[argyllcms] Re: Best way to proceed?
- From: "Alastair M. Robinson" <profiling@xxxxxxxxxxxxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Sun, 01 Jun 2008 11:46:19 +0100
Hi,
Ben Goren wrote:
I don't think that'll help me. When I've adjusted curves, etc., in
Photoshop to get the ColorChecker close to spot-on, the wood hasn't
gotten any closer -- it's just shifted to other equally-worng colors.
Another possibility which springs to mind. How about creating two
profiles, based on a target in the image, before and after you've
manually adjusted the image to get the wood right.
In theory, you should then be able to link from one to the other to
create a transform that will duplicate your manual adjustments, which
you can then apply to the other pieces.
It's not as straightforward to do in practice, unfortunately, because
input profiles don't have reverse tables (Is this a feature of input
profiles in generally, or Argyll's in particular?).
I think the second profile will have to be generated as an output
profile, which means editing the .ti3, changing "INPUT" to "OUTPUT",
changing "XYZ_RGB" to "RGB_XYZ", swapping the first three columns of
values with the second three columns, and adding an extra patch with
RGB=100,100,100, XYZ=suitable white point - just to test it this end, I
used 82, 85, 75 - which is the white from one of my printer profiles.
Oh, one other thing - increment NUMBER_OF_SETS by 1.
This sounds complicated, but it's fairly easy to do in a spreadsheet.
Again, I can't predict how accurate or useful this approach will be -
just thinking aloud once again ;)
All the best,
--
Alastair M. Robinson
Other related posts:
