Performance of fuzzy based clustering algorithms for the segmentation of satellite images — A comparative study

Segmentation is the process of partitioning or classifying an image into some meaningful classes or clusters or segments. The segmentation process based on any one characteristics of the image such as texture, color or intensity. The images received from the satellite contains huge amount of information to process and analyze. It is possible to extract the or identify the object or regions of interest in the image from the segmentation results. The segmentation process is very useful to the subsequent image analysis. Many approaches have been proposed for the segmentation of satellite images, but fuzzy based approaches are most popular and widely used because they have a good performance in a large class of images. In this paper, the fuzzy based clustering approaches Fuzzy-C-Means (FCM) Clustering, Possibilistic C Means (PCM) and Possibilistic Fuzzy C Means (PFCM) are compared and the performance of these algorithms were tested with number of satellite images.