Fuzzy clustering of hyperspectral data based on particle swarm optimization

The unique capabilities of hyperspectral images in expressing the properties of phenomena of earth surface, guides the researches of this branch toward developing methods that so soon as possible decreases the need of interference human factor in processing data. However clustering is one of the most applicable methods in many of propounded processing in hyperspectral data. Nevertheless, paying attention to high dimension of these data, the traditional clustering such as FCM for these data has low efficiency and usually is trapped into local optima. The techniques of population based clustering because of random search, can overcome many problems of traditional clustering methods. One of these techniques which is inspired from group bird´s behaviour or fish is particle swarm optimisation (PSO). In this paper a hybridized method based on combining FCM and PSO is utilized. The result of using this method on hyperspectral data, in two spaces data and feature i.e. PCA shows its high ability than fuzzy clustering.