By Dave Popham,
Coppin State University/CoSIDA Computer Committee
Not that long ago, most SID’s sent photos for their media guides to their printers or service providers, who were responsible for placing the images into the publications file so that they would output properly on press.
Today, a significant number of sports information shops produce their media guides in-house, especially the text pages, using page layout software such as QuarkXpress, InDesign or PageMaker, and the responsibility for placing the images in the publications file has shifted to the SID.
Compounding the issue is that most images today are color, which means they’re generally in some flavor of RGB. In the case of those used in text pages, they must be converted to grayscale to meet NCAA guidelines for media guides.
The quickest and easiest way to get a grayscale image from an RGB image in Photoshop is to select Grayscale from the Image>Mode pull-down menu.
Although it’s fast, anyone who has done a straight conversion to grayscale on numerous images has probably been disappointed with the end result more than a few times, and that usually means spending more time making adjustments with Photoshop’s various tools to get a satisfactory image or using a less-than-ideal image.
There are several other ways that will give you more control over the results.
An RGB file actually is three grayscale images—a Red channel, a Green channel and a Blue channel—and your computer uses the information contained in the file to display it on your monitor in color. Bringing up the Channels palette (the default location is in the same palette group as the Layers palette) and clicking on each channel allows you to see that for yourself.
When Photoshop performs a standard RGB-to-grayscale conversion, it takes approximately 60 percent of the Green channel, 29 percent of the Red channel and 11 percent of the Blue channel and blends them together to make a grayscale image.
The key to outdoing Photoshop’s default conversion is to use the method that will best combine the information in those channels to create the final grayscale image rather than let the software impose its own will.
Most of the images in media guides are head shots and action shots, in which the main subjects are people. As a rule, the best channels to utilize for people are the Red and Green channels. Blue usually will contain a significant amount of noise, and it’s best avoided in many cases.
Alternative Methods for Converting Color Photos to Grayscale
1. Use an Existing Channel
Sometimes the best grayscale already exists within the image. Examine each of the channels, especially the Red and Green channels. You can quickly scroll through each channel using keyboard shortcuts—Cmd(Mac)/Ctrl(Windows) + 1 for Red, Cmd/Ctrl + 2 for Green and Cmd/Ctrl + 3 for Blue. To get back to the composite RGB image, hit Cmd/Ctrl + ~.
Photos shot under fluorescent lighting that have a strong green cast often have a very usable Green channel. If you find one that works, select that channel so that it is displayed, then make the conversion using the normal process—Image>Mode>Grayscale. A dialog box will ask if you wish to discard the other channels. Click OK. (Note: When you try to save a converted image that had an RGB color profile embedded, Photoshop will sometimes try to save the grayscale image with the RGB profile, which can create problems with how the image will output. Look at the bottom of the Save As dialog box next to Color. If the Embed Color Profile box is checked and there is an RGB profile—i.e. Adobe RGB or sRGB—the image should be assigned a grayscale profile before saving.
Use the Image>Mode>Assign Profile command from the pulldown menu. The Dot Gain 20% profile works well for most images that are going to press.)
2. The Lightness Channel
There is another grayscale channel that can be coaxed from all images, but it first requires a conversion. Convert the image to the Lab mode by choosing Image>Mode>Lab Color. In this mode, the color and luminosity components are located in separate channels, with the L channel holding the luminosity information. Select the L channel by hitting Cmd/Ctrl + 1. This channel often makes a good grayscale image, although it can be somewhat light and flat at times. If you decide to use the L channel, use the same method described above when using one of the RGB channels.
3. The Channel Mixer
This method allows you blend percentages of the Red, Green and Blue channels and can generate excellent results. There two ways to use the Channel Mixer—a) select Image>Adjustments>Channel Mixer under the main pull-down menu or b) click on the Adjustment Layers icon located at the bottom of the Layers palette and choose Channel Mixer. Unless you’re strapped for RAM, use adjustment layers so that your image that can be edited later if necessary. Click the Monochrome checkbox in the bottom left of the dialog box and make sure that Preview also is checked so you can see the changes. The rest is fairly intuitive, as each channel has a slider that allows use of a percentage of that channel, and the results are visible.
To maintain the tone of the original, the percentages should add up to 100, but this is a rule that can be broken, although going much above 100 will likely blow out the highlights. Using adjustment layers also allows you to vary the opacity of the layer, change the layer blend mode and use a layer mask. For example, you can create a Channel Mixer layer that provides good tone for your main subject, mask out the background with a layer mask, then create another Channel Mixer layer with tones optimized for the background. Once you’ve got the image you want, you’ll need to convert the final version to grayscale using the Image>Mode>Grayscale method and clicking the Flatten Image button in the dialog box.
4. Calculations
The Calculations dialog box, which is opened by selecting Image>Calculations under the main pulldown menu, can be somewhat intimidating, but it’s a very useful method to blend channels, although it allows the use of only two channels at a time. Set the channel you want to use as the main image to blend into under Source 2 in the dialog box, and set the image to be used for blending in Source 1.
When choosing channels, it’s helpful to consider the main subject and background as two different parts of the image, since the goal is to separate them tonally. For example, if the main area of interest is a light tone, look for a channel that has a dark background that can be utilized when blending. What makes Calculations so powerful is that you can set the opacity and the blend mode and mask out parts of the image you do not want affected by clicking on the Mask box located in the Blending section and choosing a channel to use for the masking. Keep in mind that in a mask, darker areas block the effects of the blend.
5. Maximum Control and Flexibility—Stacking Channels As Layers
All of the previous techniques can provide quality results, but many high-end production artists and photographers use a method that provides the most flexibility and quality. Because this method is the most time-consuming and requires hand-tweaking each photo, it’s best reserved for those handful of images where you want maximum quality, such as shots of your president, AD, head coach or top athletes, especially ones that will be used more than once. In the RGB image, make the Channels palette active, go to the Options submenu on the palette (look for the arrow on the top right of the palette) and select Split Channels.
This creates three separate images—one of the Red channel, one of the Green channel and one of the Blue channel. Choose the one that most resembles the final image you want (usually the Green channel) for your base image. Since the resulting images are in Multichannel mode, you’ll need to convert the base image to grayscale. Under the main pull-down menu, select Image>Mode>Grayscale. Then select the Red channel, select all (Cmd/Ctrl + A) and copy (Cmd/Ctrl + C). Go back to the base image you created and paste (Cmd/Ctrl + V) the Red channel, which will create a new layer on top of the original layer.
Do the same thing with the Blue channel if you wish to use information from that channel. From that point, you can experiment using the opacity slider and blend modes in the Layers palette, as well as layer masks, to vary how much of the image above contributes to the bottom layer.
All of the above methods will usually generate better grayscale images than a straight conversion. The image will probably still need editing with either Levels or Curves, but you will be starting with a better image.
And best of all, if you have a batch of images that were taken using similar lighting and exposure settings (i.e. head shots), the first four can be automated using Photoshop’s actions feature (with the exception of layer mask edits).