Image segmentation using modified SLIC and Nyström based spectral clustering

Image segmentation is a fundamental and challenging problem in the field of computer vision. In this paper, an efficient two-stage image segmentation method is proposed which takes advantage of modified SLIC segmentation and Nyström based spectral clustering. With the modified SLIC approach utilized in the first stage, Nyström based spectral clustering method is used to cluster the segmented regions instead of the pixels in the image to bring the final result. Therefore, the memory requirement and the computational complexity are significantly reduced. To verify the proposed algorithm, it is applied to images of different characters and compared with six other famous image segmentation approaches. Experiment results show the effectiveness and the robustness of the proposed method.