How to enable channels in Photoshop. What are color channels used in Photoshop?

Color correction in Photoshop

Sofia Skrylina, teacher information technologies, Saint Petersburg

CompuArt No. 10’2011 described methods for diagnosing image tonality and tools for tone correction of photographs. As we promised, this issue will look at tools for color correction. All examples in this article use the RGB color model.

Diagnosis of color shift

Before you begin to correct a photo, you need to determine the color that prevails over other colors, that is, the color shift of the image. A histogram, sorted by color channels, and color samples will help you with this. Moreover, these same tools will inform you about the results of the correction. In addition, when correcting, you should remember that the color shift is noticeable in light and gray tones, but it is not visible in the shadows.

As a rule, for an image that has a color shift, in addition to color, it is necessary to carry out tone correction. Therefore, before you begin to remove color shift, balance the image tones.

Image histogram, sorted by color channels

Let me remind you that the palette bar chart(Histogram) is called using the command Window(Window) -> bar chart(Histogram). To arrange the graph by color, you need to select the item from the additional palette menu, which opens by clicking on the button View all channels(All Channels View). And to display color graphs, select the item Show channels in color(Show Channels in Color) - fig. 1.

The histogram shown in Fig. 1 shows that red color predominates in the image: the graph of the red channel is shifted relative to the others towards the highlights. For the convenience of comparing all channels, they can be shown in the same coordinate system (Fig. 2). To do this you need from the list Channel(Channel) select item Colors(Colors).

Note. A color channel is a grayscale image that reflects the luminance distribution of the corresponding base color. An RGB image has three color channels: red, green, and blue. Halftone images in the red, green and blue channels are superimposed on each other to form a color image. This result is usually called the combined RGB channel.

Color tests

To analyze the color shift of an image, it is also convenient to use color samples (marks) that are set by the tool Color standard(Color Sampler).

Tool properties panel Color standard(Color Sampler) contains only two parameters:

• list Sample size(Sample Size), which controls the size of the sample (the default size is Dot(Point Sample));
• button Delete(Clear), which removes all color samples from the image.

To manage color proofs, use the following rules:

1. To move a color swatch, hover your mouse over it, it will change to an arrow , and then simply drag the swatch to a new location.

2. To move the color proof while the adjustment window is open, click Shift key, move the pointer to the sample and move it.

3. To delete a color proof, drag it outside the document window. You can also press and hold Alt key(on Mac OS - Option key), when pointing at the sample, the pointer will take the form of scissors, clicking the mouse will delete the sample.

4. To delete a color sample while working with the adjustment dialog box, press and hold the Alt+Shift key combination (in Mac OS the Option+Shift combination), when you hover over the sample, the pointer will change to scissors, clicking will delete the sample.

5. To delete all color samples, click the button Delete(Clear) on the tool properties panel.

When working with labels, remember the following numerical values ​​of the color components of the RGB model:

• gray color should have equal or close to equal values ​​of all three components - R, G, B;
• white color has the maximum or close to the maximum brightness value of all components;
• black color has zero or close to zero values ​​of all three components;
• basic colors should have a higher value of the corresponding component compared to the others: for red - R, for green - G, for blue - B;
• mixtures of two basic colors should have a greater value of two components compared to the third: for yellow - R and G greater than B, for purple - R and B greater than G, for cyan - G and B greater than R.

In Fig. Figure 3 shows an image with two color samples marked on it. In the palette Info(Info) displays the numerical values ​​of the three components R, G and B of the established samples, from which the color shift can be judged.

Let's try to determine the color shift using established samples. The image contains two color marks. The first is installed on the steps, which should be gray, that is, have equal brightness values ​​of the three basic colors. And in the example we have a predominance of red (R = 110) with a reduced content of green (G = 81) and blue (B = 84). The second mark is placed on a lilac leaf, which should be green - this means that the second component of G should be larger than the others. And in the example we have a lack of blue and green colors: R = 140, G = 139, B = 96. Therefore, the purpose of the correction is to reduce the content of red.

Once the goal is set, actions must be determined to remove the color shift. A color wheel diagram will help you with this.

Color wheel diagram

The main principle of color correction is to adjust the color balance, which is based on a color wheel showing the relationship between the primary colors of two models - RGB and CMYK (Fig. 4).

For color correction, use the following three rules derived from this scheme:

1. Decreasing the content of one color increases the content of the opposite and vice versa. Opposite colors are colors that are connected by a line passing through the center of the circle. For example, by increasing the content of red, we decrease the content of blue, and vice versa.

3. To increase content a certain color, it is necessary to reduce the content of colors adjacent to the opposite one, and vice versa. For example, to increase the red color, the content of green and blue should be reduced.

In Fig. 3, we determined the goal of correction - reducing the content of red color. Based on the above rules, you should act like this:

• according to rule 1, you need to increase the content of the opposite blue color;
• according to rule 2, the content of neighboring yellow and purple colors should be reduced;
• according to rule 3, it is necessary to increase the content of green and blue colors adjacent to the opposite one.

When the goal of correction is determined and the actions to achieve it are clear, it is necessary to select the right tool. Also, during the correction, it is better to try to remove the shift according to two or three rules, so that you can compare the results obtained and choose the best one.

Channel Tone Correction

Channel tone correction is carried out in dialog boxes Levels(Levels) and Curves(Curves). Unlike tone correction, you must first determine the color channel in which the tone correction will be performed. And the further technology is the same as that of tone correction.

a

b

Rice. 5. Examples: a - increasing the red color content in the Levels (left) and Curves (right) dialog boxes;
b — reducing the red color content in the Levels (left) and Curves (right) dialog boxes

1. To add color, you need to lighten the image in the selected channel (Fig. 5 A):

2. To reduce color, darken the image in the selected channel (Fig. 5 b):

Levels

In the example in Fig. 3 it is necessary to reduce the content of red color.

1. If in the window Levels(Levels) from the list Channel(Channel) select item Red(Red) and try to move the gray slider to the right to reduce the red content, then the color marks will show a decrease in the red component, but the image will become darker. This is due to the additivity of the RGB model. This means that the shift needs to be addressed differently.

