is likely blown, and blue is nowhere even close to the sRGB blue primary. These images do not approach pure green and they never show pure red. The Autochromes all generally fall within a particular limited range.

If we can assume that the colors used in Autochrome were uniform, then we can select primary colors that would pretty much give us an 'Autochrome look' to images. Be aware that it is not my intention to precisely characterize Autochrome, for that would be a large university- or industrial-level multidisciplinary research project, and while I would be happy to accept a grant for pursuing this research further, I really just want to make pretty pictures. I want to choose primary colors that are good enough, that give my images the right 'look’.

I went about estimating the Autochrome gamut by two methods. First, I got estimates from the color gamut diagrams, as seen above. Stacking the images in ImageJ, I was able to get both the total and the average gamuts using the "3D Project” tool:

Note that since these images were in the sRGB gamut, we have an outside triangle made up of accidental colors: this is the sRGB gamut triangle. These are due to calibration colors I inserted in the images, as well as artifacts due to low-quality JPEG compression and variation in digital capture. However, we can clearly see that the example Autochromes do not fill the sRGB color gamut, and likely exceed it along the red-green axis. Interestingly, note how the line between orange and green slopes in the bottom diagram: I suspect that orange is farther out of the sRGB gamut than the green primary, and that even possibly that the green primary is within sRGB.

To get a good Autochrome look, we probably need to go with a tighter triangle rather than a looser one; also, there is evidently very much variation in the digital captures of these images. Some images have strong blues, but most do not, even where we would expect to find them. I would think that in those cases, the blue colors come from the digital camera or post processing, and not from Autochrome.

I did an alternative method of estimating gamut. For each chromaticity diagram, I plotted the smallest area triangle that would contain all of the colors in the image. Then, I averaged together all of the gamut triangles, and took the resulting area that encompassed roughly 70% of the total gamut: it formed a very neat, regular triangle, which leads me to think I’m on the right track:

I then plotted these diagrams on a grid, allowing me to read the approximate chromaticity numbers for each primary color:

We can read the CIE xy numbers off of this grid, and create an ICC profile estimating the Autochrome gamut. Here I show two gamuts — both surrounded with red triangles — one shows my total estimated Autochrome gamut, which exceeds sRGB, and another one is bound to sRGB.

Please note that while I worked to estimate the Autochrome gamut, I did nothing to determine a white point for the medium, nor did I do any work to reproduce the pointillistic or impressionistic effect found in Autochrome slides due to their colored film grain, nor did I attempt to reproduce the tonal response or gamma of Autochrome. I only attempted to find the primary colors used in the process.

Using ICC profiles for creative color control

I created two ICC profiles. Click the following links to download:

You may also be interested in an older Autochrome estimate of mine. It is smaller and has a blue primary that is less violet than my newer one:

Old Autochrome

Put the files into whatever folder where Photoshop or your software accesses ICC profiles. On my Mac, it is ~/Library/ColorSync/Profiles.

The Autochrome.icc file uses Autochrome-like primary colors while remaining in the sRGB color gamut; you can use this to process images for display on the Internet. The "Autochrome Wide.icc" file has a gamut that exceeds sRGB; while the extra colors typically can’t be displayed on most monitors, they may be able to be printed, depending on the gamut of your printer: I wouldn’t suggest using the Wide gamut if you intend to display the images on the Internet.

Once the profile is in your system, you can edit image files with them. Using Photoshop, your image colors, no matter how much manipulation you do to them, will remain within the Autochrome gamut.

Please be aware that these ICC profiles are editing profiles, not device profiles, and are not what the ICC calls a "three component color encoding standard,” for it isn’t a standard at all, nor does it characterize any device. Only use this profile while editing your file, do not display files with these profiles on the Internet, and be careful when making prints from files with these profiles assigned. Far too many web browsers or consumer grade printers assume that image files are in the sRGB gamut.

Here is how these profiles are used:

In Photoshop, do either Edit->Convert to Profile or Edit->Assign Profile to use the Autochrome gamut. You may notice that your colors change either dramatically or slightly.
Edit your photo as you normally would edit it— adding curves, sharpening, saturation, vibrance, whatever you want. Your image will remain in the Autochrome gamut.
When you complete your edits, use Convert to Profile to put the image back into a standard color gamut: most typically, you’ll want to convert to sRGB for maximum compatibility. Again, this is an important step, for these ICC profiles are intended only for editing files, not for display.
The images may be a bit dark, and you may want to brighten them after the conversion to sRGB.

Please note that these ICC profiles only change the primary colors used. If your image lacks much color and is already within the Autochrome gamut, you may find that converting to Autochrome does little or nothing:

The image on the left is the original sRGB, the one on the right was Autochromed. Since the gamut of this image is rather small, there is little or no apparent change in the image. However, the following image shows a striking difference:

The weak Autochrome blue is quite evident, and it shifted to violet. Also, much of the detail on the tree on the right is now lost, due to the color being driven out of gamut.

Please note that we can either Convert or Assign a profile. This has a distinct difference:

Converting to a profile causes the pixels’ color values to change; if a color is out of the new profile’s gamut, it will be adjusted, often causing loss of texture and detail. Typically, any color that is already in the new profile’s gamut will not be changed visually, but the numbers used to describe that color will change. For example, the red sRGB primary color is (255,0,0), but the same shade of red in ProPhoto is (179,70,26).
Assigning a profile does nothing to the pixels, but they are interpreted according to the new profile. An sRGB value of (255,0,0) is interpreted as a the pure red primary color; after assigning it to Autochrome, the value will still be (255,0,0), but instead will be interpreted as being the Autochrome primary orange color. This technique will change your colors, but is the quickest way to make an Autochromey-looking image. This also has the advantage of preserving all of the original detail and texture in your image.

Here are both methods:

Note that Assign kept the texture in the tree, but did change all of the colors, as well as making them less saturated. Be sure to convert your images back to sRGB when you are done processing them.

Using orange, green, and violet primary colors means our color wheel is rather unfamiliar:

The colors in this color space aren’t as brilliant as those found in sRGB, but you cannot get the same effect simply by desaturation. Any color space becomes quite distinctive when you push it to the limits, when you try to represent something that is near or exceeds the bounds of what is possible. If you push this color space too much, you will get large areas of orange and pink and violet, which is the Autochrome ‘look’.

However, I must admit that after playing with a number of images in this color space, I think that I made it too broad, and the primary ‘blue’ color too violet. I’m not sure as yet. I think that I may prefer my Old Autochrome ICC profile.

The sCMY colorspace

The attempt to replicate the Autochrome color space was an interesting exercise, and led me to consider limited color spaces in general, ones which would give photographers and graphic artists a new way of controlling the use of color. After all, when you have an ICC profile fixed to a particular palette of colors, the use of color is strictly controlled: you can’t go beyond the fixed limits.

The sCMYcolorspace uses dull cyan, magenta, and yellow primaries — dimmer tones, but of the same hue and saturaton as the sRGB secondary colors. Since this is an additive color space, the secondary colors are brighter than the primaries, so this does not work the same way as does the subtractive CMYK color space used in commercial printer. However, the color channels do have similarities.

In sCMY, the red channel is cyan; the green channel magenta; and the blue channel yellow: the opponents of the sRGB primary colors. And so, the green channel is black where we would expect to see lots of green ink if we were to print the image, and it is white where we would not expect to see any green printed.

This colorspace is unique in that there are no pure saturated red, green, or blue colors whatsoever; we do see pale tints of these at best.

How useful is this colorspace? For general use, not very. However, if there is a need or desire to limit the colors used, it can be interesting.

Due to the inverse quality of this colorspace, we cannot assign sCMY to an image and expect it to look anything close to being natural; instead, you typically want to convert to this profile before editing.

You can download sCMY by clicking here.

This exercise in editing color spaces can be explored much farther by the artful selection of primary colors, or alternatively, selecting bright secondary colors and deriving the primary colors from them.

"photographs very rarely turn out good likenesses"

2011-11-28T11:00:00.001-06:00

The first thing that caught my attention was a portrait of mother that hung over the writing table; a photograph in a magnificent carved frame of rare wood, obviously taken abroad and judging from its size a very expensive one. I had never heard of this portrait and knew nothing of it before, and what struck me most of all was the likeness which was remarkable in a photograph, the spiritual truth of it, so to say ; in fact it looked more like a real portrait by the hand of an artist than a mere mechanical print. When I went in I could not help stopping before it at once.

"Isn't it, isn't it?" Versilov repeated behind me, meaning, "Isn't it like?" I glanced at him and was struck by the expression of his face. He was rather pale, but there was a glowing and intense look in his eyes which seemed shining with happiness and strength. I had never seen such an expression on his face.

"I did not know that you loved mother so much !" I blurted out, suddenly delighted.

He smiled blissfully, though in his smile there was a suggestion of something like a martyr's anguish, or rather something humane and lofty ... I don't know how to express it; but highly developed people, I fancy, can never have triumphantly and complacently happy faces. He did not answer, but taking the portrait from the rings with both hands brought it close to him, kissed it, and gently hung it back on the wall.

"Observe," he said; "photographs very rarely turn out good likenesses, and that one can easily understand: the originals, that is all of us, are very rarely like ourselves. Only on rare occasions does a man's face express his leading quality, his most characteristic thought. The artist studies the face and divines its characteristic meaning, though at the actual moment when he's painting, it may not be in the face at all. Photography takes a man as he is, and it is extremely possible that at moments Napoleon would have turned out stupid, and Bismarck tender. Here, in this portrait, by good luck the sun caught Sonia in her characteristic moment of modest gentle love and rather wild shrinking chastity….

—from A Raw Youth, by Fyodor Dostoevsky (1821-1881), translated by Constance Garnett

Additive versus Subtractive Color

2011-11-14T15:15:00.001-06:00

THE COLOR WHEEL is a basic tool for representing the relationship between colors. It is basic in that it only attempts to show the relationship between particular hues, and it is a tool only insofar as it is useful and not misleading. But we must recognize its limitations.

This wheel is only valid for the sRGB color space, since it explicitly uses the primary colors of that space; it may or may not be useful for other color spaces. Also, be aware that since this color space uses only three primary colors, it won’t be able to portray the entire color gamut of human vision. The wheel illustrates the principles of additive color, where we mix together various colors of light to get other colors; additive color works well enough with back-lit computer monitors and televisions, and also digital projectors. But this wheel is definitely not valid for mixing paints, watercolors, or other materials with color. It only works with light.

Additive mixing is conceptually simple

Mixing colors with light is an unusually pure and simple process that follows relatively basic mathematical laws. A computer scientist or programmer, or a digital photographer, not experienced with the manifold nuances of paints and pigments, can get a good representation of the world’s color with relatively little effort. But not so with the artist who works in oil paint: color mixing is a far more complex phenomenon.

Digital photography uses additive color — each phosphor on the computer monitor adds together with its companion phosphors to create a particular color. Adding together colored lights gives us brighter resulting colors. This addition is implicit in the color wheel above: with red, green, and blue as our primaries, mixing them together gives us our bright secondary colors of cyan, magenta, and yellow. If, for whatever reason, you had a computer monitor with primary colors made up of the cyan, magenta, and yellow phosphors of the wheel, you’d get a secondary colors which would be brighter and also less saturated: the secondary color made by mixing yellow and magenta would be a bright pink, and not red. But if we were to change the set of phosphors used in a monitor, we still should be able to accurately (and fairly easily) predict how our colors would mix.

Subtractive mixing

Paint is far more complicated. Light falls on paint, and some of it is absorbed (usually turning it into heat) while some of it is reflected, which leads to the color that we see. Leaves are green because their chlorophyl reflects green light — and they absorb much red and more blue light. So real-world objects subtract out some of the light falling on them, reflecting back colored light. Computer monitors have three, fixed colors which mix together, while real-world pigments reflect or absorb colors throughout the spectrum: this makes the subtractive colors of the painter much more complex.

For this reason, exceptionally pure and saturated paints are necessarily going to be rather dark: if a paint reflects only the far end of the color spectrum, near violet, then by necessity it will reflect only a tiny fraction of the total light falling on it. Fluorescence helps here, because normally invisible ultraviolet light is shifted in frequency to visible light — but this only works if the radiation falling on our paint has ultraviolet in it.

The colors should look the same...

Digital cameras have precisely three different kinds of color sensors, and the typical human eye also has three kinds of color receptors (plus another more noticeably used in low light). A camera and the eye collapses a broad range of hundreds of distinguishable light frequencies to merely three stimuli or numbers. But consider what this implies: various colored objects may absorb and reflect all sorts of differing frequencies of light, but they all may appear to be the same color to the human eye under a given lighting condition. To complicate matters, digital cameras and human eyes don’t have matching color sensitivity: they are fairly close within the range used in photography, and under good lighting conditions such as daylight or bright incandescent, but they don’t match for all colors or under all lighting conditions. They most certainly won’t match under poor quality light, such as found with plasma discharge lamps (including fluorescent lamps) and LED lighting.

Therefore you cannot say that mixing a yellow paint and blue paint — even if they appear to perfectly match the colors on our color wheel — will give you a gray mixture. You need to know which yellow, and which blue paint are used. After all, a wide variety of reflectances will give you same yellow or blue color. You cannot use a color wheel like the one above to predict the outcome of mixing paint.

A good theory of subtractive color mixing?

If we could measure the reflectance of each pigment, and measure the spectrum of light falling on the pigments, then we could predict, with much bother and mathematics, what the resulting color mixture would look like, although still we would have to take into account whether the paint is wet or dry, or what kind of brushes and brushstrokes were used, plus the opacity of the paint, plus the thickness of the paint, plus the texture and color of the canvas or board or paper on which the paint is applied.

With additive color we can reliably produce white, black, and shades of gray by mixing together all three of our primary RGB colors in equal quantities. White, of course, can’t be mixed by any combination of colored paints, nor can we reliably get pure black by mixing together a few colors. The color of the canvas or backing, as well as the opacity of the paint, will limit the brightest and darkest colors that can be produced. As it so happens, the white and black paints used by artists does not evenly reflect or absorb light uniformly, making the process of creating tints or shades not completely straightforward — mixing a color with black may cause the color to shift towards warmer or cooler tones if not done carefully.

The solution, the difficult, laborious solution to subtractive color mixing is experience, or otherwise having access to sample swatches — a very large number, thousands, of color mixes — although experience is definitely needed if you want to do good work.

Minimal palettes of colors

A good, comprehensive, predictive theory of subtractive color mixing is complicated to the extreme, far more complicated than the already complicated theory of additive color mixing. To reduce complexity, in photography, we limit ourselves to three primary colors, knowing that not all the colors visible to the human eye can be represented in their full glory, and knowing that we accept these limitations in order to practice our art. In the realm of painting, artists (especially beginning artists, or those painting in oil in plein air) often limit themselves to six or so colors, plus one or two kinds of black paint, and one or two kinds of white paint.

The idea of three primary colors works well with additive color, but even though it is limited, it is still used almost universally in digital photography: plus we have the advantage of being able to use mathematical primaries that are out of the range of human vision, expanding the gamut of colors we are able to accurately represent. But the use of three primary colors in painting is disastrous; following this misguided theory could cause an artist to severely limit his palette unnecessarily — although, if that is his artistic intent, that’s fine, it just isn’t a good theory of color. And so, a painter must select a set of paints, and typically this set is more than three colors. If the color of any given paint cannot be mixed with the other paints in the set, then that is a primary color. An artist may use additional colors, such as burnt sienna, which are located within the chosen gamut, but using these is largely a convenience and not necessity.

Three numbers suffice to describe any given color of light or pigment under controlled conditions. But three colors are never sufficient to make up a complete system of colors.

Sadly, most commercial and much digital printing technology uses only three primary colors — cyan, magenta, and yellow — plus black, which gives us a limited gamut of possible color, and which is highly dependent on the whiteness of the printing paper used. While adequate, painters can do far better by adding other colors which cannot be mixed by these primaries. Typically speaking, a gamut made of six bright, saturated colors is adequate for creating a minimal palette, with the colors being fairly visually equidistant to each other, along with black and white. An older split-palette theory of color mixing used three pairs of primary colors, with one of each pair mixing cool and the other which mixed warm. Oddly enough, even though the split palette is somewhat deprecated in painting, a form of this is used in high-end digital printers: pastel primary colors are used to get better color in highlights, and only a few digital printers use good red, greens and blue inks which are otherwise not mixable.

While painters may envy the purity and simplicity of color mixing available to photographers, they must not assume that the overly simplistic tools of the photographer — like the digital color wheel — apply to their art. Likewise, photographers may have a certain smugness over their control of color, but instead they ought to be humbled by the skill required by the artist when mixing colors, as well as the range of the artist’s palette, which far exceeds that found in digital printers.

Color Spaces, Part 3: HSB and HSL

2011-10-14T17:48:00.000-05:00

IGNORE THIS ARTICLE: the information presented here is problematic, difficult to apply in Photoshop, is error prone, and obsolete. Do not read any farther. This information is of no importance to photographers.

OK, not really. This article is about color systems that ought to be useful, that ought to be implemented — somehow — in Photoshop, but they still have plenty of traps for the unwary.

Artists, unlike those who have an affinity for mathematics, may find the RGB color system a pain. They perhaps object to the idea of reducing such a subjective concept as color to mere numbers, or perhaps they object to the unintuitive process of mixing the colors together in RGB. Photoshop is such a pain. It must have been invented by nerds.

For example, the RGB color for a particular tone of orange is R=180, G=95, B=6. But this is ridiculous; orange, a mixture of red and green? And with some blue thrown in also? No, that's wrong. Orange is a mixture of red and yellow, and there ain't no blue it it at all: but for sure that dim tone of orange has some black in it. And what's with these RGB values going from 0 to 255??? That number 255 makes no sense at all. Now, if we have to use numbers (and why can't we use good names for colors instead?), then why not use something more understandable like 0% to 100%?

Also, it seems that some particularly pure colors are missing from RGB. Tyrian purple, intense electric cyan colors, and rich, deep tones are nowhere to be found.

A painter would rather describe color in ways other than RGB. Here are some artists' color terms:

Hue is the basic quality of a color: is the color basically red, orange, yellow, chartreuse, green, aqua, blue, violet, or magenta? These hues are arranged in a circle around a color wheel, which shows how these colors relate to each other, and also how new colors can be obtained by mixing other colors. Now there is some controversy as to what precise set of colors ought to be used to obtain a minimal palette: some say that the three primary colors (red, yellow, and blue) is the minimum number of colors for a basic full-color palette; although white, black, or both are sometimes added to the mix, depending on whether the artist is using paint or watercolor. Other artists say that six colors (plus black and/or white) reproduces nearly the full range of color. Since these colors can mix together to produce nearly all known colors, any extra colors in the palette are simply a convenience to the artist; they aren't strictly needed. But no matter what color system is used, mixing together adjacent colors on the color wheel will produce novel colors, while mixing opposite colors together will give a grayish, blackish, or muddy tone.

Tints are pastel colors, made by mixing a color with white.

Shades are dark colors, made by mixing a color with black.

Brightness, value, or intensity tells us how close a color is in luminous quality to either white or black, or how bright a color is compared to the brightest available version of that color. Aqua is a brighter version of teal. Azure, dark midnight blue, and Dodger blue are basically only different in brightness.

Tones are colors that are mixed with both black and white (or gray) paint. These colors are muted, toned-down versions of the base colors.

Saturation, chroma, or colorfulness is how pure a particular color is. Tones, shades, and tints are less saturated than pure hues. Slate gray is less saturated than azure. Carmine is more saturated than copper rose, even though they are approximately the same hue and brightness.

Other qualities include transparency, fluorescence, glossiness, and so forth,

A good, artistic color system will take these usages into consideration. As it so happens, there does exist the HSB (Hue, Saturation, Brightness) color system, and it is implemented (albeit poorly) in Photoshop. This color system works in a more artistic manner than RGB.

Instead of three numbers for the red, green, and blue channels, HSB provides alternative numbering:

Hue specifies the kind of color, and it follows the arrangement of colors in a color circle. The assignment of colors is thus:

Red = 0°
Yellow = 60°
Green = 120°
Cyan = 180°
Blue = 240°
Magenta = 300°
Red = 360°, which is the same thing as 0°.

Brightness tells us how bright a color is, compared to the brightest, most saturated value of the color, and brightness goes from 0% to 100%. Anything with a brightness of 0% will be black, while 100% brightness gives us the most saturated bright color available. All the colors in the wheel above have 100% brightness.

In this image, hue goes from 0 degrees on the left, to 360 degrees on the right. Brightness goes from 0% on the bottom, to 100% on the top. But note that HSB assigns 100% brightness in a way that isn't uniform; some colors are much brighter — visually — than others. 100% yellow is much brighter than 100% bright blue.

But note that brightness seems to act like a shade, where you mix a particular color of paint with black. The brightness value of 80% then is like a mixture of 80% color and 20% black, while 0% brightness uses 100% black paint.

Saturation tells us how pure or intense is a particular color. Saturation tells us how far a color is from neutral.

Hue goes from 0 degrees on the left, to 360 degrees on the right. Saturation goes from 0% on the bottom, to 100% on the top. As saturation goes to 0%, the colors go to the same pure gray color.

Saturation seems to act something like a tone and shade combined. First you mix together the pure color and white paint: for example, 40% saturation would mix 40% color paint with 60% white. Then you mix the resulting color with black in the amount specified by its brightness.

Here we mix together brightness and saturation with pure red:

Brightness goes from 0% on the left to 100% on the right, while saturation goes from 0% on the bottom to 100% on top. Hue is fixed to zero degrees. This image shows the following characteristics:

Pure red is in the upper right hand corner.
Pure white is in the lower right hand corner.
Pure grays are along the bottom.
Pure shades of red are along the top.
Pure tints of red are along the right hand side.
Pure black is found along the entire left side.

Photoshop offers HSB as an option when specifying and measuring colors. The HSB selector is circled below in red:

The color selector square on the left of this shows us all possible hues with all saturations at a fixed brightness. The slider to its right can adjust for the brightness, keeping saturation and hue fixed.

The HSB system is useful because it works somewhat like an artist mixing bright primary colors with black and white. It also presents a natural ordering of color as is seen in the spectrum and color wheel.

Note that brightness, as defined by the HSB color system, is defined according to whatever color you happen to be using, and so 100% brightness gives us the pure color (whenever saturation is 100% also).

The HSB system is also called HSV, with V=Value.

An alternative system, HSL (Hue, Saturation, and Lightness) retains the H and S dimensions but substitutes Lightness that gives us white whenever it is 100%. For example, here is hue plotted against lightness:

Hue goes from 0 degrees on the left to 360 degrees on the right, while lightness goes from 0% on the bottom to 100% on the top. Pure saturated colors are found at 50% lightness. Adjusting lightness is like tinting a color when greater than 50%, and shading a color when less than 50%.

Plotting lightness versus saturation is this:

Lightness goes from 0% on the right to 100% on the left. Saturation goes from 0% on the bottom to 100% on the top. Shades and tints of red are along the top edge of the image, with pure red being in the middle of the top edge. Pure white is along the right hand side, pure black along the left edge, and tones of gray are along the bottom.

Photoshop allows us to select colors based on HSB, but we cannot edit images in that colorspace as we can when we convert an image to the Lab or CMYK colorspaces. But there is an optional component in Photoshop (available on the original disk) that we can install to get the HSB and HSL channels. From the Filters menu item we select Others and then HSB/HSL to get this dialog box:

Please note that after conversion, the image actually remains in the RGB colorspace. Since Photoshop really doesn't support HSB images, you have to remember what color space you converted to in order to go back.

The original image is on top, and the ‘converted’ image is on the bottom. Although you can edit your photos after the conversion, you can't see what the final result is like until you do the reverse conversion.

This simply changes the appearance of the color channels:

From top to bottom:

Red channel = Hue
Green = Saturation
Blue = Brightness or Lightness

As we could expect, the brightness channel looks fairly normal and it has the least noise of the three channels. From information theory we could predict that much more real information is contained in Brightness, while the noisy saturation channel tells us that the information there is minimal and unreliable.

When I originally discovered this colorspace, I thought it could be very useful. For example, back in the days before Photoshop had a ‘Vibrance’ function, I wanted to selectively saturate an image so as not to overexpose or underexpose any of the color channels; this is something that the standard Saturation tool can do all too well. I thought that the saturation channel of HSB could be used as a mask.

But black can have any saturation at all according to HSB. This is counter-intuitive. Just look at the sky in the photo: it is very dark, it ought to have low saturation, but the noisy saturation pattern shows us otherwise.

I had the bright idea that HSB could be useful for noise reduction. Blurring the color channels is a popular and accepted method for reducing chroma noise, so why can’t I do it in HSB?

I just had to do this once to see that it was a big FAIL. I didn’t think about what I was doing. It is interesting to see that HSB thinks that the black color of the chart has a green hue, which is seen bleeding over the colors.

I’ve been unable to come up with any good use for the HSB channels. Whatever I wanted to do, there was a better way of doing it in another colorspace.

