Photoshop CS5 Bible

Photoshop Fundamentals: Working in Different Color Modes

Adapted from Photoshop CS5 Bible (Wiley Publishing)

By Lisa DaNae Dayley, Brad Dayley




Photoshop provides several color modes that help when working with images. Ultimately, color is just different intensities of light at different frequencies. However, that data needs to be translated into a quantifiable form that can be understood by Photoshop, the monitor, printers, and ultimately you.

What follows gives you an overview of the different color modes available for use in Photoshop and why to use them. It also helps you understand the bits/channel settings for the image modes.

Understanding the Different Color Modes

A color model is simply a method to translate the light captured in an image into a digital form that the computer and other devices can understand. Each color model breaks the light into one or more channels and then assigns an intensity level of each channel for each pixel in the image. Photoshop provides several color modes that match the most common color models.

Depending on what you are doing with an image, you want to use a specific color mode that provides the best management of the color. To set the color mode of an image, select Image > Mode and then select the mode you want to use from the main menu in Photoshop. The following sections discuss each of the color modes and what they are for.

Bitmap

The bitmap color mode contains only one channel with only two possible levels, 0 and 255. The translation is that a bitmap is a black and white image without color and even without shades of gray. In essence, the image becomes a series of black dots on a white background. This may not sound very useful. However, there are several good uses for these types of images.

The most common use of the bitmap mode is outputting the image to a black and white laser printer. Laser printers create images as a series of black dots on the page. So working with the image as a bitmap lets you make changes to a version that appears exactly how it will when it is printed by the laser printer.

Tip: Bitmap images that are printed on low-resolution laser printers often end up darker than you would expect. Be sure to lighten the image before printing it to a low-resolution laser printer.

The bitmap dialog box shown below lets you set the resolution and method to use when creating the bitmap from the image.



The options for the bitmap mode in Photoshop.

You should set the resolution to the same resolution that you will be using to output the image to the printer. The following methods can be selected from the drop-down list in the dialog box:

  • 50% Threshold: This sets every pixel in the image that is more that 50% gray to black and every pixel that is 50% gray or lighter to white. This is by far the simplest pattern; however, the end effect is very choppy, and the image typically doesn’t look very good unless you are trying for a special effect, for which it would be better to use the Threshold tool in the Image > Adjustment menu.
  • Pattern Dither: The pattern dither uses a pattern to mix black and white pixels together, which results in the appearance of different shades of gray. The problem with this method is that the pattern shows up in the image, so the effect is not very smooth.
  • Diffusion Dither: This uses an error-diffusion method of converting an image into a series of dithered pixels that are less structured than the pattern dither. The diffusion dither method produces an image similar to a mezzotint, which often is the best option for printing on low-resolution laser printers.
  • Halftone Screen: This uses a series of dots of varying sizes and spacing that trick the eye into believing it is seeing a continuous tone. When you select the halftone screen option, an additional dialog box pops up that enables you to select the frequency, angle, and shape of the halftone dots. Typically, the best shape to use is the round shape because it is closest to what the printer generates. The frequency depends on the resolution of the printer. The higher the frequency, the better resolution you get in the image; however, if you set the frequency too high, the patterns overlap when printed and the results do not look good.
  • Custom Pattern: This allows you to select a custom pattern, either one of those included with Photoshop or one of your own. The custom pattern can be used with similar results to the pattern dither. Typically, this option is used only if you have a specific pattern that you want to show up in the image.

Grayscale

The grayscale mode contains only one channel, but unlike the bitmap mode it can have intensity levels from 0 to 255. The grayscale mode is useful if you are outputting the image to a noncolor printer.

Tip: You can create a grayscale image from a single RGB channel by selecting that channel in the Channels panel and converting the image to grayscale. Only the selected channel is converted to grayscale. This can be useful for special effects as well as utilizing the detail of a specific channel.

Another advantage of grayscale is that viewing and adjusting the image in grayscale reduces the overhead of dealing with three color channels. Photoshop is much faster at performing complex operations on grayscale images than on multichannel images. Having a single channel also makes it easier to make adjustments to an image because you only need to worry about adjusting that channel.

Caution: To change an image from color to grayscale, Photoshop takes a composite intensity for all three channels and reduces it to the single grayscale channel. This results in a loss of the original color channels. Therefore, make sure you have a backup copy of the file before you save it again.

Duotone

The duotone mode uses on contrasting color of ink over another to produce highlights and middle tones in a black and white image. Duotones typically are used to prepare images for printing. Using the Duotones Options dialog box, you can add one, two, three, or four inks to create a monotone, duotone, tritone, or quadtone image in Photoshop.

