Fuzzy clustering algorithms incorporating local information for change detection in remotely sensed images

In this paper wk have used two fuzzy clustering algorithms, namely Fuzzy C-Means (FCM) and Gustafson Kessel Clustering (GKC) for unsupervised change detection in multitemporal remote sensing images. In conventional FCM & GKC no spatio-contextual information is taken into account and thus the result is not so much robust to noise/outliers. By incorporation of local neighborhood informationthe performanceof the algorithms is enhanced. In this work we have used two different techniques for incorporation of local information. Change detection maps are obtained by separating the pixelpatterns of the difference image into two groups. To show the effectivenessof the proposedtechnique, experiments are conductedon three multispectral and multitemporal remote sensing images. Results are comparedwith those of existing Markov Random Field (MRF) & neural network based algorithms and are found to be superior. The proposed technique is less time consuming and unlike MRF does not need any a priori knowledgeof distribution of changed and unchanged pixels (as required byMRF).