There are a number of other technical reasons why HSB and HSL are bad colorspaces:

HSB and HSL are not absolute, but rather depend on what color space you are working in, like sRGB or Adobe RGB. An HSB number like 0° Hue, 100% Saturation, 100% Brightness does not name a concrete color, but rather the primary red color of whatever colorspace you are working in.
Look at the images above which show hue. Note that the colors do not flow smoothly into each other, as we would like. Yellow, cyan, and magenta appear to be ‘spikes’, with same phenomena to a lesser degree happening to red, green, and blue. As it happens, the colors are distributed linearly between the primary and secondary colors, and do not flow in a more visually-uniform manner.
Take a look at the image of hue versus saturation above. At 100% saturation, the areas of red, green and blue tones appear to be much broader than the secondary colors. But go down the image towards lower saturation: the secondary tones yellow, cyan, and magenta now appear to be much broader than the primary colors. It appears that saturation shifts colors. This color shift is also apparent in the hue versus lightness image.
Black can have any hue and any saturation. Attempts to edit an HSB image (as seen in the Colorchecker image above) can lead to strange artifacts whenever black is involved.
Brightness and luminance do not describe the relative brightness of each color. Pure blue is considered to be just as bright as yellow, even though visually it appears to have a tenth of the absolute brightness.
Saturation is problematic in HSL. A pale light yellow is said to have the same saturation as does a rich green color.

HSB and HSL, along with another colorspace called HSI, were invented in the 1970s as a way of selecting colors on a computer in a more artistic manner than RGB. However, these are simplistic mathematical transformations of RGB color spaces — and are not based on the properties of human color vision — and so have problems as seen above. More information can be found here.

Cynthia Brewer of Pennsylvania State University wrote:

Computer science offers a few poorer cousins to these perceptual spaces that may also turn up in your software interface, such as HSV and HLS. They are easy mathematical transformations of RGB, and they seem to be perceptual systems because they make use of the hue-lightness/value-saturation terminology. But take a close look; don’t be fooled. Perceptual color dimensions are poorly scaled by the color specifications that are provided in these and some other systems. For example, saturation and lightness are confounded, so a saturation scale may also contain a wide range of lightnesses (for example, it may progress from white to green which is a combination of both lightness and saturation). Likewise, hue and lightness are confounded so, for example, a saturated yellow and saturated blue may be designated as the same ‘lightness’ but have wide differences in perceived lightness. These flaws make the systems difficult to use to control the look of a color scheme in a systematic manner. If much tweaking is required to achieve the desired effect, the system offers little benefit over grappling with raw specifications in RGB or CMY.

RGB and CMYK are output-based color systems, which are closely related to actual, physical output devices such as computer monitors and printers. RGB also closely reflects how cameras capture color information. Editing in these color spaces, while not necessarily intuitive, is recommended so as to take full advantage of the output media. But as we have seen, there is very much value in using a more intuitive color model that reflects human vision. The Lab colorspace (fully supported by Photoshop) is particularly useful and is based on human vision.

HSB and HSL are problematic color spaces that do not live up to their promise. They were invented by computer scientists who needed a simple color system that worked well with the low-performance computer hardware available at that time. HSB (also called HSV) was invented by Alvy Ray Smith, who later co-founded the Pixar animation studio. HSL was invented by George H. Joblove (who subsequently did animation in the films Terminator 2, Spider-Man 3, the Chronicles of Narnia, and many others) and Donald Greenberg (who previously worked on the Gateway Arch; he subsequently won many awards in the computer graphics field, and has trained many of the prominent animators working today.) Both HSB and HSL were first published in the August 1978 volume of SIGGRAPH's Computer Graphics newsletter. As these men all had quite remarkable careers, we perhaps should not judge them so harshly because of their color system inventions.

The problems found in HSB and HSL are also found in more serious color standards. The Natural Color System (NCS), which is the official color system of Sweden, Norway and Spain, is based on the notions of blackness, saturation, and hue based on opponent colors, But it has the defect of assigning the same brightness to its base colors of red, green, yellow, and blue. Also, NCS had the flaw of assuming that red and green are opponent colors, which is not true. NCS is based on the philosophical theory of phenomenology — and not directly on human psychology — which may have led the designers to make some mistakes.

Three numbers suffice to describe color, and so we can build three-demensional color models. The RGB system forces all colors into a cube, while HSB and HSL force all colors into a cylinder. But as we see, these systems are not visually uniform and have plenty of problems.

Mathematics is good only insofar as it reflects reality. Bright, saturated yellow is brighter than bright, saturated blue. Magenta is the opponent color to green. Black and white have no hue and no saturation. A mathematical model of color ought to consider these facts and model them well. It appears that the best color model today is the Munsell color system, which arranges the colors in a visually uniform, but irregular solid. It does not force colors into a regular geometric solid such as a cube or cylinder. Albert Munsell was an artist and art teacher, and he wanted to be able to communicate color to his students in a more uniform and logical fashion than just using color names, and he began by organizing his colors by visual appearance. Originally, he thought that he could organize all the colors into a sphere, but eventually he found out that an irregular shape better reflected the visual ordering of the colors. Here is an example of slice taken through the Munsell color solid:

**Munsell *value* (vertical) and *chroma* (horizontal); hue 5Y and 5PB**
	12	10	8	6	4	2	0	2	4	6	8	10	12
10							255 255 255
9						228 228 250	232 232 232	243 227 207	250 227 178
8					190 201 239	200 200 222	203 203 203	215 200 181	221 200 154	227 200 126	233 199 97	237 199 63
7			142 176 241	154 175 225	164 175 210	173 174 195	179 179 179	188 173 155	194 173 128	200 173 101	205 172 72	210 172 29
6	79 150 244	101 150 227	116 149 213	128 149 198	138 148 182	146 148 168	150 150 150	161 147 129	167 147 103	173 146 75	178 146 42
5	46 124 214	72 123 199	89 123 185	101 123 171	111 122 156	120 122 142	124 124 124	134 121 103	141 121 77	146 120 48	150 119 9
4		38 97 172	59 97 158	74 97 144	85 96 130	93 96 116	97 97 97	108 96 77	114 95 52	119 94 25
3			26 72 133	45 72 120	58 72 106	67 72 92	70 70 70	81 71 55	87 70 33
2				20 49 93	35 49 79	44 49 66	48 48 48	57 48 34	63 47 6
1	5PB				13 28 56	23 28 45	28 28 28	37 27 9			5Y
0							0 0 0

[source]

This is a visually-uniform system; and there ought to be a visually equal difference between adjoining color patches. (However, note that the presence of the gray background causes the appearance of some irregularity in this image.) Because the system is visually uniform, the outline of the colors are not uniform. Saturated blues will always be darker than saturated yellows, as is shown here.

In the Munsell system, there is no upper limit to the chroma, or the brightness and saturation of color. Bright new pigments or fluorescent dyes can be added to the outside edge of the Munsell solid as needed.

I started this article with a warning that the information contained here was problematic, error-prone, and obsolete. The HSB and HSL color systems, although supported somewhat in Photoshop, are problematic color systems at best.

There is still a need for an artistic, more natural system of color in our editing software; but unfortunately HSB is not the solution. The commercial product HVC Color Composer allows the user to select color based on the Munsell color system. What makes this system novel is that it will automatically limit colors based on color gamut and human vision; it is also visually uniform. Unfortunately, since it is neither available on OS X Lion or on Photoshop CS5, I cannot evaluate the product here.

The best free alternative is to use the Lab color space found in Photoshop. Since this system arranges colors into a cube, it still has a component of unreality. Using Lab, it is very easy to drive a color out of gamut or out of the bounds of human vision. I also suspect that Lab has a slight color shift for dark colors. But Lab is otherwise very useful and logical, and I use it quite often.

For more information on color spaces, see my articles:

Color Spaces, Part 1: RGB
An RGB Quiz
Color Spaces, Part 2: CMYK
Part Two of "Color Spaces, Part 2: CMYK”
A CMYK Quiz

Deblurring

2011-10-11T10:42:00.002-05:00

This is a technology preview of Photoshop software that can remove motion blur from an image.

A Visually-Uniform Digital Color Wheel

2011-10-07T18:11:00.000-05:00

A WHILE BACK I published a color wheel, made according to the digital sRGB color standard. This standard color space is used with most digital cameras, and is the standard used by the Web and high-definition television. You can see this wheel by clicking here. But this wheel is wrong: rather, I present a new color wheel which is more visually uniform:

Most artist's color wheels, as found in art stores and Internet searches are quite misleading. The idea of a color wheel is to show the relationship between the colors. Of great importance is the idea of color opponency: if you mix two colors opposite from each other on the wheel, you will not get a third color, but rather gray. But these color wheels do not work for digital technology. Red, yellow, and blue — the primary colors typically used in these wheels — are not the primary colors used in digital photography.

But like the artist's color wheels, my old color wheel is also misleading:

The primary and secondary color relationships shown here are correct. The primary colors of red, green, and blue are opposed to the secondary colors of cyan, magenta, and yellow, and if you mix these opposed colors in Photoshop you will indeed get a neutral gray. But the tertiary colors look wrong. Orange appears to be closer in brightness to red, being much darker than yellow, and likewise sky blue looks darker than it ought to be. There is hardly any differentiation between the green colors.

We are looking for the brightest colors we can as our primaries. Setting R=255, with G and B both equal to 0, we will get the brightest and most saturated pure red that can be represented by the RGB color system. The opponent color to red is cyan, which we get by setting R=0 and G and B both equal to 255, and this is the brightest and most saturated cyan we can represent in our system. When we mix these colors together, we get a neutral gray color. By default, Photoshop will average all these colors together to gray. As it so happens, the average color number ought to be 127.5, but since fractions are not possible when using Photoshop in 8-bit mode, we get either 127 or 128; but this is close enough.

Click here for an overview of the RGB color system.

The tertiary colors are halfway between adjacent primary and secondary colors. So the tertiary color between red and yellow will be orange. Our red value is equal to (255, 0, 0), while our yellow value is (255, 255, 0). Averaging these numbers together, I get a middle orange tone of (255, 127.5, 0). Rounding the green value up to 128 (since we can't use fractions), I get the orange tone shown in the old color wheel above. But it is too dark, too close to red in tonality than to yellow. The color appears to be correct, but not its brightness. So I set about manually trying to come up with a color that appeared to my eye to be halfway between red and yellow. Using a number of ad hoc methods to blend colors in Photoshop, I attempted to produce a visually continuous range of tones between each primary and secondary color. As it so happens, I found that the new tertiary colors were always brighter than my original estimates, in fact, the middle RGB value for these tertiaries were quite close to each other: I estimated that they all were around 180-190. Now this meant that the tertiary values across from each other on my color wheel were no longer opponents. Mixing my new orange with my new sky blue got me a somewhat dim and unsaturated green color. But I was willing to accept this shortcoming in order to portray my tertiary colors with visual uniformity.

My mistake was assuming that the mathematically intermediate value of 128 was also visually intermediate. This is wrong; I did not account for gamma correction. The human eye sees light in such a way that doubling the intensity of light only makes a scene to appear a bit brighter. Likewise, cutting the intensity of light by half makes the scene appear to be only a bit darker. Our eyes can operate under an extreme range in lighting due to this property. The practice of gamma correction adjusts the color tones in the RGB system so that the range of brightness is more visually uniform: otherwise, most of our RGB numbers would represent only the very brightest of colors and would neglect darker shades; our coding would be inefficient.

Now gamma correction requires more than simple arithmetic, and the sRGB standard uses a non-standard, non-uniform gamma correction to the RGB data, making this whole process murky. I'll spare you the details; instead, read the article linked above. What is important is that gamma correction does not change the values of 0 and 255 in 8-bit RGB color, it only modifies intermediate values, and so our calculations with our primary and secondary colors are unchanged. Only the tertiary colors will change with gamma correction. Intermediate values, after backing out the gamma correction of sRGB, become 187.5, and so in the color wheel above, I set the values to either 188 or 187.

This still leaves us the problem of mixing colors in Photoshop. Blending together opposite tertiary colors does not give us a neutral gray. However, this has already been thought of. If you go to Edit-Color Settings, and then click More Options, we get this:

Select Blend RGB Colors Using Gamma 1.00, and the color mixing will be correct. This option subtracts out the gamma correction, allowing more accurate color blending. If we blend together our leaf green and our purple, under a gamma of 1, we will get a neutral gray color, and it will be the same shade of gray that we get if we blend together any of the opposing colors on the wheel. Since I'm not all that sure what other effects this option may have, it probably shouldn't be set at all times. But we do see that it in fact does produce more visually accurate color blends.

How does the wheel look? One problem is that the sRGB color gamut exceeds the capability of my calibrated iMac monitor, and so most of the colors here cannot be presented to my eye in their full glory. I find that I can hardly distinguish rose from magenta — how about you?

The Academy on LED Lighting Technology

2011-10-06T07:57:00.000-05:00

BETWEEN 2% AND 10% of a motion picture's budget is for lighting, and if we consider that much of that cost is for electricity and for the purchase of low-life expectancy lamps, then new technologies look quite promising. Solid state light emitting diode (LED) lamps produce great amounts of light at low power consumption, they don't generate much heat, they have a long life, they are lightweight, and they are tough, making them unbreakable and safe to work around. The new solid state lighting has many benefits, and the United States government is heavily encouraging its use.

The Academy of Motion Picture Arts and Sciences is studying the new LED lamps for suitability in motion pictures. Their preliminary research can be found at the Solid State Lighting Project page.

The results are not good. LED lamps do not present colors very well to the camera.

Generally speaking, according the presentation linked above, LED lighting is currently a poor choice for getting good color, most particularly for getting good skin tones, which are particularly harmed. The example tests show skin looking more green and sickly instead of having a healthy ruddy glow. The tests show also that makeup looks inconsistent under LED lighting, and does not blend as well with bare skin. Tans, light browns, and reds tend to be harmed significantly, while some blues are enhanced. Colors in general appear to be less differentiated under LED lighting, and complex fabrics tend to 'flatten' and have less pronounced shading and texture. The use of these lights could impose greater costs, requiring color checks before filming.

Tungsten lamps and daylight produce a flat, uniform spectrum, and so are perfect at rendering colors accurately. Film stock and digital sensors render the colors from this kind of lighting very well.

In the Academy's opinion, the discontinuous spectra produced by LED lighting is problematic. The terms 'color temperature' and 'color rendering index' are not particularly useful when discussing LEDs, because of their discontinuous color quality. A solution, they think, may not be in improving the LED lamps themselves, but rather in developing custom film stock and digital sensors that better match the LED spectra. They also think that custom discontinuous filters could be used, but this will remove much of the electrical efficiency of the lamps, making their use largely moot. These fixes are not likely to help photographers who are much less in control of their lighting, and who need more general-purpose equipment — an LED optimized sensor would do a rather poor job under daylight.

Visually, the various light sources look the same to the human eye: but the camera does not perceive light the same way and colors are rather unpredictable. I ought to note that much research is currently being done to make a visually balanced LED spectrum, but this by no means is expected to work well with photography. Every model of sensor will have a different response to the LED lighting, making color matching more difficult.

The Academy thinks that LED lighting products are being rushed to market without taking into concern the needs of quality color reproduction. They also see the need to have quality third-party evaluation of the products emphasizing cinematography.

Autochrome

2011-10-01T01:47:00.000-05:00

THE LUMIÈRE BROTHERS, Auguste Marie and Louis Nicolas, are known for being pioneers in motion pictures, having patented the cinématographe in 1895, and by holding the first commercial movie screening that year. However, they thought that “the cinema is an invention without any future,” and instead they concentrated their work on developing color photography. While photographers did produce color images by taking a series of black-and-white exposures from behind various colored filters, the Lumières (their name, appropriately, is French for ‘light’) developed a method of making a color image from a single exposure. Their plates could be easily developed with standard darkroom methods and chemicals and required no unusual equipment to view.

Autochrome plates had a layer of starch grains, dyed to three primary colors. These tiny grains acted as color filters during both exposure and when viewing. The plate used standard silver chemistry, and was developed so as to remove the silver that was exposed to the light — which is the reverse of the normal process. The completed plate then had a positive color image when viewed with a back light.

Autochrome Lumière plates were first marketed in 1907, at a cost a number of times more than standard black and white plates. This proved to be the most practical color photography method until the 1930s.

Children in Medieval Costume, by Mrs. Benjamin F. Russell, ca. 1910. George Eastman Collection.

Autochrome was used to create slides, viewed either by a slide projector, or through a hand-held viewer. It is said that printed reproductions of Autochromes — especially those made during the era when the process was used — do not do justice to the medium; the original colors are delicate, with excellent color separation and balance. But autochromes are sensitive to light, heat, and oxygen, with surviving images often being in poor condition.

While there are many differences between Autochrome and newer color photography methods, the most striking difference is that Autochrome used orange, green, and violet dyes as the primary colors of its palette, in contrast to the now-standard red, green, and blue primary colors. This inevitably and severely limits the range of colors that can be represented by this method.

But muted colors are not necessarily a bad thing. Autochrome remained popular until the 1950s, especially in France, because it produced more subtle color than newer films. The American film Kodachrome (produced from 1935 to 2009) was known for producing excessively bright color, while the Japanese Fujichrome Velvia (introduced in 1990) produces colors very often criticized as garish. The contemporary digital photography color standard also produces colors which can be excessively saturated. Can an understated, limited, and muted use of color be sometimes superior?

As it so happens, the range of colors produced by Autochrome ought to be accurately reproducible by modern digital cameras, since their native color gamut is so large. Intrigued by Autochrome, I set about devising a method of reproducing its color range. Here are some experimental digital photos, limited to use only my estimate of Autochrome's color palette:

I reduced the color gamut of these photos to my estimates of the primary colors used by Autochrome. As these primary colors are more limited than the sRGB primaries, we ought to be able to reproduce them more or less faithfully here. There are plenty of Autochrome images on the Internet, but many of them are scans from printed images, which will throw the colors off. I attempted to find only direct digital images of Autochrome slides, and used those to estimate the primary colors. Clearly there is lots of variation among my sample photos, not the least of which is changes in exposure and white balance, as well as the color response of the various digital cameras or scanners. But I did the best that I could. Here is the process I used:

I collected dozens of Autochrome images from the Internet, and put them together into a large photomosaic.
Using selection and threshold tools in Photoshop, I eliminated colors that were not saturated, changing these to a uniform neutral gray.
Using Indexed Color, I was able to reduce the color palette gradually until I was left with several clusters of colors, which corresponded to the three Autochrome primary colors, three secondary colors, and black and white.
From the sample colors found in the Color Table, I picked out the brightest, most saturated typical-looking colors from each cluster as the primaries.
Using a color conversion tool (ColorSync can do this), I determined the Yxy color values associated with the RGB numbers found in the Color Table.
I created an Autochrome ICC profile in Photoshop using these Yxy values.
Converting an sRGB image to the Autochrome profile may or may not make a big difference, depending on the brightness of the colors found in the original images. However, you can edit the image considerably to bring out the colors better, and never leave the Autochrome gamut.
I have no way of knowing if I got the colors right, and there obviously is a lot more work that can be done with this, but this limited palette still has a certain beauty and opportunity.

My processing may mimic somewhat the color produced by Autochrome, but these examples do not show the distinctive pointillist-like colored grain found in that medium.

Some Frenchmen, using original Lumière equipment and notes, are attempting to duplicate the process in its full glory. While not likely to become widely used, this is still a worthy medium for color photography.

UPDATE: see the article Using ICC Profiles for Creative Color Control for specific instructions on converting images to the estimated Autochrome color gamut. For the photographs in this article, I used the Old Autochrome ICC profile.

White Balance, Part 2: The Gray World Assumption and the Retinex Theory

2011-09-15T14:39:00.002-05:00

WHITE BALANCE is one of the most important functions of digital color photography. Understanding a little bit of theory about this function could improve your photography.

Very many beginning photographers (and I was certainly one of them) are quite disappointed with the quality of their images. Although they may find it hard to put into words just what precisely is wrong with their images, bad color or a color cast may be a significant problem. When I learned about white balance and how to correct for it, I immediately thought that my photos looked much better.

Bad versus better white balance, among other things. Saint Vincent de Paul Chapel, in Shrewsbury, Missouri.

Clearly, the color of light is often variable, and the color my camera captures is usually somewhat different than what I see in real life. However, if I am careful to set the white balance, then typically my colors look much better under most circumstances.

There are some principles that we ought to keep in mind:

Our eyes do white balancing naturally. This leads to color constancy, where particular shades of color appear to be subjectively similar under a wide variety of objectively different lighting conditions.
White balance in photography corrects for extreme changes in the color of light; it is not a slight adjustment.
For the best results, we ought to do white balancing early in the photographic process.
We cannot rely on the camera or software automatic white balance function to give good results.

White balance and the process of adjusting color is also called gray balance, neutral balance, color balance or color correction. In cinematography and video, the equivalent process is called color grading, color painting, color timing, or digital grading. Color film is manufactured to produce good colors for only specific light sources; changing film stock or using color filters is typically needed to compensate for lighting changes.

Our eyes have built-in white balancing which strongly subtracts out the color of the lighting. While this mechanism also works when we are looking at photographs, the fact that most photos are relatively tiny and low-contrast means that our eyes will adjust to the room condition more than our images. This means that we have to color-correct our photos to subtract out the color casts due to the color of light at the time of capture.

For an overview, see my article, White Balance, Part 1.

Please recall that digital cameras have a fixed white balance, a balance that is not particularly optimized for any typical lighting condition. For most cameras, this white balance is biased towards green, and is deficient in red and blue; most particularly, cameras are highly deficient in capturing blue light under incandescent lighting.

For example, here is an image taken with my digital camera using its native white balance:

Buildings in downtown Clayton, Missouri. Image presented with the camera’s native white balance.

I processed this image in Raw Photo Processor, and used the UniWB setting which does not adjust the white balance. Notice how green the image appears! White balance is an intrinsic operation of digital cameras: some sort of white balance is always applied to your image, and be aware that your camera may not always give you a good balance.

Now, I recommend setting white balance at the time of capture, either using one of the white balance presets, or manually setting it using a neutral gray card. Automatic white balance usually works rather well in daylight but can be problematic especially under incandescent lighting, on overcast days, around sunset, or in mixed lighting conditions. Setting your white balance at capture time also means that you can get better exposure and can rely more on your camera's histograms. Setting white balance at capture time is less critical when you shoot RAW, but it is extremely important to get right if you shoot JPEGs.

Although it is hard to give good numbers, I do think that the majority of photos — especially editorial photos produced for publication — ought to have a good white balance. An object that is objectively white, gray, or black in the scene ought to be objectively represented in your image with RGB values that are objectively white, gray, or black; that is, all RGB numbers are equal on a neutral subject. Click here for an overview of the RGB color system. Even if you need to reproduce a color cast, for example the color of sunset or candlelight, only slight color casts are usually desirable, otherwise you risk washing out the detail in your image, making it look flatter without texture. Knowing how to make a good white balance will also give you the ability to precisely add tones to your image for an effect.

Knowing some theory can help your photography. Understanding how white balance works will help you to make better photographs. Being able to control your color will allow you to make a photo look as you want it to look.

We first need to understand that white balance is usually a very simple arithmetic adjustment, with little to no higher math. Typically all of the red values are multiplied by a fixed factor, and the same goes for all the blue values. Now these numbers will vary by lighting conditions, the camera make and model, and there are other things that happen behind the scenes in your camera or RAW processor. But this multiplication is the basic correction for white balance. Many digital cameras, in daylight, will multiply the red channel by about 2, and the blue channel by about 1.5; under incandescent lighting, red might not be adjusted much at all, but the blue channel might be amplified 2.5 times, increasing noise greatly.

For example, here is the familiar X-Rite Colorchecker target, photographed in bright daylight:

On top is the camera native white balance, and on the bottom is the white balanced version, created from the native white balance using the curves tool:

Photoshop's simple Curves or Levels tools are sufficient for doing a white balance, although there are other clever tricks you can do for a better job. But ultimately, white balance is linear: the adjustment is the same for all of the pixels in an image.

Now sometimes you don't get good white balance for whatever reason, and you have to adjust it after the fact on the computer. On the photo above, I know that the squares on the top row of the X-Rite target are approximately neutral, and I could have adjusted my curve much more precisely than shown here. But what if you take a photo and have no idea what white balance settings you should use? What if you don't have any particular patches in your photo which are guaranteed to be neutral? Also, how do cameras perform the automatic white balance?

For a long time, when I was puzzling about how a camera does white balance (and why it fails), I assumed that the camera simply measured the overall color cast and then corrected for it. Clearly, when I took photos under incandescent lighting, the colors tend towards yellow; subtracting out that yellow (I thought) ought to produce a nicely white balanced image. Likewise, photos taken in the shade tend to look blue, and so this blue ought to be removed. Photos taken under fluorescent lighting are often too green.

The Gray World Assumption is a white balance method that assumes that your scene, on average, is a neutral gray.

Imagine you are given this JPEG image, and your task is to correct the white balance:

Old sink-hole iron mine, at Maramec Springs Park.

This appears to have an overall yellowish-green color cast.

Gray World assumes that all the colors in an image ought to average out to a neutral gray: for every blue there is a yellow, for every red there is a cyan, and for all greens there are magentas. For this image, using Photoshop's Average function, we get:

Here is how you can use the Gray World Assumption in Photoshop:

Duplicate the image layer
Do a Filter-Blur-Average on the top layer.
Place an eyedropper anywhere on the image.
Create a Curves layer on top.

