How to get mode values of RGB to represent one image

Hello

I got the sub-images and I tried to find a proper way to extract RGB from those images and I used mean value for each color ( redChannel = subImk(:,:,1); meanR = mean(redChannel(:));) and it's great for perfect image. Anyway, some images have shadow, grains or some undetermined artifacts that still remaining from segmentation step (attached image) and these effect a lot in getting mean RGB representatives.

Now, I think that getting mode (most frequent R,G,B value in sub-image) might be the better way because the images are quite constant. However, I do not know how to code it in efficient way. Can anyone please help me how to write to find a proper mode in sub-image? and if you have a better idea to get RGB, please say it.

Thanks in advance :)

Answer 1

[Image Analyst]

Explain exactly what "extract RGB from those images" means to you. Because to get the individual color channels (R, G, and B) from the images, you'd use imsplit():

[R, G, B] = imsplit(rgbImage);

I don't see what messing with the means and modes adds. imsplit() will extract the R, G, and B from an RGB image without worrying about (computing) means and modes.