A New Method for Image Segmentation

Image processing is a form of signal processing. One of the typical operations on image processing is image segmentation. In this paper, the authors use new image segmentation algorithms based on information bottleneck method. Here they are going to use three algorithms; first they introduce the split-and-merge algorithm, where an image is segmented into set of regions (input) and the intensity histogram bins (output) is obtained. The second algorithm is the histogram clustering algorithm, where the input variable represents the histogram bins and the output is given by the set of regions. Finally, the registration based segmentation for two registered multimodal images.