Your layers should look like this:

Now adjust the curves until all the RGB values are equal. You may have to change the Curves blending mode to Color, to avoid blowing a channel, but this makes adjusting curves a bit less straight-forward.

You should see that the image now looks perfectly gray — if not, then your monitor needs calibrating.

Remove the gray layer, and you have a white balanced image:

This process worked pretty well for this image. But certainly we cannot hope that this will work for all images; what if the subject is brightly colored?

A view of my parents’ back yard, in the snow.

The Gray World is not a good assumption for this image; it simply has too much blue and too little red to reconstruct a plausible image. Now it so happens that I captured this image in RAW, but had inadvertently set the white balance to incandescent lighting. While this wrong white balance is what is presented here, the RAW processor can be adjusted to use a better balance. This is a strong reason to shoot RAW instead of the compressed 8-bit JPEG format.

But not everyone shoots RAW, nor does every camera offer it, nor can we always have access to the original RAW image file. We can have some other assumptions besides the Gray World. First, it would seem that the very brightest highlights in an image might reflect the color of the light source in a rather pure way. Likewise, the very darkest shadows in an image could be assumed to be pure black.

The Retinex Theory was developed by Edwin H. Land, the inventor of Polaroid cameras and polarizing filters. Since humans can identify colors under a wide range of lighting conditions, Land theorized that the very brightest patches of light are used by human vision to determine the color of the light, to automatically compensate for this color cast. Like Land, we can identify the brightest patches of the red, green, and blue channels in an image, and adjust our colors based on these.

A dinosaur exhibit at the Saint Louis Science Center. This image has an orange color cast.

Using the Retinex Theory, we examine the red, green, and blue channels separately and put eyedroppers on the places where each of these channels is brightest.

On each color channel, I used the Threshold tool to determine the brightest patches. Dropper #1 is on the brightest part of the red channel, and dropper #2 has the brightest part of both the green and blue channels. I used 3x3 droppers, but you can experiment with using larger droppers, or single pixel droppers.

I adjusted the image with Curves so that the brightest parts of the red, green, and blue channels were all equal. The result isn't bad:

We can extend the Retinex Theory to not only identify the brightest colors, but also the darkest. We make all the bright colors equally bright, and all the dark colors equally dark. And while we are at it, why don't we force the very brightest colors to their maximum value, and why don't we adjust the darkest colors to their minimum value? This will expand the dynamic range of our image, which is usually considered good practice for commercial photography.

[However, please note that some artists state that we ought to retain color in deep shadow, since this more closely models what we actually see, and it is also pleasing. In this case, we ought not adjust the darkest shadows to be a perfect black color. A problem with this is that the RGB color models allocate very few bits to this kind of shadow detail; on the contrary, CMYK and other printer profiles, since they use black ink in combination with colors, can in fact nicely show textured shadows with color casts.]

This happens to be precisely what Photoshop's Auto Levels does:

Trees in fog, illumined by the rising sun, on the Meramec River near Eureka, Missouri.

The bottom image was corrected with Auto Levels. While a color cast may or may not be desirable with this particular image, Auto Levels often does a surprisingly good job of automatic white balance. Here are the settings I used:

Selecting “Enhance Per Channel Contrast” will identify the brightest and darkest bits of each color channel, (like we did with the dinosaur) and adjust the image from there. This option will produce the strongest correction. “Find Dark and Light Colors” will rather select the brightest and darkest pixels as a basis for adjustment — this is similar to the Perfect Reflector (or White World Assumption) algorithm for automatic white balance, which assumes that the very brightest patch in an image reflects the color of the light source, and so forces it to be white.

The Photoshop menu items “Auto Tone” “Auto Contrast” and “Auto Color” are all shortcuts to various settings in this window.

Be aware that Retinex or Auto Levels will not work properly, or at all, if any of your color channels are over or underexposed. Auto Levels, using the settings above, does nothing to this image:

Saint Patrick Church, in Rolla, Missouri.

Here the camera's white balance was set to Auto, and it failed to adjust for the color of the light. I suspect that my Nikon may use some sort of Retinex algorithm, although that can't be the whole story since overexposed daylight images usually appear to have good white balance. Typically, when I shoot indoors, I do a manual white balance in the camera.

I attempted to produce a good manual white balance of this photo. Alas, after examining this photo, I conclude that the light shining on the altar at center is undoubtably incandescent, while the light fixtures hanging from the ceiling have fluorescent lamps. The corrected result shows green and magenta color casts all over the image, which looks rather worse than the yellow image shown above. If I can, I turn off fluorescent lights before photography: blue and yellow color casts look natural, while magenta and green color casts look sickly. If I were seriously interested in improving this photo, I'd put down some masking and color-correct the altar area separately from the rest of the nave. Mixed lighting will harm every automatic white balance, and will make life difficult for a conscientious color retoucher.

There are white balance algorithms that attempt to combine aspects of the Gray World Assumption and Retinex Theory. The major problem with these is that these two algorithms are not compatible. Remember that white balance is ultimately a linear transformation of color data. A combined approach will generate curves that are not linear, and so are ultimately inaccurate white balances. Take a look at the Auto Color Correction Options shown above; the “Snap Neutral Midtones” feature (from what I've been told) will blend in a little bit of Gray World with the other adjustments. But this rarely gives good white balance and often looks a bit odd because the colors may shift with brightness.

If we have a RAW image file with strong red and green channels, and a weak blue channel, then we can probably guess that it was taken under incandescent lighting; likewise, if we have strong blue and green channels but weak red, then we can conclude that the image was mostly likely taken in open shade, while a strong green but weak red and blue channels is probably daylight. If you have looked at tens of thousands of images you start seeing obvious patterns.

These Photoshop histograms show data from two photographs of the same object, taken under widely different lighting:

These show the channel structures of two RAW images, uncorrected for white balance: the top is from an image taken under incandescent lighting (which is deficient in blue light), while the bottom image was taken outside during an overcast day (and so is deficient in red light). Recall that the histograms show the darkest bits on the left side, and the brightest on the right. Looking at the histograms we can conclude that the top image has a greenish-yellow cast, while the bottom image has a bluish-green color cast.

This observation gives us another method of white balance, called Color by Correlation, which makes assumptions about typical light sources. It attempts to determine if a particular image was plausibly taken under a wide variety of light sources. It is highly implausible that the first histogram above was taken in the shade or on an overcast day, but if we assume that we have incandescent lighting, that histogram becomes quite likely. We can even analyze individual colors: can we be really confident that a bright blue object (according to the RAW channels) can be found under incandescent lighting? Unless it is a light source within the image, bright blue is impossible. If we go over a broad range of colors and ignore outliers in our data, then we can estimate the quality of light in an image.

This type of method can confidently estimate a range of lighting conditions that could plausibly produce a given image. But unfortunately, we cannot be satisfied by a range of plausible white balances, for an in-camera automatic white balance algorithm must produce one and only one result. Color by Correlation, by design, uses only a discrete number of specific light sources, and not the full range of all theoretically possible lights. The Law of Parsimony or Occam's Razor may lead a software designer to severely limit the choices for light sources: after all, nobody uses green lights to illuminate the inside of a home, so why even consider it? Right? And so I would expect that this method will select a white balance based on one of the limited number of light sources used in its model.

I suspect that Adobe Camera RAW's Auto White Balance feature uses a variant of Color by Correlation: very often it will select a white balance equal to one of the standard white balance options, most notably Incandescent. Considering the wide range of color adjustments you can make in ACR, it seems highly unlikely that it will so often choose a standard preset so precisely if it weren't using this method. ACR Auto will never produce a horribly bad white balance — as we so often see in the other methods described earlier — but it very often produces a mediocre white balance. On the other hand, Retinex Theory, when it works, it works very well, but when it doesn't work it is horrid, while the Gray World Assumption can give us nearly anything.

There are other white balance algorithms, but I don't know enough about them to comment, and I've come up with a few automatic methods which tend to generate results ranging from OK to dismal. From what I've read, Color by Correlation is the best automatic method, but you can do better yourself.

I hope this exercise demonstrates that we cannot rely on automatic white balance. According to a major scholarly article, “Automatic white balance is an under-determined problem and thus impossible to solve in the most general case.” There are technical methods that could improve automatic white balance significantly, but they would require modifications to existing camera designs.

Manual white balance is something photographers ought to consider when doing their best work, as well as being aware of the quality of light in their image. As I mentioned earlier, even though precise white balance is typically needed for commercial work, an understanding of the principles involved can help those who want to move away from neutral balance for artistic purposes.

Here are some hints for getting excellent colors and white balance:

Have uniform lighting on your subject. Use a good incandescent light source like the sun or bright incandescent light bulbs. Avoid fluorescent lights, and especially avoid mixing incandescent and fluorescent lights together. The newer light emitting diode (LED) lighting also have poor spectra for photography.
You get the best color and best white balance when you use your camera's native ISO; this is usually, but not always the lowest setting.
When shooting outdoors in the sunshine, you can hardly go wrong if you set your camera to its fixed Daylight white balance, and this will retain the pleasing warmth of late afternoon light. When indoors, using the Incandescent or Tungsten setting may work well; but be aware that gas discharge lights (which include fluorescent lamps and sodium vapor lamps used for street lighting) will generally make your photos look poor and they may not be fully correctable.
Do a white balance as early in your workflow as possible. The best time is when you take the photo. Do a manual white balance at your subject, with the white balance card facing the camera.
Photograph a white balance target within your scene to double-check your settings.
If you don't have a white balance target, examine the scene closely for objects that appear to be neutral in color; you can use these for white balance in the camera or later in post-processing. Be aware that white paper or fabrics may have ultraviolet dyes to brighten them, and so this may throw off your white balance.
Color calibration will not only get your colors looking right, but this will also increase the accuracy of your white balance.
The best white balance after the fact is done using the camera's original RAW channels. RAW conversion software may use these channels directly for white balance.
If you have a lot of time on your hands and are mathematically proficient, you can create your own color transfer function to convert your camera's RAW channels into a precisely color-calibrated image for specific lighting conditions. This would also facilitate accurate white balances. That is what I did in the Colorchecker photo above. See the article Examples of Color Mixing.
If you color-correct in Photoshop, use 16 bits and a broad color space like ProPhoto or Wide Gamut RGB, even if you have to convert to a smaller color space later.
If you have three objectively neutral colors in your image: one dark, one light, and one medium, you can edit your curves to make a very precise and uniform white balance. However, if your curves are very curvy instead of being straight, then be aware that your neutral colors may not be illuminated by a uniform color of light, or your colors may not be as neutral as you think.
When adjusting white balance using Curves, you typically only need to modify the red and blue channels.
The Lab color space at first seems to be a good space to do color correction, since it separates luminosity information from color. However, it is not precisely uniform: doing white balance adjustments in Lab will throw severe color casts into the shadows.
Another bad space for doing white balance is the standard sRGB color gamut. Editing in 8 bits also will give you more problems. But this is very common; many cameras only produce 8-bit, sRGB images, and some software packages can only process 8 bits per color channel.
The best book for learning color correction is Dan Margulis' Professional Photoshop. Although this book is slightly dated and notoriously difficult, if you diligently work through it several times you will learn a lot. Dan isn't always right, but this book is definitely worth the effort: you will thoroughly learn how to best use the basic tools of Photoshop.
Examine the separate color channels of your image to see if there is sufficient color information to do a good correction. If all three channels exhibit good detail and have a wide range of tones that do not go all the way to black and to white, then you likely can do an excellent white balance. If one of your color channels is nearly all black or white over large areas then there is little hope for a good white balance; we see this in the snow picture above, where the red channel is largely black. Sometimes it is possible to reconstruct parts of a channel by using data from other parts of the image and by careful use of the painting tools.
Don't rely on your eyes. Your monitor may be defective, your eyes may be tired, and changes in room lighting can cause variation between images. Rather, measure objects in your image that are objectively neutral in real life and then modify these colors so that they are objectively neutral in your color space (typically when R=G=B). Even if you want an overall color cast, precisely add this color in reference to an objectively neutral object; this is essential when you have to put the same color cast on a number of different images. By using these objective techniques, someone who has color-deficient vision or who has a terrible monitor or bad working conditions ought to be able to do excellent color corrections.
Rely on your eyes. Don't trust any automatic method of white balance: if you think the colors are bad, then they most likely are bad. Also, ‘perfect’ white balance, for some images, simply looks wrong. In dim lighting, cameras will not produce images that have tonalities that we see with our eyes, due to the Purkinje Effect. An overall color cast is perfectly acceptable or even desirable for many images.
If a large neutral gray object filling your screen does not appear to be neutral gray to your eyes, then you probably ought to calibrate your monitor. Mac computers have a built-in calibration that often works rather well, and there are hardware devices that can do a more thorough job. However, calibrating your images is more important than calibrating your monitor, especially if you are going to share your photos on the Internet or submit the photos for printed publication.

Anniversary of the Daguerrotype

2011-08-19T00:57:00.000-05:00

ON AUGUST 19^th 1839, the government of France released the patent on the Daguerrotype to the world, which made photography commonplace.

The promise of photography had been around for decades: Joseph Nicéphore Niépce had been working on his heliograph process since 1793. His problem was that his photographs quickly faded, and his oldest surviving images date from about 1825 or 1826. Starting in 1829, Niépce partnered with the prominent painter Louis-Jacques-Mandé Daguerre to further develop the heliograph process.

Daguerre and Niépce worked on the physautotype, a process where lavender oil dissolved in alcohol was applied to a silver plate to produce an image. This process required that the plate be exposed in a camera obscura for many hours. After Niépce's death in 1833, Daguerre discovered that exposing a silver plate to iodine vapors before exposure, and then mercury vapor afterwards, could produce a plate far more sensitive to light. Daguerre patented his process in 1839, and then gave his patent to the French government in exchange for a pension for himself and Niépce's heirs.

There was a boom in professional photography. By the 1850s, photography studios could be found in nearly every major city.

1 Bit Depth Cameras

2011-08-01T16:19:00.001-05:00

PLEASE EXAMINE these two photos:

Not too much difference. The second photo has perhaps slightly greater contrast, saturation, and sharpness.

Now both these images came from the same RAW file, but had different processing; I reduced these images to this size using Photoshop's Bicubic algorithm.

Now let's zoom far into a small part of the original images:

The second image has one bit per color channel! Each of the red, green, and blue color channels is either fully on or off, and so we have only 8 possible colors per pixel versus millions of possible colors in the top image. But when we reduced the image in size, these pixels were averaged, and so appear normal in the final image. By averaging, we can obtain every possible color in the sRGB gamut.

Lousy, horrible image quality, when we pixel-peep at 100% resolution; but it looks just fine at the resolution shown on the screen.

But we really don't have to do pixel averaging to get a plausible image:

The pixels on this image are black and white only — no shades of gray. This image is interesting, but it has poor image quality; but if we had more and smaller pixels, an image like this would appear to the eye to be a continuous grayscale of high quality.

The notorious megapixel wars of a few years ago had new models of cameras coming out with ever higher pixel densities, at the expense of higher noise. When I got my Nikon D40 back in 2008, I specifically chose that model because it had the best balance between noise and pixel density. Since I was going to produce full-resolution images for the book Catholic St. Louis: A Pictorial History, I needed a minimum number of high-quality pixels for full-page images. Now I could have used a higher pixel density camera, but at that time I had problems with image downsizing, which harms sharpness.

The problem with high megapixel, small sensor cameras is noise; but noise is not a problem if you don't look too closely: and clever image manipulation can make this noise invisible or unobjectionable in the final image. Let's not forget that the single best predictor of image quality is sensor size. You can get a decent small image with a small sensor, but a large sensor will deliver high image quality with both small and large images.

What this little experiment tells us is that we don't need high quality pixels to get a high quality final image, if we have enough pixels.

Suppose some manufacturer could cram 250 megapixels into a typical ASP-C sized sensor as found in consumer grade DSLR cameras. If you zoom into your image at 100% resolution, the image might appear to be only noise, but by clever processing you could produce excellent large images with superb detail.

We can learn a few lessons from old film photography. The crystals of silver halide embedded in film change chemically when exposed to light. Please be aware that light, when examined closely, appears to be made up of little discrete chunks called photons. Whenever a number of photons would hit a piece of silver halide, no matter how large or small a crystal, the crystal would change. And so, large crystals are more sensitive to light than smaller crystals, since roughly the same number of photons would likely change either size crystal. The crystals that absorbed photons would then remain in the negative, and would appear black, while the crystals that did not absorb photons would be washed away during developing, and so only the transparent film remain.

So in a certain sense, digital photography is more analog than film photography, and film is more digital than digital sensors. A film grain is either on or off, with no intermediate steps, no shades of gray. In developed film there is either a grain at any given part of a negative, or there isn't: it is all black or white. Only when we look at the negative from a distance can we perceive an image.

Film grain has many advantages. Because the grains have random shapes, we do not see aliasing in film images — we don't see digital jaggies or interference patterns. By mixing various sizes of grains, we can expand the dynamic range of film: some grains will be more likely than others to be exposed, and so we can get both good shadow and highlight detail, as well as a gradual fall-off in highlights, instead of an abrupt cutoff in highlights as we see with digital cameras (Some Fuji digital cameras had two sizes of sensors to avoid this a bit). Film grain can also add perceived texture as well as sharpness.

Digital photography, following the long example of film technology, could utilize very large numbers of tiny pixel sensors of varying sizes and shapes, and these sensors would register only on or off signals: they would be 1 bit sensors, just like film grain. Although pixel peeping on such an image would be painful, we should expect to obtain superior final images, with much higher dynamic range, good fall-off on highlights, and nice texture and sharpness without digital artifacts. We also would have better control over digital noise, particularly in the shadows.

One interesting advantage to having super high pixel densities is that camera manufacturers could include sensors of various color sensitivities, greatly expanding the color gamut of the sensor, as well as provide far more accurate color rendition. These extra color sensors could emulate the Purkinje effect, and be able to accurately capture an image that corresponds to how humans see in dim lighting, particularly at dusk. Some sensors could be sensitive to infrared, and others to ultraviolet. We have so many pixels in our camera that it wouldn't be much of a waste to assign a certain number of them for odd uses: it would hardly cause a degradation for most images. Various sizes of pixels could be scattered across the sensor, giving better dynamic range; random shapes and sizes could produce images less prone to aliasing, as well as providing more sharpness.

Sensor design would be challenging, for various trade-offs will need to be made: what would be a good balanced design? Blue sensitivity is poor in the standard Bayer design, and red sensors could be boosted in number also. Why can't we also include a number of panchromatic sensors as well, instead of relying mainly on green for luminance? What percentage of the sensor area should be dedicated to big sensors, how many tiny pixels should be included? Can small pixels be superimposed upon big pixels? If varying shapes of pixels are used, what shapes are best; pointy or round? This design offers great potential. Manufacturers could even incorporate Foveon-style sensors, where multiple colors are sensed at the same location. They could even get very clever, by changing the size of the sensors according to depth.

RAW conversion for such a sensor would be difficult — or rather interesting — and would likely require a much faster on-board computer in the camera. But for a hypothetical 250 megapixel camera, we would have precisely one bit per pixel, and so the RAW file size would be only about 31 megabytes in size before compression: not terribly much larger than what we have today. JPEGs produced by this kind of camera would not be full resolution, but then we don't want full resolution, we need to have some sort of averaging of the pixels. But even a lightly processed TIFF file would produce a superior image; up close, we would perceive something that would look like photographic grain, which is much more organic than digital pixels.

This kind of camera would give us unprecedented control over the RAW conversion process. You want an infrared image? You got it. You want ultraviolet? Easy. You want to trade off color accuracy for dynamic range? You want to bring up detail in both the highlights and shadows? A one-bit depth camera can be very useful.

Coming Soon

2011-07-07T14:48:00.002-05:00

IMAGINE A light source, which can automatically match the color of ambient light.

So many times I've taken interior photos, with the room illuminated by tungsten lighting; however, some light comes through the windows, but that looks strongly blue and so harms the photo. For my best work, I adjust the color of the window lighting and the artificial lighting separately in Photoshop, but this is time-consuming and error-prone. Big-budget cinematographers will put colored gels over the windows to get the light colors to match, or they will put gels over their lights. But the use of gels has the unfortunate side-effect of decreasing the light intensity, requiring even brighter and hotter light sources, much to the dismay of the actors.

Rift Labs is developing a three-color lighting panel which has variable color temperature. Using low-power, three-color light-emitting diodes, their forthcoming light panels will automatically be able to match any lighting situation. Or, any color of light can be dialed-in, and the brightness of the lighting does not change.

So photographers and cinematographers can have a portable, battery-operated, low-cost lighting system which can quickly and conveniently produce the color of lighting needed. A good use of this would likely be for fill-in lighting, an augment to ambient lighting. Portable, battery-operated white-light LED panels have appeared in the last couple of years, causing much excitement, but Rift is taking this to the next level.

The organization is making these lighting units open-source, and provide the circuit diagrams and controller software free of charge; if their concept proves usable, we ought to see many similar designs produced at low cost in the near future.

I noticed that they are having trouble finding the correct method of producing a uniform illuminance under a change of color temperature. This is likely to be problematic for them, since the sensitivity of digital cameras to various color temperature conditions is variable across manufacturers and models of cameras. See here. They may need to adapt their system to include color calibration data specific to particular kinds of cameras. Likely this will merely require a custom color transfer matrix — nine numbers — that will provide closer mapping between what the camera sees and what the light source delivers. This color matrix could either be encoded as a camera model database in the device firmware, or the user could type in the numbers if they know them. A more clever method could be a built-in algorithm for color calibrating the camera: this little feature could be a strong selling point.

I also wonder about the color quality produced by this kind of unit. Since it uses only three colors — red, green, and blue — it may not operate equally well under all color gamuts used by digital cameras. Please recall that it is impossible to mix all possible colors from only three primary colors. However, if they concentrate on getting sRGB right, then this unit ought to operate well for 99% of uses. But there will still be metamerism problems: colors that look identical to the human eye will look different to the camera, and this kind of three-color lighting will make metamerism failure even more prominent. They may find they get better performance if they also include some other colors in their LED array to provide a more uniform spectrum.

This kind of lighting could be of great benefit for those who shoot RAW. Digital cameras have a fixed white balance: this raw data is processed to produce images correctly balanced according to the color of the light. However, lots of the RAW data is thrown away during this processing, which leads to increased digital noise, particularly in the red and blue channels. The Rift Labs product can be adjusted to produce a magenta light which will accurately produce an accurate RAW light balance: no data is thrown away, and the noise level is reduced to its practical minimum. This is likely only of interest to specialists, but it does show how useful this device can be.

UPDATE: Click here to read an article on color rendition under LED lamps. You may expect color shifts with using these kinds of products.

Contrast, part 1: Black and White

2011-07-01T19:00:00.009-05:00

A FRIEND ASKS, “What is contrast?”

Simply, contrast in photography is what differentiates one thing from another because of differences in tone. A contrasty image will have lots of black and white tones, while a low-contrast image will have lots of medium gray.

The English word ‘contrast’ comes from the Latin contrastare, meaning ‘stand against’, and so in a contrasty image, some elements will strongly stand against others, instead of blending together.

Here is a more-or-less visually uniform grayscale image that I'll be using as the basis for examples:

It goes from black on the left, to white on the right. I threw in a little bit of noise to illustrate my examples better. I used this contrived image instead of a real photograph in order to more accurately show what is going on. Please don't get too bored with this repetition.

Here is this grayscale's histogram in Photoshop:

The grayscale illustrate how many pixels are in each range of tone; with black pixels on the far left, and white pixels on the far right. The noise adds lots of jitter to this histogram, but please note that it is otherwise fairly uniform going across; on average, there is just as much black as there is white, and just as much middle gray as either of those.

Photographically, we can use the word contrast in various ways. Global contrast — which is what we are mainly concerned with here — tells us how the total range of tones are divided up in an image; how much strong black and strong white are found in an image? Since our example image is uniform, we can say this has neutral global contrast.

Here I use the Brightness/Contrast tool in Photoshop:

I made this image with this Photoshop setting:

Setting the slider to 100 contrast gives us simultaneously more dark and more light tones, at the expense of midtones. Take a look at the histogram: we have more pixels at the far left and right.

Setting the slider to -50 gives us low contrast:

Now notice how we have more midtones, at the expense of fewer dark and light pixels. The histogram shows the pixels more clumped together in the middle:

Here are all three together:

High contrast on top; low contrast on bottom.

Now I don't normally use the Brightness/Contrast tool, since I can get more control over the final results when I use the Curves tool.

