On Mon, 27 Apr 2009 22:56:00 jcup...@gmail.com wrote: > RGB isn't the best colour space for this, but it would sort-of work.
Another issue is that RGB is (usually when encountered "in the wild") the sRGB colour space. sRGB as well as most other RGB colour spaces is *not* a linear colour encoding per channel. They have been encoded with a geometrical function with an exponent of (in the case of sRGB) about 2.2. This is a historical effect to compensate for the non-linearity of CRT (cathode ray tube) based screens. If you want to be at least "sort of" linear, you need to undo the gamma compensation first: c' = c ** (1 / gamma) gamma: the gamma coefficient, usually in the case of sRGB = 2.2 c': linearised RGB colour channel c: un-linearised (s)RGB colour channel Now you can perform quantitative sensible computations on the R'G'B' colour tuples, like comparing colour distance as the Euclidean distance mentioned in John's post. Guy -- Guy K. Kloss Institute of Information and Mathematical Sciences Te Kura Pūtaiao o Mōhiohio me Pāngarau Room 2.63, Quad Block A Building Massey University, Auckland, Albany Private Bag 102 904, North Shore Mail Centre voice: +64 9 414-0800 ext. 9585 fax: +64 9 441-8181 eMail: g.kl...@massey.ac.nz http://iims.massey.ac.nz _______________________________________________ Image-SIG maillist - Image-SIG@python.org http://mail.python.org/mailman/listinfo/image-sig