Datasheet

38
c h a p t e r 1: DIGITAL IMAGING BASICS, WORKFLOW, AND CALIBRATION ■
numerical representation is currently CIE L*a*b* (CIELAB, essentially the same as
Lab in Photoshop). Profiles are generated by capturing the color rendering of specific
devices and mapping the results to the numbers required to generate colors on that
device. As an example, your monitor uses a certain set of RGB numbers to generate a
color that can be defined as a specific shade of red in Lab. When you want to display
that shade of red (as defined in Lab) on another monitor, you will likely need a differ-
ent set of RGB numbers. If the other monitor has been profiled, you can derive the new
set of RGB numbers by running the Lab numbers through the profile for that monitor.
As such, profiles allow for some method of translating colors for different devices by
referencing into and out of Lab.
Profiles are also used for generic color-editing environments called workspaces.
Adobe RGB, Colormatch, and sRGB are examples of standard RGB workspaces.
An RGB workspace is defined in a way that allows for easier color editing, and one
unique property of this is that neutral gray is always defined with equal values in red,
green, and blue. One of the common activities in color management is transforming
color numbers from a capture device into workspace numbers for editing and then
transforming those numbers into numbers appropriate for a printer or display device
for viewing.
Most of the problems associated with color management concern the mismatch
between color spaces with different gamuts. Color gamut is the volume of colors
defined in a given color space. The real world represents a huge volume of possible
colors. Digital cameras are capable of encoding a somewhat smaller volume of colors.
The Adobe RGB workspace is smaller still, and almost every output device can repre-
sent a smaller volume than Adobe RGB. Overall volume is not the only factor. There
are other mismatches. Some possible colors in one workspace don’t exist in another,
even if the overall volume in the second space is greater. Imagine trying to fit a square
peg in a round hole—color spaces can be thought of as having 3D shapes based on
different regions of colors. The graphics industry has adopted various methods of
translation or gamut mapping between the different color space shapes using ICC pro-
files in an attempt to mitigate the problems.
Most of the time, photographers make a bigger problem out of this than is nec-
essary. People photography is mostly concerned with the color of skin. As it turns out,
skin color is easily defined in just about every color space with which we can work.
Conservative color spaces such as sRGB are perfectly adequate for people photogra-
phy, and translating between sRGB or Adobe RGB and any type of output we need
is not a problem. Problems typically arise with attempts to match garment colors or
other products that may be of primary interest to an advertising client. Some products
might contain colors that are outside the range of Adobe RGB but just inside CMYK
for a magazine ad. Sometimes the dyes and colorants used to manufacture these items
do not reproduce accurately no matter how well the system is color managed. Most of
the time, the solution involves color correction in the final output space and not color
management. The odd color mismatch in a manufactured product can usually be tol-
erated; however, if you can’t get good skin color, you are in trouble.
592120c01.indd 38 6/17/10 9:51:25 AM