Every so often you run across an article that offers some unfortunate advice on an important subject – perhaps it offers outdated advice or suggests a less effective technique. I saw such an article today on methods for converting color digital images to black and white in Photoshop. Sometimes seeing such an article encourages me to write something about the same topic as a counterpoint – and that is what led to this post.

I cannot claim to be the master of this conversion process, and I recognize that there are several ways to do it that are both effective and different that what I’ll describe. In addition, this will be more or less an overview rather than a detailed tuturial. IN any case, I want to offer some ideas that might be a bit more contemporary, flexible, and powerful than those I read about in that article.

Should you simply use the black and white mode on your camera, or should you capture in color and convert during the post-processing stage? The answer to this has several parts.

First, if you intend to produce a very high quality photograph you will almost certainly want to shoot in RAW mode – and RAW capture on DSLR cameras always includes full color data. Even if you select a monochrome option on your camera and see a monochrome display on the LCD and perhaps even see a monochromatic image when you open in your proprietary RAW conversion software, by definition a RAW file includes all of the color data. Whatever it looks like, it really does still include all color data. So, in a sense, the question is moot.

Beyond that, even if you could capture only black and white luminosity values you probably would not want to. There would be no advantage in throwing away picture data in order to get a purely black and white file, and there are considerable advantages in keeping the full color data for your post-processing work. The latter option is comparable to bringing the full color original scene into your “darkroom” and having the ability to make very careful and precise decisions about things that black and white film photographers had to calculate in the field with little or no opportunity to reconsider later. In addition, you can apply many of the same processes that black and white film photographers might apply (such as using color filters) in much more precise and sensitive ways. For example, with film you might use a red filter. In Photoshop you can apply a similar red filter in post, but you can also scale the effect or even apply it to only a portion of the image, and such techniques provide you with much greater creative control.

So, I strongly recommend capturing a full color RAW image in the camera, and then keeping the image in color during the RAW file conversion process. Here the main idea is to essentially protect the data by avoiding clipping at either the white or black ends of the histogram scale and so forth. Keeping the file in 16-bit form take it to Photoshop. (I use ACR so that I can import the file directly into a smart layer in Photoshop without an intermediate TIFF file stage.)

Once the file is in Photoshop there are a number of ways of going to black and white. You could simply desaturate the image, but I strongly recommend that you not use that approach since it provides very little control over the image quality. If you are using an older version of Photoshop you could use the Channel Mixer adjustment layer – actually you can still use this approach in CS4. (Before CS4 this was the tool that most photographers probably used for black and white conversion.) The quick description of the approach is that you select the monochrome option in this box and then manipulate the three color sliders until you get a pleasing result. It is possible to do effective black and white conversions this way in CS3, but the newest version of Photoshop provides a more effective alternative

With CS4 it is generally a much better option to use the Black and White adjustment layer. I won’t explain all of the possibilities here, but here are a few. First, a popup menu at the top of the Black and White adjustment window lets you select from a number of preset “filters,” some of which correspond to the familiar filters you may know from black and white photography: red, yellow, green, and much more. In many cases one of these choices will make a fine starting point for your conversion. Beyond that, this adjustment layer provides six color sliders rather than the three in the Channel Mixer. You could either set up your conversion manually using these sliders or you could start with one of the preset filters and then fine tune it for your image. There is also a “tint” control that you can use to, well, tint your photograph if you are so inclined. In earlier versions of Photoshop I learned to use the Channel Mixer tool quite effectively – but now that the Black and White adjustment tool is available I never find myself going back to the old method.

In order to really control the image you will likely want to do more than just convert the whole thing using a single setting. One very powerful approach with the Black and White layer (and this also works with the Channel Mixer) is to mask separate areas of the photograph and convert differently – something you cannot realistically do in-camera. Consider a scene that includes somewhat dark vegetation and a somewhat washed out sky with interesting clouds. In one Black and White layer create a mask so that the conversion affects only the area of vegetation and perhaps apply a green filter to lighten the plants. Then create another Black and White layer with a mask that constrains the effect to just the sky and apply a red filter to increase contrast between clouds and blue sky.

At this point your photo is probably moving in some interesting directions, but to really get the tonal values where you want them you’ll almost certainly want to make some additional changes to make blacks black and whites white and everything in between look as good as possible. While you could sort of do this with a Levels adjustment layer, that provides far less subtle and flexible control than using a Curves layer – so I strongly recommend the latter instead. The Curves layer will let you do anything you can do with Levels (e.g. adjust the white and black points and move the gray value), but it also lets you fine tune these effects much more precisely. (A big ‘thank you’ to Charlie Cramer for setting me straight on the power of using masked curves adjustments!)

The idea here is to use the Curves layer to accomplish three things: get blacks to print black (not some almost black shade of gray), to get whites to be white or at least very close to perfect paper white, and to increase contrast in the mid-tones. A simple starting point is to move the black and white ends of the curve in to deal with the “black” and the “white” issue, and then to create a shallow S-curve to increase contrast a bit in the middle. There are, of course, many subtle variations on this idea, but I’ll leave it to you to play with that.

There is one more important twist on the use of curves. While you could simply apply one curve to the entire image, you probably don’t want to do that. It is much more effective to mask specific areas of the photograph and apply the ideal curves to those areas. For example, the curve that might work for your cloudy sky is probably quite a bit different than the curve that might work for dark and low contrast foreground vegetation. (There are some interesting tricks to apply when making the selections to create the masks, and also some subtleties concerning how to set the values along the curves that I won’t go into here.)

This post is not meant to be a definite tutorial or “how to” on this subject, but I hope that it may help some people think about and apply some effective techniques for creating better black and white conversions. If you have questions about any of this please (join the web site and) post a comment and I’ll try to respond.

*