2. According to rule 3 of the color wheel, the content of the green and blue components should be increased. Alternately selecting the blue and green channels, move the gray tones slider to the left. At the time of correction, pay attention to the change in the values ​​of both samples in the palette Info(Info). For the first test (step), it is necessary to achieve equal values ​​of all three components (R, G, B). And for the second sample (lilac leaf), the value of green color (G) should exceed the values ​​of other colors.

3. During the correction, the value of both samples changed. The first sample indicates the ideal gray color: R = 110, G = 110, B = 110. In the second sample, the green color content is greater than the other components: R = 140, G = 164, B = 122.

In Fig. Figure 6 shows an example of correction with changed color sample readings.

Using the Dialog Box Curves

Using the dialog box Curves(Curves), remove the color shift in the image of St. Michael's Castle.

1. Let’s place two marks on the images of the cloud and the gray wall (Fig. 7). The cloud should be white, but we have an underestimated content in the green and red components: R = 190, G = 189, B = 255. The second sample should have equal values ​​of all three basic colors, but in fact there is a predominance of blue: R = 135, G = 137, B = 180. The purpose of the correction is to reduce the blue content.

2. According to the rule of 3 of the color wheel, it is necessary to increase the content of red and green colors. To do this, the correction straight line should be converted into a convex curve in the corresponding channels.

Rice. 7. The original image of the Mikhailovsky Castle with color marks applied to it

3. As a result of the correction, we have the following sample readings. First sample: R = 239, G = 237, B = 255. Second sample: R = 179, G = 177, B = 180 (Fig. 8). The first sample indicates a slight predominance of blue, which is quite normal for the color of a cloud. The second sample has almost equal values ​​of all three components. So the result of the correction can be considered satisfactory.

Using the Dialog Box Color balance

Dialog window Color balance(Color Balance) allows you to perform color correction in different tonal ranges of the image independently of each other. The areas of influence of the tool are set by three switches: Shadows(Shadows) Midtones(Midtones) and Sveta(Highlights). The correction is carried out by moving the slider towards one of a pair of opposite colors on the color wheel: cyan or red, purple or green, yellow or blue (Fig. 9).

This dialog box is called by the command Image(Image) -> Correction(Adjustments) -> Color balance(Color Balance) or the keyboard equivalent of Ctrl+B (in Mac OS - Command+B).

Using the dialog box Color balance(Color Balance), let’s color-balance the image of the Kazan Cathedral shown in Fig. 10.

1. We will install two samples: the first in the area of ​​​​the base of the dome, the second in the shadow between the columns. Both samples correspond to the gray color, only the first is lighter than the second. Both indicate excess blue color:
R = 168, G = 169, B = 190 (first sample) and R = 19,
G = 19, B = 45 (second sample).

2. According to rule 1 of the color wheel, to reduce the content of blue color, you should increase the content of the opposite color yellow color.

3. In the window Color balance(Color Balance) check the box Midtones(Midtones) and move the slider towards the yellow color. Let's adjust the color in the same way by first checking the box Sveta(Highlights) and then Shadows(Shadows).

4. During the correction, we obtained approximately equal ratios between the components in both color tests: R = 188, G = 188, B = 189 (first sample) and
R = 26, G = 26, B = 26 (second sample) - fig. eleven.

Changing the color balance using the command Photo filter

Dialog window Photo filter(Photo Filter) (Fig. 12) imitates a technique where a color filter is placed on the camera lens, changing the color balance and color temperature of the image. You can select a filter from a list of ready-made ones, or you can select it yourself, depending on the specific image.

Remove excess green color from a photo of children using Photo filter(Photo Filter).

1. Excess green color is visible to the naked eye, and this is also indicated by the image histogram (Fig. 13).
2. According to Rule 1 of the color wheel, to remove green, you need to increase the amount of purple.
3. Select a purple color from the list and set the filter density to 65%. Notice the changes that the histogram shows. Now the green graph does not move out of the stack of other graphs (Fig. 14).

Attention! The magenta color selected from the list of ready-made filters does not correspond to the magenta color of the RGB model with the maximum brightness of the red and blue components: R = 255, G = 0, B = 255. It has a different shade with an admixture of green: R = 227, G = 24, B = 227. Therefore, when using pure magenta, the filter density value may be different.

4. As can be seen in Fig. 14, the image darkened during correction. Let's lighten it a little and increase the saturation (Fig. 15).

This article provides a far from complete list of color correction tools. Behind the scenes there are automatic and semi-automatic correction, tools for online photo processing and correction in Lab mode. They will be discussed in one of the upcoming issues of the journal.

In this Photoshop tutorial, I'll show you how to cut out complex objects from the background using the Channels palette. (Channels) and the calculation function (Calculations). This method of extracting objects from the background is very convenient because it saves a lot of time, unlike using other selection tools.

Let's look at the effect of this technique in practice and see for ourselves.

Why use channels to mask an image?

IN Photoshop program there are different selection tools: magnetic lasso, magic wand, quick mask, etc... why use channels for this purpose? The point is that using channels will save a lot of your time and make your work more accurate.

Most users of the program who retouch photos and process their own photos and photos of friends find it difficult to understand that the Channels palette can be used to create a Layer Mask (Layer Mask) because they just don't know what it is.

I won't explain what channels are, that's a topic for a separate lesson, but let me remind you that channels divide an image into three different brightness levels, corresponding to the three colors red, green and blue (Red, Green, Blue), which constitutes RGB mode. Let's look at the action of channels using the example of this lesson.

This is the result of the selection we should get:

Step 1.

Open the image of a tree branch on a blue background from the lesson resources in Photoshop and duplicate it (Ctrl+J). Rename the duplicated layer "Branches" layer. On this layer we will create a Layer Mask (Layer Mask) to hide the blue background on it by cutting out a branch from it.

Step 2.

You know you could use the Pen to highlight branches from the background? (Pen Tool) or Magic Wand (Magic Wand), but this selection technique will take a lot of time, effort and will lead to the formation of uneven edges of the object. Using Quick Mask (Quick Mask) also quite a troublesome task in our case.

Since we have a uniform background, it is appropriate to use the selection method using alpha channels. Even if you don’t understand anything yet, just follow all the steps with me in order and everything will become clear.

On the "Branches" layer, go to the "Channels" palette (Channels), which is located next to the Layers panel (Layers) by clicking on its tab.

Step 3.

