This study describes a meta-analysis of three color palettes for protans and deutans, respectively proposed by Tol (2012), Krzywinski et al. (2012), and Ito et al. (2013). Their three color palettes were defined using standard red-green-blue (sRGB) tristimulus values, and they were designed to help color-deficient people distinguish the colors of graphs, maps, and other visual representations. However, color differences between the component colors of their palettes and the chromatic distribution of the component colors were unspecified by these researchers. Without comparative studies, it is difficult to compare the performance of these palettes and their visual effectiveness for color-deficient people. Hence, this study provides a meta-analysis of their color palettes. Protanopic and deuteranopic color perception and color differences were used for this analysis, along with a combination of a dichromatic simulation method and a uniform color space, such as CIELAB and CIELUV. This study elucidates a problem for analyses that rely on dichromatic simulation methods calculated using the reduced stimuli planes in the RGB color space, and proposes a way around such simulations. The results of the meta-analysis reveal that the three color palettes have problems in terms of the uniformity of color difference and the chromatic distribution of the component colors.
Takashi Sakamoto, "A meta-analysis of color palettes for protans and deutans: The problems with using dichromatic simulations" in Proc. IS&T 23rd Color and Imaging Conf., 2015, pp 138 - 142, https://doi.org/10.2352/CIC.2015.23.1.art00024