i recently heard about the topic called "content based image retrieval systems". i am trying to learn about the algorithms used in this topic. I already found 2 algorithms. 1) AverageRGB- which calculates the average values of red,green and blue channel of each pixel in an image.
2) GlobalColorHistograms - which create a 64 buckets of colors using the 2 significant bits of RGB colorspace.
Here is my question.How is it that we are getting 64 distinct colors by taking 2 significant bits of RGB channels.What will be the value of the significant bits? i have looked at different places but couldn't understand anything.
Start by splitting the RGB channels in R
, G
, and B
. Given that you will pick only 2 bits, the only possible values are 00
, 01
, 10
, or 11
. No matter the values you have in each channel, these are the only four values you can obtain in a given channel. Supposing you found all the four values in each channel, then there are 64 possible 6-bit colors that you can build from that.
R: 00 01 10 11
G: 00 01 10 11
B: 00 01 10 11
With R
at 00
you get 16 colors: 000000
, 000001
, 000010
, 000011
, 000100
, ... Then more 16 starting with 01
and so on, basics of combinatory.