You see in the palette that opens three layers of channels for each of the three primary colors of the image: red, green and blue (Red, Green, Blue), which are called alpha channels. The top layer of the "RGB" channel is a combination of all three alpha channels at once.

For high-quality selection of an object, we need a channel that will have the best contrast between the background and the branch. Go through all three channels one by one and select the one with the most contrast. In our case the best option will be blue (Blue) channel because the contrast here is stronger between the background and the branch than in the other two.

Click on the blue channel, activating it.

Step 4.

With the blue channel selected, go to the Image-Calculations menu (Image>Calculations). This function will help us mix the alpha channels and show the preliminary result. Using the Calculation tool (Calculations) we will create a new alpha channel based on the three existing ones. Complete the settings from the screenshot below:

Let me explain what happens to our new channel with these settings. For source 1 (Source 1) The blue channel of the background layer and its duplicate is taken as source 2 (Source 2) Only the blue channel of the “Branches” layer is accepted, then they are mixed using the Multiply mode (Multiply) between themselves. The result of blending is a more contrasting image, which is the new alpha channel (Alpha 1). We are faced with the task of creating a sharp contrast of black and white colors in the image, in otherwise, we will not be able to efficiently cut out the branch from the background and hide the background on the layer mask.

Step 5.

In the previous step we increased the contrast of the image, but this is not enough for our purpose. Let's increase it even more. On the "Alpha 1" channel layer (Alpha 1) again go to the Image-Calculations menu (Image>Calculations). This time we change the channel mixing mode to Overlay (Overlay) and perform the remaining settings of the option from the screenshot.

The resulting blending resulted in another "Alpha 2" alpha channel, in which the light areas became lighter and the dark areas darker, increasing the contrast between the background and the subject.

Step 6.

Even after two calculations, we cannot select the branch because the sky has gray areas and we only need black and white colors.

On the “Alpha 2” channel layer, go to the menu Image-Adjustments-Levels (Image-Adjustment-Levels) or press the combination Ctrl+L. Now we will get rid of the gray tint of the background, making it completely white. In the Levels dialog box, move the white Highlights slider (Highlight) left to 167 until the background is white.

Then, move the black Shadows slider a little to the right (Shadows) to 13, making a clear contrast between black and white.

Step 7

Now we have done everything we set out to do and are ready to select the branch and create a layer mask. Since we need to hide the background on the mask, and not the branch, we invert the colors in the image by pressing Ctrl+I on the keyboard. Now, load the branch selection by pressing Ctrl and clicking on the “Alpha 2” channel icon. A running stroke appears around the white branch, indicating that the selection has been activated.

Step 8

With the Alpha 2 channel still selected, click on the RGB channel to return the color to the image and return to the Layers palette. (Layers).

Step 9

All that remains is to hide the unnecessary background on the mask. To do this, go to the menu Layers - Layer Mask - Show Selection (Layer>Layer Mask>Reveal Selection) or click on the Layer Mask icon (circle in square) at the bottom of the layers panel. Once you create the mask, your background will be hidden on it. To see the result of background removal, turn off the visibility of the background layer by clicking on the eye of its thumbnail.

If you enlarge the image (Ctrl+), you will see how accurately the branches are selected in this way.

Step 10

The masking is done, and now we need to check the quality of the selection against a uniform background. Create a layer (Ctrl+Shift+N) above the background layer and fill (Edit-Fill or Shift+F5) color it in any color to see errors in the image.

Problem with tint.

If you want to use this image on another background or in a collage, then there is a problem with the presence of an unnecessary blue tint on the branches, which must be eliminated. To give you an idea of ​​what I'm talking about, I've chosen a different stock image for the background.

Removing unwanted tint in Photoshop.

There is an easy way to remove unwanted tint from an object using Hue/Saturation adjustments. (Hue/Saturation) and decreasing the layer mask. If the shade is located at the edges of the object, it will be enough to reduce the mask by 1 px and the problem will be solved.

Since we already have a mask on the "Branches" layer, we can't create another one. But there is a way out. We'll create a mask on the group layer later.
Select the “Branches” layer and place it in a group by pressing Ctrl+G on your keyboard.

Load a selection mask on the "Branches" layer (Ctrl+click on the mask icon), then go to menu Select-Modify-Compress (Select>Modify>Contract), set the compression to 1 px. Next, create a mask on the layer with the group and to do this, go to the menu Layers-Layer Mask-Show Selection (Layer>Layer Mask>Reveal Selection). In the Layers panel, you can see your new mask on the layer with the group.

If your image is low resolution, 1px compression may remove fine detail from the subject. In this case, you need another way to remove the unwanted tint.

After we cut down our mask, the blue tint is still present on the branches. Let's try to remove it using the Hue/Saturation correction. (Hue/Saturation), discoloring blue and cyan tones or changing this shade to another.

Create a Hue/Saturation adjustment layer (Hue/Saturation), by clicking the black and white circle at the bottom of the layers panel and selecting the one we need from the list of layers. In the options dialog box, select blue tones (Blues) and desaturate them by moving the Saturation slider (Saturation) at -94. Then, select blue (cyans) tones and also bleach them. After completing the settings, create a Clipping Mask on the adjustment layer for the “Branches” layer (Alt+click between interacting layers).

Finally, take a look at the result before and after applying compression to the mask and creating an adjustment layer.

I hope you found something new for yourself in this lesson and will now successfully use this method of selecting complex objects.

In this tutorial, we'll learn how to increase the contrast and color saturation of an image in unique and creative ways by applying blend modes to individual color channels in a photo. If you use Photoshop, you know that we usually select blend modes in the Layers palette because they are most often used to change how a layer interacts or blends with the layer(s) below. Here we will go further and learn how to apply the same blending modes not to the entire layer, but only to one of the individual ones RGB (Red, Green And Blue) channels that Photoshop uses to create a color image. How will we apply blend modes to color channels? You'll see that it's actually very easy thanks to Photoshop's Apply Image(External channel).

