On 3/16/2010 6:13 PM, Ken Norton wrote:
> No, the scans were not ICC profiled, although I may get around to doing that
> one of these days. But you are correct--it wouldn't have changed a thing,
> after all, K4 is still purple on the target. But that doesn't count, though,
> because the pigments or dyes in the print of the target are well within the
> range of acceptance of all films and sensors. The films and sensors failed to
> literally see the purple flowers the same way the human eye saw them. No
> amount of ICC profiling will fix that. The films and sensors in this case are
> "color blind."
>
The problem is clear. The solution of going to an E-1 or to specific
films, not so attractive.
There is a more general solution available in post. Later version of PS
include a color match function, which was, I believe included for
exactly the sort of issues you are talking about. The problem of the
flowers and dress that should match are exactly what many color
production people have to control.
In you examples, you were kind enough to include an IT8 target. Using
the K row of the IT8 as source and the flowers as target, I got the
following result. <http://www.moosemystic.net/Gallery/Others/AG/Purple.htm>
Looks pretty good to me, maybe even a bit more violet than the E-1. Not
having he flowers here, I can't say which might be more accurate. Using
a mask, it could be fine tuned with opacity adjustment. Working with a
larger pixel version would make selection easier and perhaps results
better. I could also have worked from E-1 colors to A1. I tried that
approach with the dpreview examples, with very good results.
As long as a source with the proper color and enough range of luminance
for the target is available, Match Color does an excellent job.
A. Violet Moose
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|