Color image segmentation based on watershed and Ncut of improved weight matrix

A new color image segmentation method combining twice used watershed and Ncut of improved the weight matrix algorithm is presented in this paper. It preprocesses an image by the twice used watershed algorithm to form segmented regions that preserve the desirable discontinuity characteristics of image. The segmented regions, instead of the image pixels are then represented by using the graph structure and considered as the input image of Ncut algorithm. In addition a new weight matrix is designed in this paper according to the image color and space information. Then the Ncut method is applied to perform globally optimized clustering. Because the image clustering uses the segmented regions, instead of the image pixels, the new method can effectively reduce the computational complexity of traditional Ncut method by using secondary watershed algorithm. The new weight matrix also has certain self-adaptability. Through a large number of experiments using color natural scene images, the results show that the proposed method has superior performance and less computational costs compared to the traditional Ncut algorithm and the method combining the mean shift (MS) and Ncut.