A novel approach for color image segmentation is studied

which is based on segmentation of subsets of bands using mathematical morphology followed by the fusion of the resulted segmentation "channels".For color images the band subsets are chosen as the RG

RB and GB pairs

whose 2D histograms are processed as projections of a 3D histogram.The segmentation in 2D color spaces is obtained using the watershed algorithm.The final color image segmentations are combined by a region splitting-merging process.CIE(L*a*b*) color space is employed to measure the color distance.The proposed approach results in improved performance and running time.