Use the following steps to configure the monotone, duotone, tritone, or quadtone options from the Duotone Options dialog box shown below:





The options for the Duotone color mode and the Overprint Colors dialog box in Photoshop.

Note: The Duotone color mode option is available only for grayscale images. If you are using a color image you need to convert it to grayscale before changing to duotone.

  • Select the type of tone from the Type menu.
  • Click the blank swatch for each ink you need to specify.
  • Select the color to use for that tone.

    Typically, you should use black for the first ink in the list.
  • Click the curve for each ink you need to configure to launch the Duotone Curve tool.
  • Use the Duotone Curve tool to adjust the tone curve for that color of ink.

    Typically, you do not need to adjust the curve unless you want a specific effect from that tone of ink.
  • Click the Overprint Colors button, shown above, to bring up the Overprint Colors dialog box.
  • Adjust the colors used when one ink is printed over another ink by selecting the colors for each of the overlapping options.

Indexed Color

The indexed color mode contains a single channel with a single set of indexed colors. Converting an image to indexed color reduces the image to a set of most important colors. The remaining colors are given an index between 3 and 256 in a color lookup table. Instead of using the level value of the pixel to define the intensity of the channel, it points to an index value in the lookup table for the indexed image.

To change the color mode of an image to indexed color, select Image > Mode > Indexed Color from the main menu. To view the table of indexed colors, select Image >Mode > Color Table from the main menu to bring up the Color Table dialog box shown below. The following are settings that can be modified in the Indexed Color dialog box:





The options for the Indexed color mode and the Color Table dialog box in Photoshop.

  • Palette: This lets you select which palette to use when choosing the color to be placed in the index. You can choose a palette based on Exact, System, Web, Uniform, Local, Master, or Custom colors. The Local option selects colors local to the image. If you are working with multiple images, the Master option selects colors from a master of all images. When selecting local or master palettes, you can use a Selective, Adaptive, or Perceptual method of choosing the colors. The Selective method tries to preserve the key colors in the image. The Adaptive method simply preserves the most common colors. The Perceptual method intelligently selects colors that will provide the best transitions, rather than just the most popular ones. You also have the option to select System colors for images that will be viewed only on a computer. The Web option selects only Web-safe colors for images that will be used on Web pages.
  • Colors: This specifies the number of colors to use in the color lookup table. The minimum is 3 colors and the maximum is 256.

    Tip: When creating Web images with file formats such as .gif and .png, using fewer colors results in smaller images.
  • Forced: This lets you force the conversion to keep certain colors in the image. The default is to force only black and white to be kept. The Primaries option protects eight colors: white, black, red, green, blue, cyan, yellow, and magenta. The Web option protects the 216 colors in the Web-safe colors. The Custom option allows you to preserve a specific palette of colors that you create. When you select the Custom option, a dialog box launches that allows you to specify the colors that you want to preserve.
  • Transparency: This specifies whether to preserve the transparency in the image.
  • Matte: The matte option allows you to specify a matte to use when working with transparency in the image. If there is no transparency in the image, this option is inactive. If the transparency option is selected, the translucent areas in the image are filled with the matte color. If the transparency option is deselected, the translucent and transparent areas are filled with the matte color.
  • Dither: This specifies the method Photoshop uses to calculate replacement colors for colors being discarded from the image. The None option simply selects the closest color in the lookup table, which can sometimes result in harder edges but is typically the best option to use. The Diffusion option dithers the color randomly, creating a more naturalistic effect. The Pattern option dithers in geometric patterns, which is usually the least desirable because the patterns show up in the image. The Noise option mixes pixels throughout the image instead of just the areas of transition.
  • Amount: This specifies the percentage of diffusion to use when dithering. This option is available only when you select the Diffusion dithering option. Lower values decrease the size of the file but result in harsher color transitions.
  • Preserve Exact Colors: This turns dithering off for areas of solid color when the Diffusion dithering option is selected. This option is not available unless the Diffusion dithering option is selected. Using this option helps your images look better even if you have to use dithering.

RGB Color

You likely will use the RGB color model most often. The RGB color model is used by computer monitors and the human eye. Photoshop processes its wide range of vivid colors quickly. RGB is the model that most closely matches the human eye for two reasons. First, RGB uses three colors similar to the receptors in the human eye. Second, RGB is additive, meaning that as you add more color, you get more light, in the same way that more color results in the eye seeing white.

In the RGB model, colors are divided into three channels of red, green, and blue. Each channel has an intensity level range between 0 and 255. Each color is made up of a combination of intensity levels from these three channels, resulting in the possibility of over 16.7 million different colors.

RGB provides by far the most vibrant use of colors of all the color models and is supported by most file formats. The one downside to RGB is that it contains more colors, especially the brighter ones, than can be printed. This can result in loss of detail in areas of your images when they are printed. The solution is to use the CYMK color model if you are going to have your images professionally printed.

