How to count each colour pixel of an image?

1 view (last 30 days)
I am trying to quantify the value of of each matter of the brain using matlab. I have segmented the image and result as shown below.
  2 Comments
Akira Agata
Akira Agata on 30 Sep 2018
Edited: Akira Agata on 30 Sep 2018

Is it a result of k-means clustering? If so, you can count each color pixel easily by using the output variable of kmeans function.

muhaimin mahfudz
muhaimin mahfudz on 1 Oct 2018
Edited: muhaimin mahfudz on 1 Oct 2018
yes it is. I see, in conclusion i can just obtain the values from there as it is the value of each segmented color?

Sign in to comment.

Accepted Answer

Image Analyst
Image Analyst on 30 Sep 2018

Take the histogram of your segmented/classified image

counts = imhist(classifiedImage);

Each class will have a certain value, like gray matter is 1, background is 0, white material is 2, etc. imhist() simply counts the pixels in each class.

  3 Comments
Image Analyst
Image Analyst on 1 Oct 2018
imhist() will do it. You pass in the classified image, not an RGB image, but the image you got from kmeans. See attached demo.
Image Analyst
Image Analyst on 2 Oct 2018
Do you still not understand? If so, give me the script you used to create that image along with the original image.

Sign in to comment.

More Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!