How can I detect color variation in melanoma images ?

Hey all,

I am really new to image processing and Matlab and I am trying to detect color variation in skin cancer cells. I have read a lot of articles and answers here but I couldn't come up with a solution. The algorithms provided in other questions didn't help find the solution.

I have tried to find them manually, but i'm not sure of it and i'm positive that there's a better way of doing it. I am searching for pixels that are red, blue, brown or black by using RGB codes and counting them, however RGB codes are very specific and I am worried this method will miss pixels that are closely colored to these colors because they don't have the same RGB code.

Here's a piece of my code:

cropedImage=imcrop(shiftedImage);
count = sum(cropedImage(:, :, 1) == 105 & cropedImage(:, :, 3) == 105 & cropedImage(:, :, 3) == 105);
count1 = sum(cropedImage(:, :, 1) == 0 & cropedImage(:, :, 3) == 0 & cropedImage(:, :, 3) == 0);
...

and I add up all the counts at the end but this way I have to create a lot of variables and there's a very good chance of missing other pixels.

PS: I cropped the image using imcrop but its not necessary in my case since there's only one lesion per image in my case.

I'm quite new to the field, any help is greatly appreciated

Thank you very much

Chris

[Sivakumaran Chandrasekaran]

please contact us at [e-mail]. we can assist you. also mail the images