5
$\begingroup$

I have a dataset of images representing human skin. How do I plot the skin color distribution in HSV color space and YCBCR color space. However, I'd like ignore the "value" or "brightness" in both color spaces and analyze rest of the layers.

edit

The below is the code, I found to plot the distribution for ycbcr but don't understand what it means. Can anyone explain it please?

chroma = zeros(256);
cb = imycc(:,:,2);
cb = reshape(cb, 1, numel(cb));
cb = round(cb);

cr = imycc(:,:,3);
cr = reshape(cr, 1, numel(cr));
cr = round(cr);

for i = 1:length(cb)
   chroma(cb(i), cr(i)) = chroma(cb(i), cr(i)) + 1;
end
surf(chroma)
$\endgroup$
6
  • 1
    $\begingroup$ why don't you just extract Ihsv(:,:,1) or Ihsv(:,:,2) for further processing $\endgroup$
    – lennon310
    Feb 28, 2014 at 17:40
  • $\begingroup$ yes, after extracting h and s layers as you mentioned.. how do I plot the distribution $\endgroup$
    – 0cool
    Feb 28, 2014 at 17:41
  • $\begingroup$ The code is pretty clear, it just count the number of (cb(i),cr(i)) combinations under the assumption of cb, and cr value all locates within the range of 1-256 (or 0-255 in your case) $\endgroup$
    – lennon310
    Feb 28, 2014 at 17:54
  • $\begingroup$ Okay!, when in the case of normalized values, how should I do it? $\endgroup$
    – 0cool
    Feb 28, 2014 at 18:01
  • $\begingroup$ chroma(round(cb(i)*256), round(cr(i)*256)) = chroma(round(cb(i)*256), round(cr(i)*256)) + 1; $\endgroup$
    – lennon310
    Feb 28, 2014 at 18:16

1 Answer 1

0
$\begingroup$

Have a look at File Exchange Submission - RGB/HSV Distribution on visualization of RGB/HSV distribution in a given image:

hsv_distribution(inputImage, 5)

It should do what you're after as the result is like:

enter image description here

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.