Graphics Programs Reference
In-Depth Information
value of 95 and black value about L20 (set the host software info palette
to read LAB or LCH). When shooting RAW data, the gray balance and
white and black patch exposure check has to be inspected after the RAW
data is processed. Once the RAW data is processed, use the Photoshop
info palette and examine the LAB values using a 5 ¥ 5 sampling. Attempt
to get these values for white and black patches above.
The proponents of this approach believe that a single profile made
from a correctly photographed target can be used in all shooting situa-
tions once the user creates a profile using these techniques. Others dis-
agree and suggest that profiles need to be made for each scene.
You can imagine the difficulties of shooting a target as described
earlier on location in a myriad of lighting situations. These experts rec-
ommend that you can be far more causal and simply place the target in
the scene, only really having to be concerned with glare on the target,
and of course, correct exposure. The idea is that the surrounding light-
ing and conditions affect the capture so the target should be affected by
these environmental conditions as well. I have seen that in some cases,
shooting a target in a controlled studio environment can be used outside
that environment. I have also seen cases when the camera profile has
failed miserably. Consequently, I have effectively profiled cameras by
casually placing a target into the scene.
For those photographers who do work in controlled studio environ-
ments, shooting the target as just described and building a custom camera
profile seems to provide a far better chance of success compared to the
photographer that needs to shoot in every conceivable situation. If you
find that you need to build a profile for each scene, even lighting may
be impossible. However, the idea here is we are profiling the scene and
the lighting. The issue becomes how well does the software that will build
the camera profile handle a large lighting imbalance on the target. The
resulting profile could be unusable. The debate, which is ongoing, is this:
Can a user profile a camera in a very specific controlled condition, and
use that profile elsewhere?
The idea is we are describing how a camera captures color. In reality,
we are profiling the RAW to color conversion rendered into scene-
referred data. That initial process is going to be different for each RAW
converter. On the other hand, do we need to profile the scene itself,
which is highly influenced by so many factors? As you can see, success-
fully profiling a digital camera is no easy task!
Building the Profile
Actually building the camera profile can be as easy as building a scanner
profile. The resulting quality and usefulness of the profile, however, is
often questionable. To build a camera profile, you need to load the image
of the target into the software, crop the target so the software knows the
boundaries, and load the TDF file. For many products, this operates in
