He provided his sources: Color Blindness Simulator and this page listing color blindness frequencies.
Color blindness is not pure mathematics.
Statistics works with samples. He gives only percents. Moreover a color has a few attributes. There is no one red color. Some reds are recommended for people with colorblidness, but other not.
When I have disponible time, I’ll compute for comparison, some numerical caracteristics( lightness, chroma, saturation, etc) for T colors, and colors used by different graphics libraries as colorblind friendly.
1 Like
In short, what you would like is nothing less than an empètesec palette.
Here you can find how is seen red, green, and orange by people with CVD:
Tableau is a renowned data viz company, with a long history: https://www.tableau.com/blog/examining-data-viz-rules-dont-use-red-green-together
Colour contrast perception is tricky, it also depends on the size of the colour patch in your visual field. When coloured objects get very small hue differences becomes less important and the perceptual contrast becomes dominated by lightness differences in the colours. (A.B. Poirson and B.A. Wandell 1993. “The appearance of coloured patterns: Pattern color separability.” Journal of the Optical Society of America 12. pp 2458-2471.)
Below about 0.5 degrees of viewing angle, there is a severe loss of colour discrimination.
(Note that a 10mm screen object viewed at a distance of 600mm subtends about 1 degree of visual angle). The reason for this is that the colour cones in your retina have finite spatial separation and the blue cones are relatively sparse. In the limit, for very small objects, you will not be able to get three colour cones in your eye to simultaneously process light from a very small colour patch, effectively inducing a degree of colour blindness whether you are colour blind or not. (L.C. Thomson and W.D. Wright 1947. “The colour sensitivity of the retina within the central fovea of man.” Journal of Physiology (London) 105. pp 316-331.)
Line and scatter plots are affected by this problem. The small scale means that the number of distinct colours you can discern if they are presented as thin lines or small dots will be less than the number of colours you can distinguish if they are presented as large patches.
Another complication is that it has also been suggested that there is also a linguistic component to the identification of distinct colours. You are better at distinguishing colours if you can give them names.
(B Berlin and P Kay, “Basic Color Terms” University of California Press, 1969 2.)
Two other papers that are probably very relevant here:
-
Maureen Stone. “In color perception, size matters”. IEEE Computer
Graphics and Applications, 32(2):8–13, March/April 2012. -
Maureen Stone, Danielle Albers Szafir, and Vidya Setlur. “An engineering model for color difference as a function of size”. In 22nd Color and Imaging Conference. Society for Imaging Science and Technology, 2014.
Good luck. It’s complicated!
5 Likes