Here I apply Curves to make the image appear to be basically the same as the Brightness/Contrast 100% setting:

Note that the histogram looks pretty much the same. But also notice that the curve is a bit complex:

Here is how we interpret the curves:

The bottom axis shows us the tones of our original image; while the vertical axis shows us which tones are changed by Curves.
Note there is a diagonal line going from the lower left to upper right; any time the curve touches this line, that means that the corresponding tones are not changed. In our example here, pure white, black, and middle gray are not changed. Whatever is white, black, or middle gray in the original image will remain unchanged in the final image.
Whenever the curve drops below the diagonal line, that means the tones are brightened. Whenever the curve raises above the line, those tones are darkened. In this example, the light tones between middle gray and white become brighter, while the dark tones between middle gray and black become darker. The image has greater global contrast, because we simultaneously get more darks and more brights.
In this curve, 70% white in the original image becomes 80% white in the final image; likewise 30% white (= 70% black) becomes 20% white (or 80% black). 0%, 50%, and 100% remain unchanged.

The Curves box shown here is not quite the same as found in Photoshop's default settings. Here are my custom settings:

Now let us define a new term, ‘differential contrast’. Take a look at the curve above: notice how it is steeper at the midpoint, and less steep at the endpoints of black and white? If you were climbing a hill with that slope, you would likely have the most difficulty in the midsection, where the curve is steeper. What this means is that there is more differentiation in the midtones; visually they appear more distinct from each other. Also note that there is less differentiation between the various dark tones and the light tones.

Whenever we get more differential contrast in one range of tones, inevitably we will get less differential contrast in other range of tones. As a photographer with Photoshop, you have control over adjusting which range of tones get the greatest differential contrast. You ought to consider adding differential contrast to those most important tones in your image, by making sure that the steepest part of your curve extends over that range of tones. But this can only be done by sacrificing other ranges of tones.

From what we can see with the curve above, it is apparent that the standard Photoshop Contrast tool adds differential contrast to midtones, at the expense of both shadows and highlights.

If we adjust the curve so that the line is steepest at the endpoints, and less steep in the middle, then we will get a low contrast image; it will have greater differential contrast in the shadows and highlights at the expense of the midtones. The overall global contrast will be less because we will have less dark and less light, with a lot more middle gray.

From our little experiments here, we can see that the Photoshop Contrast tool is a bit like Curves, but only where the endpoints and middle are fixed. We don't have to keep our endpoints fixed in Curves. For example, I can get much lower contrast in Curves:

This image does not go from pure black to pure white, rather, its range is significantly reduced.

Note that the histogram does not go all the way to the edges; indeed we are using only half of the dynamic range available. In our curve, 100% white is transformed to only 75% white, and likewise for black. Our total global contrast has been reduced greatly. But unlike the curved curve shown above, the differential contrast between tones is the same across all tones, since our curve is straight here.

If we make the curve completely horizontal, we lose all contrast:

Our histogram is now simply a spike at middle gray. Our image has no contrast whatsoever.

If we make the curve vertical, we get maximal contrast:

Please note that I added some noise in the image to simulate texture.

Actually Photoshop does not let us make a true vertical line, but this is pretty close. On the histogram, we have spikes at pure black on the left and pure white on the right. We have lots of differential contrast at the middle gray, with none for darker and lighter tones.

Take a look at the data found in the Histogram boxes in the images above. Note that in all cases, the Mean value is close to 127.5, which is medium gray. We did not change the overall brightness of the image with our adjustments. The next value, ‘Std Dev’ (or Standard Deviation) is a measure of how much the pixel values deviate from the mean. A large standard deviation is high contrast, while a low standard deviation indicates low contrast.

One of the great powers of Curves is that we can increase or decrease differential contrast in whatever range of tones we desire. In our examples above, the middle point is always fixed, and so differential contrast is adjusted only for the midtones. With Curves, we can adjust the contrast in the highlights:

Note that here we have greater distinction in the brightest tones in the image (you can see lots of my added noise there), and we have essentially created a low-key image.

But this distinction in the highlights comes at the expense of losing lots of our shadow detail. The tall spike at the leftmost part of the histogram unfortunately does not give us a good impression of how much of the image is lost in black without significant texture: roughly 40% of the pixels are pure black.

We can also increase visual detail in the shadows at the expense of the highlights:

We now have a high-key image.

Here our curve and histogram are reverse of what we had before, and 40% of our highlight detail is lost as pure white.

OK, let's try these techniques with a real image. This is a low-contrast, dim, black and white image of a monument found in Sylvan Springs Park, in Saint Louis County, Missouri, USA.

Bleh. Looks poor, and here is its histogram:

Most of the image's pixels are on the dark side of the histogram. Unless this is an exercise in conceptual art, most photo editors at magazines would reject this image: most would probably like to see a broad histogram, with significant contrast and a full range of tones from black to white.

Merely using the Contrast tool in Photoshop does not help:

However, boosting the brightness here would help a lot.

Instead, I will adjust the image using curves. First I put in an initial curve, which will expand the range of the significant part of the image:

Straight line curves are equivalent to using the Levels tool in Photoshop.

Then I add a strong curve to the most significant part of the image, which I think are the tones around the carved lettering:

The steepest part of the curve, and therefore the greatest amount of differential contrast can now be found at the tones surrounding the text “1939”. I don't care to see texture in the darkest part of the grass or in the shadows between the rocks; I think that overexposing the patch to the left of “Sylvan” is also acceptable.

I added some sharpening, and the photo is basically complete.

Sharpening is a kind of local contrast enhancement, which is a powerful technique for improving images; but that is for another day.

Adding contrast — and using curves — in a color image is far more problematic, since we can easily change the hues of our colors. That is also a topic for later.

Examples of Color Mixing

2011-06-20T21:18:00.003-05:00

I PURCHASED ADOBE Photoshop CS3 a number of years ago for one and only one reason: color management.

I was the owner of a Minolta Dimage 7 camera, and was greatly disappointed in its photos. Naturally, I blamed the camera and had huge buyer's remorse. One of the many problems was that the camera took photos in a non-standard colorspace. According to the camera review linked above:

Looking over the D7's images I couldn't help but feel that certain colours seemed under-saturated (mostly greens and blues)...

At this stage it was clear to me that the DiMAGE 7 was shooting in its own colour space...

This is not documented in the DiMAGE 7 manual. I feel it should be made very clear to users, there's certainly a chance that the average user will simply load images directly from the camera using a card reader and never use the Minolta Image Viewer. These users may well end up disappointed by the D7's colour.

The average user won't know what colour space they're in, indeed most users don't even calibrated their monitors. However, at a consumer level, most of you will be viewing this web page and all the digital photographs you ever deal with in the sRGB colour space.

I admit that at the time I didn't know much about color spaces (and really I still don't). My Dimage images looked poor for various reasons, and the processing utility that came with the camera produced very dark, unacceptable images. Despite doing some research, I really didn't know how to correct for these problems, most notably the color space problem.

Then at Christmas 2006, I received a copy of Adobe Photoshop Elements. By that time I had already purchased a newer, somewhat inferior camera, but it produced great images right out of the box. But I tried using the Adobe Camera RAW software in Elements to process my old Dimage photos — and behold! — they looked great. The Adobe software knew all about the Dimage color space problem and correctly converted my images to the Web standard sRGB colorspace, and solved other problems too. By the next summer, I upgraded to Photoshop CS3 to gain access to the more advanced features in Adobe Camera RAW.

To the left is a typical JPEG image I got from the Dimage. On the right is an image I shot in RAW and processed with ACR.

So I was a new owner of Photoshop. But I didn't know how to use it much beyond the excellent RAW conversions. Many of the functions seemed to be useful for only producing odd special effects; although I didn't know it at the time, I was using a chain saw instead of a scalpel.

For example, I thought the Curves and the Channel Mixer functions were only used to produce special effects as seen here. Perhaps you can, but that is neither the intent of the tools nor is it their highest and best use. In the bottom image, I randomly adjusted the sliders in Channel Mixer.

Thanks to Dan Margulis' excellent book, Professional Photoshop, Fifth Edition, I learned a few tricks on the right use of some of the Photoshop tools. For example, Dan showed how to brighten green foliage using Channel Mixer:

Following are the adjustments:

Searching around the web, I see that most people use the channel mixer as a kind of mysterious saturation or desaturation tool. Above we see how it is used to brighten a color. It is often used as a black and white conversion tool: all you need to do is check the 'Monochrome' box and adjust the sliders until you get the contrast you desire in your conversion.

For a while I've been interested in the Purkinje effect, which describes the shift in colors we experience as light gets dimmer. I was prompted to do this in response the the poor quality of photos I took around dusk: how, I thought, could I reliably and repeatedly correct my photographs to look something like I remember seeing it? My earlier efforts at coming up with a Purkinje correction are here and here.

This correction is a two-step process. First I adjust the relative tonalities of the colors in the scene. When light gets darker, the eye becomes more sensitive to blue-green light, and simultaneously gets less sensitive to red colors. In particular, foliage has a strong red color component in daylight which makes them appear yellow-green; under dim lighting, foliage looks relatively darker. I also noticed that I had to adjust the white balance towards a sky blue color.

In the scene above, I distinctly recall the foliage being quite dark, while the fountain (seen in the lower right corner), the limestone edging on the sidewalk, and the columns appeared much brighter than the surrounding areas. In a Luminosity layer, I used the channel mixer to both boost the brightness of the blue channel and to darken the red channel:

I could not just use this channel mixer indiscriminately across the image; the yellow light in the window in the tower was also darkened unrealistically. In this image, I applied the channel mixer with masking so that it did not change the brightest tones in the image.

I got these numbers in the channel mixer by comparing charts of light response of the dark adapted eye versus the sRGB primary color response to light. Measuring the response values across the color spectrum and doing some statistics, I was able to determine which mixture of color components best simulated the eye's night-adapted response. Here we are getting very close to the intended use of the channel mixer.

The most precise use of the Channel Mixer is for converting between various RGB color spaces. Please consider how I started the article: my old Dimage camera had an odd color space, which if not handled properly, would produce disappointing images and poor color rendition.

Every digital camera has a native color space, and it isn't the Internet standard sRGB or Adobe RGB. Every brand of digital camera has a native color gamut that is whatever. The electronics within the camera do the conversion between the camera's native color space and sRGB; or, if you shoot in RAW, the software on your computer (like Adobe Camera RAW, Lightroom, or Apple Aperture) does it for you after the fact. Digital cameras are sensitive to the various colors in a manner which is quite unlike the human eye's response. Practically speaking, a digital camera is limited by cost, manufacturing technique, and long-term stability; while the human eye is precious beyond cost, has the most complicated manufacturing known, and repairs itself. Therefore it is unrealistic to expect that cameras will see color precisely as is seen by human eyes.

As it turns out, it is very difficult to get precise color capture capability on a digital device, and forget about trying to capture color precisely as the human eye sees it, especially with exceptionally pure colors. (Technically speaking, digital cameras do not meet the Luther-Ives Condition, which determines if a camera can capture colors accurately.) There are cameras that do have better color rendition than consumer or pro-level cameras but a) you can't afford them, b) you don't want to spend the time getting them to work right, c) these cameras only work in controlled laboratory conditions, and d) you won't be able to display your accurate colors on a computer monitor or print them on a printer. So we are looking for good enough colors.

So we take a digital camera that has a color space of whatever, and we use the computer to convert the images to a standard color space like sRGB. Here is an example photo taken with my Nikon D40. I shot this X-Rite ColorChecker Passport in daylight using RAW capture. I set the white balance using the neutral card in the bottom photo. Here I used Nikon View NX2 software to convert the RAW image to JPEG in the sRGB color space:

I held up the target in daylight coming through my office window, and compared it to this image on my screen. The colors look pretty accurate. But this is not the image as the camera captures it. There is some significant processing going on to produce this final image. Here is an approximation of how the camera really sees the target:

I used the free and excellent RAW Photo Processor for Mac OS X. I set the white balance to UniWB, set the gamma to 1.0, and saved the image as raw pixels without a color profile loaded. By looking at this image we can see that the camera is very sensitive to green light, less sensitive to blue, and not very sensitive to red light. All cameras have a fixed, native white balance which does not correspond to any common lighting condition. Because digital cameras capture more than 8 bits per channel — usually in the range of 10 to 14 bits per channel — these extra bits are used to prevent banding artifacts when the software does this kind of processing. For daylight, the camera must amplify the blue channel by about 50% and approximately doubles the signal in the red channel. Under incandescent lighting, the red and green channels are usually fine, but the blue channel must be boosted strongly, which typically causes lots of noise.

Let's do a white balance. I used RAW Photo Processor to apply the white balance which I captured when I took the photo:

Cameras don't see light like humans do. In a camera, twice the light intensity generates twice the RAW signal; with humans, doubling the intensity of light makes a scene look only somewhat brighter. Likewise, halving the intensity of light only makes a thing look slightly darker. And so digital images are designed to provide extra sensitivity to the darker tones at the expense of brighter tones; this will allow us to more efficiently use our image data by allocating more data to the important midtones. To correct for this, we have to brighten the midtones in the camera's image, while keeping the very darkest and brightest tones unchanged.

Here I applied a brightness (or gamma) correction in RAW Photo Processor. This is not the precisely same brightness curve as is found in the sRGB standard, so the tonality of this image doesn't quite match the Nikon-processed image. But it is good enough for our purposes.

The first thing I notice is that the colors are disappointing. They look a bit flat and unsaturated compared to the first image. But this is actually a good thing. It tells me that my camera's native color space — whatever it is — is broader than the sRGB color standard. This flatness tells me that the colors could be potentially much brighter, if only I use a large enough color space, one larger than sRGB. In the camera native color space, the bright red color on our X-Rite target is merely mediocre.

To do this conversion from the camera native color space to sRGB in Photoshop, we imitate the processing used by the camera manufacturers by using Channel Mixer.

Suppose you take a given standard color, and take a photograph of it, being careful to do a good white balance and exposure. Good standard colors ought to produce a particular sRGB color value if exposed correctly; for example, a pure red the same color as the sRGB primary red color ought to give you R=255, G=0, B=0. But since your digital camera uses a different color system, it gives you another value, such as R=200, G=58, B=27. Your digital camera likely has a primary red color that is brighter, more saturated, and more pure than what the sRGB standard can describe.

So what we do is to boost the camera's red value somewhat, and then subtract out any green or blue that happens to be contaminating the red. So for any given Camera RAW value, we can get the equivalent sRGB values.

I got test data from DxOMark, a company that measures performance data for digital cameras and lenses. Here is an excerpt from the dataset:

Go to http://www.dxomark.com/index.php/Camera-Sensor/All-tested-sensors/Nikon/D40, click on Color Response, and click CIE-D50. This gives you daylight color data. They also provide CIE-A data, which models incandescent lighting.

Please note the White balance scales: this shows how much we should boost the various RAW color channels to get good white balance under D50 daylight conditions. Please note that D50 is much bluer than midday sunlight at mid-lattitudes, but is a good enough approximation for our use here.

To convert the RAW data to sRGB, we put the DxOMark Color matrix data above into the Channel Mixer:

Please note that due to rounding errors, the numbers in each Output Channel do not total to 100%. Undoubtably this will shift our white balance slightly. Generally speaking, be sure that each channel mixer output channel totals to 100% to avoid a change in white balance. However, the final image looks pretty good:

The colors are fairly close to the Nikon-produced colors.

You can use the same sort of process to do your own camera calibrations under all kinds of lighting conditions. Please note that the DxOMark color matrix data assumes the use of something like the channel mixer to convert from the camera's native color space to sRGB. For example:

sRGB Red = (1.64 x Camera RAW Red) - (0.61 x Camera RAW Green) - (0.02 x Camera RAW Blue)

If you have a target with known sRGB (or other color space) values, you can convert an image to these colors by comparing the delivered colors with the known colors. I use Microsoft Excel to come up with estimates for the channel mixer values, using the Solver tool found in the Analysis ToolPak add-in. This is a rather complicated step, and requires some knowledge of statistics.

Please note that this mathematical transformation between color spaces is not exact, but is rather statistical in nature; the conversion matrix merely gives good color conversions on average. A severe channel mixer setting will also cause much more noise in your image.

"The False Photographer"

2011-06-13T01:07:00.000-05:00

ON THE LIMITS of photography:

This weakness in civilisation is best expressed by saying that it cares more for science than for truth. It prides itself on its "methods" more than its results; it is satisfied with precision, discipline, good communications, rather than with the sense of reality. But there are precise falsehoods as well as precise facts. Discipline may only mean a hundred men making the same mistake at the same minute. And good communications may in practice be very like those evil communications which are said to corrupt good manners. Broadly, we have reached a “scientific age,” which wants to know whether the train is in the timetable, but not whether the train is in the station. I take one instance in our police inquiries that I happen to have come across: the case of photography.

Some years ago a poet of considerable genius tragically disappeared, and the authorities or the newspapers circulated a photograph of him, so that he might be identified. The photograph, as I remember it, depicted or suggested a handsome, haughty, and somewhat pallid man with his head thrown back, with long distinguished features, colourless thin hair and slight moustache, and though conveyed merely by the head and shoulders, a definite impression of height. If I had gone by that photograph I should have gone about looking for a long soldierly but listless man, with a profile rather like the Duke of Connaught's.

Only, as it happened, I knew the poet personally; I had seen him a great many times, and he had an appearance that nobody could possibly forget, if seen only once. He had the mark of those dark and passionate Westland Scotch, who before Burns and after have given many such dark eyes and dark emotions to the world. But in him the unmistakable strain, Gaelic or whatever it is, was accentuated almost to oddity; and he looked like some swarthy elf. He was small, with a big head and a crescent of coal-black hair round the back of a vast dome of baldness. Immediately under his eyes his cheekbones had so high a colour that they might have been painted scarlet; three black tufts, two on the upper lip and one under the lower, seemed to touch up the face with the fierce moustaches of Mephistopheles. His eyes had that "dancing madness" in them which Stevenson saw in the Gaelic eyes of Alan Breck; but he sometimes distorted the expression by screwing a monstrous monocle into one of them. A man more unmistakable would have been hard to find. You could have picked him out in any crowd—so long as you had not seen his photograph.

But in this scientific picture of him twenty causes, accidental and conventional, had combined to obliterate him altogether. The limits of photography forbade the strong and almost melodramatic colouring of cheek and eyebrow. The accident of the lighting took nearly all the darkness out of the hair and made him look almost like a fair man. The framing and limitation of the shoulders made him look like a big man; and the devastating bore of being photographed when you want to write poetry made him look like a lazy man. Holding his head back, as people do when they are being photographed (or shot), but as he certainly never held it normally, accidentally concealed the bald dome that dominated his slight figure. Here we have a clockwork picture, begun and finished by a button and a box of chemicals, from which every projecting feature has been more delicately and dexterously omitted than they could have been by the most namby-pamby flatterer, painting in the weakest water-colours, on the smoothest ivory.

I happen to possess a book of Mr. Max Beerbohm's caricatures, one of which depicts the unfortunate poet in question. To say it represents an utterly incredible hobgoblin is to express in faint and inadequate language the license of its sprawling lines. The authorities thought it strictly safe and scientific to circulate the poet's photograph. They would have clapped me in an asylum if I had asked them to circulate Max's caricature. But the caricature would have been far more likely to find the man.

— G.K. Chesterton, “The False Photographer”, from A Miscellany of Men.

“Prime Subjects of Photography”

2011-05-20T23:28:00.000-05:00

IF YOU ARE interested, please see my new series of challenges at the Digital Photography Review website, called the Prime Subjects of Photography.

My first challenge, Unconventional Portraiture, is currently in the voting stage, while the current challenge, Photography of Flora, is generating an astounding amount of interest. This will be followed by:

Sports Photography
Cityscapes (urban landscapes)
Night Photography
Environmental Photography (taking a portrait of a person in the context of their work or home)
Food Photography
Street Photography (photos in public in an urban area)
Modern Landscapes (landscape photos with extreme sharpness and focus, with little Photoshop afterwards)
Abstract Subjects
Child and Youth Portraits
Wildlife Photography (large animals in the wild; later challenges will concentrate on small wild animals and insects, and birds)
Concert Photography
Architectural Photography
...more challenges planned...

The challenges in the series have a particular structure, with easier subjects coming first, with these being reflected by similar yet more difficult subjects later on. Unconventional Portraiture came first, because it has few to no rules which ought to be followed, while Classical Portraiture, which is far more difficult, comes later.

Typically, human subject challenges alternate with nature studies and inanimate objects. More specialized challenges come later in the series. The entire series will have 33 challenges, and each challenge includes basic hints for good photography for that particular subject.

The dpreview website is largely suited to beginning photographers. The challenges on this site serve rather well as a learning tool.

Zillions and Jillions

2011-05-03T14:39:00.008-05:00

PLEASE CONSIDER the absolutely most minimalist camera possible. This camera will have precisely one pixel or light sensor location, and it will generate a photograph that has a bit depth of exactly one binary digit: black or white. This camera can produce precisely two images:

This maximally minimalistic digital camera is actually useful. These are incorporated into proximity switches; devices that answer the question is something there? You might find them on conveyor lines in factories, or on automatic door openers.

Instead of just one pixel, let's consider a camera with four. Here are all possible images taken with this sort of camera:

With just a 2 by 2 pixel array, at one bit depth, we are able to take 16 different photographs.

We can easily calculate the total number of individual photographs that can be possibly taken by a digital camera capable of displaying only black or white:

1 bit camera = 2 photos
2 bit camera = 4 photos
3 bit camera = 8 photos
4 bit camera = 16 photos

We add one bit and we double the number of possible photos we can take with our camera. This number gets very big very fast. Suppose we take a very low end digital camera, 500x500 pixels = 250,000 total pixels. Even if we limit this camera to using only black and white pixels, we end up with a huge total number of possible images:

Total images possible =

2 x 2 x 2 x ..... (multiply a total of 249,999 times)
= 2²⁵⁰⁰⁰⁰
= 3 followed by 75,257 zeros, plus a bit more.

By comparison, the total number of elementary particles (electrons, protons, etc.) in the entire universe is estimated to be a 1 followed by 80 zeros, or 10⁸⁰.

Suppose you have an entire universe of particles, and then you give every particle its own universe of the same size, and in each of those universes, you assign a similarly sized universe for each elementary particle inside them, and repeat the process nearly a thousand times. That's how many unique photographs you can take with your cruddy 500x500 pixel 1-bit depth digital camera.

We can get large numbers when we examine matter, but information is another class of being in itself, vastly larger than mere matter. It is for this reason that philosophers have posited that information resides in a realm above, beyond, or outside that of mere matter.

The following ought to convince you that an image 500 pixels on a side with 1 bit pixels can be quite rich:

My fellow Saint Louisians ought to recognize this scene. 1-bit images are generally quite useful, and have been used for decades in copying machines.

Even a 1 bit depth digital camera can produce an astounding number of different images. But suppose we add color — the number of possible images becomes even more staggering. Take my lowly Nikon D40 camera; it has about 6 million sensors, and if I shoot JPEG, I get 256 possible values per sensor. The camera has about 3 million green sensors, and 1.5 million sensors each of red and blue; full color is mathematically estimated for each pixel location.

So for each pixel location, we multiply by 256. The total number of images = 256^6016000 which is approximately equal to a 1 followed by 14,487,972 zeroes. This is a mere pittance compared to the Seitz D3 digital scan back, which can capture 500 megapixel images with 48 bit color per pixel:

Total = (2⁴⁸)^{^500000000} = 1 followed by 7,224,719,896 zeros.

I think it is safe to assume that like snowflakes, no two photographs are alike.

But this vast potentiality of photography ought not get us puffed up with pride in our creativity. Alas, the vast, overwhelming majority of these theoretically unique photographs look like this:

Uniform random noise. A trillion monkeys, each generating a trillion random images per second for a trillion years, will likely never once produce anything that looks like a photograph. You could call this an ‘image’, but only in the most general terms. At best, images like this are only an exercise in conceptual art, and then only for the first photographer who does it. And I just did it.

[Note: to generate a random noise image such as this in Photoshop, be sure to start with a 50% gray image. If your starting point is a white image, then half of your final pixels will be white, and if you start with a black image, then half your pixels will still be black. Using the Filter->Noise->Add Noise... function, add 50% Uniform noise. Be sure to turn off ‘Monochromatic’. I find I get better results if I do this for each color channel independently.]

We call an image like this ‘random’ because it doesn't look like anything in particular. Perhaps a supremely intelligent being can look at this and be able to immediately discern the specific algorithm used by Photoshop to generate this. But we can't; we are finite creatures, and so the perception of noise is very much a human phenomenon. Perhaps we can imagine some figure in this noise, but that is tenuous at best. Actually, the entire concept of randomness is problematic: see my article here.

But we are good at perceiving some image if the noise isn't too great, especially if we are familiar with the subject:

That is an image with a 1:10 signal/noise ratio. Can you guess the subject? What does your gut say?

I've never before attempted to estimate signal to noise ratios for digital images. Here, I've created a series of images with varying relative amounts of noise:

Clearly, high ISO images, and heavily manipulated images, can have very high relative amounts of noise, much more than I suspected.

A 1:4 signal-to-noise ratio is roughly the limit I'm able to handle, while surprisingly a 1:1 ratio isn't horribly bad for some purposes. But please note that noise found in actual digital images is not uniform; rather it lurks most of all in the shadows, or rather, the signal-to-noise ratio is smallest there, for highlights have a large absolute amount of noise, even though it is small relative to the signal.

