Sepia Scenes #10: "Sepia Tutorial Part I - Color Conversions"

---

This post wasn't originally scheduled to come out until after the first of 2009, but I sent a preview of it to Mary the Teach, and she got so excited by the idea I decided to reschedule it. This is part one of a series of "How I Did It" articles. I say "How I Did It" rather than "How To" because like most photographic -- or for that matter artistic -- methods, there's no one "right" way. Pretty much everybody who participates in this meme already has a conversion process they use, and all of them work. My aim in doing this series isn't to convince those people that my way is "better" than theirs. My aim here is to give those who might want to participate, but aren't sure how to get from point-a to point-b one method of doing it that they can try out and adapt to suit their own particular styles. And if the current players get some ideas from it as well, that's a bonus.

I plan to discuss two broad topics here in the next few weeks. First, how do we go about getting a good sepia image? Second when is sepia the best choice? Since the "how" question is much more easily answered, and probably of more immediate interest to most of those who will read this, I'll tackle that one first. There are a couple of basic ways to get an image. The first -- and probably the most common -- is to shoot the original image in color and then convert it. The second way is simply to shoot it in monochrome using a sepia effect. (Most digital cameras have this capability, but I haven't heard anybody say they did it this way so far.)

Since converting from color is the most popular method among the participants here, I'll examine that process first. Converting from color has a couple of significant advantages. First, and most obviously, it gives you the ability to turn any image to sepia whether it was originally shot yesterday or 50 years ago. I've found that a great many shots I've taken that I wasn't excited about take on a whole new dimension when converted to monochrome -- be that black and white or some other variant. But there's another advantage to converting from color that's more subtle. By adjusting the color balance of the original image prior to conversion, we can improve -- sometimes drastically -- the tonal and dynamic ranges of the finished product. Finally, conversion is probably the easiest method for obtaining a good sepia image. Most of us know how to go about shooting a good color photo, and it's the original shot that will ultimately determine the quality of the final image. Shooting in color is pretty much a WYSIWYG operation. The scene you see before you through the viewfinder is going to be faithfully recorded on media of some kind just as you see it (with all the usual caveats about exposure, depth of field, etc.) Shooting in monochrome isn't so much difficult as it is unfamiliar. All those colors you see through the viewfinder are going to be reduced to shades of gray (in a generic sense at least) so you have to mentally remove them from the image. So it's not simply a matter of finding a scene you like and clicking the shutter. It becomes an exercise in visualization, and it takes a little practice and a little experimentation to be able to look at a scene and know how it's going to look with virtually all of the colors removed. By using a conversion method, you get to see step by step how it's going to look, and you can adjust as you go rather than having to anticipate in advance.

So what are you going to do in your conversion process? Simply put, all you're really doing is removing certain colors and replacing them with a (much) smaller set of colors. How you go about this depends a lot on the software you're using. Some programs make this a one-click operation, others make you work for it a little, still others make you work for it a lot. And I don't use all of those programs, so everything I tell you here in terms of "How I Did It" (bonus points if you know where that reference came from) is based on the software I use. The good news for you is that I'm hopelessly antiquated in my setup which mainly consists of an embarrassingly old version of Adobe Photoshop (Version 5.5. There. I've said it. Are you happy now?) So anything I was able to do, you should be able to do -- and probably easier.

In outlining the steps I took, I've tried wherever possible to avoid references to specific menu options or anything else specific to Photoshop or any other specific piece of software. Because whatever software you use, the concepts are the same. I do have certain shortcuts I take for most posts because ... well partly because I'm lazy and partly because I don't like reinventing the wheel. But for this post at least, I took very careful notes of each step. You can adjust the numbers and percentages as you like -- there isn't really a "right" way to do this -- all I'm doing here is giving you the adjustments that yielded these images. If you want them toned differently, lighter, darker, yada-yada you can adjust for it when you do your own.

I started with these three images all take in February of 2001 during a trip to the Blue Ridge Mountains (specifically Valle Crucis, NC if you're interested). All of them were shot on 35mm color negative film and scanned from the negatives so the original versions are pretty close to exactly what came out of the processor with no adjustments for anything.

The process for all three images was essentially the same:

1. I created a heavy "red shift" by pushing the level of the reds in each range (shadows, midtones and highlights). The reason behind this step is to create a better tonal range in the monochrome version. If you've ever shot black and white film with a red filter, you've done the same thing. Adding red actually deemphasizes and lightens the reds when the image is converted, while at the same time emphasizing and darkening the complimentary colors (the greens and blues in an RGB image like this). The result is better contrast and better tonal range in the monochrome image. (For Photoshop users, I did this with the "Color Balance" option. You could also use the red compensation curve or the Channel Mixer, though these are a little trickier to master. For non-Photoshop users, you should have a similar option somewhere in your menus, but I'd have no way of knowing what it might be. Sorry, you're on your own on that one.)

2. I desaturated the images 100% to dial all the colors down to gray. Note that I didn't convert to a greyscale image, because that would remove all the color data from the file completely and I want that data available later. PS/PSE users have an actual "Desaturate" option on the menu. Other packages, I don't know about. (You may be able to convert to greyscale and still apply a tint using some other option if you can't simply desaturate the colors.)

3. Once the image is rendered in greyscale, I adjusted the brightness and contrast as if I were going for a good black and white image. In greyscale all you have to enhance the image is contrast, brightness and tonal range. So what makes a good greyscale image will usually give a good toned monochrome image (sepia or otherwise).

4. Once I was satisfied with the greyscale image, I added red and yellow back in until I got the toning I wanted. (Once again I did this with the Color Balance option, once again the other methods I mentioned could be put to the same use. You can also save this scheme using the "Variations" option so you can recall it later. And once again, non-PS users will have to find the equivalent options. Sorry, you're on your own on that one.)

Next Time: Shooting it in monochrome to start with!

---

Stumble This!