Macbeth Colour Checker and Illumination temperature

[tomczak]

I'm trying to understand how to interpret the RGB values of Makbeth colour checker in different colour spaces. I think what I'm missing is the illumination that is supposed to produce such reflection off the chart. Would it be one of the CIE standard illuminant series: C? D? What practical illumination am I getting with, for instance, Colour Match transformation: I don't understand what illumination it is trying to match the test image to?

I found a few references below, but I'm still not sure. Would anyone have a clear understanding of it?

http://www.mambo.net/cgi-bin/TempProcessor/view/113
http://scien.stanford.edu/class/psych22 ... 5/joanmoh/
http://www.brucelindbloom.com/index.htm ... erRGB.html
http://en.wikipedia.org/wiki/Standard_illuminant

[jsachs]

Each color space has its own white point as defined the profile.

[tomczak]

Sure, but what what do I get from Colour Match if I offer it a Macbeth photographed at sunset? I guess what I don't understand is what is the physical checker illumination white point (or spectrum) that either Colour Match or the published theoretical RGB values in different colour spaces are trying to mimic.

[jsachs]

The Macbeth patches are defined in terms of CIE XYZ coordinates. From these coordinates, their appearance in any color light can then be computed. By transforming according to the gamma, white point and primaries of the color space, theoretical RGB values can then be computed.

[elied]

Maciej, are you familar with this document? Especially section 5 and the charts in section 3.

[elied]

Soory, forgot the link.
http://www.babelcolor.com/download/RGB% ... hecker.pdf

[tomczak]

Cheers, Elie!

[tomczak]

If the Colour Management is disabled and the Working Colour Space is set to 'None', the Match Profile still works. What gamma, primaries and white point does it use it this case to calculate theoretical R'G'B' values for patches?

Also, what's the reference for the CIE XYZ values used in the calculations? Would it originate from 1976 xyY values published by Gretag, or perhaps the recalculated from the 2005 published L*a*b* values? The reason for asking is the article that Elie pointed to earlier in this thread.

[jsachs]

A working color space of none is treated the same as sRGB.

[jsachs]

The reference values are the older xyY supplied by Gretag.