[argyllcms] Camera profiles for artificial lighting

From: Stephen T <stwebvanuatu@xxxxxxxxxxxx>
To: ArgyllCMS <argyllcms@xxxxxxxxxxxxx>
Date: Mon, 27 May 2013 05:01:47 -0700 (PDT)

Hello ArgyllCMS users,

I am pretty happy with my sunlight camera matrix profiles. In sunlight/daylight 
and with appropriate white balance they give satisfactory accuracy (low delta 
Es) and the colours look good.

In artificial lighting, my sunlight profiles are inaccurate and the colours can 
look bad.

I am now making additional profiles for artificial lighting: tungsten and 
fluorescent. I can't get low delta Es from colprof however.

What could be the problem(s)? Assume the input image and white balance are 
good. 

I am thinking that the white balance (equalizing of raw photo RGB) is not 
satisfactory to "correct" strange light sources???

So a more detailed look-up table model is required to accurately transform to 
the Profile Connection Space (I am using a ColorChecker SG with D50 reference 
data)???

I would like to better understand the limitations of camera input profiles with 
strange illuminants.

Stephen,

[argyllcms] Camera profiles for artificial lighting

Other related posts: