Black and white photography used to be a limitation of film, rather than a conscious choice the photographer made. Today, it's become a style or genre; some photographers ( Sebastiao Salgado ) still shoot black and white exclusively, but most make decision on a photo-by-photo basis. Like sepia toning, B&W can make a photo look old and timeless, and for some viewers, lend it class. There are several ways to convert a color photo to black and white, so this article examines their strengths and weaknesses.
First, we'll examine a color photo, at right, then we'll test different conversion methods against it. This is a good candidate for B/W treatment because the photo already has a lot of white and bright gray throughout the frame, and plenty of contrast between the clouds, sky, and some of the trees in the background. We'll start with two attempts that don't produce very good results, in the hope that these help the reader see why the preferred method is better.
Finally, we'll look into toning the image.
The first thing most people try is converting between color modes, from RGB to Grayscale, and then back again if you'd like to save the results as a jpeg file. This requires very little effort, and protects the photographer from having to make any hard choices, but the resulting quality is low.
This is done by selecting the Image menu, Mode submenu, and then Grayscale. In general, you'll probably want to change back to RGB Color after throwing the color information away, because so many file formats demand a full-color image.
The result is to the left. Although the detail is preserved fairly well, this just doesn't look very good, tonally. The image has lost a bit of contrast; the grass to the right is all the same tone after the conversion.
Clearly, this is not an ideal way to make B/W images.
Saturation refers to how vibrant a color is; another commonly tried ( and commonly given up on ) technique is to use the Hue / Saturation tool in Photoshop, and setting the saturation as low as possible ( -100 ). Basically, this removes all the color from the image, leaving behind a map of the luminosity in the scene.
The result still fails to impress, but for different reasons.
Contrast in the de-saturated photo looks more like the original, although the difference is subtle in most of the frame. More obvious is the slightly low-key feel that results. A lot of the grass is colorful, but dark; now it's just dark.
Look at the trees in the middle-ground; they're pretty hard to distinguish in the de-saturated image; individual trees stand out more in the color mode conversion. The tones aren't adjusted to try to reflect what color they used to be when you simply de-saturate the image.
Again, this isn't a great way to convert to a B/W image, most of the time.
The Channel Mixer is a much more powerful option, uniquely well suited to digital captures. And yet, this conversion method is popular with ex-film shooters because, like much of Photoshop, it mimics a common black and white film technique - using a red filter in front of the lens.
Note: the CCD or CMOS chip in digital cameras and scanners are color blind; they see luminosity, but not chromaticity. To overcome this, color filters are placed in front of each pixel on the chip. Bayer sensors use an alternating RGBG pattern ( meaning there are twice as many green pixels as red or blue ) and then reverse engineer a full-color image, much like the human eye; while Foveon's x3 sensor uses all three color filters at each pixel at different layers, more like film.
Ultimately, rather than using a red filter in front of the lens, you're using a red, green, and blue set of filters behind the lens. RGB files are encoded as a series of pixels with a red, green, and blue component; Photoshop represents these as channels. ( See graphic and links. )
This shows why black and white film veterans used red filters. Look at the dramatic effect on the sky, especially useful for landscape work. Contrast in the red channel is very good all around, except that some detail is lost in the in the green hillside to the left. The green channel is more like what the scene looked like in real life, and closest to the color version of the photo, but the red is more dramatic.
In most cameras, the green channel will be the cleanest - least noise - because there are twice as many samples. The blue channel is usually darker, because the color filter itself is darker and blocks more light; for the same reasons, there tends to be more noise in the blue channel. In this case, though, we see the sky burnt out, with one of the clouds disappearing almost entirely.
Like its name implies, the Channel Mixer lets you adjust how much influence each channel has on the image. As useful as the Channel Mixer is in color photography, it's also the ideal way to convert to black and white.
How much of each channel to include is a matter of personal taste. When the percentage from all three channels totals 100, the image will match the brightness of the original; you can make the frame brighter by exceeding 100 % or darker by falling short.
Notice how the trees in the middle-ground, to the left, are more distinct, but the sky is dark and dramatic, in stark contrast against the clouds? This is clearly better than de-saturating the image or converting to grayscale. Let's review:
The difference is subtle, but can be important. Remember, we're only talking about how to convert a color image to black and white; there's plenty more you can do to make it sing. This is just a starting point.
After a black and white print is made, it can be "toned" in the wet darkroom to a particular color, or a mixture of dyes to augment the black in the print.
