As I’ve mentioned in several posts I’ve been going through some awesome tutorials by David Nightingale (http://www.chromasia.com). David has quite a few tutorials relating to the “creative process”. He walks you through his thought process on sample images and explains the choices he makes — ranging from basics like cropping and minor cloning to advanced masking and application of adjustment layers. I can’t recommend the tutorials highly enough.
To create the image above I used many of the things I learned from David. He’d probably be horrified and have thoughts along the lines of “You’ve read/watched my tutorials and this is all you got out of them!” Nonetheless I’ll briefly describe what I did to get this final image.
To start off, let me say that for this image I did not crop nor did I clone or do spot healing edits. There is some work which could be done in these areas but I was focused more on trying to execute a certain “look”. I’ve learned (from the tutorials and via experimentation on a half dozen images or so) that it’s rarely best to convert an image to black and white using a straight grayscale conversion or desaturation. My main objective was to practice getting a “nice” B+W (or toned monochrome image — that’s technically what my final image is) using the channel mixer in Photoshop.
Here are the layers I ended up with:
The first thing I did was duplicate the background layer and change the blend mode to “soft light” at 72% opacity (played with the opacity until I liked it). This gave this image a soft glow. It brightened the skin significantly. I did not want this effect on the eyes, hair, and scarf and you can see that I masked the background layer back into these areas.
I then used two channel mixer adjustment layers to get a monochrome image. The first layer used values R:+43, G:+39, B:+18. I decided that the background needed different look so I used the B+W with blue filter setting with R:0, G:0, B:+100. Where did I get the numbers for the channel mixer layers? I experimented until I liked it. I won’t go into how the channel mixer works(because I can’t remember it all anyway) but the settings you choose can have drastic effects on your images. As a general rule of thumb you want your RGB values to add up to around +100 for a “normal” looking image. I leave further theory/study up to the reader until I understand it better myself.
The first curves adjustment layer lightened the background to give more separation between it and my daughter. The second slightly brightened and added contrast to the eyes, lips, and teeth. The third curves adjustment brought up the midtones and was masked only into the eye sockets and a small shadow area between the chin and lower lip. This evened out the levels in the face.
Up to this point I was working with a true black and white image. I wanted to tone the image a bit and I used the “Curves 3” layer to do that. I brought up the reds a bit in the shadows and midtones and decreased the greens and blues in the midtones by adjusted the red, green, and blue channels individually. There are endless toning possibilities using curves. Of course, tools like Lightroom have many presets available that are often satisfactory but using curves gives you ultimate flexibility — you can change the colors infinitely and even narrow your adjustments to particular parts of your histogram. Here is what the toning curves looked like:
The “Curves 2” layer darkened the image and was masked in selectively to add some vignette.
Here’s the image without the toning curves layer. There’s no right or wrong answer but I’m curious what others like — toned or full B+W? Honestly I have a hard time choosing between the B+W and the toned image — I like both quite a bit. I think the toned version works well with my daughter’s brown eyes and brunette hair color.