4 ms·

It's definitely an interesting topic and a nice high level overview of the problems of colour and colour spaces. I like the 4 point colour wheel in some way, b

It's definitely an interesting topic and a nice high level overview of the problems of colour and colour spaces. I like the 4 point colour wheel in some way, but I don't entirely see the benefits. From a computer vision perspective, this representation still has problems (as do all colour spaces). Look at how massive the green space is. There's 4-5 greenish slices that are close to each other, yet only 1 slice for yellow and 2ish for red. Red is the colour humans see "best", physiologically it should have the largest representation, whereas blue should be smallest. From the simpler side (i.e. a basic colour wheel application), I'm not sure if this makes it much easier or faster for me to find a colour. In any case, the article is very nice high level overview of a fascinating problem.

AceJohnny2 14y ago
"look at how massive the green space is..." This is why, in non-symmetric color coding systems, Green is given the largest share. For example 16-bit image formats give 5 bits for Red and Blue, but 6 bits for Green. The Bayer color pattern pattern used on CMOS sensors (basically all sensors on consumer devices) has interleaved rows of Green and Red and rows of Blue and Green: twice as much green as the other colors! Finally, green represents more than 50% of Luminance! From an evolutionary perspective, it makes sense to be more sensitive in the green spectrum: we were surrounded by green plants. Better catch the nuances to better identify predators or prey!
evincarofautumn 14y ago
I thought green was the colour humans saw best, if best means brightest. For instance, when converting from RGB to greyscale the usual weights are 30% red, 59% green, and 11% blue. (Though I always found it kind of unfortunate that these weights are wrong for people with nonstandard colour receptors.) In that sense, green should (seem to) be the largest because we perceive more colours as greenish than as reddish or bluish.

AceJohnny2 14y ago

"look at how massive the green space is..." This is why, in non-symmetric color coding systems, Green is given the largest share. For example 16-bit image formats give 5 bits for Red and Blue, but 6 bits for Green. The Bayer color pattern pattern used on CMOS sensors (basically all sensors on consumer devices) has interleaved rows of Green and Red and rows of Blue and Green: twice as much green as the other colors! Finally, green represents more than 50% of Luminance! From an evolutionary perspective, it makes sense to be more sensitive in the green spectrum: we were surrounded by green plants. Better catch the nuances to better identify predators or prey!

evincarofautumn 14y ago

I thought green was the colour humans saw best, if best means brightest. For instance, when converting from RGB to greyscale the usual weights are 30% red, 59% green, and 11% blue. (Though I always found it kind of unfortunate that these weights are wrong for people with nonstandard colour receptors.) In that sense, green should (seem to) be the largest because we perceive more colours as greenish than as reddish or bluish.