Multilevel fuzzy partition segmentation of satellite images using GSA

Image segmentation in satellite image processing became an interesting topic in the past few years due to its utility in various fields such as natural calamity detection, object tracking, defense etc. A satellite image consists of several ill-defined and ambiguous areas. So, to deal with ambiguity in satellite image, this paper presents a fuzzy partition and maximum entropy based multilevel thresholding approach. The main objective of multilevel thresholding is to divide an image into several classes. This paper focuses on finding the optimal values of multiple thresholds such that entropy of fuzzy partitions in an image is maximized. Gravitational search algorithm (GSA), recently developed metaheuristic approach is incorporated to obtain optimal threshold in lesser time and results are compared with other metaheuristic techniques such as ABC, PSO and GA.