I am trying to analyze the color intensity of an area on a small lizard as seen by an avian predator. Is there a way to get a quantitative output of RNL color maps so that I can compare how intensely colored the area is as compared to others of the same species? Thanks!
The colour maps are ideal for generally showing the distribution of measured colours in RNL colour space. The video guide shows you how to combine colour maps and to perform pairwise comparisons of the degree of overlap between any two colour maps (or combined maps).
This method is excellent for testing colour matching when you have complex colours or continuous gradients. However, if the patch you’re measuring is a fairly consistent colour, then comparing averages is probably your best (and most standard) approach.
One easy way to measure average colour is to convert the cone-catch image to RNL chromaticity, then measure your ROI (so you have average X, Y, Z etc…) Once in a spreadsheet or R you just need to calculate the Euclidean distance between any two XYZ coordinates to get the “Delta-S” colour difference for your chosen Weber fractions.
e.g. for a dichromat this is:
DeltaS = SQRT( (x1-x2)^2 ), which is the same as ABS(x1-x2)
A trichromat is:
DeltaS = SQRT( (x1-x2)^2 + (y1-y2)^2 )
A tetrachromat is:
DeltaS = SQRT( (x1-x2)^2 + (y1-y2)^2 + (z1-z2)^2 )
Alternatively, there is a tool built into the toolbox to do pairwise calculations, first run “Batch multispectral image analysis” (don’t measure pattern – just leave that switched off if not necessary), then run “Colour JND difference calculator” on the cone-catch output.
Cheers,
Jolyon