Today’s tip comes from from Chapter 26, “Masking Essentials,” of Deke’s video course Photoshop CS5 One-on-One: Mastery from lynda.com. And because this course is all about mastery, Deke takes on tougher masking challenges in this chapter, delving into such deep techniques as exploiting the native components of an RGB image to create highly exacting alpha channels and masks.
Some of you may recognize this image (care of Stas Perov of the Fotolia image library) from Deke’s Photoshop Top 40 videos, starting with Feature #33: Calculations. But this is a more detailed analysis, which includes compositing the masked image against a blue sky background.
We’ll focus on the segment entitled “Making an alpha channel,” where Deke demonstrates how the art of masking lies in using the image to select itself.
The goal is something like what you see immediately below: to use the image data to create a mask with detail, finesse, and high contrast that cleanly separates the flying hair and the figure from the background.
“How does that work?” you may ask. Well, you can assess and use the contents of the Red, Green, and Blue channels of your image to create an alpha channel, which can be further refined to create the final mask.
- The first thing to do is to activate the Channels panel. When creating and editing masks, 9% of your activity happens there.
- It’s where you create new Alpha channels (Alpha channels are extra channels in the file, not the R, G, and B channels that control the color in your image.)
- It’s where you develop your channels, e.g. exaggerating contrast and finessing details
- It’s where you store masks for later use
You begin a new mask by evaluating the contents of the individual color channels in your image. You’re looking for a channel with good contrast as a basis for your mask. When working with portraits, some general tendencies apply.
- The Red channel is where the skin tones reside and it contains the brightest skin values.
- The Green channel often exhibits the best detail.
- In the Blue channel, skin tones aren’t as bright and detail is typically lower.
- Once you’ve determined a channel to use as a starting point, duplicate it by dragging it down to the Create New Channel icon (the upturned page icon at the bottom of the panel) and rename your new channel as needed.
- To create an effective mask, you may need to invert the channel, since masks work by the principle of “white reveals and black conceals.” The result may resemble the image below.
That’s really just the beginning. You’ll probably find that the resulting channel doesn’t have enough contrast in some areas, and may not have enough density in others to be an effective mask. Where to go from there?
- In some cases, you may be able to resolve it by applying the Levels command to punch-up the contrast, but sometimes that causes a loss of detail.
- Another approach is to combine information from multiple channels with the Image > Calculations command.
In the ensuing two segments of this chapter of Photoshop CS5 One-on-One: Mastery, Deke shows how to use the Calculations command, including the Add and Subtract blending modes along with the Offset and Scale values.
What if you want to use these masking techniques with a CMYK image? You can’t mask as effectively using CMYK channels as you can with RGB channels. (Deke explains why in the video.) But, all is not lost. You can create an RGB version of the file to build your mask, then transfer the mask back to the original CMYK image.
- Select Image > Duplicate from the menu bar
- Select Image > Mode > RGB Color from the menu bar
- Create your mask
- Arrange the image windows to show the RGB and CMYK files next to each other
- Drag the mask channel in the RGB file to the CMYK file to copy it.