The Channels panel in Photoshop: It’s always there, sitting in the background, ready for us to use, but what exactly are channels, and what can we do with them? This article will attempt to shed some light on channels and their role in Photoshop.
Channels are grayscale images built into all Photoshop files that store different types of information: color, selection storage, and spot color. When you open an image (or start a new file) in Photoshop, color information channels are automatically created. For example, when a typical photograph is opened, it’s an RGB file. In the Channels panel (Window>Channels), you’ll see a channel for each color (Red, Green, and Blue), plus a composite RGB channel. The composite channel is the image that we edit.
By default, the channels are displayed as grayscale images, although there’s a preference to change the channels to display in color (go to Photoshop CC [PC: Edit]>Preferences>Interface, and check on the Show Channels in Color option). Generally, there’s no real advantage to having the channels display in color; in fact, as we’ll see later, there is an advantage for viewing the channels as grayscale.
In the traditional printing world, a color image is “separated” into pieces of film. These pieces of film help determine how much of each ink color is applied to the paper on the printing press. To some degree, we can think of channels in the same way: an RGB image could be “separated” into three pieces of film.
If we look at each channel of an image, you can see the relationship of the channel to the image. For example, in this photo let’s focus on the red roof of the barn. In the Red channel, the area of the roof appears almost white; indicating that red is strong in those areas.
The roof area appears much darker in the Green and Blue channels, which indicates that there’s very little of those colors in the roof.
If we were to apply Levels (Image>Adjustments>Levels) to just the Red channel and lighten the brighter areas of the channel, the red of the roof would look even more intense—but there would also be more red in all areas of the photo. This illustrates the relationship between individual channels and the composite image.
The number of channels in an image will depend on the mode of that image. RGB images have four channels (three colors, plus the composite), CMYK images have five channels (four colors, plus the composite), Grayscale images have one channel (since there’s no color information), and Lab images have four channels (Lightness, a, b, and the composite). As we’ll see, these separate color channels can be used to adjust the overall image.
Working with Selections
Channels can be used to assist in making selections. Once a selection has been made, you can save that selection for future use by creating something called an alpha channel. With an active selection, go to the Select menu and choose Save Selection. You have the option of naming the channel, but it’s not necessary. When you click OK, a new channel will be created called “Alpha 1” (if you gave it a name in the Save Selection dialog, you’ll see that name instead of Alpha 1).
An alpha channel “records” your selection in black and white: white represents your selection, and black represents the areas that aren’t selected. In this example, the alpha channel looks like a white silhouette of the subject in the photo.
If you’ve worked with layer masks, this should look familiar since masks share the exact same concept of using black and white. In fact, it’s quite common after saving a selection to turn that selection into a layer mask by loading the selection and clicking on the Add Layer Mask icon (circle in a square) at the bottom of the Layers panel.
You can also see (and take advantage of) a similar black-and-white look when you use the Select and Mask function to fine-tune a selection. In the View drop-down menu in the Properties panel, change the view to Black & White (K) to see a “preview” of your selection as a layer mask/alpha channel. (To create an alpha channel when using Select and Mask, set the Output To method to Selection and click OK. Then use Select>Save Selection to create an alpha channel.)
One quick warning about saving selections: If you improve upon your selection and decide to save an updated alpha channel, be careful. When you use the Save Selection Command, it always defaults to New Channel, so rather than updating the existing alpha channel, it would create a new one. To avoid this, in the Save Selection dialog, choose Alpha 1 (or the name of your alpha channel) from the Channel drop-down menu, and then use the Replace Channel option.
If you save an image as a PSD, PSB, or TIFF file, the alpha channel will be included in the file. To reload the selection, you have two options: go to the Select menu and choose Load selection. In the dialog, select Alpha 1 (or whatever you named the alpha channel) and click OK. Or, you can go to the Channels panel, hold down Command (PC: Ctrl), and click on the alpha channel. Either way, a selection will be made, and the alpha channel remains in place for future use.
An RGB image can have up to 56 alpha channels, but keep in mind that the file size increases with each additional channel. To see the impact of alpha channels on the file size, look at the bottom-left corner of the Photoshop window where the Document Size is displayed. If Document Size isn’t displayed, use the drop-down menu (circled) to select it. The size displayed shows the current file size and the size with the alpha channel(s).
You can use standard selection shortcuts while working with alpha channels; for example, in this photo, two separate selections were made and an alpha channel was created for each selection. To load the selection of both alpha channels, first load a selection by holding down Command (PC: Ctrl) and clicking in the Channels panel on the first alpha channel. Then, hold Shift-Command (PC: Shift-Ctrl) and click on the second alpha channel. In the world of selections, the Shift key will add to an existing selection, so in this case, clicking on the second alpha channel will add to the selection that was created when you clicked on the first alpha channel.
Similarly, if you wanted to select everything in a photo except for the area in one of the alpha channels, you could try this: Press Command-A (PC: Ctrl-A) to select the entire photo, and then hold down Option-Command (PC: Alt-Ctrl) and click on the alpha channel that represents the area you want to remove from the pixels. In the world of selections, the Option (PC: Alt) key will subtract from an existing selection.
Where to Click?
There’s an important distinction between clicking on an alpha channel in the Channels panel, and using the Eye icon beside the alpha channel. To view and work on the alpha channel, click on the thumbnail in the Channels panel. To return to the view of the image, click in the Channels panel on the RGB thumbnail.
