Color Correction Basics: The Info Palette
Like a tape measure in a carpenter's toolbox, the Info Palette provides necessary and useful information in your color correction toolbox. By itself, it does not make any physical changes, but is an integral part of making successful color changes with other tools.
No matter how well your monitor is calibrated, your eyes can trick you when it comes to neutralizing subtle color casts. It is also difficult to determine the makeup of certain color casts. Is the cast red or is it magenta? Is it blue or is it cyan? It is difficult to determine the composition of a color cast with sight alone. The info palette is your reality check on determining actual colors in your image.
The Info Palette provides information for several Photoshop tools, but this tutorial covers just its use for color correction.
Setting Up The Info Palette
Whenever you are doing color correction, the Info
Palette should be open and visible on your desktop (Window/Show Info)
before using any
of the image adjustment dialogs such as Curves or Levels. Once one of the
adjustment dialogs is open, you cannot make the Info Palette visible if it
is not already showing on the desktop.
The Info Palette shows you critical information about the values and location of pixels under your cursor on the image. You control the information displayed on the Info Palette with the Info Options dialog box which opens from the fly down menu on the Info Palette.
The
Info Options dialog, shown to the right, allows you to set the dialog for
two different color readouts. Most users who work with color images in RGB
mode, set the first color readout to RGB and the second color readout to
CMYK, as shown in the Info Options dialog here and the Info Palette above.
If you are working in grayscale, you can set one of the color readouts to
either Grayscale or Actual Color. Actual Color is a useful readout because
it displays information for whatever mode you are working in.
The Grayscale readout displays values as a percents, as does the CMYK readout above. I find it useful to also use the RGB readout when working in grayscale because it shows the numeric value of the sample (all three channels simply display the same value).
You can also set the Ruler Units in this dialog, which makes it a useful place to quickly change your ruler settings without going all the way to the Edit/Preferences/Units and Rulers... dialog.
What Do The Color Readouts Mean?
If you have your Info Palette set up as the one above, it shows you RGB numeric values and CMYK percentage values of the point beneath your cursor on the image. RGB numbers represent intensity values running from 0 to 255 for each of the three color channels. CMYK percents represent percent of cyan, yellow, magenta, or black making up the color under the cursor.
No matter what your monitor displays, you can always move your cursor over a given point in an image and determine the values that make up that color in the Info Palette. In the Info Palette shown above, green predominates at a value of 146. The pixels under your cursor should appear green.
If you are working in one of Photoshop's color working spaces such as Adobe RGB and you move your cursor over a value you know to be neutral in your image―black, white, or gray―each of the three color channels should display the same value. If one of the RGB values predominates, chances are you have a color caste. The Info Palette makes this color caste evident even if it is too subtle to be seen on your monitor.
Color Sampling Using the Info Palette
You can use the Info Palette to store and display
sampled points from
your image. If you select the Color Sampler Tool from the toolbox and use
it to click on a spot in your image, the values for that spot will be
stored and displayed on the Info Palette. You can store up to 4 values
using the Color Sampler Tool.
Storing sampled values is an extremely useful feature when trying to color correct an image. A useful technique is to sample and store a black shadow value, a white highlight value, and a neutral value. You can then use image adjustment tools to change the color values of your image while monitoring these sampled points.
The Info Palette just above has three sampled values: #1 a shadow, #2 a highlight, and #3 a midtone. The sample points are visible on the image itself and are labeled #1, #2, and #3 to match the sampled information in the palette.
In sample #1, the shadow value, note the blue value of 28 versus the lower values for red and green. This higher blue value likely indicates the shadow has a blue caste. If you know the shadow value should be a neutral black, then using one of the adjustment dialogs like Curves or Levels to lower the blue shadow value to 11 (and the green value to 11) will neutralize the shadow.
Likewise, the higher R values in samples #2 and #3 likely indicates a slight red caste for the overall image. Raising the G and B values of sample #2 to the R value of 253 will give you a neutral white and equalizing the #3 values at a midpoint value of 127 will give you a neutral gray midpoint. Simple corrections like this to remove color casts will frequently clear up most of the color problems for a given image.
Used In Conjunction With Image Adjustment Tools
The Info Palette has a useful feature when used in
conjunction with one of the image adjustment tools such as Curves or
Levels. It shows the before and after color values of the pixels beneath
the pointer and
color samples.
This palette displays the corrections to neutralize the image discussed above. You can see both the original values and the values after a color correction using Curves. The shadow values of #1 have neutralized by lowering the green and blue values to match the lower red value. The highlight values of #2 have been neutralized to white by raising all the values to 255 to produce pure white. The midtone values have been set at a neutral gray of 127 for each color channel.
Sample Size
When using the eyedropper to test or sample colors
on your image, you
should set the eyedropper's sample size in the eyedropper menu. You can
select among a single sample point, a 3x3 pixel average, or a 5x5 pixel
average. I tend to favor a 3x3 average with the thought that a single
point is frequently not representative of a color I am trying to capture.
You should also be aware that this sample size only applies when the image
is displayed at 100% viewing size. The number of sampled pixels increases
if you sample at less than 100%.
Copyright 2002 Michael W. Rollins