| Author |
Comment/Response |
Clusty
|
 10/19/07 4:00pm
Well I am not 100% sure but this is how I think you can do it:
You compute the histogram of the image and rescale it in such a way to get values from 0..1. This probability function we call p.
Then you sum up all pixels as
-\sum_{i} p(x_i)log(p(x_i))
P(x_i) is how many pixels have color x_i (rescaled from 0..1)
Hope it helps
URL: , |
|
