Image Segmentation Based on the Mean-Shift in the HSV Space

The mean shift based on the HSV space is proposed for the image could be exactly segmented in the color image. Because the correlation of HSV was very little, H, S and V separately are segmented based on the different step size. Because the correlation of the RGB space is strong, the drawback being the image which could not be separately segmented by R, G and B was overcome. It is proved by experiment that the algorithm could get better result.