Weighting neutrals for input profiles

[Graeme Gill]

Is there any reliable way to preprocess the images or ti3-files or to instruct colprof
to minimize the error in neutrals?

Hi,
I guess I'm not sure what you mean. Do you mean that the profile
fit to the measured data is poor ? What stats. do you get on that at the
end of colprof (or profcheck) ?

If you mean the neutrals at the end of your color space conversion chain,
then this is quite a different question. What is your color conversion chain ?
What is the nature of the material you are scanning, and what output space
are you converting to ?

Graeme Gill.

[Nyberg Åkerström Wolmar]

I'm creating input profiles for a set of scanners and I haven't been able to figure out how to avoid colour shifts in the (near-)neutrals. I'm using patches from a ColorCheckerDC (all but the S-column) and a QPCard 201 and so far get the best results from the shaper curves since the LUT:s didn't generalise well.



Is there any reliable way to preprocess the images or ti3-files or to instruct colprof to minimize the error in neutrals?



Kind regards,
Wolmar