Title: A modified fuzzy C means approach for segmenting the input flood images captured by synthetic aperture radar
Authors: N.V.S. Natteshan; Sureshkumar Nagarajan
Addresses: School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, India ' School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, India
Abstract: Image segmentation is a process of separating a homogenous area into a heterogeneous area. In this work the fuzzy C means algorithm is utilised for segmenting the flooded areas in the given input image based on intensity of images. But there is a problem of speckle noise occurring in the images due to backscattered echo from earth surface. Hence a fuzzy discontinuity adaptive weight based non-local means filtering is being used to eliminate the speckle noise. In the fuzzy C means segmentation algorithm a modification is proposed which is done as two steps namely the quantisation and aggregation. A modified fuzzy C means algorithm is required in order to reduce the convergence rate of the algorithm. This research work particularly focuses on the application of the fuzzy C means clustering algorithm for segmenting the flood occurred regions and further processing of the flood occurred regions will be carried out.
Keywords: fuzzy C means; de-speckling; clustering; quantisation; aggregation; segmentation.
International Journal of Intelligent Enterprise, 2023 Vol.10 No.2, pp.180 - 191
Accepted: 01 Jul 2020
Published online: 05 Apr 2023 *