Compute CIE delta E 2000 color error metric
[de00, errComponents] = deltaE2000(Labstd,Labsample, KLCH )
ImagEval comments --
The CIE delta E standard is a fundamental tool of color science. The
metric is described in a wide variety of textbooks. It has gone
through a series of refinements over the years. This code is the
CIEDE2000.
K is a 3 vector, in the form [kL, kC, kH]. k =[1 1 1] for the basic
conditions (uniform surround field with L*=50, 1000lx illuminance,
homogeneous stimulus >4 degree visual angle, color difference between
0-5 CIELAB units). In the textile industry kL is commonly set to 2.
Default k is [1 1 1];
The Labstd and Labsample matrices contain corresponding pairs of CIELAB
values. The difference between these is the error measure. The CIELAB
values are computed from CIE XYZ measurements using standard lab
equipment.
ImagEval added a new return term (errComponents) that contains the
separate luminance, chromatic, and hue error terms. These components,
which are scaled and combined
The original authors and a reference source for this code are cited
below. The comments in the original source code follow:
Original comments --
Compute the CIEDE2000 color-difference between the sample between a
reference with CIELab coordinates Labsample and a standard with CIELab
coordinates Labstd
The function works on multiple standard and sample vectors too provided
Labstd and Labsample are K x 3 matrices with samples and standard
specification in corresponding rows of Labstd and Labsample The optional
argument KLCH is a 1x3 vector containing the the value of the parametric
weighting factors kL, kC, and kH these default to 1 if KLCH is not
specified.
Based on the article:
"The CIEDE2000 Color-Difference Formula: Implementation Notes,
Supplementary Test Data, and Mathematical Observations,", G. Sharma,
W. Wu, E. N. Dalal, submitted to Color Research and Application,
January 2004.
Code is available at http://www.ece.rochester.edu/~/gsharma/ciede2000/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%