Remote sensing image segmentation using feature based fusion on FCM clustering algorithm

Abstract Image segmentation of heterogeneous comparable objects lying beneath the earth’s surface is a fundamental but challenging research area in remote sensing. Learning approaches are used sensing image to improve accuracy at expense time and large amount data, their performance need be finely classified due information diversity constraints. In this work, we proposed an novel feature based fuzzy C -means-extreme learning machine (FBFCM-ELM) algorithm for which classification on entropy, intensity, edge features performed such way that it updates intensity value preserve most local characteristics while still being able clearly distinguish image’s boundaries by assigning pixel values each cluster peak cluster’s sub-histogram. Using FBFCM, extracted as reliable samples ELM training. Undetermined segmented pixels obtained using trained classifier. Experiments over number images confirmed method yields better RGB image, evidenced observable details, edges, improved appearance resembles ground truth outperforms state-of-the-art algorithms.