image classification using color variation

Hi All,

I have many coloured images some of them contains some color (red,blue,brown,and black)what I need to know how can I detect the color variation in the segmented image for image classification using color variation where if there is color variation in the then image will be belong to first class else the image will be belong to second class ,could anyone help me by to this job please?thanks in advance.

Answer 1

[Image Analyst]

You need to compare the delta E (the color difference) locally. So you scan the image with a window and get the mean color in that window. Then you look at the delta E inside that window and find it's standard deviation. If it's high, the colors vary a lot. If it's low, there in not much color variation. See attached demo.