Suppose we are willing to accept a 4:1 signal-to-noise ratio, which means that 1/5 of our bits are just noise. For this image, which is 500 pixels on an edge, with 48 bits of color, that means that about 10 bits are noise. We can calculate:

Number of acceptably noisy images equivalent to a reference image = (2¹⁰)^{²⁵⁰⁰⁰⁰} = a 1 followed by 752,576 zeros.

2% total noise is virtually undetectable by the human eye, unless you look very closely. Also, slight adjustments or rotations of images are hard to see:

For our 500 x 500 pixel full-color image, I estimate that there are approximately 4 trillion largely undetectable variations for every reference image. For larger images, the number goes up to values never heard of even in government economics.

OK, so I assert that of all theoretically possible images, the relative number which would be recognized as photographs is vanishingly small. Fortunately, since we are dealing with almost unimaginably large numbers, this isn't an issue. Roughly estimating the number of photographic images possible ought to be doable.

Patterns make an image. Either there are significantly large adjacent patches of pixels that are very similar, or there are similarities between pixels even though they are widely separated. Or, we can find a pattern that repeats on various scales.

Look at this image:

It is pretty obvious that we have symmetry of the left and right halves, as well as the top and bottom halves. Recall that the total number of possible variations of a 500x500 pixel, 1 pixel depth image is represented by a 3 followed by 75,257 zeros. Because we have mirror symmetry, we basically repeat a 250x250 pixel image four times:

Total number of images:

= 2^{250 x 250} = 2 ⁶²⁵⁰⁰
= about 1 followed by 18184 zeros

This is an enormous figure, but is only 1/(10⁵⁶⁴³³) fraction of the total number possible. We will be able to see the mirror reflection in nearly all the images produced: we would not perceive the symmetry if the image is all black or all white, or if the pattern is symmetric to begin with, but this is a very small proportion of the total.

With this 1 bit depth image, I find that I can only detect symmetry when I have greater than a 1:1 signal to noise ratio. We also should be able to detect other variations, such as changing the location and orientation of the axes of symmetry — all we have to do is ensure that our axis of symmetry is not too close to the boundary of the image.

The eye can detect global patterns such as seen above, and also local patterns, where there is some correlation between adjacent pixels:

Within a 500 x 500 pixel image, we can produce a total of 400 x 400 = 160,000 different images of a black 100x100 pixel square; our numbers go up if we can accept slight variations in the color and size of the square and its orientation. But if we are willing to accept some noise, as we see here, we can get zillons of possible images.

No two photographs are alike, and even if you take multiple shots with an ordinary camera under controlled conditions, there is no chance in your life that your resulting images will be exactly the same. This can be helpful: if you take multiple shots, you can blend them together to greatly reduce the amount of noise visible in the final image. Super-resolution techniques can also use multiple images to construct a higher-resolution final image, and can even remove diffraction artifacts.

The sharpness of lenses is a major limiting factor for producing unique photographs. The blurring found in some optics means that adjacent pixels are often strongly correlated, even if they are supposed to capture a high-contrast edge. Chromatic aberration and the mysterious purple fringing also lead to greater correlation between pixels, reducing the originality of photographs. Same goes with noise reduction software. Extreme, high quality bokeh, found in excellent portrait lenses, can reduce the out-of-focus background detail to a nearly uniform blur; this lack of detail leads us to focus our attention on the subject of the photograph. With extreme blurring and the use of the Photoshop Threshold tool, we can even reduce our images to match the 1-bit camera demonstrated at the top of this article.

Despite, or rather because of, the huge variation of possible photographs, it is relatively easy to detect unauthorized duplicates of images. Forensic analysts can detect duplicates with overwhelmingly high certainty, even if the original image was severely altered. Likewise, the use of the Clone tool in Photoshop — this copies one part of an image to another part — is easily detected if a large enough area is cloned, even if it is visually blended well, because these kinds of correlations within an image can not be practically attributed to chance. Never in a jillon years can we expect something like that to happen on its own.

If two photographers with two different cameras each take a photograph of the same scene, they will be different from each other in a huge number of minor (or even major) ways, so much so that we can be absolutely certain that the two images are in fact different. On the contrary, with a reasonably complex scene and good image quality, it seems that we ought to be highly confident that these two photographs were in fact photographed at the same time, and we also ought to be able to detect if the scene was artificially recreated at a later time, or was adjusted in Photoshop.

I think this discussion of seemingly impossibly large numbers tells us that photography, and digital art in general, is an incredibly rich and humanly inexhaustible medium.

[Click here for a discussion of names for long numbers. As it so happens, practically the only time these names for long numbers are actually used are in lists of names of long numbers. The word ‘zillion’, even though it is almost meaningless, is a good enough name for the quantities we are discussing here.]

Photography as an Art

2011-04-12T17:30:00.000-05:00

THE GREATEST OBSTACLE the modern photographer encounters is his adherence to an idea that the camera "holds a mirror up to nature," that it is "true to nature." If that were so, photography would be for all times contained among the sciences and debarred from art. For nature is never art, nor does nature as a whole ever affect us as art. In art we are dealing strictly with the mental and emotional faculties more or less developed in each individual. These faculties respond when, on a flat surface such as paper, we find certain emotional and intellectual records of things we have seen or experienced in nature. And it is the manner in which these records are made that affects us as art. Every stroke, touch, spot, and patch of light and dark governed by the mind and hand of the artist interprets first an emotion, second a meaning. In this lies the province of art. The "mirror of nature," as expressed by photography, is a cold, impersonal, undesirable tracing of certain facts reproduced by pure science — heartless, uninteresting. Its value is wholly scientific, and it deals with only one kind of truth. There is nothing impressionable or impressive about it. Pictorial art is strongly emotional. It exists to give pleasure and at the same time knowledge; not such knowledge as the dissecting sciences impart, but the kind inherent in music, poetry, literature, religion.

Nature in itself has nothing to do with art; it is only the quarry, the reservoir out of which material for art can be taken. It is plain, then, that "true to nature" cannot refer to the comprehensive truth, but that of necessity selection of truths must be resorted to in any event. This being so, the phrase "holding a mirror up to nature" is evidently meaningless from the standpoint of art, and "true to nature" must be understood as referring to a phase of nature of which we have become conscious.

— Art principles in portrait photography (1907), by Walter Beck, p. 18-19

Frequently Asked Questions

2011-04-07T17:15:00.003-05:00

HERE ARE A FEW questions photographers often get from the public.

Q: What kind of camera should I get?

What do you want to do with it?

Oh, I don't know. I think that I would like to get into photography. It might be a fun hobby; I've always been creative.

I would suggest getting a cheap one, a camera that has just the basics and manual override. A camera store can help you with this. Get a camera cheap enough so that you won't get buyer's remorse or get too upset if you drop and break it. Also get some inexpensive photo editing software for your computer, as well as an introductory book on photography. Play with the camera and software, and follow the examples in the book to see if you like photography. Maybe after six months you will find that you want to upgrade your camera, or you will find out that you aren't interested in pursuing it further. In any case, your investment is insignificant.

Q: What kind of filter do I need to protect my lens?

None.

You have to very careful with the kind of filter you use, for cheap filters will optically degrade your images, sometimes severely. If you do need one, B+W filters are highly regarded.

But if you spent so much on a lens that you worry about damaging it, then you ought to worry about image quality degradation — otherwise a cheaper lens would suit your purposes. Be careful in handling it, and a rigid lens hood and lens cap is your best protection against damage. I would use a filter only if it is needed creatively (like using polarizers to cut reflections or darken blue skies), or if you are in a salt-air environment.

Q: I want to get my wife the ultimate anniversary present. What is the best camera you can buy?

Might I suggest the Nikon D3x? It is very popular with professionals and takes superb photos. You will need a variety of lenses to go with it. I would suggest getting the 400mm f/2.8G lens for taking photos of wildlife, the 200mm f/2G for indoor sports pictures, the 14-24mm f/2.8G, a wide angle lens, particularly suitable for taking photos indoors, the 135mm f/2.0 DC and the 85mm f/1.4 for taking portraits, the 60mm f/2.8G Micro lens for close-ups, and the 28-300mm zoom lens for general purpose use.

Please click the links above to purchase. My associate will provide you with excellent service.

Q: I have a bunch of 512 MB CF cards. I want to buy a new camera that will use these.

Why do I hear this question so often?

The card you use in a camera is really not that important. Buy whatever card the new camera uses. You will find that the new cards are cheap, and are likely faster and more capacious than your old cards.

Q: How do I get good bokay?

Perhaps you mean bokeh? The word bokeh comes from Japanese, and is used to describe the quality of background blur in an image.

These factors contribute to getting a stronger background blur: getting closer to the subject, opening up the lens aperture, using a longer focal length, and using a camera with a larger sensor. Now the quality of the background blur can be harsh or creamy, and this depends of the lens design itself; some lenses, especially those made for portraiture, will have better bokeh than others, and will be priced accordingly.

Q: My wife has a Nikon D3x camera and 7 lenses that she wants to sell because they are too big and heavy to be practical. Where can I get a good price for used equipment?

You can sell your equipment to Adorama or B&H Photo. These large dealers are likely to give good prices for high-end equipment.

Q: Thinkin about turnin pro. What camera do pro's use? Is $200 too much to charge for a wedding?

Professionals use whatever camera they want to use. And $200 is way too much for you to charge for a wedding.

Q: What DPI setting should I use for my photos?

The dots per inch setting does not matter at all, and it will have no effect whatsoever when you print out your photos or display them on your computer. All that matters is the pixel dimensions of your image. Too few pixels will look terrible if you attempt to make a large print of an image.

DPI does matter when you submit your photos to a commercial publisher. Typically book and magazine publishers will specify a fixed DPI setting: usually 300 dots per inch. When you submit your photo, they will not resize your photo at all, they will merely crop it to fit. For this reason, the publisher will specify both an image size and a DPI size, and your camera must have enough pixels to ensure that your image will cover this size. They will typically specify a size larger than they will use, to allow for bleed off the edge of the paper and to give them flexibility of layout. As a photographer, this is actually very helpful, for you process your image as you want it to look: resizing photos can degrade quality greatly.

Q: What kind of camera do I need to take good indoor pictures of my kids?

Are you having problems?

Yeah, my current camera is slow and my kids are fast. If I do manage to get a photo, either the photo is really grainy, or the kids are blurry, or the flash goes off and the picture doesn't look all that great.

I would suggest getting a low-end DSLR. Most large electronics retailers and camera stores will have fairly inexpensive cameras from Canon, Nikon, and other manufacturers that will solve your problem.

DSLRs in general produce superior results because they have larger sensors than compact cameras. This reduces the problem you see with grain and blur and reduces the need for a flash. Because these cameras have a mirror and separate viewfinder, they are able to incorporate a much faster focusing mechanism than what is found in compacts.

The trade-off is that these cameras are bulkier and heavier, so I would recommend going to the store and trying it out yourself. Usually, people get used to the larger size fairly quickly and the quality improvement is usually worth it.

The lens that comes with your camera is probably adequate. However, you may want to spend about $100 more and get a fast lens, like a 50mm f/1.8. This will let in more light and will allow you to avoid blurring and avoid using the flash in dimmer light.

Q: Why are my interior photos yellow?

That is due to the relatively yellow color of lighting used inside.Your camera has a built-in automatic white balance which attempts to correct for the color of light, but it doesn't always work well. Take at look at my article White Balance, Part 1.

You might try manually setting your camera to incandescent, tungsten, or fluorescent; just remember to set it back to automatic when you are done.

For really good work, you can calibrate the camera to match the lighting precisely. For this you need a target that is guaranteed to be neutral in color. Click here for an inexpensive card.

Q: I want a compact camera with excellent image quality but I don't want a DSLR.

Well, that will be a problem. DSLRs, due to their design, will produce better...

I just told you that I don't want a DSLR! Don't you listen? Or are you just stupid? I want a compact camera that will take as good pictures as a DSLR.

Your options are limited — it could be quite expensive —

Money is no object.

Well, in that case, might I recommend the Leica M9? Like DSLRs, it incorporates superb interchangeable lenses, among the best in the industry, and it has a full-frame sensor like those found in the highest-quality DSLRs. Because of its compact size and silent shutter, it is prized by photojournalists who need to be discrete. And may I suggest an assortment of lenses to go with it...

Um — well — sorry. I'm afraid that is a little beyond my budget. Can you recommend something else?

I think the Sony NEX-5 or the Fuji X-100 would be perfect for you. Not quite as flexible or high performance as a DSLR, but both of these will deliver excellent image quality similar to a DSLR in a compact size. You might also investigate the Micro 4/3rds cameras, which have smaller sensors than DSLRs, but offer much greater quality than typical compacts.

Q: Black and white photography is obsolete. Why don't you get it?

What are you talking about?

Color is where it's at now. Black and white was merely a product of historical forces, ignorance, and technological compromises, and so it has no relevance to us today; black and white photography is something that ought to be abandoned and forgotten.

This sounds familiar. See my article Black and White.

Q: Why do my pictures look washed out?

What do you mean by washed out?

Well, I'm not really sure. But the colors are disappointing. They are pale. The sky is often white instead of blue.

It sounds like your photos are overexposed — that is really common. Take a look at your camera's manual and look up the exposure override function. If your colors look pale, then try adjusting the camera to expose the picture less — -2/3 E.V. would be a good start. You can also try to increase the saturation in the camera, but don't do too much of it. Take a look at my article Three Opportunities for Overexposure.

Unfortunately you now risk underexposing your shots — your pictures could look dark and muddy instead of washed out. See if your camera has a histogram feature: you want to be sure that your histograms don't show too much or too little exposure.

Q: Help! I was told to set my camera to Adobe RGB so that I can capture brighter colors, and now my colors are all muted!

Well, don't do that then.

Most cameras and computers use the sRGB color standard, which can capture about 35% of all colors. Adobe RGB is useful since it can capture 50% of all colors: but if you don't know how to handle managed color workflows on your computer then I wouldn't bother. See my article Imaginary and Impossible Colors.

Q: Should I use RAW or JPEG?

RAW is technically superior, but you are forced to do more work on the computer. Do you want to do no work at all on the computer, and will you do all your work in the camera? Or do you want to spend extra time sitting in front of your computer making your images look right? If you are going to do photo editing anyway, then shooting in RAW is useful since it gives you more flexibility in adjusting white balance and exposure.

Be aware that Adobe Camera RAW, as found in Photoshop and Lightroom, will not produce the same results as the camera's own JPEG. For this reason, I usually use Nikon's View NX2 software to do my RAW conversion before moving the image into Photoshop.

Q: What kind of camera do I need for sports?

If you are shooting sports outdoors in the daytime, an inexpensive point-and-shoot camera with a decent zoom lens should be adequate. If you are shooting sports indoors or at night, you will likely need a DSLR with a fast telephoto lens. This will be quite expensive to do well, and it would be far cheaper to hire someone to do it for you.

Q: What settings do I use to take pictures at night?

I can't tell you that, but your camera and your eyes can tell you. If you are far away from artificial lighting, set your camera's white balance to daylight, otherwise set it to tungsten or incandescent. Put your camera on a tripod. Set the self-timer, and let the camera make whatever exposure it wants. You may find that the camera can't focus, you will have to figure out the camera's manual focus feature. If the camera tells you it is too dark, and you have a manual exposure feature, then use it. Adjust the exposure time until the image on the screen looks good to you. Your exposure could be several minutes long or longer. See the Wikipedia article Exposure Value for hints.

Q: Can you recommend a good lightweight tripod?

I can recommend a good tripod or a light tripod, but not a good light tripod. If you are serious about providing a solid platform for your camera, I would recommend getting a heavy tripod, and a tall one at that. You don't want to lean over all the time to take pictures, and a high camera angle is sometimes superior. A heavy tripod will likely be both sturdier and much more stable than a light one. Having extreme stability is important when taking long exposures, such as at night or in dim interiors.

I've had bad experiences with light tripods. Vibrations from people walking across the floor, a truck rumbling outside, the wind blowing, or even the weight of the camera itself can ruin a photo.

Three Opportunities to Underexpose

2011-04-02T15:27:00.002-05:00

SPRING IS FINALLY here in Saint Louis, and we have spring's floral finery waiting for the camera.

When you photograph brightly colored flowers, be aware that you risk overexposure:

Overexposing bright colors is easy with digital photography — and also when using film for slides. This red color really isn't all that bright compared to pure white, for it is only a bit more luminous than a middle gray. But digital cameras can only record so much red before they become overwhelmed. Blue skies are often overexposed, as we see in the article Three Opportunities for Overexposure.

As we see here, even a slight overexposure can ruin an image; adding contrast or saturation (common settings in a camera that usually make images look better) will make matters worse. Just by looking at this photo you should be able to tell that it is overexposed: we have a large, nearly featureless areas of bright primary and secondary colors, namely the red tulip as well as section of the yellow flower on the right.

The color of each pixel in this image is determined by three numbers, each of which range in value from 0 to 255, with the numbers determining the amount of red, green, and blue light at each pixel. For an overview of this RGB color system, see the article Color Spaces, Part 1: RGB. Variation of the RGB color numbers across the image gives us detail: but here, much of the red tulip has its red value pegged at 255. So the flower looks flat and largely void of detail.

Overexposure in and of itself is not necessarily bad, and it is often unavoidable, but you likely will want to avoid overexposing even a single color channel on your main subject, like this tulip. If your digital camera offers three separate color histograms, you may want to check it to avoid this problem. Here are the overexposed portions, shown masked in black:

I took another photo nearby, using better exposure:

I took these photos the other day at the Missouri Botanical Garden in Saint Louis. You can see more of my photos taken that day here.

The red channel just barely kisses 255 in the brightest red areas, and so this flower, unlike the one above, is not overexposed. But this image is still disappointing. While we do see detail and texture in the flower, it still looks somewhat flat, especially compared to the adjoining tulips. The yellow parts of those tulips look more interesting and textured. What is going on here?

Here are the three color channels of this image:

Please note that nearly all of the red tulip is black in the green and blue channels. As we would expect, there is good detail in the red channel. Take a look at the yellow parts of the tulip on the left; it shows good detail in all three channels, and so it has nice texture in the final color image.

We see texture when there is variation of the RGB color channels across an area. If there is little variation around a point, we see little texture at that point. Variation in color shows texture, but we have little to no variation in color with the red tulip. Variation in luminosity or brightness is typically more important when you want to show texture. See my article Luminance is More Important than Color.

A rough model of luminosity is:

Luminosity = 30% red + 59% green + 11% blue

If you want to show lots of texture or detail, then this calculated luminosity has to strongly vary over an area. With our tulip the majority of the green and blue color values are zero — seen as black in the above illustration — and so these color channels do nothing to give us texture. The texture in the red channel is attenuated by more than two thirds, and so the image looks somewhat flat.

Underexposure then is not one thing, but rather is three things, for we need to be concerned with the exposure of all three color channels. Typically we do not expect to see detail in black or white areas of a photograph, but certainly we would like to see texture in mid tones. But as we see here, we cannot count on that. Because two of our channels are underexposed, we lost texture.

I might add that the color of the tulip is wrong. It was not that particular shade of red, rather it was more intense, closer to scarlet, closer to the far edge of human color perception.

Just by looking at the RGB color numbers, we see that G and B are zero, and so all color is determined by R alone. This means that the color of the tulip is fixed to the color of the sRGB red primary. The sRGB standard is used by most digital cameras and most computer monitors.

The sRGB color standard can display only about 35% of all possible colors, as illustrated here:

This image originally appeared on Wikipedia. Source and attribution: http://en.wikipedia.org/wiki/File:Cie_Chart_with_sRGB_gamut_by_spigget.png

The horseshoe shape shows the limit of human color vision. The sRGB color gamut is represented by the triangle in this diagram. Our tulip color is somewhat near the lower righthand corner of this chart. Of course, this chart itself is not accurate in color, because it uses the sRGB standard, and so the outlying colors, outside of the triangle, cannot be displayed.

In photography, whenever you get one or two RGB colors at 255, while the other color is zero, then you know that you are out of gamut. We even see this in our first, overexposed photo. Red is largely fixed at 255, while blue and green are zero.

One solution is to capture the image using a color standard that is larger than sRGB. We won't be able to see these extra colors here, because lots of web browsers and computer monitors are limited to the sRGB standard. We can use a wider color gamut when printing, especially if use a higher-end printer than uses more than three color inks.

As it happens, I took this image using Camera RAW, and my Nikon can capture a very broad range of colors. During post-processing on my computer, I can convert the RAW file to sRGB, or I can convert it to the ProPhoto colorspace, which can represent more colors:

Image originally appeared on Wikipedia. Source and attribution: http://en.wikipedia.org/wiki/File:CIExy1931_ProPhoto.png

Notice the much larger triangle here, and also note that the red primary goes all the way to the limit of human vision; undoubtably our tulip can be accurately captured.

Here I edited the photo after converting the RAW image to the ProPhoto colorspace.

The gray here approximately indicates the areas which cannot be represented by sRGB. If we look at all three color channels, we can see that there is good texture in the tulip:

While we cannot show the ProPhoto image on the computer screen in all its glory, it may be printable, depending on the kind of printer we use. The final image ought to have better texture than what we had before.

We could process this image for computer display if we are willing to desaturate the red in the ProPhoto image. This would give us better texture at the expense of color.

Here I used the good green and blue channels from the ProPhoto color gamut to add texture back into the tulip. The flower is no longer intensely red, but I think it is otherwise a slight improvement.

Fluctuation in Fluorescent Lighting

2011-03-07T16:20:00.002-06:00

I'VE OFTEN HEARD that fluorescent lamps will quickly fluctuate in color as electric current flows through them. I decided to do a little experiment to see if this is true. I took a number of photos of a compact fluorescent bulb in rapid succession. The exposure and white balance of the camera was fixed, and the shutter speed was 1/1250 second. Here are four of these photos:

I saturated the colors here to make them more evident, but they do noticeably fluctuate. If like me you often severely manipulate or saturate your images, you will increase the likeliness of seeing these color shifts. Long exposure times will average out these fluctuations.

The human eye is rather well adapted to changes in the color of lighting from yellow to blue, like we find outdoors on a sunny day or with incandescent lighting. However, fluorescent lamps will produce problematic and sickly-looking green and magenta color casts, especially if there is any natural or incandescent lighting in the scene. This effect is particularly noticeable by the camera. When I am confronted with fluorescent lighting, I will turn it off if possible before starting photography.

Accurate Color Rendition

2011-03-04T00:49:00.000-06:00

AN ARTIST ASKS how to get accurate color in photography. This is a problem. Fairly accurate color rendition is possible, but is not easy to achieve.

Your first goal is to be sure you get accurate white balance. Many digital cameras have the ability to do a manual white balance on the camera, which is achieved by shooting a specially-made target which is guaranteed to be very close to a neutral color. If you place the target at the subject, facing the camera, you can get a very good white balance as long as the lighting conditions are fairly uniform.

I used to have color problems with my digital cameras, particularly, I would not get good red colors, rather they turned out magenta. I got an X-Rite Colorchecker target and used that to calibrate my cameras' colors. It works very well, but you have to shoot RAW, as well as create separate profiles for each lens, ISO, and lighting condition combination. But keep in mind these calibrations are only accurate for a limited number of colors.

Do not expect perfect color rendition, even with calibration. Digital cameras are designed to produce adequate color with an acceptable amount of noise. Some recent Sony cameras, I'm told, produce more accurate colors at the expense of more digital noise. You might want to check http://www.dxomark.com: look up Color Response and the "Sensitivity metamerism index" to see how accurate the colors actually are for a particular brand of camera.

Use your camera's native ISO sensitivity. Boosting ISO will harm color rendition severely: again, look at the DxO mark data to see the degree of harm. Typically, the lowest ISO setting on a camera is the camera's native ISO, but not always; some manufacturers have ‘extended’ settings on both the high and low ends of the ISO scale.

The quality of light is very important. Daylight and incandescent lighting are known to have good, continuous spectra. Using these light sources will give you better results than fluorescent lights, or sodium and mercury vapor lamps. When you use a light with an odd spectrum, your eyes and camera will likely see the colors rather differently. In my experience, fluorescent lights produce magenta and green color casts variably across the image, and so I try to avoid photography under these kinds of lights. According to the Academy of Motion Picture Arts and Sciences, the new LED lighting produces severe color shifts, most critically in skin tones, and so is not recommended for quality work.

Be aware that dark parts of the image, due to nose, will be less accurate. Common techniques used to brighten shadows will also harm color there.

There are highly accurate multispectral cameras, but they are expensive and designed for laboratory use only. The results from such a camera cannot be accurately displayed on a computer monitor.

Cameras produce images within a limited color gamut. The sRGB standard, used by most digital cameras, can only represent about 35% of the range of color that can be seen by the human eye. If you shoot RAW, or set your camera to Adobe RGB then you can capture more colors, but unless you know what you are doing, you will end up with images that look pale and unsaturated: the opposite of what you want. Even if you use a wide color gamut, your monitor and printer will likely not be able to show you those extra colors. Colors that come from rare and expensive pigments or dyes, such as the colors produced by Murex shells, typically are at the extreme of human color vision, and may be out of a digital camera's gamut.

