Simple Linear Iterative Clustering (SLIC) and Graph Theory-Based Image Segmentation

Abstract

With the extensive application of deep acquisition devices, it has become more feasible to access deep data. The accuracy of image segmentation can be improved by depth data as an additional feature. The current research interests in simple linear iterative clustering (SLIC) are because it is a simple and efficient superpixel segmentation method, and it is initially applied for optical images. This mainly comprises three operation steps (i.e., initialization, local k-means clustering, and postprocessing). A scheme to develop the image over-segmentation task is introduced in this chapter. It considers the pixels of an image with simple linear iterative clustering and graph theory-based algorithm. In this regard, the main contribution is to provide a method for extracting superpixels with greater adherence to the edges of the regions. The experimental tests will consider biomedical grayscales. The robustness and effectiveness will be verified by quantitative and qualitative results.