Working with channels in Photoshop is too broad a topic to cover in this article. I'll touch on them briefly here and we'll move on, but I encourage you to read the individual articles on RGB channels if you're less familiar with them.
For this tutorial I'll be using Photoshop CS6, but earlier versions will work as well. Here is the original photo:

The first thing we must do before starting work is to create a copy of our image. If you look at the screenshot of my Layers palette, you'll see that the original photo is on the Background layer:

Duplicate the background layer using the keyboard shortcut Ctrl+Alt+J. Photoshop will not only create a copy of the layer, but also open a dialog box New Layer(New Layer) where we can specify the name of the layer before it is added. Name the layer, for example “ Apply Image”, then press OK:

If we look at the Layers palette now, we can see a copy of the image located on a layer called “ Apply Image”, right above the background layer. It's important to give your layers meaningful names, otherwise we'll end up with a bunch of layers with Photoshop-generated names like “ Layer 1″, which tell us nothing about the contents of the layer:

As I mentioned at the beginning of the article, we most often select blending modes in the Layers palette, since they usually apply to the entire layer. Drop-down menu Blend Mode(Blend Mode) is on the left top corner layers palette. For example, I'll change the layer's blend mode Apply Image With Normal(Regular) on Soft Light(Soft light):

We changed how the layer Apply Image interacts with the background layer below. Soft Light(Soft light) – one of contrast blending modes, because it increases the overall contrast of the image, as we can see in our photo. Color saturation has also been slightly increased:

I'll change the blend mode back to Normal(Normal) to return it to its original state:

So, if changing the blend mode on the layers palette is so important for blending layers, where are those individual color channels, and how do we use blending modes with them? Regarding the first part of the question, if you look at top part layers palette, you will see that it is grouped with two other panels - Channels(Channels) and Paths(Outlines) - and each panel is presented as a tab at the top. Click on the tab Channels(Channels):

You will switch to the panel Channels(Channels) where you can see individual Red, Green And Blue the color channels that make up your image. RGB The channel above isn't really even a channel, but the result of combining the Red, Green, and Blue channels, or in other words, what you see as a full-color version of our image (each color in the image is a specific combination of red, green, and blue):

We can select an individual color channel simply by clicking on it. I'll select the Red channel:

Selecting the Red channel temporarily disables the Green and Blue channels and allows us to see only this in the window. Photoshop renders color channels as grayscale images, and each channel looks different. Below you can see what my red channel looks like in the document window. If you compare this black and white version to the original full color photo, you will notice that areas that contained a lot of red in the full color version are lighter in the black and white version, while areas that contained little or no red appear darker :

Next, I'll select the Green Channel from the panel. Channels(Channels), which will temporarily disable the Red and Blue channels:

Currently, the Green channel appears in the document window as a grayscale image. Please note that it is very different from the Red Channel. Again, if you compare the channel to the original image, you will notice that areas that contain a lot of green will be lighter in it, and vice versa:

Finally, I'll select from the panel Channels(Channels) only the Blue channel, which will disable the Red and Green:

And now we see in the document window the Blue channel, which again gives us a picture that is very different from the other two channels. This time, the more blue an area contained in the original version of the image, the lighter it will be in the grayscale version, while areas containing little blue will be darker. When we select an individual color channel in the dialog box Apply Image(Outer channel), which we'll get into shortly, we need to remember that these are actually different grayscale versions of the image, with different brightness values, than the image we see initially:

To switch back to the full color version of the image, click on the channel RGB at the top of the panel Channels(Channels). This will turn all three channels back on:

So we're back to viewing the full color version of the image again:

A full-color version of the image appears again in the document window.

Apply Image function (External channel)

Now that we know where the color channels can be found and what each of them looks like as a grayscale image, let's get back to the second part of the question - how can we apply blend modes to them? You may have noticed that there is no option Blend Mode(Blend Mode) at the top of the panel Channels(Channels), as was the case with the panel Layers(Layers). In fact, we don't even need open panel Channels(Channels), so let's switch back to the Layers palette by clicking on its tab:

To apply a blend mode to a single color channel, we use the function Apply Image(External channel). Go to menu Image(Image) and select Apply Image(External channel):

A dialog box will open Apply Image(External channel). It might be a little intimidating if you've never used it before, but what we'll be doing here is quite simple. In fact, we will only use two options − Channel(Channel) and Blending(Overlay):

In the drop down menu Channel(Channel) we select the color channel we want to use. Selected by default RGB, this, if you remember, is the same composite RGB channel that we saw at the top of the panel Channels(Channels) (it mixes the Red, Green and Blue channels to create a full color image). In the drop down menu Blending(Overlay) we select the blending mode we want to use. If we leave the option Channel(Channel) set to RGB and simply select the blend mode from the menu Blending(Overlay), we will get the same result as if we set the blend mode in the layers palette. For example, I'll select the blending mode Soft Light(Soft light) for option Blending(Overlay) ( Channel(Channel) set to RGB):

And here we can see that the image is no different from what it was when I selected the blend mode Soft Light(Soft Light) in the Layers palette earlier. We get the same contrast and saturation boost:

Now the fun begins. For option Channel(Channel), we can choose from three individual color channels. I'll leave the Blending option set to Soft Light(Soft light), but I'll change the meaning Channel(Channel) with RGB on Red(Red), so I will overlay only the red channel:

This time, we get a different result. We can still see the contrast boost with the blend mode. Soft Light(Soft Light), but since it only overlays the halftone image of the Red Channel that we saw earlier, we get a different effect. The woman's skin looks much lighter than before. The same can be seen in her hair, along with the red T-shirt, as well as the red, orange and yellow pieces of her jacket. Basically, anything that contains warm red tones is now lighter. Likewise, cool areas that don't contain red, such as the blue and green parts of the jacket, appear darker than before:

Let's see what happens if I select the Green channel (the blend mode is still Soft Light):

With the green channel selected, we get a different effect. Now the areas containing green color became lighter, and red and blue became darker. The most obvious differences are in the skin, which is now darker and more detailed than with the red channel:

And finally, the Blue channel:

Here we see a third variation of the effect, blue areas appear light and warm tones appear dark. We wouldn't be able to achieve these effects (at least not without a lot of time and extra effort) if it weren't for the individual color channels that Photoshop gives us control through the Apply Image(External channel):

Of course, we are not limited to using just the blend mode. Soft Light(Soft light) when working with channels. We can use any blending mode from those that we usually select in the layers palette. I'll leave the option Channel(Channel) set to Blue(Blue) and change Blending(Overlay) with Soft Light(Soft light) on Overlay(Overlap):

