We have a dataset with RGB values. For example,

175 91 53

And the label is red. What can you get out of this feature? or somehow determine the color? Here is a link to the example above (PS color does not look like red, but we consider it red):

https://convertingcolors.com/rgb-color-175_91_53.html

Closed due to the fact that the essence of the issue is not clear to the participants of Kromster , Viktor Tomilov , user192664, 0xdb , Yaroslav Molchan 16 Nov '18 at 12:50 .

Try to write more detailed questions. To get an answer, explain what exactly you see the problem, how to reproduce it, what you want to get as a result, etc. Give an example that clearly demonstrates the problem. If the question can be reformulated according to the rules set out in the certificate , edit it .

  • do you need something like 175> 91 + 53 = color is red? and so for each color check? - TEA
  • 207, 147, 108 - also red, but the amount is greater than R - potemin
  • Give the definition of red;) - MaxU
  • The data set is somehow crookedly compiled. Blue, green, yellow, red, white, orange. Maybe you should consider what it looks like more. Or just knn some kind of use. - potemin
  • Maybe it will help. r-bloggers.com / ... "Utilizing K-means to extract colors from your favorite images" - passant

1 answer 1

Translate the color from the RGB view into the HSV model and match the hue value with the color name (you will need to specify the ranges).

At low saturation, the color will be gray (white, black - depending on brightness)