Hierarchical segmentation for color images

Abstract

It is a challenging task in the field of computer vision to obtain the hierarchical segmentation for a color image. This paper focuses on the problem of efficiently performing hierarchical segmentation and improving the segmentation precision. We first apply multiscale normalized cut to image pre-segmentation step to replace Voronoi algorithm or watershed transformation. In addition, an improved ultrametric measure method is introduced and employed to merger regions. Finally, the hierarchical segmentation results are obtained and one can get the best segmentation result by using threshold. Experimental results show that the proposed method is able to provide accurate segmentation result with high computational efficiency.