Like Soft Light, Overlay(Overlay) – a contrast blending mode, but with a stronger, more intense result:

This is what the same mode gives us Overlay if we change the Channel value from Blue to Green:

Well, this is what the Red channel looks like in Overlay mode:

Red channel in mode Overlay(Overlay) looks too contrasty and saturated, but if necessary we can reduce the effect by simply lowering the opacity of the overlay. You will find the Opacity option right below the option Blending(Overlay), it works exactly the same as the opacity slider on the layers palette. The default value is 100%. I'll lower it to 60%:

By lowering the opacity, we brought back some detail in the shadows and highlights:

If at any point you want to compare the result with the original version of the image, simply uncheck Preview(View) on the right side of the dialog box. This will hide the effect and show you the original photo in the document window. Check the Preview checkbox again to display the effect again:

Over time, especially when working with photos of people, you will notice that the blend modes Soft Light(Soft light) and Overlay(Overlap) will give the best results, but there are others useful modes, such as Screen(Screen) and Multiply(Multiplication). Screen(Screen) brightens the image while Multiply(Multiplication) - darkens. Try using them with each of the three color channels to match best option, then adjust the result by decreasing or increasing the opacity of the overlay. For example, in the image below I have selected the green channel, blending mode Multiply(Multiply), and lowered the opacity to 40%:

The result is a darker, more detailed image:

When you are happy with the result, click OK to close the window Apply Image(External channel). Now you can compare what you got with the original image by hiding the layer Apply Image on the layers panel:

So now you can apply blend modes to individual color channels using the Apply Image(External channel) in Photoshop! We wish you creative success, stay with us and follow the updates on the site.

Layers on this moment are the most powerful tool for working with photographic images. When you edit a photo, whether in Photoshop or Paint Shop Pro, you can't do without multiple layers of the image. Sometimes more complex layers are used - custom layers, effects and masks based on layers. Layers are everywhere. And the question involuntarily arises: how did users manage without them before?

The answer is simple - using channels. You can easily spot a Photoshop veteran by the fact that he uses the Channels palette as often as the Layers palette. But this should not be regarded as old-fashioned. Rather, this is evidence of professionalism, since the use of channels opens up wide opportunities.

So, what is a "channel". The standard definition - "a two-dimensional array of information, usually 8-bit" - is unlikely to clarify the situation. Therefore, let's take a closer look at the two main types of channels - color information and alpha channel - from the point of view of their practical application.

Color channels

The easiest way to see color channels in action is to use Photoshop's Channels panel. Open regular 24-bit photographic image in RGB. In the channel palette you will see 4 layers, each with its own icon: RGB, Red, Green and Blue. If you click on RGB, you will see the image that you usually see - composite and full color. For each of the other channels, a grayscale version of it is displayed. The keyboard shortcuts Ctrl+1, 2, 3 allow you to view each of the color channels separately, and Ctrl+~ - a regular composite image.

The full spectrum of RGB colors is created from the red, green, and blue color channels, where they are represented in a gray scale.

To understand what information is displayed in the Layers palette, it is worth observing the channels when working with a test image of a rainbow gradient (in other words, an image of all the colors of the rainbow). In a grayscale image that represents a channel, each of the pixels can represent one of 256 values. You will see that the red part of the rainbow when viewed in the red channel will be white. The yellow stripe of the rainbow will be white in the red and green channels, but will be black, i.e. absent, in blue. In fact, the rainbow image shows that the full spectrum of 16 million colors can be obtained by combining values ​​from 0 to 255 for red, green and blue (256 x 256 x 256). In our image, the yellow stripe corresponds to large values ​​of the red and green channels and 0 to the blue channel. Photoshop works with an image not by individual pixels, but by channels. In this case, three eight-bit layers - red, green, blue - are superimposed, and we get the final image.

Of course, not all images are created in RGB, but this is not a problem since the channels are very easy to customize for different needs. We can convert our image to a different mode using the Image > Mode command. For Bitmap and Grayscale there is only one channel, which represents 256 values ​​from black to white. In Lab mode there are three channels: A (color value between green and red), B (between blue and yellow) and L (luminance value). Separating color from its brightness can be very useful. Select the luminance channel and convert it to a grayscale image. The result will be much better than if you converted the scanned RGB image.

The Lab color model can be very useful. But still the most important, after RGB, is CMYK. This model is print oriented. Convert the rainbow image to CMYK. The first thing that catches your eye is the sudden change in some colors. The fact is that CMYK does not support many pure RGB colors. The second difference is that four color channels have appeared: Cyan, Magenta, Yellow, Black (turquoise, purple, yellow, black). Look at the yellow stripe of the rainbow in the yellow channel - you will see that it is represented by black.

In the CMYK palette, the channels represent the ink colors used in multi-layer, four-color printing.

This mode differs significantly from RGB because Photoshop has to work with ink colors that combine with each other in a subtractive manner (summing all the colors together results in black). In RGB, on the contrary, the principle of addition is applied, and the superposition of all colors will give white. But, by and large, all color models are similar to each other. And Photoshop can recreate the full gamut of any of them with no more than four eight-bit channels, each of which has no more than 256 values ​​(or 65,536 values ​​if you really want to and go to Image > Mode and choose a 16-bit channel).

Channels are the primary tool for working with color, so it's worth looking at the Channels panel occasionally to learn how Photoshop creates different colors. This is especially true when working with a CMYK model that is designed for printing. When working with RGB, what you see on the monitor does not always correspond to what will come out when printed. Although the values ​​of each channel reflect this. Moreover, if you want to replicate an image, you need to consider factors such as dot blurring and gray component replacement. This means that accuracy in working with color channels is very important, since color separation subsequently depends on it.

Channels are useful for more than just CMYK printing. You will find their use in RGB as well. Let's say you can edit not the entire image, but a separate channel. When adjusting color, it is worth looking at individual channels to identify defects in the scanned image. For example, if you find a blurred or misaligned area in the red channel, you can correct this issue by using a sharpening filter or adjusting the levels.

And in order to create some special effect, you can apply an artistic filter to one of the channels. In any case, you can immediately see the result of your actions on the composite image - just make the RGB channel visible. Just remember to select the desired color channel again later if you continue to work on it.