If you want to be able to measure color in the real world accurately, perhaps you can invest in a spectrophotometer. Pantone makes one, but it only outputs the closest Pantone color, so it is not strictly accurate, but I would guess it is far more accurate than a digital camera. Alternatively, you can invest in a collection of Pantone color chips: you can visually match colors. Munsell chips are also available, but are also very expensive. If you only want to match one particular color, then your job is much easier. Even if you can't see it accurately on your monitor, you still will be able to print it if you use spot colors or if you use one of the excellent wide-gamut desktop printers.

A Black and White Conversion

2011-02-18T00:24:00.001-06:00

THERE ARE SOME occasions when a photo can be improved by converting it to black and white. Typically, I find these are photos taken at night, particularly with high ISO under sodium vapor lighting. Here is another example:

The shadow detail here is weak, and all my standard techniques for lightening dark detail looks unsatisfactory. But after converting the image to black and white, I was able to put severe curves on the image, as well as high levels of local contrast enhancement.

You can bring up large amounts of shadow detail in a monochrome image.

White Balance, Part 1

2011-01-27T00:47:00.000-06:00

PLEASE CONSIDER THIS photo of my living room:

(Or rather, this is a photo of the lounge at the Ritz-Carlton in Clayton, Missouri.)

Like very many interior photos — taken without a flash — this has a yellowish color cast. Undoubtably this yellow cast is due to the color of the lighting. Obviously incandescent light has a slightly yellower color of light than daylight. So there is no surprise that incandescent photos appear yellow.

Perhaps you think that the white balance feature on your digital camera is simply a minor adjustment. Perhaps it corrects for the slightly yellower color of incandescent lighting. Certainly I thought so.

Here is a white-balanced version of the photo above. I adjusted the photo so that a white object under this lighting was measurably neutral: that is, the red, green, and blue color values of the white object were equal after adjustment.

It looks a bit better: the yellow color cast has been removed. A small difference, but I think it improves the photo a bit, for it brings out the variety of colors better, and it is closer to what I remember seeing. I recommend always using a neutral white balance unless you have a specific reason not to do so.

Now consider this photo:

This is very yellow. But this so happens to be the scene assuming a daylight white balance. Were I to take a photo of a white object under bright daylight with this balance, it would look white. Incandescent lighting is actually far more yellow/orange in color than is daylight.

You may not be aware that your eye does an automatic white balance: when you look at a scene, your eyes attempt to subtract out the color of the lighting, and your eyes do a pretty good job of this, up to a point. Cameras work the same way; they also attempt to automatically subtract out the color of the light.

Keep in mind that perceptibly slight changes in the color of light may translate to a radically different objectively measurable difference in color.

Here is a photo taken outdoors, with a Daylight white balance:

And the same photo, with the color balance set to incandescent lighting; it is the same white balance I used in the nicely-corrected interior photo #2 above:

We see that the camera's white balance corrects for extreme differences in the color of light, not merely minor differences. How the eye corrects for white balance is a mystery, and the technology behind a camera's automatic white balance feature is ultimately imperfect, which is why I do a manual white balance whenever I'm doing my best work.

White Balance and Noise

Digital cameras have an intrinsic, fixed white balance. The color data captured by the camera sensor is adjusted by the camera's computer according to the white balance setting. Here are our sample pictures again, showing what they look like when using my camera's native white balance:

The images are greenish because my digital camera (and likely yours too) has twice the number of green light sensors than it does either red or blue, and also each green sensor is more sensitive under most lighting conditions than the others. For more details, see the article What Does the Camera Really See? For an overview of the color system used by digital cameras, read Color Spaces, Part 1: RGB. Digital cameras generally are rather insensitive to blue light.

To correct for the green snow scene above, the camera must amplify both the red and blue color channels to match the green, and whenever you amplify a signal, you also amplify noise.

For the interior scene, the camera must amplify the red channel a little (since incandescent lighting has plenty of red light), while it has to amplify the blue channel a lot, creating plenty of noise.

Here is an extreme example of this sort of noise amplification:

I took this a number of years ago with an inexpensive point-and-shoot camera. While the red and green channels have a lot of noise, due to the camera being set to ISO 1600, the blue channel at the bottom is particularly bad. The blue channel, due to incandescent white balance, was greatly amplified relative to the other channels, and so it shows plenty of noise.

Generally speaking, if you want low-noise interior photos, you are asking a lot of your camera, and you will likely spend a premium to get this. Read this for details: One Easy Rule for Quality Images.

White Balance and Exposure

If you don't set your white balance correctly, you risk bad exposure if you shoot JPEG. Please note that I define good exposure by taking all three color channels into consideration; see the article Three Opportunities for Overexposure for details. If even one of your three channels is significantly overexposed, you will get shifts in highlight color. In an extreme case, here is the daylight photo set with an incandescent white balance:

This is a selection from Nikon's View NX2 software, showing the three color histograms at this white balance. Each graph has the dark pixels to the left, and the bright pixels on the right: we see here that the red channel is underexposed, white the blue channel is overexposed. Due to poor white balance, we irrevocably lost both highlight and shadow detail.

The histograms seen here are similar to the three color histograms found on many digital cameras. Like many photographers, I usually check the color histograms to make sure that my photos are exposed properly. In order to use as much of the camera's dynamic range as possible, I try to expose the images as much as I can without overexposing any one of the three histograms. However, were I to use this process with a poor white balance, I would inevitably get a severely underexposed image. In the image above, reducing the exposure to preserve the blue channel highlights would lead to a severely underexposed red channel:

Now you might just want to have this ‘look’ in your photo, and that's fine. But if you instead plan on ‘fixing in Photoshop’, forget it.

Please note that you will see the same problem if you shoot incandescent lighting with bluer white balances — Daylight, Cloudy, and Shade — except your red channel will likely be greatly overexposed while your blue channel will be underexposed.

As a general rule, you will get the best exposure if you use a neutral white balance. You can expose the image longer with less noise and less chance of clipping highlights if you set your white balance precisely.

But please remember that cameras have fixed intrinsic white balance, as seen above. You get a greater risk of overexposing or underexposing color channels when you shoot JPEG images, because the camera throws away lots of its original sensor data when performing a white balance — and you risk throwing away good useful data if you set your white balance wrong. For this reason, I shoot RAW images (which retain all of the original sensor data), because I can adjust the white balance on my computer after the fact. The risk of bad exposure is lessened — although not eliminated — when you shoot RAW and so you still ought to be careful at the time of shooting.

The trade-off is that RAW files need to be processed on your computer to produce an image usable for either printing or displaying on the web. I find the trade-off acceptable, although I know that many photographers do not.

Some authorities state that since a camera has a fixed intrinsic white balance, then the camera exposure histograms ought to show the RAW color channels. I think this is an excellent idea. Some photographers attempt to do this by forcing the camera to use a white balance that does no color adjustment at all: their histograms in this case are correct. The UniWB method of using the camera's own intrinsic white balance was developed by Iliah Borg and others, and the method is described here. It is not for the faint-hearted, since all your photos will turn out green, and you will have to correct for white balance on your computer. You can however get better exposure. At the very least, this is a good learning tool, if not really practical.

Note to camera manufacturers: please show the RAW histograms when shooting RAW! Also, give the ability to zoom in to the brightest pixels on the histograms, for overexposure in digital photography is worse than underexposure. When you show blinking pixels, be sure that they blink when even one of the channels is overexposed.

When to White Balance and When Not to White Balance

As I mentioned before, I recommend always setting your white balance precisely unless you have a specific need to do otherwise.

Have a look at A Digital Color Wheel. Colors opposite from each other are called opponent colors: a color balance biased towards one edge of the wheel will aways be at the expense of the opposite edge. If your image is too blue you will not get enough yellow, and too much green will mean too little magenta. If the white balance is in the center you will trade off quantity of color for quality of color; a well-white-balanced image will look richer in color content, and you have less risk of having your colors go out of gamut. You can get better results in saturating the colors if your white balance is precisely in the middle.

Sometimes you may want to capture a scene as you remember seeing it; but don't forget that your eye already does strong white balancing. So if you want to capture the warm glow of candlelight, set your white balance to somewhat warmer than neutral. To capture the cold mood of a snowy day, set your white balance to a somewhat cooler balance than neutral. Although your eye does do white balancing, this mechanism doesn't work well under dim lighting, although I am uncertain as to what the actual relationship might be. This is worth further research.

Here is a photo where I did not want the camera to subtract out the color of the light:

A fantastical beast, at the City Museum in Saint Louis. Click the image twice to view it on black.

Sometimes you may want an image with a strong overall color tone, but you may get better results if you first convert the image to black and white, and then add a tone afterwards, not relying on white balance.

Mixed Lighting

Your eyes not only do an automatic white balance, but they adjust this white balance variably across the scene while you are looking at it. When you take a photo of a scene, and reduce that panorama down to a tiny, low-contrast image displayed on a page or on a screen, this automatic white balance hardly operates, which is why you have to get it right in the camera. A severe problem occurs when you have mixed lighting of multiple colors, for example, when you shoot an interior, illuminated with incandescent lighting, while also having windows to the outside appearing within your photos. Invariably your windows will look fine with the interior too yellow, or the interior looks fine while the scene outdoors is very blue. The photograph just doesn't look as you see it in real life.

With architectural interiors, I will use daylight white balance for the windows and incandescent white balance for the interior, and then composite these versions of the image. The results are good, even though this is a tedious process. Big-budget cinematographers will put large yellow-colored gels over the outside of the windows so that the color of the transmitted daylight matches the lighting used in the interior.

Far more problematic is when fluorescent lighting is used in an interior. Not only do these lights have an odd color — typically they are simultaneously more yellow and more green than daylight — but fluorescent colors are not constant across brands of lamps. They do not provide a continuous full spectrum of light, and even more problematic, they change color as they quickly flicker 50 or 60 times per second. Almost invariably, if you attempt to white balance fluorescent lighting, you will get strong shades of the opponent colors green and magenta throughout your image. These green/magenta color casts will be considerably increased if you also have daylight or incandescent lights in the scene. If at all possible, I will turn the fluorescent lights off.

Also see the article: White Balance, Part 2: The Gray World Assumption and the Retinex Theory.

Challenge: "High Photographic Modernism"

2011-01-11T14:04:00.005-06:00

I AM NOW a Challenges host on the dpreview.com website. My first challenge is called High Photographic Modernism, which invites images made in the style of the famous Group f.64.

This is a challenge of straight photography: images ought to have great depth of field, extreme sharpness and detail, little over or under exposure, and generally speaking technically precise with a simple composition — and no special effects. The photos must be black and white, although some toning is acceptable.

To enter this challenge, you need to sign up at dpreview. You may enter up to three images from Tuesday, January 18th, 2011 through Monday, January 24th, 2011. Voting then occurs the week afterwards. When the challenge is finished, you will get an email with your results. Winners receive fame and glory!

For the following week, the theme will be the New Pictorialism. This is inspired by a style of photography popular a century ago, which used classic techniques from painting to make images that were beautiful with a dreamlike quality.

Focal Length

2010-12-26T16:24:00.001-06:00

PHOTOGRAPHY IS NOTORIOUS for its many numbers that a photographer needs to know about. Focal length is one of those numbers.

Most inexpensive compact cameras have an easy-to-use zoom feature, and casual photographers can merely set the zoom to whatever they want without worrying about any confusing numbers. But confusion can occur if they use a camera with interchangeable lenses, for then they need to learn about focal length.

Fortunately for beginners, the kit lens that comes with most inexpensive interchangeable-lens cameras is adequate for most purposes. These cameras may even come with two lenses; for example, 18 to 55 millimeters and 70-200 mm. All you really need to know is that large numbers zoom onto distant objects, while smaller numbers capture ‘more of the scene’.

It's easy. If you want to get the whole scene in your photo, you set your lens to 18 millimeters. If you want to zoom in, you set your lens to 55 mm. But then a friend asks you to take a photo of her, using her camera. You stand about ten feet away, and taking note of the millimeter markings on her lens, you set it to 18 millimeters and then look through the viewfinder — and you are surprised that she appears smaller in the viewfinder than you would expect. She suggests that you zoom in a bit, using a setting of about 30 millimeters. So 18 mm on your camera is the same as 30 mm on her camera. As it so happens, a nearby photographer is taking a photo of the same scene: his camera is large and he tells you that he is using a 55mm lens — but he too is taking in the whole scene, for 55 millimeters is a wide-angle lens for his camera. You learn that focal length settings are not necessarily commensurate between cameras.

A pinhole lens

Light usually travels in a straight line through air, and so we can construct a very crude, but workable, lens just by making a small hole in an opaque surface. Light will travel in a straight line from an object, through this pinhole, where it reaches its destination, which may be light sensitive film or a digital camera sensor.

The focal length of the pinhole lens is merely the distance from your sensor to the pinhole. To illustrate the angle of view of this pinhole lens, draw a line which is the length of your sensor: say, 35 millimeters wide. Perpendicular and centered on this line, draw a dot, which represents your pinhole. Draw a straight line from the edges of the sensor through the dot: this shows the angle of view of your pinhole lens. If you bring the pinhole closer, the view gets wider; and draw it farther away, and the angle of view gets narrower. You should see that for any given focal length, a larger sensor will give you a wider angle of view. Using trigonometry, you can calculate the angle of view for any combination of sensor size and focal length. Suppose you have two cameras, one with a sensor twice as wide as the other: doubling the focal length of the pinhole lens on the larger camera will give you precisely the same angle of view as the smaller camera.

Now take a glass lens, and focus it on some object very, very far away, and note the size of the object projected on your sensor. Then take a pinhole lens, and move it closer or farther from the sensor until its projected image is precisely the same size as the image formed by the glass lens. The distance from the pinhole to the sensor is the effective focal length of the glass lens. An 18 millimeter glass lens projects the same size image as would a pinhole located 18 millimeters from the sensor.

But please note that this equivalence between a glass lens and pinhole lens only works when the distance from lens to the object is much greater than the distance from the lens to the sensor. A regular camera lens, after all, is not a tiny dot like our pinhole lens, but rather is made of multiple thick chunks of glass. If you focus a glass lens upon a subject very close by — like when using a macro lens to focus on a small insect — then its effective focal length will change considerably. Click here for more details.

Also note that this equivalence only works when a glass lens produces a rectilinear image — where straight lines in the scene translate to straight lines on the image. Fisheye lenses are a bit more complicated since they produce so much distortion.

Equivalent focal length

Serious photographers use seriously large cameras. This is for the simple reason that large camera sensors — either digital or film — naturally produce cleaner, sharper, more detailed images. Click here to see why. Now photojournalists also want good picture quality, but they also lug cameras around all day long, and so they need a camera that is a good compromise between weight and image quality. Photojournalists are typically the most commonly-seen type of professional photographer — and amateurs, in imitation, started using similar equipment, which included the 35mm film format. Vast numbers of amateur-grade, interchangeable-lens 35 millimeter film cameras were produced, most notably by the same manufacturers who made the photojournalist cameras.

People became quite used to the sizes of lenses for these cameras. For example, a 50mm lens produced an image that looked rather normal — not too zoomed in, and not too wide. Lenses in the range of say 105 millimeters or larger were good for portraits, while 30 millimeter or smaller focal lengths were good for architectural interiors. Now, please recall that these focal length sizes are for 35 mm film; a medium-format camera would use longer focal lengths for the same purposes, while an inexpensive consumer camera would use much shorter focal lengths.

Eventually the manufacturers of photojournalist cameras went digital; alas, however, due high cost, the digital sensor size was smaller than the beloved 35 millimeter film. Because people were so familiar with the focal lengths used by 35 millimeter cameras, manufacturers stated equivalent focal lengths. So an 18 mm lens used with the new digital sensor is said to be equivalent to (that is, provides the same angle of view) a 27 mm lens used on a 35 mm camera. A 35mm lens on these digitals is equivalent to a 50 mm lens on a 35 mm camera. Is this helpful, or confusing?

Because the 35mm format was rather standard, digital cameras with sensors smaller than 35 mm are often called cropped-sensor cameras, although this terminology can be rather confusing to beginners. I find that beginners often get hung up on the marketing term ‘crop factor’. A 20 mm lens on a camera with a crop factor of 1.5 will provide the same angle of view as a 20 mm x 1.5 = 30 mm lens on a 35 millimeter film camera. Now this terminology is likely only useful if you are very familiar with the old 35 millimeter cameras and their lenses, and is otherwise confusing.

If you are a beginner, I would suggest you forget all about equivalent focal lengths and crop factors. Instead, find out the size of your sensor, in millimeters. For example, many consumer digital SLR cameras have a sensor that is about 30 millimeters across on the diagonal. A wide-angle lens will have a value that is less than this measurement, while a telephoto lens will be much larger than this value. A normal lens — for this sensor — will be equal to this size or perhaps a bit larger.

A Digital Color Wheel

2010-12-21T11:23:00.001-06:00

MOST COLOR WHEELS you find at art stores, or images of wheels you find with Internet searches aren't too helpful for digital photography. While they may illustrate the visual order of the colors, they aren't too helpful if you want to mix colors digitally. They may even be quite misleading. So I created my own color wheel using the primary colors found in the sRGB standard, which is used by digital cameras, computers, and high-definition television.

This color wheel shows the correct relationships between the red, green, and blue colors that are primary in the sRGB color system, as well as their opponent or secondary colors of cyan, magenta, and yellow.

These primary and secondary colors are the brightest and most saturated colors that can be generated from the sRGB color system. The coding in each color circle gives you the formula for generating the color: for example, cyan is GB, which means that Red=0, while Green and Blue = 255. Halfway in between the primaries and secondaries are bright tertiary colors. These tertiaries are coded with lower-case letters indicating half a given color: for example sky blue is coded gB, meaning Red=0, Green=128 and Blue= 255.

Some old color wheels use red, yellow, and blue as primary colors; others use green, purple, and orange as primaries. This is misleading for computer use since they don't give us a good idea of opponent colors. In this color wheel, if you mix together equal portions of colors opposite to one another, you will get a middle gray color; mixing together blue and yellow gives you a gray where the red, green, and blue values all equal 128.

If your images have a color cast, you can achieve white balance by moving towards the opposite color. An image that is too yellow needs more blue, an image that is too green needs more magenta.

UPDATE: My use of a value of 128 for the tertiary colors is not correct, since 128 is NOT the middle tone. It is for this reason that the wheel does not appear to be visually uniform: the tertiaries appear to be somewhat dark. Updated wheel can be found here.

sRGB Colors Out of Gamut

2010-12-19T06:38:00.002-06:00

YOU OWN AN inexpensive desktop color printer. You have a digital camera, and you want to make prints. You print your images, and your final photos are disappointing. Does this sound familiar?

This is the color gamut problem: inexpensive desktop printers — those with four ink colors (cyan, magenta, yellow, and black) — cannot reproduce all of the colors that are produced by a digital camera. The best way around this is to get a printer that has more ink colors — but these can be expensive. And so the best alternative is to process your images to make the most of your printer's limited color gamut.

Here are the three primary colors in the sRGB color system:

In the wide strips, we have one of the pure sRGB primary colors going from a value of 0, which is black, to 255 which is the brightest pure color that can be represented by the sRGB system.

Do you see the blue line at the bottom of each strip? This is the color gamut limit of four-color commercial printing presses and inexpensive desktop printers (this color space is abbreviated CMYK, after the four ink colors used: cyan, magenta, yellow, and black). Everything above the lines cannot be accurately printed — which is most of the image. Note that all the bright primary colors are out of the CMYK gamut. The narrow strips on the right are an approximate representation of the colors you will get from an inexpensive printer. Note that greens and blues are particularly poor and relatively unsaturated.

When we mix colors together in sRGB, we still see the same problem:

Here, red goes from zero on the left to 255 on the right; from the bottom, green goes from zero to 255 on the top. Red and green mix together to make yellow. The areas surrounded by the blue lines are colors that are within the CMYK color gamut! Reds, oranges, greens, and some leaf green colors cannot be accurately portrayed by CMYK, in fact, most of the colors in this mixture cannot be printed.

Other color mixtures are hardly better:

Red going across, blue going up. Again, most of the image is not accurately printable.

Green across, blue going up. This is somewhat better, but you still can't print decent primary colors.

Things do get better when we have mixtures of all three colors. Here are mixtures of two colors; in each, the third color is set to 50% of its maximum value:

The top image has dark blue mixed in, the middle dark green, and the bottom a dark red. The printable color gamut is expanded by the addition of the third color. If we had a pure grayscale image, then all the gray tones will be printable.

Real-world photos typically don't have too many pure, saturated reds, greens, and blues, and so the out-of-gamut problem may be a bit less prominent that what we see here. But most images will have at least some colors that can't be printed:

In this image of a snowman the out of gamut regions are seen on the right, painted in green. If you were to print this image on a four-color printer, these color regions would look a bit flat and unsaturated. You will lose detail also.

But CMYK gives as well as takes away. Even though we can not print the bright red, green, and blue primary colors, CMYK has its own primary colors: cyan, magenta, and yellow, which are typically brighter and more saturated than what you have with sRGB. You could process your images to take advantage of these colors.

To get an overview of these color systems, you may want to take a look at some of these articles:

Color Spaces, Part 1: RGB
An RGB Quiz
Color Spaces, Part 2: CMYK
Part 2 of "Color Spaces, Part 2: CMYK
A CMYK Quiz

When processing for print, you want to emphasize the colors the printer can print, while toning back the colors that are out of the printer's gamut. Following is a relatively simple process where you can make the most of your images in Photoshop.

Convert your image to a wide-gamut color space. Typically Adobe RGB is used. If you shoot RAW, Photoshop's Adobe Camera RAW (ACR) program can select this color space upon import — this would be useful for best quality. Some cameras can shoot JPEG images in Adobe RGB, but I would suggest not using it unless you really know what you are doing. Select the menu item Edit, Convert to Profile...

Turn on the Gamut Warning feature in Photoshop:

If your target printer has a color profile installed in Photoshop, go to the Custom... menu and select it instead of CMYK. Following shows an image with the gamut warning on; here I have the warning set to gray, but you can change the color to be more visible on a particular image.

Select the Image, Adjustments, Hue/Saturation... menu item:

Note that the drop-down list has both the RGB and CMYK primary colors. For each of the color classes which are out-of-gamut, adjust the Saturation and Lightness sliders until the Gamut Warning turns off. You can be as careful or as sloppy as you want here, by adjusting the slider on the bottom. You can also select individual colors with the eyedropper tool. (The middle part of the slider shows the colors that will be fully corrected. You can adjust the outside parts of the slider for good blending.)

Here I brought the red bow tie into the CMYK color gamut by darkening and desaturating the color range; but be aware that there may be more than one way to bring it into gamut, with some being better than others. Were I being more careful, I would have done this on a layer with a mask so as not to also desaturate the snowman's smile. Then I corrected the the blue part of the image to bring it within the CMYK gamut. In other images you may also have to tone down the bright primary green colors.

Next we can enhance the printer's primary colors. We go through the same process as before, but we work with the cyan, magenta, and yellow color ranges, increasing brightness and saturation. The major improvement we can make here are with the yellow colors, which I was able to saturate and brighten considerably. You can brighten and saturate until the Gamut Warning turns on; then you've gone too far. (However, it is OK if some small parts of your image are out of gamut... you just don't want too much over a broad area, otherwise you will lose detail.)

This is a bit of a leap of faith, since you most likely cannot see the final results of your editing: it is out of your monitor's gamut. If you have areas of your image that ought to have lots of bright cyan, magenta, or yellow ink, you can place an eyedropper tool on the spot and measure the CMYK values directly. If a spot is supposed to have a very bright yellow component, the brightest you can get, then that spot, after your processing, ought to be rather close to having 100% yellow ink.

Purists may insist that all this manipulation is ‘inauthentic’ but in reality this scene greatly exceeded the color gamut and dynamic range of my digital camera; in fact this is a blend of three separately exposed images. So we are justified in making the colors of the snowman as bright and as saturated as we are able to. Likewise, if we are printing a full-color image to a narrow-gamut CMYK printer, we are justified in printing as much of a full-range of color as we are able.

There are many ways of accomplishing a goal in Photoshop, and this one is particularly straight-forward. The most visually accurate color corrections can be made using the Lab color space. Also of use is the Vibrance tool, layers, masking, and most notably Levels and Curves.

When I am preparing images for commercial press, I eventually manipulate the images directly in the CMYK color space, making the most of that limited range of color. Unfortunately we cannot do the same with desktop printers, since they use different inks than are found in commercial presses, and so they typically require the image to be in RGB format. The Gamut Warning feature is the most powerful tool for this purpose.

A CMYK Quiz

2010-12-08T14:22:00.003-06:00

HERE IS A sample image, which shows the four channels of a CMYK image. Use your knowledge of CMYK to determine some facts about this image.

If you haven't read them yet, you may first want to read these articles: Color Spaces, Part 2: CMYK and Part Two of "Color Spaces, Part 2: CMYK".

This sculpture of an acorn is found in Wydown Park, in Clayton, Missouri.

I am convinced that a thorough knowledge of the channel system of digital images is essential for good photography. By looking at a color photograph, you ought to be able to imagine which each color channel ought to look like, and by examining the color channels, you ought to be able to determine what colors are represented.

