A study on color space selection for determining image segmentation region number

When image segmentation is treated as a problem of clustering color pixels, the finite mixture model with EM algorithm can be used to cluster color space samples. With estimated mixture model parameters, we adopt the BYY model selection criterion to determine how many regions should be segmented on a given color image. In this paper, we experimentally investigate the effect of choosing different color space for determining a reasonable region number based on the BYY criterion.