The channel merging function provides ample opportunities. For example, using the Apply Image command, you can overlay information from any channel of another image of the same dimensions onto the RGB channel of our image and customize this effect by changing the opacity and color mixing mode. The Calculation command allows you to select two layers and create a new layer, selection, or document. If you use the subtractive color mixing mode, you will select areas of the image that have changed compared to the original.

With the introduction of the Channel Mixer command and custom layers in Photoshop 5, the ability to combine channels has expanded. The channel mixer feature is used only for individual images and allows you to customize each channel by adding information from other channels. With its help, you can create special effects, swap channels, and correct color defects in scanned images. It's also ideal for creating "tinted" and high-quality halftone images (and you can control how each color channel is converted to grayscale).

Channel blending is very helpful when converting color images to grayscale.

Adjusting colors using channels will undoubtedly come in handy. But the most common way to use them is to work with secretions. Often, image elements are much clearer in a single channel than in a composite image. For example, selecting a bear in the bear.psd training file is much easier in the blue channel.

Alpha channels

Once you've created a selection, you'll probably want to save it. This can be done in the alpha channel. To do this, use the “Selection > Save Selection” command or the “Save Selection as Channel” icon in the Channels panel. After this it appears new channel, in which selected pixels are displayed in white, unselected pixels in black, and blurred pixels at the border of the selection in shades of gray.

By double-clicking on the channel icon, you can rename it and set the color it will appear in the composite image. You can work with the alpha channel like any other, for example, adjust levels or apply filters. Moreover, you can edit this channel using a brush - for example, paint over defects. When you are satisfied with the result, use it. To do this, convert the alpha channel into a selection using the Selection>Load Selection command, the Load Channel as Selection icon in the Channels panel, or simply by Ctrl-clicking on the channel icon. If you have multiple alpha channels, you can create more complex selections. For example, you have selected and converted an image of a person and bushes in the background to alpha channels. You can select only the bushes if you select them and load the channel as a selection using the "subtract" option. Keyboard shortcuts will speed up this procedure: click on the icon while holding Shift+Ctrl - this will add a new area to the selection, Alt+Ctrl+click - subtract from the selection, Alt+Shift+Ctrl+click - will give the intersection of two selections.

Alpha channels can be saved and loaded as selections.

Once you've converted the alpha channel to a selection, it will immediately appear on screen and you can transform it, copy it to a new layer, or edit it in any way. The flickering dotted line along the edge of the selection does not very accurately indicate its boundaries - you need to take this into account. It only displays alpha channel pixels that have gray intensity below 50%. For most selections this makes little practical difference. But the advantage of the alpha channel is that it allows you to work with 256 levels and thus create complex, variable transparency masks.

How can you use these masks? Here are some examples (it's actually easier to do than it looks when you read it). Click on the "Create New Channel" icon and create an empty alpha channel, then apply a gradient to it and select the composite channel again. Ctrl-click on the alpha channel icon. Now apply any artistic filter to the created selection. Its effect will depend on the halftone values ​​of the gradient mask. It will seem that the photograph is gradually turning, for example, into a painting.

The alpha channel can be used as a variable opacity mask.

Another example: create a copy of the alpha channel from a text selection and apply a Gaussian blur to it. After that, subtract the original channel (Alt+Ctrl+click) and you will get a new channel that will show only the noise surrounding the text. Now select this area in the composite or any of the color channels. You can adjust the levels so that the highlight glows, or even remove original text and the glow will appear to be part of the photo.

You have the foundation, now experiment - you can create many effects using shadows, extrusions, vignettes, etc. Surely, if you get serious about this, you will need the ability to save alpha channels with your file. But Photoshop's capabilities in this regard are surprisingly limited - the program supports saving the alpha channel in a tiff file or in Photoshop itself (but the number of channels in a file, including color ones, is limited to 24). This won't pose a problem if you only work in Photoshop. But the presence of an alpha channel can significantly affect the export of a file to one of the standard formats. For example, when saving for the Web, in the dialog box that appears, you can use the alpha channel as a mask to adjust the quality and, accordingly, the size of jpg and gif files. Using the alpha channel, you can draw attention to a certain part of the image that will be of the highest quality. The quality of other areas of the image will deteriorate, but this will significantly reduce the overall file size. Alpha channels are also useful when working with transparency. Create an alpha channel that covers the area that you don't want to be visible in the web version, and Ctrl-click to select that area. Now call the Transparent Image Export Wizard and select gif. The selected area will become transparent in the final file, and you will be able to avoid making irreversible changes to the original. The wizard allows you to export the image to a png file that supports 24-bit color and 8-bit transparency. If you create an alpha channel with radial gradient, then you will receive a vignette. Unfortunately, few programs support 32-bit png files with transparency. The only exception is, perhaps, Director 8.5.

With the ability to add spot colors in CMYK, alpha channels are also useful for high-quality printing. Select the area of ​​the image that you want to print as a color fill, and then use the New Spot color command from the Channels panel menu. When you set a color, click "Custom" and select a color, for example from the Pantone library.

Please note that adding a color fill is not as simple as it seems. The image on the screen, as you know, does not always correspond to what is printed. Ink colors often cannot be represented in RGB and Photoshop simply shows what it thinks the image should look like when printed layer-by-layer. If you save your file in DCS 2.0 format and load the EPS composite image into a professional publishing program, you can color separate the four primary colors and custom color files you created.

The custom color channel allows you to display ink colors that are not included in the CMYK palette.

There's no doubt that channels play a very important role in photo editing, from adjusting colors and highlights to working with images for the web and high-quality prints. But time moves on, and layers often produce better results than channels. Having created a selection, copy it to a separate layer - and you can work with it independently, and, if necessary, return to the original version. The same goes for many other effects that were previously created using channels. Today, custom layers, effects, and layer-based masks do a better job of accomplishing this task.

The canal system was also improved. So, although in many areas channels are gradually being forced out of use, they are unlikely to disappear completely. If only for the simple reason that layers are created based on channels. Strictly speaking, a layer is a set of independent color information channels and alpha channels, one of which controls transparency, and the second acts as a layer mask. The same goes for custom layers, which are essentially alpha channels through which color is adjusted.

