peter dalgaard <pdalgd <at> gmail.com> writes: > > > > On 25 Jul 2015, at 21:49 , Atte Tenkanen <attenka <at> utu.fi> wrote: > > > > Hi, > > > > I have tried to find a way to find opposite or complementary colors in R. > > > > I would like to form a color circle with R like this one: http://nobetty.net/dandls/ colorwheel/complementary_colors.jpg > > > > If you just make a basic color wheel in R, the colors do not form complementary color circle: > > > > palette(rainbow(24)) > > Colors=palette() > > pie(rep(1, 24), col = Colors) > > > > There is a package ”colortools” where you can find function opposite(), but it doesn’t work as is > said. I tried > > > > library(colortools) > > opposite("violet") and got green instead of yellow and > > > > opposite("blue") and got yellow instead of orange. > > > > Do you know any solutions? > > Not directly, but a few hints: > > First read up on "complementary colors" in Wikipedia. In particular, note that the traditional color > circle does not satisfy the modern definition of opposite-ness. E.g. red paint mixed with green paint is > brown, not black or grey. > > The construction of the color circle is simple in principle: red, blue, yellow go at 0, 120, 240 degrees, the > other colors on the circle are formed by mixing two primaries in varying proportions: green (at 180 deg) is > an equal mixture of blue and yellow, violet (at 60 deg) of blue and red, orange (at 300 deg) of red and yellow. > Blue-green (at 150 deg) would be half blue, half green, alias three quarter blue, one quarter yellow. Etc. > > The tricky bit is that the above mixtures are subtractive mixtures (mixing paint rather than light beams) > and I don't know how to make a subtractive color mixture in the additive RGB space that we usually work in. > Maybe there are tools in the colortools package? > > -pd > > > > > Atte Tenkanen
To start with, you should be specifying your "colors" or lights actually in an additive color space like CIE 1931 xy, https://en.wikipedia.org/wiki/CIE_1931_color_space which you can do in the colorspace package. But this is based on an average observer and the results are unlikely to match a given individual's vision. On top of that, decisions made when this norm was specified are such that it deviates from human vision for short wavelengths so that you would be better off using a corrected version like that proposed by Judd in the 1950's or for the most recent suggestion see ww.cvrl.org under New CIE XYZ functions transformed from the CIE (2006) LMS functions best, Ken -- Kenneth Knoblauch Inserm U846 Stem-cell and Brain Research Institute Department of Integrative Neurosciences 18 avenue du Doyen Lépine 69500 Bron France tel: +33 (0)4 72 91 34 77 fax: +33 (0)4 72 91 34 61 portable: +33 (0)6 84 10 64 10 http://www.sbri.fr/members/kenneth-knoblauch.html ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.