The CMYK color system represents inks printed on a page, and includes the colors cyan, magenta, yellow, and black. Each channel represents one color of ink. No ink is placed on the page where the channel is white; and where the channel is black, we have 100% ink coverage. For example, a bright cyan-colored object will be black in the cyan channel, and white in the other channels. Where we happen to have roughly equal quantities of cyan, magenta, and yellow ink, the CMYK system will subtract those colors and replace them with black ink. So K (black) will dominate the shadows.

Here is your task:

Identify each color channel in the image above.
There are two colors of flowers in the image. Identify the colors. We have taller flowers which can be seen in front of the acorn, and shorter flowers of a different color in the foreground.

Unlike my last quiz, I won't give you any clues. Use your knowledge of nature and the channel structure of CMYK to determine the answers.

The Problem of Resizing Images

2010-12-04T23:26:00.005-06:00

IMAGINE YOU HAVE a peculiar boss at work. He wants to make sure that you are at your desk working forty hours per week. So once a day (seven days a week!) at precisely the same time every day (since he is extremely methodical), he peers into your tiny cubical to see if you are at work. You are never there, and he is quite upset. You will hear about this at your next annual review, nine months from now. Sadly, it appears you won't be getting a raise.

Well, he looks into your cubical precisely at midnight every day. Despite graduating with honors in a top M.B.A. program, he really isn't all that bright, and he lacks a life outside of work. As a matter of fact, you do work 8 a.m. to 5 p.m., Monday through Friday, and you are always at your cubical during those times. But boss marks you down as being absent 100% of the time.

Well, since you actually do your required work, bossman decides to check up on you four times a day. Quite methodically, he appears at your cubical at midnight, 6 a.m., noon, and 6 p.m. As it so happens, you take your lunch at noon, and he just barely misses seeing you every time. You are still absent 100% of the time, in his mind.

Boss is still puzzled. With apparently too much time on his hands, he checks on you eight times a day. Midnight, 3 a.m, 6 a.m, 9 a.m., noon, 3 p.m., 6 p.m., and 9 p.m. He finally sees you! Since he sees you 2 out of the 8 visits he makes, Monday through Friday, he estimates that you are working at most 2/8 x 24 x 5 = 30 hours per week. He is disappointed, but at least you get to keep your job.

Note that if he visited your cubical three times a day, at midnight, 8 a.m., and 4 p.m., he'd see you twice (the first time just as you got there) and would estimate a working time of 2/3 x 24 x 5 = up to 80 hours per week. But four times per day gives zero. Clearly the frequency of his visits can change the results dramatically.

Your boss's boss likes what he is doing, and asks that he get more data so that he can present an impressive chart at an upcoming meeting. Your boss now checks your cubical 12 times a day. He visits at midnight, 2 a.m., 4 a.m., 6 a.m., 8 a.m., 10 a.m., noon, 2 p.m., 4 p.m., 6 p.m., 8 p.m., and 10 p.m. He sees you working 4 times per day, Monday through Friday, and so he estimates that you work up to 40 hours per week. If he visited your cubical 24 times a day, or 48 times a day or more, he may (hopefully) notice that his increasing visits didn't give him much more useful data: he would always get a result close to 40 hours per week.

Slightly different

Let's consider a slightly different scenario. Your boss is always on top of the latest scientific theories. He read that the natural sleep-and-wake rhythm of human beings, when not exposed to the cycle of the sun, is 25 hours a day. Always ready to implement these newest findings, and since he apparently never sees any sunlight, your boss now lives out a 25 hour day, although who knows when he actually gets any sleep. Instead of checking your cubical once a day, he checks it once every 25 hours. If he sees you at your desk, he gives you credit for the entire 24 hour day (unfortunately, he has yet to convince his boss to move all the employees to this new scientific schedule). So the first day, he checks for you at midnight, the second day at 1 a.m., the third at 2 a.m., and so forth.

Although he finds you working sometimes four days in a row, he is infuriated that you are (apparently) taking two week (and longer!) vacations at regular intervals. He does estimate that in the long run you are actually working on average 40 hours per week, but he is worried about all the important conference calls you must be missing.

No Common sense

This hypothetical boss, despite being diligent, lacks common sense. This lack of common sense, while being reprehensible in a human being, is quite the norm with digital cameras and with computer software such as Photoshop, although we must credit computer technology with also being diligent. It's hard — no, impossible — to program a computer with common sense, and so we ourselves must make up for what computers lack if we want good results.

Better yet worse

I was delighted when I upgraded from a nice but lowly point-and-shoot camera to a decent, yet inexpensive, DSLR model. Immediately I noticed how sharp my new photos were, as well as having much less noise, even in low light. But there was a problem, and I couldn't quite put my finger on it. With my old camera, when I was processing images for the Internet, I would simply resize them and add sharpening. Even though there were various re-sizing algorithms available in Photoshop, none seemed to make much of a difference. I did put a lot of effort into using good sharpening algorithms, which made my photos look much crisper without obvious artifacts. But this did not work well with my new camera.

My old process did not work all of the time with my new camera — and the maddening thing was that my results were quite inconsistent — some of my final images looked fine, some were terrible (nature photos were typically the worst). Formerly, when I reduced the size of my images, I had Photoshop set to use the Bicubic Sharper algorithm, which Photoshop says is “best for reduction”, but I found that the new camera's images looked quite rough. So I changed it to use regular Bicubic. This required quite a bit more sharpening than I had used before, and I started using better algorithms that would reduce the bright artifacts I was now seeing, especially around distant leaves on trees and along certain edges. Sometimes I had to manually retouch out some of the sharpness. To me, this is not acceptable, so I started asking around for advice. As it turns out, Photoshop gets it wrong, and does not implement its resizing algorithms correctly.

Hit a brick wall

Digital cameras have ranks and files of pixels, arrayed across their sensor, in precise order, just like the smart-yet-foolish boss in our allegory above. Precisely every x micrometers, a different sensor captures light, just like precisely every y hours the boss would check up on his subordinate.

In the story, you arrive at work at a regular interval, but your nosy boss, if he didn't check up on you frequently enough, would get a wildly inaccurate estimate as to when you actually were present. Only when he checked up on you many times in a day did he get an estimate that was accurate enough.

A similar same thing happens in a digital camera. If there is an underlying, repeating pattern in the scene, the camera may get a wrong estimate of what the scene looks like if you have an inadequate number of pixels to capture enough detail. We see this on initial capture. Here is a section of a photograph, showing textured carpet:

The camera did not have enough resolution to capture the repeating texture of the carpet adequately. So we end up with ugly artifacts, here shown by the odd pattern. There were no curves in the texture of the carpet, but the pixels on the camera, being spaced too far apart, got a strange signal. It was not sampling the texture frequently enough. This pattern is called the Moiré effect or an interference pattern, and is a special case of aliasing.

Not only do we see this on initial capture, but this can be a severe problem when we are downsizing an image. Downsizing in business ruins lives, while downsizing in digital photography ruins images. If there is a repeating pattern in an image, we can get bizarre patterns upon downsizing if we end up with fewer pixels than a particular pattern requires. Here is a detail of a larger image:

A brick wall. We have a classic, repeating pattern, which will test Photoshop's ability to resize.

First we use Bicubic Sharper, which Photoshop tells us is best for reduction:

Ugh. Bad pattern. Just like the boss who was checking on you at 25 hour intervals (and thinking that you were frequently taking 16 day vacations), we see some bands of the brick wall where the lighter white mortar predominates, and other bands where the dark brick predominates. Also, the rest of the image looks rather rough.

Please note: these sample images are intended to be viewed at 100% resolution. If you are viewing these images on a mobile device, they may be further resized by your device, not giving you an accurate representation.

Now let's try Bicubic:

The repeating pattern is still there, but the rest of the image looks a bit better, if soft. Now normally I'd add sharpening to an image like this, but the pattern on the bricks just looks unprofessional.

Bicubic Softer does not help:

Now lately I've been using the Bilinear algorithm for resizing. The final images, to my eyes, look crisper than Bicubic, yet less rough compared to Bicubic Sharper. Let's try Bilinear on the brick wall:

Interesting. The pattern changed, and maybe it is somewhat less obvious. But still bad. I do like how the rest of the image turned out though: it would hardly need any sharpening at all.

For the sake of completeness, let's try the Nearest Neighbor resizing. As Photoshop says it is best when we want to ‘preserve hard edges’, and since the building has hard edges which we want to preserve, it should look fine, right?

Nope. Blech. Looks like a zebra.

Note that the big problem we are seeing is due to the regular pattern of the subject, such as the brick wall, coupled with the regular pattern of pixels on the digital image. We do not see Moiré patterns with film cameras and prints: the chemical film grains are irregular shapes and sizes.

But fortunately there is a good general theory to help us out. The Nyquist sampling theorem states:

If a function x(t) contains no frequencies equal to or higher than B hertz, it is completely determined by giving its ordinates at a series of points spaced 1/(2B) seconds apart.

Very roughly speaking, if we take a picture of something that has a regular pattern, if we don't allocate more than two pixels for a repeating element, then we will get the Moiré pattern. But actually it is slightly more complicated than that, since we have three colors of pixels, at slightly different locations on our sensor. In the photo of the carpet above, there is much more Moiré in the red and blue channels compared to green, as we do have twice as many green sensors.

There are also other mathematical effects involved to complicate matters, such as the Nyquist theorem assumes the frequencies are perfect sine waves. A pattern with hard edges, such as the bricks, actually are equivalent to somewhat higher sine wave frequencies. So some authorities state that for hard-edged repeating patterns such as these bricks, and with a Bayer Array (where we have separate photo-sites for each color channel), we ought to capture at least three (or maybe up to four) pixels per repeating pattern element to avoid aliasing.

We find the exact same thing when downsizing an image. If the final resampled image does not have more than two pixels capturing each element of a repeating texture in the original image, we will get a Moiré pattern. Because of the complications given above, maybe we need a little bit more, like 3 or so pixels just to be safe. So if our bricks, about 10 pixels apart vertically on the original image, are roughly reduced 1/5 or less in size, then they will definitely show a bizarre pattern, since we are allocating two or fewer pixels per brick. This is what we see in the photos above. I didn't get any Moiré effect when I downsized the image to either 50% or 33% (5 or 3.3 pixels per brick) — and just started getting Morié at 26%, which is about 2.6 pixels per brick.

This is analogous to what is done in the audio recording industry. Young, healthy human ears can hear frequencies up to about 22 kHz, and audio engineers will sample the audio at more than twice that frequency, 44.1 kHz, to avoid audio artifacts like we see in our aliased images.

Boss tries harder

Your boss still wants to keep track of you, but because he has other duties, he attempts to automate the task. He installs a sensor at your cubical door. Whenever you are in your cubical, the sensor records that fact. At the end of a fixed period of time, the sensor resets itself and sends a signal to your boss's office whether or not you were in your cubical anytime during that period. He sets the sensor to send him data every day at midnight. He successfully finds out that you are in the office every weekday. Under your boss's old system, he knew precisely when you were at your desk at a given moment in time, but the new system, while it is less specific, gives him more useful information. In effect, the sensor blurs the boss's data a bit, but he gets better results. With one sample per day, he gets better results than visiting your cubical four times. Were he to sample the sensor more times per day, he would get a much better idea of your attendance than if he were to visit your cubical the same number of times. Maybe he'll find something better to do with all the time saved.

As it so happens, digital cameras incorporate anti-aliasing filters to combat Moiré patterns. This softens the image a bit, but it can lessen the effect that we see in the carpet photo above. Consumer grade compact cameras tend to have heavy anti-alias filters, DSLRs have weaker ones, while medium-format digital camera backs may have none. With the higher-grade cameras, it is up to the photographer to either avoid or correct for these imperfections — although with more pixels, this will be less of a problem.

Blurring is the key

This softening is the key to downsizing images. According to the Nyquist theorem, our samples need to be more than double the frequency of the original signal to avoid artifacts, but when we make an image smaller, we greatly increase the frequency of our patterns. So what we need to do is to blur the image first — before downsizing — so that the Nyquist theorem still holds for our final image. In more technical terms, an image needs to be put through a low-pass filter before being down-sampled — the high-frequency components of the image have to be eliminated first by blurring.

I started getting the ugly artifacts when I reduced the image below 2.6 pixels per brick, and so to eliminate them we need to run the image first through a low-pass filter, which will get rid of any detail 2.6 pixels in size or smaller.

Photoshop does not blur an image prior to downsizing, not even the newest Photoshop CS5. That is why we get these digital artifacts. I would think that this would be fairly easy to implement.

How an image ought to be blurred prior to downsizing is a mathematically complex subject, and certainly the optimal blurring algorithms are not found in Photoshop. But we could experiment with Gaussian Blur, although choosing the Gaussian radius may be a bit problematic.

OK, so we want to be sure that we don't have any frequency components of our bricks being any less than about 2.5 pixels per brick in the final image. I initially choose to apply a Gaussian blur with radius 2.5 before downsizing. This is a quite naïve start, and so I did blurs in various steps:

Radius = 2.5. Just for fun, I used the Nearest Neighbor resizing algorithm, which gave us the horrendous zebra stripes seen above. It doesn't look too bad, does it? I added 50% Photoshop Sharpen to these images to make them look a little better. Better sharpening is called for however.

Here are other Gaussian blur radii:

Radius = 1. We still have severe aliasing.

Radius = 1.5. Still some aliasing.

Radius = 2. Some very faint aliasing; otherwise this is a good image.

Radius = 3. Too soft.

Ok, for sure we can get rid of aliasing when it obviously appears on an image like this one. But this may not be optimal, for the final image appears a bit too soft. One trick I've used is to blend together two copies of an image, reduced using different algorithms. In this case, I'd select the anti-aliased part for the bricks, with a normal downsize for the rest of the image.

However, anti-aliasing may help images even without an obvious pattern such as this. I recall that I often get poor resizing results, particularly with distant leaves against the sky, and along certain edges. Perhaps using even a soft blur will help with these images.

But we really ought to be using better algorithms than Photoshop offers. Very many algorithms are implemented in the free ImageMagick command-line utility, and in-depth discussions are here and here. For downsizing, they recommend the Lanczos algorithm for photographic images. It properly does blurring before reducing, although it does not use the optimal blur algorithm, for the sake of good performance. Using that software, I resized the brick building:

Lanczos still has a bit of Moiré, so I'm a bit disappointed. Otherwise it looks pretty good, and is much better than any image found above.

I tweaked the processing a bit and got this:

I blended the above image with a version that I blurred before downsizing. I masked out the building in the unblurred layer, giving us this composite.

Apparently there are some other, better algorithms available, but they are computationally expensive, or difficult to fine-tune optimally. However, whichever resizing algorithm you use, it is important to sharpen the image afterwards to bring back some crispness to the image.

Part Two of "Color Spaces, Part 2: CMYK"

2010-12-01T02:22:00.005-06:00

IF YOU TRY to invent a new language, like for example Esperanto, you won't get far if your new language has nouns but no verbs. Likewise, if you invent a new color system, you won't get far if you don't include the common primary colors, as well as black and white. Your color system doesn't have to cover every conceivable color, just like every human language does not need to have words to describe quantum physics. You just need the basics.

The CMYK color system is used by commercial presses, as well as by inexpensive desktop printers. CMYK is not a very broad system of color, but it has the basics, and is suitable for most printing purposes. Using cyan, magenta, yellow, and black inks, these printers can output a smaller range of color than even the sRGB standard (used by most cameras, computers, and HDTV), which in turn only displays about 35% of all possible colors. But CMYK can print all the basic classes of colors, with a nice, continuous gradation between these colors.

I am convinced that a thorough knowledge of the color structure of images is needed for quality photography. At a bare minimum, a photographer ought to know about the three color channels delivered by the camera — red, green, and blue — and how the RGB channels work together to represent color. You should just be able to look at an image, and imagine with your mind's eye how each of the channels ought to look. And by looking at black and white representations of the channels, you ought to be able to estimate roughly what the various colors are in the image. Using the “by the numbers method”, you ought to be able to know if your colors are correct just by examining the RGB values — even if you are color blind. See my article, Color Spaces, Part 1: RGB for an introduction to this color system.

But it is nice having printed output, instead of just viewing pictures on a screen. If you are fortunate someone might be willing to pay you to print your photos in a book or magazine, or you may make prints for clients. If you want to do an excellent job with printing, better than typical, then having an understanding of the printer's color channel structure is also essential.

RGB output assumes three primary colors on a brightly lit screen — you illumine red, green, and blue lights which mix together to produce a broad range of colors, including black and white. The more light illuminating the screen, the brighter the picture. CMYK, on the other hand, places four colors of ink on a page, and the more ink on the page, the darker the image. Fortunately, once you know RGB, moving to CMYK is is quite similar. See my article, Color Spaces, Part 2: CMYK for details.

The key is the opponent color system. Some colors, when mixed, produce other colors, like green and red lights shining together will produce a yellowish light; or when you mix cyan and magenta inks together you get blue. However, when you mix opponent colors together, you get gray. The RGB and CMYK color systems use colors that are opponent to each other: red is the opposite of cyan, and so on, and so the red channel will look quite similar to the cyan channel, while green will look similar to magenta. The major difference is the black channel in CMYK, which will have much of the shadow detail (by the way, RGB has very little color information in the shadows, and CMYK beats it in that department).

Let's examine how colors mix in the CMYK system (see the RGB article for analogous images):

Here, we simulate an increasing amount of cyan ink moving across the image from none on the left to 100% coverage on the right. Likewise, we have an increasing amount of magenta ink from 0% at the bottom to 100% at the top. No yellow or black ink is shown. We have white at the lower left-hand corner, and somewhat purplish-blue at the upper right hand corner: if we draw a diagonal in between those corners, we have a purplish-blue color going from fully saturated, to a pastel, to white. Along the upper side of the image we a gradation between magenta, purple, and blue, and the right-hand side we have cyan merging into blue.

Please note that this is simulating ink on a page. The red outlined region, in CMYK, is actually outside of the color gamut of the sRGB color system used by this image. When I converted the image from CMYK to sRGB, Photoshop chose the closest sRGB color to represent what was found in CMYK. As it so happens, we can get better, brighter, more saturated cyan inks than what can be shown on most computer monitors: what you are seeing here is actually a bit duller than can be printed. (Perversely, if you were to print this image, some of these sRGB colors are themselves outside of the CMYK gamut, and the quality would degrade even further. Color management is complex, and frustrating.)

Most critical for photographers is the sky-blue colors along the middle of the right hand edge. These colors are outside of the gamut of sRGB — but are well within the range of CMYK. When skies are particularly deep in color, such found at high altitudes, or during a brilliant, clear winter day — especially when you use a polarizing filter — your sky will be out of the sRGB gamut, and will exhibit lots of noise (and this noise will be exaggerated by JPEG compression). Examine your red channel: if it is black, then you know the sky is out of gamut; but if you carefully process your photograph, starting with a RAW image and never entering sRGB, you still might be able to get a clean printed sky.

Here we have cyan ink going across, and yellow ink going upwards. There is no magenta or black ink. These two colors mix together to produce green. We have various colors of leaf-green going along the top edge, and ocean-green along the right edge.

Our outlined gamut warning areas show that we can get a bit better yellow on printed output than we can on a display. Recall that yellow is the opponent color to blue, and digital cameras do a very poor job of capturing blue colors, particularly at low light levels, which may translate to a somewhat poor yellow. But if you capture an exceptionally clean image in the blue channel, you can convert your RAW image to CMYK and use the high-quality ink to get a slightly wider range of yellow in your final image, especially good pastel yellows which are hard to come by in sRGB.

Far more problematic are the green colors: CKMY does a much better job with certain shades of green compared to sRGB. Again, this is a simulation, and were I to have printed the original CMYK file, the color differences would be rather striking. Especially problematic are most shades of ocean green, as well as some shades of leaf-green. This is an interesting observation: CMYK technology, which is quite venerable, does a better job with the natural colors of the sea, sky, and land compared to the computer standard sRGB color system. Also, flesh tones — especially for Scandinavians and Africans — can easily go out of the sRGB gamut. The computer standard was developed before digital photography became widespread — when computer graphics were more concerned with simple business and scientific diagrams — and so was not fine-tuned for common natural colors. However, sRGB does a better job with pure bright reds and blues.

Magenta going across, yellow going up. These mix together to produce red at the upper right hand corner. They don't mix to produce a really good, bright red as we see in sRGB. But we see that CYMK produces better yellows, and some oranges.

Cyan across, black up. Here we finally mix in some black tones, and clearly CMYK is the winner with cyans — and especially dark cyan colors. Recall that sRGB will often throw blue skies out of gamut, which the bright primary cyan ink and the black channel here makes up for quite nicely.

Magenta versus black. CMYK wins with dark magenta tones. Again, remember that you really can't see the actual effect of ink blending on your monitor — the real result is darker and richer.

Generally, RGB color models can be poor because they don't allocate much information to the shadows. There are very few variations of colors that are darker than the blue primary — only about 1% of all allocated colors, which can be seen in another article. Shadows in general tend to be poor, and the number of dark colors are severely limited. CMYK makes up for this by allocating much of its gamut to dark colors.

Yellow versus black. If you studied the previous charts, you probably guessed what this looks like.

Here I mixed 100% each of the three colors across the image, while I added black going up. You ought to notice on the lower half of the image that the mixture of the three colors is not precisely gray, rather it has a slight reddish tone, which tells us that the cyan ink is a bit deficient. (Cyan and red are opponent colors — more of one means less of the other.) With RGB, you merely make all three values equal if you want a pure gray tone, but with CMYK, cyan has to be a bit stronger. For this reason, doing a white balance in this color system is a bit more complex.

There are two reasons why printers have a black ink. One is that a CMY mixture is dark gray at best, and the other is the fact that too much ink on a page can cause smearing or other defects. The black ink can nearly replace all of the colored ink in the darkest shadows, using merely 1/3^rd of the amount of total ink.

In the article Imaginary and Impossible Colors, I showed how three numbers are sufficient to describe any color visible to the human eye. CMYK uses four colors, which means there is often more than one way to specify the same color, by trading-off CMY for black. Printers consider the black plate to be the most important, and photographers creating CMYK separations of their photographs ought to study the trade-offs very carefully. You can also manipulate the K channel in Photoshop — it is an excellent place to add sharpening, local contrast, and nice steep curves for rich shadow detail, but you might inadvertently remove some color.

If you are sending your images to a desktop printer, do not use the CMYK color system. Rather, process your images in a wide-gamut RGB color space, such as Adobe RGB or ProPhoto, and set Photoshop's gamut warning to either CMYK or preferably the printer's own ICC profile. This will allow you to get the rich, deep colors, and to fully express the colors of nature, but avoid bright reds and greens which cannot be printed. Be aware that out-of-gamut colors will either translate to noise or to flat, muddy colors. Understanding CMYK will let you know what to expect. If your printer uses more than four colors, then you are quite fortunate as you can get richer, purer colors — install the printer's ICC color profile in Photoshop and process your images using that profile as your gamut warning. Be sure to use a wider-gamut RGB colorspace for your processing. Since you will be working with colors which likely can't be displayed on your computer monitor, you ought to take the leap of faith that your images might actually look better when printed than what you see on the screen — just keep a close eye on the numbers and on the gamut warning.

If you are sending your images to commercial press, you will want to study CMYK further, or just take your chances and let the pre-press folks do the conversion for you.

You don't have to convert your image to CMYK in order to see what amounts of inks your image would use. You can configure Photoshop's Info panel to display CMYK values for the eyedropper tool. If you are attempting to set a particular color to the brightest possible CMYK red value, you can set an eyedropper on the color and keep an eye on the CMYK values as you adjust your image: you want to set magenta and yellow near 100% with cyan and black low.

CMYK values are also used to correct an image for good skin color. Humans of all races have a cyan channel that is less than the magenta channel, and a magenta that is less than yellow. Black can be nearly any value depending on race. Be very careful not to adjust skin tones so much that they go out of gamut — that is particularly noticeable.

Read part one of this article on CMYK: Color Spaces, Part 2: CMYK
And here is my article on RGB: Color Spaces, Part 1: RGB
If you are confident that you understand CMYK, try this: A CMYK Quiz
For color spaces based more closely on human vision, see this: Color Spaces, Part 3: HSB and HSL

Color Spaces, Part 2: CMYK

2010-11-29T01:23:00.006-06:00

THREE NUMBERS SUFFICE. If you desire to mathematically describe or represent any color seeable by the human eye, the simplest and most well-ordered models will include exactly three numbers, no more and no less.

But please notice that I wrote that three numbers suffice. Normally we think of color theory in terms of mixing colors: for example, computer monitors typically have three kinds of dots, each a certain precise shade of either red, green, or blue. Various mixtures of these color dots at various intensities will produce all the shades of color viewable on the screen, from dark gray or black, to white, and with a rainbow of colors throughout. Alas, although we can accurately characterize every known color by three numbers, we cannot mix all known shades with three primary colors. Three colors do not suffice, and this is the color gamut problem. If you choose three colors for your primaries, then no matter which colors you chose, there will still be colors that you are unable to mix.

For more information, see my article on imaginary and impossible colors.