Photoshop, like any other photo editor, creates an image on the screen by processing values ​​from the color and transparency channels of the layer. And so, one by one, all the layers from the background to the foreground. Of course, in reality, working with an image is much more complicated - the program calculates color mixing modes, opacity, settings, etc. But it's based on a step-by-step mathematical process. The program does not see the image; it works with the data array of each channel. Any photo editor “thinks” in terms of channel categories. This is a habit worth adopting as a professional user.

In this article we will talk about CMYK channels in Photoshop. What are they for and how to work with them? How are they different from RGB channels? In the last article, we found out that RGB channels reflect the work of the RGB color model in Photoshop. Channels for CMYK perform the same task, but in a different color space. You will learn about what this space is, how to effectively work with CMYK channels and do color correction for printing in this article.

In my opinion, I managed to create the most understandable and simple article, which clearly demonstrates the principles of working with CMYK channels and the basics of CMYK color correction.

• Part 1:

CMYK color model

A lot has been said about the CMYK color model even on Chronofag.ru. Therefore, I hope that everyone is familiar with the theory of the CMYK color model. This article is more practical. Unlike RGB, CMYK displays colors on the screen in a form in which they can be printed in a printing house. Working in CMYK is an essential attribute of a print designer. Therefore, it is especially important for every beginning designer to understand the principles of his work.

The CMYK color space is based on the principles of mixing inks during printing. While RGB channels subtract color information, CMYK channels add color information. With the maximum value of the RGB channels (R 255 G 255 B 255), we get white color (prism effect). In CMYK it’s the other way around: the more information on the channel, the darker the paint. In RGB - zero values ​​- black. Which is logical, there is no information about light, which means there is no light. In CMYK, on ​​the contrary, there are zero channel values ​​(C 0 M 0 Y0 K 0), which means the color is white, since there is no paint.

The principles of nature lie in both color models. Principles of RGB light emission and principles of CMYK light absorption. One problem. Modern printing has failed to display exclusively (in cyan, magenta and yellow). And even if it were possible, spending three layers of 100% paint to extract color is not rational and expensive, given that most of the printing is black text. Therefore, in CMYK there is an additional fourth channel K (black). However, you can read more about the features of the CMYK color model separately.

Principles of interaction of CMYK channels

Only practice is better than theory. You'll find the Channels panel in Windows > Channel. can be done via Image > Mode > CMYK (Image > Mode > CMYK)

It is very simple to understand how an image is constructed in CMYK channels. After all, each of us painted with paints in childhood, mixing different shades with each other. And most of us certainly know that mixing red pigment with yellow produces orange. When blue and yellow are mixed, green comes out. This is why it is so easy to understand the principles of CMYK, because everyone has experience in painting with paints.

CMYK channels work exactly the same. Only if in the gouache palette we had 10 or more paint pigments, in CMYK there are only four. Cyan, Magenta, Yellow and optional black. Black ink should ideally be used only in areas where three CMY inks are not powerful enough to produce a rich black. Therefore, the black channel in CMYK is contrasting.

Moreover, you can easily simulate CMYK channels by creating your own image using masks. In this article we will do such a trick. But first, let's take a simpler example.

Let's take for example a pure yellow channel and a gradient, where on one side we have Y100, and on the other side Y0. On other channels the values ​​are zero.

Add the same purple gradient to it and blend it together. It is logical to assume that on different segments of the gradients we get combinations of different shades of paint. On the gradient we have M100Y100 - red, in the middle M50Y50 - a less saturated red. The less paint, the less saturated the color.

Let's add a blue gradient at the bottom. What happens when you mix cyan and magenta? Dark blue. And different shades of saturation.

Now let's move the blue gradient to the center and get a combination of all three channels, where point 1 is the darkest shade that can be obtained by combining three colors together.

So far, our gradients have been in a row. And the saturation of colors changed in proportion to each other. Now let's expand them and put them on top of each other in a completely random order. I chose Oval as a more convenient shape.

Several things are clear from this picture:

1. The fewer channels, the purer and clearer the color.
2. The more strange compounds there are in the percentage of paints, the more unattractive the color.

Everything is based on these two simple rules. The more ink, the darker the color, the more confusing the digital values, the duller the shade and the greyer/blacker the printed graphics. Let's select several points in our image and take a color proof from them using the Eyedropper Tool.

In sample 1, 2 and 3 we have two channels and pure colors. In sample 4 there is no obvious gain in any channel and the color is dirty.

Mixing paint is the basis of working in CMYK. Novice designers are accustomed to perceiving paint as a color on a monitor. They find it difficult to visualize digital CMYK values ​​as if they were real ink on paper. They choose a color from the Color Picker panel and don't look at the numbers. The result of such work is the same. Pale print, gray colors.

Mixing paint on gradients

Gradients pose particular challenges. It is not always clear to a novice designer by what logic CMYK paints are mixed. When I started working in Corel Draw the gradient results shocked me. I chose a gradient from red to black and got amazing results. Red on one side, black on the other, and muddy in the middle.

But if you understand what happens on CMYK channels, it is much easier to understand this problem. After all, if you know that the red color is a mixture of the Magenta and yellow channels, and the black paint is on a completely different channel, then the problem of the gray transition is obvious. Here's what's happening on the channel.

The loss of red paint from the middle is the source of the problem. If we take a color proof from the center of the gradient, we will get the values ​​C0M50Y50K50. What does a color look like with these values? This is exactly what it looks like.

But if we had drawn the same gradient on paper, there would have been no problems. After all, on an intuitive level, everyone understands what colors need to be mixed to get a gradient from red to black. You need to take red and black paints and mix them together, reducing the pigment of the black paint. Red paint should be present 100% in each section of the gradient.

In our case, the red paint decreases in proportion to the arrival of the black paint. And it’s not surprising that the gradient turns out faded. But if we adjust the values ​​of the black side of the gradient, and add a yellow and magenta channel to it, the gradient will become bright and saturated.

The situation is changing in the same way on the channels. Now we have the Yellow and Magenta channels as solid colors. And black is a decreasing gradient. At each segment, black is mixed with 100% red, resulting in a rich shade.

Simulating CMYK channels

It is quite possible to create your own image using CMYK. And now we will do this with you. Select absolutely any photo and convert it to CMYK (Image > Mode > Cmyk (Image > Mode > CMYK)).

