Studies on Color Space Selection and Methods of Segmentation Quality Evaluation

The selection of color space is one of the determinants of the image segmentation quality, the segmentation results will be more accurate if the appropriate color space is adopted. In this paper, five kinds of color space RGB, HSI, YIQ, YCbCr, Lab are compared by three types of evaluation methods. The third method of image segmentation quality evaluation is proposed in this paper. The results shows that RGB space for image segmentation is the best choice, and the evaluation results by the method of this paper is consistent with the final experimental results.