For reasons of cost and practicality, most color devices use just three colors, and these provide a limited gamut of colors. Most computer monitors and High-Definition televisions use the sRGB color gamut, which can display about 35% of possible colors. Expensive high-gamut monitors can approach 50% of possible colors. Generally missing in these color output devices are rare colors such as scarlet and Imperial purple. Good cyans and some greens are also missing, but the system is generally adequate for most uses.

In the RGB color system, red, green, and blue lights are mixed together to provide a wide range of colors and shades. But we cannot use only red, green, and blue inks on a page to produce a similar range of colors. See my article, Color Spaces, Part 1: RGB, for examples of how these additive colors work together: for example, if you shine a red and green light together, you will get a bright yellow color, but if you mix red and green paints together, you will get a dark muddy mess. You cannot get a bright color by mixing RGB inks. Mixing saturated colored lights together will always produce a brighter color; mixing saturated colored paints together will always produce a darker color. So when we put ink to paper we have to use a subtractive color system, which chooses pure light primary colors for mixing.

Recall the discussion in the RGB article about the opponent color relationships. These are opposite color pairs, which produce shades of gray when you mix them, and not a unique color.

Red is opponent to cyan
Green is opponent to magenta
Blue is opponent to yellow

And since we are working with ink on paper, I might add:

White is opponent to black

The three primary colors in the RGB or additive color system are red, green, and blue, while the three primary colors in the CMY or subtractive color system are cyan, magenta, and yellow. RGB and CMY are therefore opponent to each other.

Consider the following image:

We have red berries against a blue sky. In the RGB system, the red berries will be bright in the red channel, and dark in the green and blue channel, since pure reds have little to no green or blue in them. If we examine the color channels separately, we see this:

Red berries are almost white here, because in the RGB color system white is a strong color, while black means the absence of a particular color. Since the berries are nearly pure red, they are white in the red channel, and black in the other channels. Since we have a nice blue sky, the sky is suitably lightest in the blue channel; and since midday blue skies tend towards cyan and not magenta, the green channel is brighter than the red.

The CMY color system works a bit differently. White indicates an absence of ink, while black means that a particular ink has 100% coverage. So a part of an image where all three channels are white means that no ink is put on the page, and so the white color of the page shows through. The same image in the CMY color system is this:

Red berries have no cyan color in them, and so are white in the cyan channel. Magenta and yellow ink mixed together makes red, so the berries are dark in both those channels. Likewise, blue skies have little or no yellow in them, and so the sky in the yellow channel is light, and is dark in the cyan channel, meaning there is lots of cyan ink there. Cyan plus magenta equals blue, and since our cyan channel is darker than the magenta, the blue sky will properly be a greenish blue shade and not purple.

Please note that the RGB and CMY channels look nearly identical. Working in the CMY color system is hardly different than working in the RGB color system because of the opponent colors used. Please note that the channels are not identical because the printing and television industries use slightly different color standards. But they are close.

Recall the discussion above about limited color gamuts, and how three primary colors cannot produce the full gamut of colors visible to the human eye. Computer monitors really have it easy, since they have powerful back-lighting which can produce bright colors much brighter than the artificial illumination typically found indoors. But poor printed pages do not have that advantage: the brightest tone available will always be the paper itself, and that paper will be duller than the room lighting. And so, printed output will generally have a poor color gamut.

But we can expand the color gamut if we add more colors of ink. Full-color printing always adds at least one additional color to expand the gamut, and in commercial printing, that color is black:

The standard cyan, magenta, and yellow inks used in the printing industry really don't mix together well to make a good black, rather they look muddy. Another problem is that commercial printers have what is called an ink limit: some presses just can't have too much ink on the page without causing problems, and so printers will insist on a limit to the total ink coverage on any given spot on the page. Some shoddy printing may even have an ink limit of 240%, which means that you can't mix together full coverage of our three colored inks, since that would gives us a 300% coverage, which is over the ink limit. 100% black ink will replace 300% colored ink, which is quite a savings, and the black ink looks much better than an equal mixture of colors. Adding black expands the color gamut of the printed page, and does it while ensuring a cleaner press run with less chance of smudging ink.

Here is our image in CMYK (where 'K' means 'key' or black):

The twig is dark brown, and much of its tone now comes from the black channel, as do the shadows. See also how Photoshop removed ink from the color channels: shadows there are now a medium gray.

The CMYK color gamut is considerably smaller than the sRGB gamut most often used in the computer industry and by digital cameras. However, the CMYK gamut is not completely contained within sRGB: printing can produce better cyan, magenta, and yellow, whereas sRGB produces better red, green, and blue.

We can expand our color gamut by adding colored ink. In the printing industry, these are called spot colors. If you don't think that the standard color mixtures are good enough, you can pay the printer to add spot colors. Be aware that this can be quite expensive, and is typically only used for the finest work. Were I to use an accurate spot color for the red berries, they would then become white in the CMY channels — since most of the color would be transferred to the new spot channel.

Cheap computer printers use the CMYK color system. Quality computer printers will have more than three colors, and there are some models that use ten colors. But if you use a desktop color printer to output photographs, be aware that you will be paying several dollars per page for the ink alone. Your costs may be fifty times higher than what a commercial printer charges in bulk.

For a further discussion of CMYK, click here for part 2.
If you think you understand CMYK, then take A CMYK Quiz.
For an overview of the RGB color system: Color Spaces, Part 1: RGB
For color spaces more natural to artists, see Color Spaces, Part 3: HSB and HSL.

Quick Tips for Food Photography

2010-11-25T11:00:00.003-06:00

Shoot quickly — food fresh out of the oven or refrigerator looks better.
Use natural sky lighting. Food often lacks definition, so a small or fairly distant window can produce good shading. Generally, you want the light to provide sharp, well-defined shadows to enhance the texture of the food. You may have to use fill-in reflectors, otherwise color and texture will be lost if large areas of shadows are too dark. Aim for a 1-to-2 E.V. range between large lit and shadowed areas: of course, dark shadows under a plate for example are completely acceptable, just not on the main parts of the food itself.
Avoid using the camera's own flash. Avoid mixing natural and artificial lighting, unless both have a close color balance. Authorities in food photography state that it is difficult to use artificially lighting well: when they do use it, they prefer small, distant light sources to provide sharper shadows.
Set your exposure and post-processing so that you get good highlights on the food: having a full exposure range will also bring out the colors of the food (in Photoshop, using Levels or Curves in RGB mode will enhance color). Be sure that you don't overexpose too large of areas because you might get muddy color shifts. It is OK to overexpose specular highlights.
The color of food is very important to make it look appetizing. Be sure to do a good color balance. Contemporary food photograph seems to prefer a slightly cool color balance, while traditional food photography preferred slightly warm: both look good, as long as the color balance is close to neutral.
Use props to good effect, such as tablecloths, utensils, glasses, napkins and shakers. But be aware that the food itself is the main subject and shouldn't be overwhelmed with secondary items.
Contemporary food photography uses very shallow depth of field, and prefers lenses with excellent bokeh or background blur. This is tricky to do right, for you have to judge the correct focus point. While I think this effect is attractive, perhaps it is a bit overdone. Some use tilt/shift lenses — or even bellows cameras with these motions — in order to precisely control the plane of focus.
Food photography is essentially still-life photography. There is an immense body of work in still-life, particularly with painting. Do some research and use still-life theory to good effect.
Your image may not look like you remember seeing it, due to the dim-light adaptation of the human eye. In particular, your image and texture may look a bit flat. In this situation, food photos may benefit from having the blue color channel blended into the image to give greater contrast to specific colors. See my articles on the Purkinje Correction.
Get low. Typically, we look down on food at about a 45 degree angle; this might not be best for getting a good shot. Get a bit lower.
Check your background. Be sure it doesn't detract from the food, which is your main subject. Classical still life preferred a black background, while contemporary food photography likes a white or pastel background, completely out of focus. You don't want the eye to be distracted by the background in most cases. Alternatively, your photo may only show the table top.
Food benefits from extreme lens sharpness. Macro lenses are particularly prized for this sort of work. Use a sturdy tripod and focus carefully. In post processing, use good techniques to preserve and enhance sharpness.
To give a good perspective, most food photographers use a slight telephoto lens for this work, and set their camera several feet away from the food, and six feet would be better. If you have a stylist, be sure there is plenty of room for working between the camera and the subject. The wider angle the lens, the more area you have to control for your photo: but some have employed wide angles and great depth of field to portray an entire kitchen along with the food.
Food styling — that is, preparing the food itself to look good in photography — is a an advanced specialty, and can be quite involved. I recommend the book Food Styling: The Art of Preparing Food for the Camera by Delores Custer.
My food photography can be seen in the book Thursday Night Pizza, by Fr. Dominic Garramone. Click here to see larger photos of the pizzas: these photos were taken from directly above with no photo styling, per instructions from the publisher. Otherwise I used natural sky lighting, reflectors, and accurate white balance. I used an antique Nikkor 55mm f/3.5 Micro lens for sharpness, with the camera being located about six feet above the pizzas.

Photoshop Wishlist #1

2010-11-10T13:42:00.000-06:00

I AM CURRENTLY evaluating Adobe Photoshop CS5 on my computer, and have 18 days left until the trial copy expires. For the most part, I am delighted by the product, and see many improvements over my old CS3 version. It does not require that much additional computer power — and sometimes it uses even less, since it uses the graphics processor and memory to do tasks once reserved for the main processor.

Photoshop is a venerable, highly developed and nuanced product, and like any complex, actively developed system that's been around for a long time, has many features which see little use nowadays, as well as the refinement to be able to do important things very, very well.

However, a highly developed system may find it difficult to adapt to new conditions, having been optimized for previous conditions. Photoshop has its roots as a raster image processor primarily for graphics arts professionals, and is well-known as a good platform for doing digital art, with its excellent support of many paintbrush-like tools for creating images from scratch. But it is also used in photography, as its name suggests. I am beginning to see some limitations of its photographic capabilities, and one major limit is that images are always strictly bound to an output medium.

For most Photoshop users, this limit means that you edit your images in the sRGB color space, with eight bits per color channel. That isn't too bad, and this is an obvious approach for 90% of all users: after all, that is the standard format used by most cameras and Internet web browsers. Certainly you would want to edit a file in the format which the camera delivers and what your computer can display. Photoshop does things the way it ought to be — right?

I see some problems with this. Each color channel has a maximum value of 255, a minimum value of 0, and we can use only integer steps between: 1, 2, 3, and so forth, with no intermediate values. This lack of precision is of little consequence to most users, and if you do need greater precision — for example, if you are applying severe curves to your image — then certainly you can use 16 bit mode (as I do) to increase the number of possible values. This extra precision helps avoid digital processing artifacts such as banding, and also lets you get better shadow detail.

CS5 has a great improvement over CS3 in that it allows far more operations on 32 bit images, giving us great precision in image manipulation; I haven't tried it yet, but look forward to experimenting with it.

But that isn't good enough. I'd like to see fractional RGB numbers. I want RGB values greater than 255. I want negative RGB numbers. But this is madness! You cannot display an image with RGB values greater than 255! And what on earth are negative RGB values? Those are clearly impossible, there is no such thing as negative light!

But remember that I stated that in Photoshop images are always bound to a specific output medium, which for most photographer users is probably 8 bit sRGB. While clearly I do eventually want an 8 bit sRGB image, while I work on processing an image, there may be times when my intermediate files will be out of that gamut. And I do process my images mainly in the wide ProPhoto gamut — or in the ultra-wide L*a*b colorspace — with 16 bits per channel to overcome the limits of sRGB, at least temporarily.

Do not think of processing images as a step-by-step process, where each increment produces a superior image. Sometimes you have to make an image look worse before you can make it look better. I propose making images so bad that they are impossible to print, or even view accurately on your computer monitor — at least temporarily.

For example, when I apply a severe curve to an image, anything that ought to go over 255 is set to 255, and so we lose information and image detail. However, if its value ought to be 300, I want it to be 300, even though it is out of the gamut for the time being. If I tell Photoshop to make an image twice as bright, I want the entire image to be twice as bright, without worrying about losing highlight detail. I will deal with the gamut when I need to deal with it, which is when I'm preparing the final image for print or web display.

I often add together multiple images to make a final image. What I have to do is apply an opacity to each layer (which is like doing division) to get my final result, but certainly there must be severe rounding errors, and we are losing tremendous amounts of detail in the shadows as a result, which is a bad thing, and especially since digital photography is known for often having terrible shadows. What I would like to do is be able to add together images with impunity. Image addition, which is called Linear Burn in Photoshop, has a maximum value of 255, but if the final value of all this addition ought to be 500, that is what I would like to see.

Generally speaking, I would like to see in Photoshop a pure kind of image algebra, where we can do all sorts of operations on images in a way that follows the standard rules of arithmetic, such as add, subtract, multiply, and divide, as well as other more obscure operations such as exponentials. To do this accurately, we can't have the hard cutoffs of 0 and 255, nor should be we limited to mere integers.

This brings us to negative RGB numbers. These in fact can represent real colors. For example, if you work in a narrow-gamut color space similar to sRGB, and you want to represent a real color outside of its gamut, you can mathematically represent this if you are willing to allow at least one RGB number which is negative or greater than 255. So a negative RGB does not mean negative light, but rather that it is merely an out-of-gamut condition. If we are allowed to use negative numbers — and numbers greater than 255 — then we will be able to represent all colors while still using a system that is otherwise identical to our narrow-gamut color system. This system will remain relative to a particular gamut, while not being limited to that gamut.

This has many benefits to a careful Photoshop user. If you work in the ProPhoto or Adobe RGB color spaces, and I know many people do, how then do you know that a particular color is out of sRGB gamut? Certainly you can turn on the Gamut Warning feature (I use it all the time), but how can you create a mask for this sort of thing? Can you tell, just by looking at an RGB value, that it is out of gamut? By using large and negative numbers, we can then precisely identify what is out of gamut simply by the numbers: is it greater than 255 or less than 0?

I often attempt to brighten shadows, and try to add lots of local contrast so that dark areas of an image still appear to be dark to the eye, yet in fact are not all that dark, and instead show lots of detail. This is often impossible to do well due to the arithmetical rounding errors found in low RGB values, which is a contributing factor to noise. Ideally, a numerical representation of RGB would give equal precision to all levels of perceived brightness, but that is not what we currently have, as can be seen in the illustration below:

Most of our current systems of numerically representing color are biased towards midtones, particularly saturated green and magenta tones, while offering a paucity of dark and bright colors. This gives us the risk of banding in our final image. Having fractional RGB numbers would alleviate this problem greatly, and though we can use 16 bit images, having fractional values would give us a better guarantee of processing shadow values — and highly saturated dark colors — to avoid rounding errors. I've noticed that 8 bit sRGB in particular handles navy blue rather poorly, which is a pity, for that is my favorite color. We always risk banding when we have large areas of dark blue, as is often found in brilliant deep blue winter skies, especially when using a polarizing filter. We see the same problem with bright yellow colors.

If you take an 8 bit image and convert it to 16 bit, Photoshop multiplies the RGB values so that they fill the new numerical representation. So a value of 255 will be converted to 32768, which is the maximum 16 bit number. In the 32 bit system, which uses floating point numbers, 255 is converted to 1.0, which is the maximum value allowed in that system: all smaller RGB values are some fraction less than 1.

Instead, I propose an alternative method. When you convert an 8 bit image to this new system, all values remain unchanged. The difference is that your values can, after processing, be greater than 255, less than 0, or some fractional number. With this system, you can be very careful, and never allow your image to go out of gamut, or you can edit to your heart's content and worry about gamut later. If you edit an sRGB image in the sRGB color space, your image may want to go out of gamut and you will never know it, except that detail will disappear.

There are a few problems with my system. First, you can't see the extra colors if you don't have a wide gamut monitor, but we already see this problem when working in the Adobe RGB or L*a*b color space. The other problem comes when we want to convert a high-precision image back down to 8 bits.

The key to working with images in any gamut is to do by-the-numbers processing, and have a thorough understanding of the channel structure of the images. Instead of merely determining if an image looks OK on your screen, you instead measure an image to be sure the colors are right. Calibrating your images is more important than calibrating your monitor.

Converting an image back down to an output format like 8 bit sRGB is more problematic, but take a look at Photoshop's own conversion options from 32 bit images.

However, doing something like this may not work well within the Photoshop product, as it would require a major redesign of many features. However, I do think that it would be quite useful for accurate image processing.

Black and White

2010-10-26T07:33:00.000-05:00

OCCASIONALLY YOU SEE, on the dpreview.com forums, a posting questioning the use of black and white in contemporary photography. The critic — almost always apparently an educated, brash young man — will declare black and white photography obsolete, for it merely was a product of historical forces, ignorance, and technological compromises, and so it has no relevance to us today; he states that black and white photography is something that ought to be abandoned and forgotten.

This is of course the error of historicism, which in its extreme view denies any universal laws or truths. An opposite error idealizes all situations according to a simplistic theory, and ignores the inherent messiness of life. Most of us bounce back and forth between these two extremes, but rather let's find the virtuous middle and attempt to find out what black and white photography is about.

I can see various reasons for either shooting black and white film or doing digital black and white conversions.

Just because you like it.
Cost, convenience, or necessity.
Nostalgia.
Technical advantages.
For aesthetics or mood.

So why should we still produce black and white photography? Let's consider these individually.

Just because you like it

OK, why do you like black and white photography? Contemplate the reasons why you find it appealing. Perhaps it is some combination of the following?

Cost, convenience, or necessity

Suppose your photograph will be printed in a newspaper, bulletin, flyer, or other inexpensive black and white medium. You may prefer your photo being printed in full color, but since that is not going to happen, you do the best you can despite this limitation.

Museum of Contemporary Religious Art, at Saint Louis University. I needed to convert my image, originally in color, to black and white for inclusion in the book Saint Louis University: A Concise History

If you shoot film, and have your own darkroom or film scanner, black and white photography remains rather inexpensive since you are able to develop and print your own quality photos. If you live in a remote area, this may actually be the most convenient solution also. Also, superb quality film cameras are available at low cost. While you can process your own color film, this is rather more expensive and difficult compared to black and white film.

Sometimes, the lighting conditions are so poor that a black and white conversion is the fastest and easiest way to produce a quality images. I will sometimes convert an image taken under sodium vapor lights to black and white, because the color of that lighting is usually unpleasant and detracts from the beauty of the image. I very often do a conversion when I use extremely high ISO or severe curves on an image, both of which produce intense noise.

Nostalgia

Do you pine for a time when style of dress and manners were better? Times that were happier, even if more difficult? Do you feel a twinge of romance when viewing those older things? Then perhaps you like the nostalgic look of black and white photography.

Doris Day, singer and actress, ca. July 1946, New York City. Photograph by William P. Gottlieb.

I must admit to being a bit undecided if this kind of nostalgia is desirable or not: on one hand, this kind of nostalgia is pleasant. But ought we not prepare for the future, where we are inevitably headed? Or rather, ought we live our life in the present, the only time we can truly see? On the other hand, escaping from the drudgery of the present by an imaginative look in the past is sometimes necessary.

We must not fall into the trap of believing the doctrine of inevitable progress, the idea that things are always getting better and better. And likewise, we must not distrust those who prefer older things; for they may not be reactionaries, but rather they might be correct. The theory of evolution implies eternal betterness, but in reality, for every advancement there are multitudes of fatal mistakes. So what we call nostalgia may very well be a rational attraction to things that were in fact better in some way.

Technical advantages

We must be humble enough to realize that older technologies might actually be better in many ways. A large format camera, with quality black and white film, expertly exposed and processed, will have a range of tones and detail that far exceeds any DSLR snapshot. I do use digital photography exclusively, since it is so convenient, but there are trade-offs.

One of the great advantages of black and white photography is the wide contrast range possible. Often in color, it is difficult getting a full range of tones from pure black to white, since your brightest significant detail may be a saturated color — you can't brighten it without losing saturation. This is particularly troublesome if your brightest color is a pure blue -- you just don't have that much room for other tones unless you do severe edits to the image. Blue skies are often a problem: you can't brighten the foreground without risking overexposure of the sky (which will damage the sky color), which is one reason why polarizing filters are so useful.

I inadvertently overexposed the sky on this image, turning it into an implausible shade of cyan: but it looks fine when de-colorized. This is Holy Family log church, in Cahokia, Illinois.

With black and white images, you only have to worry about over- or under-exposing one tone: white or black. With color images, you need to worry about three color channels, any one of which may be poorly exposed, harming the final image. With color, we have a far smaller dynamic range, which is why color images benefit from fairly flat lighting. On the contrary, the masters of black and white photography use the increased dynamic range to excellent effect.

I took this photo for a book on the Gateway Arch. This is a merge of numerous exposures, and the camera was set to automatic white balance. This is a terrible image in several ways, and the yellow sodium vapor lighting is particularly objectionable.

High efficiency electric lighting often has poor color; fluorescent lights are quite bad, due to the unattractive and broad range of green-to-magenta tones found. Sodium vapor lighting, with its narrow yellow-orange color leads to extremely poor color photos. In these cases, a black and white image may be superior.

The same series of images, but I converted them to black and white before blending. I did some additional processing on this image, such as applying curves and sharpening. In my opinion, this isn't an image I'd particularly want to see in print, but I do think it is an improvement.

Digital cameras have a linear sensor that respond to light such that twice the brightness registers as twice the signal. Unfortunately, this means that most of the sensor's data is clustered around the very brightest of objects, and there is always a great risk of losing detail through overexposure. This also means that most of the tonal scale will be represented with very little data, which leads to lack of detail and noise in the shadows. So the general advice for digital is that you expose for the highlights and post-process to improve the shadows. Black and white film technology, on the contrary, is known for having great shadow detail — you expose for the shadows and post-process to improve the highlights, and unlike digital, it doesn't have a hard cutoff at the ends of the tonal range. Black and white film is traditionally very good for photography in dim, highly contrasty lighting and is used to good effect in film noir.

A forest scene, at Gravois Trail, in Saint Louis County, Missouri. This hand-held photo is underexposed, and was shot at ISO 3200. There is hardly any visible detail, and brightening the image would reveal extreme color noise in the shadows.

The same image, converted to black and white — I discarded most of the red and blue channels. I brightened the image greatly, and applied some noise reduction and sharpening.

Digital noise is most evident in the shadows, and color digital noise is usually ugly and highly undesirable. On the contrary, black and white noise is far less objectionable, and can even improve an image, giving an impression of texture and sharpness. This is often an advantage when shooting at very high ISO, or when brightening a severely underexposed image: a terrible color image can often be dramatically improved by converting it to black and white.

For aesthetics or mood

While nostalgia seeks the better things from the past, and black and white photography may evoke that nostalgia, we must always remember that reality in all ages past is high-resolution, wide-gamut, high-dynamic range color. Would a master photographer of a bygone era have used color photography if it were technologically feasible? Was his mastery of the black and white medium merely making the best of an unsatisfactory situation? Undoubtably for many, although this is speculation. We do in fact know that color technology was eventually widely adopted, and also that black and white never went away.

Color is an important factor in beauty. Bright colors are pleasing, and there are many studies and theories of the psychology of color which assigns good, desirable effects to the various colors. However, the color black is nothingness, the color white can be blinding, and gray is dreary: black and white photography necessarily is less cheerful and pleasant than color. Since black and white is more abstract than color, it can also suggest mystery.

A statue in a cemetery - heavily processed to imply a bleak mood.

So contemporary photographers can use the dreary aesthetics of black and white to evoke a mood of bleakness, despair, and ugliness. This can invoke a kind of anti-nostalgia, seeing not the good in the past, but rather its ugliness, and so black and white photography can be used in a mocking, disparaging fashion. It can also be used with fantasy, where the dull everyday world is seen in black and white, while the fantasy world is in color.

Some conversion hints

When shooting or converting an image to black and white, it is usually essential to adjust the image to give you the full range of tones, otherwise the image may look flat. Good global contrast is essential for a good black and white image. Adjusting curves of color images is a perilous activity: you can have color shifts, oversaturation, and you can send an image out of gamut; these are hardly concerns with black and white images.

Gothic-style ornament at Washington University in Saint Louis. The straight-forward conversion on the left lacks contrast, which is corrected on the right.

The second important consideration is the conversion of colors to gray tones; there are many ways to do this in Photoshop, and there are some excellent plug-ins that improve the process. Even though you lose color information in your conversion, the various gray tones ought to imply different shades in the final image.

The color image on the left was converted in Photoshop using Image->Mode->Grayscale. This is obviously a fabricated image, since I specifically chose all colors to have the same luminance. Photoshop has many ways to convert to black and white, some may be better than others at implying changes in tone.

Generally speaking, you want to select the parts or combinations of each RGB channels which show good contrast between different objects. If the subject has stripes, you probably will want a conversion that shows the stripes in a good manner -- some conversions may not show the stripes at all. Also, for faces your conversion will have a drastic effect on showing or hiding blemishes and wrinkles. In Photoshop, the Black and White tool is very good for doing this conversion, however, a thorough knowledge of the channel structure of images and the laws of color mixing is very helpful for this.

Finally, you can add far more local contrast to a black and white image compared to color, while still making it look plausible. This final step has been used to great effect by the masters of the medium.