Now let's open the Channels panel and see what's on our channels. It is very easy to imitate the combination of paints. Naturally, you can easily and quickly turn off channels and see how the color is combined, but doing all the work instead of Photoshop is much more clear.

Press CTRL and click on the yellow channel. By doing this you have loaded the selection. That is, a cast of the mask. Selection - half mask. Therefore, switch to the workspace and create a fill layer Layer> New Fill Layer> Solid Color (Layer> New Fill Layer> Color...).

In the Color Picker window that appears, select CMYK values ​​- C0M0Y100K0

Since we had an active selection when creating the layer, the layer was created with an automatic mask. But the mask is filled in backwards. To make it the same as the image, click on the mask icon and press CTRL+I Invert.

We created the first channel manually. Create your layers for the black, blue and purple paint in the same way. Arrange the layers in the same way as in the channels. And of course, don't forget to invert each mask.

Now let's change the blending mode of each layer to Myltiply (Multiply). Myltiply mode simulates the overlay of paints on top of each other. It cuts off light pixels, leaving dark ones. And this is how the paint is mixed. The more paint, the darker the shade. And as a result, we get exactly the same image, made with our own hands.

Manually changing CMYK channels

CMYK channels are very simple. The more pigment, the darker the channel; the less pigment, the lighter it is. You can correct the information on the channel “by hand” using a simple brush. You must understand that the channel's operation is based on a regular mask, which hides some of the light information in some places and reveals it in others.

As an example, I chose a colorful sky, in which the blue channel dominates at the top, yellow in the center, and purple on the sides. But knowing the principles of how channels work, you can easily fix this in reality. basic level. Color correcting channels. It is clear that in order to remove pink, you need to lower the values ​​on the magenta channel.

Make a CTRL+click on the purple channel, and select this channel to work on its mask. Select a large soft-edged brush with white color. Set the brush's invisibility to about 25% and begin to lighten the magenta channel mask, stroke by stroke. You will see the color pink literally fall from the sky.

To remove yellow, CTRL+click on the yellow channel, and again, without losing the selection, reduce its effect.

In no case do I advocate color correcting channels this way. For color grading in Photoshop, you will find more than 100,500 different tools. There is no need to make irreversible changes to the channels. If you think about ways to correct color, then editing channels directly is the worst. But for the purposes of this article, it was important to show that the nature of the channels is a regular mask, which can be adjusted manually by adding color pigments with a brush.

Now that we have dealt with the channels, let's move on to more advanced methods of color correction.

Color correction in CMYK

For example, let's choose the same photo of the sky. Actually, in Photoshop there are 100,500 filters and color corrections to change colors without interfering with the channels. And the whole point of these filters comes down to the fact that they regroup the information on the channel. As an example, I will choose color correction - Chanel Mixer (Channel Mixing).

It is not difficult to guess that this color correction is tailored to change the saturation of channels. In the RGB color space, this filter may seem very complex because the RGB mixing principles are more complex to perceive. But when it comes to paints, everything is much simpler. In the photo we have 4 color areas. I'll take a pipette and sample the canals.

• In sample 1 we have a dark blue color. And we see that dark blue is obtained from the combination of cyan and purple paints.
• Sample 2 shows the color blue. And indeed, apart from this paint, no others are used.
• The purple areas in sample 3 are due to the prevailing purple paint.
• And the dull greenish tint in the middle was created due to the fact that a little yellow paint was mixed into the weak blue channel.

How to achieve exceptionally blue skies? It's simple. Reduce the influence of third-party paints. By reducing the magenta, we will remove the purple colors on the bottom and the dark blue on top. By reducing the yellow, we will remove the greenish tint in the center.

Chanel Mixer color correction makes this easy to achieve. All you have to do is choose desired channel, and reduce its impact. If we compare paint samples after applying the channels, we get a completely different result. Now all samples are dominated by blue paint.

Face color correction in CMYK using curves

Everything is simple with the sky. Removed unnecessary paints and that's it. But what if the photograph is more complex? If the channels contain not just shades, but details? Simply remove shades when all the details of the sky are in the blue channel. But what about more complex photography? If you know how paints interact, then it’s much easier to correct color correction. And the better the print result will be.

In the photo below we have the actor's face. The purple muzzle and overall blue color immediately catches the eye. But if we think in CMYK paints, then we understand that blue is due to the prevalence of blue paint, and redness is due to excess purple. Think gradients that intersect randomly and muddy grey-brown-crimson shades. So in this photo there are dirty mixtures that we need to clean.

If we take samples from the actor’s skin, we will get the same picture. Light gray-brown-raspberry and dark gray-brown-raspberry. And CMYK channels testify to the reasons. 37% blue 45% purple 20% yellow. So the result is almost gray, with a purple tint due to the purple channel.

It’s easy to correct the situation if you understand the principle by which paints are mixed on paper. For this adjustment, I'll select Curves and create a new color adjustment layer. First of all, let's get rid of the exposure to blue paint. It is she who creates blue and grays colors.

And the photograph began to sparkle with purple colors. The reasons for this are clear. The blue paint held back the purple. Now she is gone and the photograph is filled with a purple channel. Let's switch to Magenta and change its curve as well.

There is now a distinct yellow tint to the photo. And this is also understandable. After all, we removed the cyan and magenta channels, leaving the yellow one, which we will correct after them.

After which I slightly increased the contrast of the black paint. And here is the final result.

You can compare the results of the composition of paints on the face and glove yourself. Instead of 37% blue, we have 9% Instead of 50% purple, we have 28%. And the skin color stopped being purple.

I really hope that this article was extremely useful for you and now such an expression as CMYK does not scare you. Well, I, in turn, will continue to write articles on the topic of channels in Photoshop. Next up is Lab Color. Have a nice day and see you for the next articles.

How to convert rgb to cmyk, Convert rgb to cmyk, Convert rgb to cmyk, How to convert rgb to cmyk in Photoshop, Convert color from rgb to cmyk, How to convert rgb to cmyk in Corel, Convert an image from rgb to cmyk, How to convert rgb to cmyk in coreldraw, How to convert rgb to cmyk in Illustrator, How to convert rgb to cmyk in illustrator, Photoshop cmyk, rgb conversion.