Thank you Kai-Uwe. I suppose I have to explain myself a little more.
The idea is to create a conversion chain from a Lab value to a L’a’b’ value
using a known n-color profile. So the chain is: Lab > n-color > L’a’b’. As
some Lab values can be outside the n-color gamut of the profile, there will
be some cases where Lab and L’a’b’ will be different. Furthermore, the
n-color profile I would like to use has a D65 illuminant and off course the
Lab profile is D50-based. To take in account the chromatic adaptation, I
though the 2 transform intents should be as follow:
*Lab to n-color transform:* relative colorimetric
cmsHTRANSFORM transform = cmsCreateTransform( labProfile2, TYPE_Lab_DBL,
nColor, TYPE_CMYK_16, INTENT_RELATIVE_COLORIMETRIC, cmsFLAGS_HIGHRESPRECALC
);
*n-color to Lab transform:* absolute colorimetric
cmsHTRANSFORM transform2 = cmsCreateTransform( nColor, TYPE_CMYK_16,
labProfile2, TYPE_Lab_DBL, INTENT_ABSOLUTE_COLORIMETRIC,
cmsFLAGS_HIGHRESPRECALC );
In order to valid my results, I use in parallel of my program, an external
software that has already been approved (colorLab) to do the same conversion
chain with the same conversion intents and obtain a third Lab value (let’s
call it L’’a’’b’’). My problem is L’a’b’ and L’’a’’b’’ are not equal whereas
they should be. For information, when I use a D50-based n-color profile and
set all my transforms with absolute colorimetric intents, then L’a’b’ is
equal to L’’a’’b’’. It seems clear to me there is something I misunderstand
about chromatic adaptation (or about use of INTENT_RELATIVE_COLORIMETRIC ),
so could someone tell me where I am wrong?
Regards
Remi
Le 23 mars 2011 22:09, Kai-Uwe Behrmann <k...@gmx.de> a écrit :
> Am 23.03.11, 19:13 +0100 schrieb Rémi CERISE:
>
> Please, could someone tell me if there is something special to do when
>> using
>> a cmsHTRANSFORM with INTENT_RELATIVE_COLORIMETRIC ? The fact is chromatic
>> adaptation is not done properly with my transform and I would like to know
>> what I am doing wrong.
>>
>
> INTENT_RELATIVE_COLORIMETRIC is intented to do no adaption. Thats correct
> behaviour. Use absolute colorimetric for chrommatic adaption.
>
> kind regards
> Kai-Uwe Behrmann
> --
> developing for colour management www.behrmann.name + www.oyranos.org
>
------------------------------------------------------------------------------
Enable your software for Intel(R) Active Management Technology to meet the
growing manageability and security demands of your customers. Businesses
are taking advantage of Intel(R) vPro (TM) technology - will your software
be a part of the solution? Download the Intel(R) Manageability Checker
today! http://p.sf.net/sfu/intel-dev2devmar
_______________________________________________
Lcms-user mailing list
Lcms-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lcms-user