To view the alpha channel in the equivalent of Quick Mask view, click on the RGB thumbnail in the Channels panel, and then on the Eye icon beside the alpha channel. Your image will be displayed with a colored overlay, which indicates the areas that aren’t selected.
Using Channels to Make Selections
The other way that channels can help with selections is by starting with channels rather than saving a selection as an alpha channel. Typically, this is used when “traditional” selection methods don’t work, often on images that contain areas that would be challenging to select; for example, using tools such as the Quick Selection tool (W) will often struggle to select areas such as the sky within tree branches and leaves. Instead of making a selection and saving it as an alpha channel, we can use the base channel information as a starting point toward creating an accurate selection.
Imagine for a moment that we had made a perfect selection of the sky in this photo and saved it as an alpha channel. What would it look like? It would be a black silhouette of everything except the sky, and the sky would be white.
That’s what we want to end up with, but we can’t achieve that using typical selection methods so, instead, we start with the existing channels. There are two different ways we can approach this.
We’ll begin by looking at the individual channels to see which one gives us the best start toward the alpha channel we imagined earlier.
Step One: Click on each channel in the Channels panel to see which one is the closest—or best start—for getting us to our goal. Although it’s not always the Blue channel, often it will give you the best start.
Step Two: The next step is very important: Duplicate the chosen channel (in this case the Blue channel) by dragging the thumbnail onto the Create New Channel icon at the bottom of the Channels panel. You’ll see a new channel called Blue Copy. It’s crucially important that you work on that channel, because if you were to work on the Blue channel, you’d be altering the image itself—and we don’t want that.
Step Three: You can then use Levels to adjust the channel to get it closer to your alpha channel target. Be careful not to push the sliders too far here as you could lose small details. (It’s likely that it will take several steps to get to our goal, so be prepared to start with Levels as just the first step.) In this example, the entire bottom area needs to be black, so we could use the Rectangular Marquee tool (M) to select the bottom area and fill it with black (press Shift-Delete [PC: Shift-Backspace] to open the Fill dialog, select Black for Contents, and click OK).
Step Four: Now we need to make the dark-gray areas black, and the light-gray areas white. The simplest way to do that is by painting in Overlay mode. Click on the Brush tool (B), and in the Options Bar, change the Blend Mode to Overlay. You may also want to lower the Opacity to around 70%. Overlay mode will change the behavior of the Brush so that a Foreground color of white will only affect shades of light gray but not black. Similarly, with black as the Foreground color, only dark-gray areas, not white areas, will be affected. Remember, our goal is to move toward the “perfect” silhouette of the sky.
Step Five: The final step is to turn this channel into a selection by holding down Command (PC: Ctrl) and clicking on the channel thumbnail. In this example, we’ll add a new sky into the selected area.
Step Six: Switch to the photo of the sky you want to use, and select the entire image by pressing Command-A (PC: Ctrl-A). Copy the pixels by pressing Command-C (PC: Ctrl-C).
Step Seven: Switch back to the original photo (the selection should still be active), and click on the RGB composite channel in the Channels panel. From the Edit menu, choose Paste Special>Paste Into. This will create a new layer with a layer mask (based on the selection). Use the Move tool (V) to reposition the new sky.
Another approach to creating a selection channel from existing channels is the Calculations command. Found under the Image menu, the Calculations dialog lets you combine channels to get the best start toward that perfect silhouette. Best of all, you can experiment with different combinations of channels and blend modes to see what gives you the result you want. Just remember, chances are you’ll have to continue to fine-tune the results using the methods we described previously.
Luminosity masks (sometimes called luminance masks) provide a way to make very targeted adjustments to a photo by using very specific selections. These selections come from alpha channels that were created based on how bright or dark an area is. The term luminosity masks actually means having a series of alpha channels that you can use to load different selections. Those selections can be made to make adjustments or to mask specific areas of layers.
Channels can also play a role in making adjustments such as color correction and converting to grayscale. If you look in the Properties panel for either a Levels or Curves adjustment layer (Layer>New Adjustment Layer), you’ll see a drop-down menu that lets you choose between RGB, Red, Green, or Blue when you’re making an adjustment.
More often than not, you’ll make an adjustment to the RGB channel to affect the entire image. If you need to make a specific color adjustment, such as decreasing the amount of green in an image, choose the Green channel and adjust the curve or histogram for just that channel. For example, to slightly reduce the amount of green in this photo, we choose the Green channel in the Properties panel for Curves, and then drag down the center of the curve.
If you apply Curves as an adjustment layer, you can paint with black on the mask to hide the effects of the reduced green (make sure you change the blend mode of the Brush tool back to Normal in the Options Bar).
Another powerful way to adjust an image is with the Channel Mixer, which can be applied as an adjustment layer. This command can be used to make adjustments to the colors in a photo, or to convert to a grayscale image. The Channel Mixer works by mixing the grayscale values of each channel by mixing their brightness values.
For instance, here in the Channel Mixer with Red chosen as the Output channel in the Properties panel, reducing the Red amount darkens the Red channel, which in turns reduces the amount of red in the image.
To add red back in to the Green and Blue channels, increase those percentages in the Channel Mixer. As a guide, aim to keep the Total close to 100%.
The Constant slider adjusts the grayscale value of the output channel, with negative values adding more black, while positive values add more white.
You can also use the Channel mixer to create a grayscale image by clicking the Monochrome checkbox and moving the values on each color channel to get the result you want.
Like many functions in Photoshop, once you see what you can do with channels, you’ll start to take advantage of them on a more regular basis.
ALL IMAGES BY DAVE CROSS EXCEPT WHERE NOTED