Photoshop mimics this in Duotone color mode, which is meant to simulate a color separation print, like in a newspaper, or 1970s comic book. Photoshop allows between one and four colors of ink in Duotone mode.
This can be used to "tint" a photograph, like the one at left, which to some degree looks like a blue filter used with color print film. ( Purple can go very well with ocean and waterfall landscapes, especially near sunset. )
This is blue ink over black ( see cyanotypes, below ), but all colors are available. You can even replace the black ink entirely, but it "grounds" the image. Using a single color ink in place of black produces something like an Andy Warhol painting.
To control the mixture, you can adjust the curves for each color ink you choose. This works backwards, compared to what we're used to from the standard curves adjustment; pulling the bar down makes the image brighter ( shown below ). This takes some getting used to, but tends to be well worth the effort.
Note: A duotone needs a grayscale or monochrome image to start with, otherwise Duotone will be disabled in the Image -> Color Mode menu.
Sepia tones can be achieved as a duotone or tritone, by mixing black and yellow ink; magenta can be added for a touch of rust-like oldness. Yellow and brown also make a good mixture. Because sepia photographs are so heavily associated with the 1880s, almost any photo will seem nostalgic after toning.
Very slight changes to the hue and the curves for its "ink" can drastically change the mood of the image, so you should be able to adjust things to your taste.
Sepia toning can be particularly effective for desert landscapes, old architecture, and "period portraits," with the subject dressed in 19th century gear.
Like sepia, cyanotypes hail from an old and long-forgotten photographic process ( from the glass-plate era ) that produced bluish-cyan prints.
Today's cyanotypes, like the one pictured above, also fall under the umbrella of Photoshop Duotones. These are made with black, blue, and cyan ink; sometimes the smallest amount of green is added.
While many photographers like to create black and white conversions that are faithful to the way a particular B/W film responds, sepia and cyanotypes are so old as to give the photographer a bit of artistic freedom.
Now let's see how these treatments applied to a different type of photo look. The large expanse of blue sky is enough by itself to make the red-filtered conversion so attractive. But how do things look when there's no sky in the frame?
In this case, the Channel Mixer didn't do a much better job than changing color modes. De-saturation stands out clearly as the worst conversion method. For reference, here are the individual channels from the color images. I applied a very strong dose ( 0.3 px at 500 % ) of sharpening to highlight the noise differences ( at ISO 800 from a D60 ) between the color channels:
The green channel clearly has less noise and, in the flower at least, much better detail. If the noise were a problem ( if it hadn't been exaggerated so much by over-sharpening ), a slight blur over the red and blue channels is one approach to dealing with it. The real point here, though, is that building a B/W image though the channel mixer tool lets you avoid the noise in these channels completely, or take little enough influence from them that it's not a problem.
We've seen that de-saturation gives bad results, because it doesn't take visual differences between colors into account when it produces a grayscale image. Converting between color modes takes the way humans perceive color into account, and does a better job for it - acceptable in most cases. Most important, we looked at the differences between the color channels, seen that the red channel usually has the most pleasing rendition, but the green channel tends to be the cleanest and most faithful reproduction. The artistic control the channel mixer tool gives makes it the ideal solution for color to black and white conversions in Photoshop.
Finally, we looked into toning the monochrome image with the duotone feature in Photoshop. This page used sepia and cyanotypes, because these are my favorite ways to tone an image, but the variations are limitless. We'll leave other color options as an exorcize for the reader.
Click to download a Photoshop sepia duotone file. You'll still have to convert color photos to grayscale, and the Channel Mixer is still probably the ideal way to do it. From that point, though, you can use this to tone your photos.
Two photos were used in this article, and they deserve a brief description.
The first is a landscape from Montana; the beginning of Garnet Ghost Town Road. On a cross-country drive from Connecticut to San Francisco, I felt I had spent too much time on the freeway, and decided to follow the next road that would take me from I-90 to anywhere near Glacier National Park.
This was the next road: dirt, pockmarked with potholes, and nearly abandoned beyond the ghost town. My average speed was about 10 mph, and the road was long, narrow, and windy; a paved road would have been a much better choice.
( The dirt road comes out at Montana Highway 200 and then 83, "Swan Highway" because it passes Swan Lake, which ultimately reaches Kalispell. )
The second photo is a sphinx moth pollinating a dandelion in Uinta National Forest, Utah. Because of their speed and the way they flutter about, these are also known as "Hummingbird Moths."