Image segmentation based on fast normalized cut

Abstract

In this paper, we propose a fast image segmentation method based on normalized cut. This method apply simple linear iterative clustering super-pixel algorithm to obtain super-pixel regions, and then use affinity propagation clustering to extract the representative pixels in each super-pixel regions, Finally, we apply normalized cut to obtain segmentation results. At the end of the paper, Numerical experiments are presented to demonstrate the effectiveness of our proposed method.