CYMK Color

The CYMK color model is completely different from the RGB model in that it uses a subtractive method, meaning that the more color is added, the less light is seen. This is one of the reasons the CYMK model works so well for printing. Think about adding ink to a page; if you add all the colors, you get black, or rather a really deep brown.

Another difference between the CYMK model and the RGB model is that it is made up of four channels: cyan, yellow, magenta, and black. The black channel is necessary because adding the ink all together makes a dark brown not black, so if you want the printer to print true black, you must have a separate channel to specify black.

Which model should you use for general color image editing? The answer is RGB. The RGB model provides the widest range for tonal adjustment and correction. The scanner, monitor, and most other devices (except printers) work in the RGB model. Also, editing images in Photoshop in the RGB mode is much faster than in CYMK.

Tip: Even if you are using the RGB model, you can select View ? Proof Colors from the main menu to toggle the view to a simulated CYMK model. This way you can periodically check to see how the image will look when printed. Just remember to turn it off.

Lab Color

The Lab color model is very different from RGB and CYMK. The Lab model does have three channels, but instead of all three dedicated to colors, only two—a and b—are dedicated to color; the third—Lightness—is dedicated to luminosity.

The a channel maps colors ranging from deep green at level 0 to gray at level 127 to a rich pink at 255. The b channel maps colors ranging from bright blue at level 0 to gray at level 127 to a dim yellow at 255. The Luminosity channel maps the brightness of each pixel from dark at 0 to white at 255.

The Lab channel is additive like the RGB model, but it has only two channels of color mixing, and the levels of those channels are not mapping to intensity but rather tones of color. The tones add together to form brighter colors, and only the luminosity channel provides data to darken the tone that is created by the other two channels.

Editing images in Lab color is about the same speed as RGB and much faster than CYMK, so it is a fun alternative if you want to adjust your thinking of mixing colors.

Multichannel

The Multichannel mode separates out the channels in the current color model into spot channels. Spot channels can be used to store parts of an image that you want to print in specific inks or spot colors. For example, you can print specific inks from a Pantone library.

When you convert an image to the Multichannel model, the current channels are changed to spot channels. The channels created in Multichannel mode depend on the original color mode of the image. For example, the RGB mode gets converted to cyan, magenta, and yellow spot channels, the CYMK model gets converted into cyan, magenta, yellow, and black spot channels, and the Lab model gets converted into three Alpha channels.

Note: The spot channels overlap, so if you do not want ink from one channel to be printed on ink from another channel, the data in those areas of the channel cannot overlap.

Bits Per Channel

How many bits should you use per channel? The quick answer is 8 bits per channel, but let’s look a bit closer. What does bits per channel mean? A bit is a single item of information for a computer with a value of 0 or 1. That doesn’t mean much in terms of an image, but if you string millions of bits together, it can mean a lot.

Using 8 bits of information, we can define an intensity level of 0 to 255. For three channels, we can define about 16.7 million different colors for each channel. If we use 16 bits per channel, that goes up to over 281.4 trillion colors; if we go to 32 bits per channel, well, you get the idea.

So why not just use 32 bits per channel and maximize our information? The answer is disk space and speed. An image with 32 bits per channel takes up much more disk space and much more effort to edit on Photoshop’s part. Plus, the human eye can’t even detect all the colors in the 8 bits per channel.

And that leads to the question of why not just use 8 bits per channel, because it is more than enough for the human eye? The answer lies in what happens during adjustments, corrections, and conversions. Each time you make a correction to an image, change the levels, add a filter, and so on, you lose a little bit of the distinguishing detail. If you do enough corrections on an image with 8 bits per channel, you may lose noticeable detail in the image. However, if you are using an image with 16 bits per channel, the data lost is in levels that cannot be detected by the human eye, so when you convert the image back to 8 bits per channel, there is no data loss.

An image must be in the RGB or Lab color modes to convert it to 16 bits per channel. To change your image to 16 bits per channel, select Image > Mode > 16 Bits/Channel. After you have changed your image to 16 bits per channel, you can change it to 32 bits per channel by selecting Image > Mode > 32 Bits/Channel.

Note: An image with 32 bit per channel is considered an HDR (High Dynamic Range) image. Typically, these images are used in 3D rendering and advanced CGI animation effects.


Don‘t miss the next Photoshop article on Graphics.com. Get the free Graphics.com newsletter in your mailbox each week. Click here to subscribe.


Adapted with permission from Photoshop CS5 Bible by Lisa DaNae Dayley, Brad Dayley. Copyright © 2010 Wiley Publishing