Bradford Transform: Matrix of Constants

[JamesGoh]

Im aware that in the Bradford transform, a matrix containing certain constants is used to obtain the end result. I have given this 3 x 3 matrix below (note poor representation of matrix using square brackets :P )

Mcx = [ 0.8951 0.2664 -0.1614 ]
[-0.7502 1.7135 0.0367 ]
[ 0.0389 -0.0685 1.0296 ]

Im aware it is a matrix of constants, but what does it represent and what is its purpose ? Is it used strictly for Bradford transforms ?

 

[John Seymour]

James,

The Bradford transform is used to convert from XYZ values to LMS values. XYZ values are the tristimulus functions that were officially standardized in 1931 (I think). These functions are a linear combination of the spectral response of the human eye. At the time, there was not sufficient knowledge to be able to define precisely the response of the eye, but there were a few color matching experiments that established different linear combinations of the spectral responses.

The LMS functions (long, medium, short) are a fairly decent approximation to the response of the eye. The Bradford transform is the way to get from one to the other.

One common use of this transform is white point normalization. We have a color measurement under one light source (sunlight, for example), and we wish to know what the measurement would be under another light source (like an incandescent bulb). The color measurement is convert to LMS, normalization is done, and the results are transformed back to XYZ.


This is a quick response. I would be happy to fill in more detail if this is of use to you.

John Seymour

 

[JamesGoh]

thanks John. Will look into it and get back to you

cheers