Conference Held at Dr. M.G.R. Educational and Research Institute, University, India Multi Scale Analysis and Change Detection in SAR Images Based on Modified MRF Energy Function for Disaster Management

In this approach it classifies image change detection process for analyzing disaster with the help of synthetic aperture radar images. It determines difference between two SAR images taken at different times. it detect difference through pixels by pixels based on change and unchanged regions by using fuzzy c- means clustering with a novel Markov Random Fields(MRF) energy function. Its main objective function is to detect real changes between two images, less time consuming, remove dot spots and patches from the images. It also modifies the member of each pixel using MRF energy function through which neighbor of each pixels and their relationship are much concerned. Index Terms—Fuzzy c mean clustering, image change detection, Markov random field (MRF), synthetic aperture radar (SAR). I. INTRODUCTION Many researches is based on remote sensing. Today, many problems in remote sensing are related to image change detection, it perform characterization and localization of areas that have evolved between two times from sequence of observation. It detects changes in images based on Markov random field technique and fuzzy c mean clustering algorithm. It gets more information by grouping similar elements together. Example such as remote sensing medical diagnosis, video surveillance, military purpose, risk management, disaster analysis updating maps and monitoring vegetation . The Images that are generated by synthetic aperture radars (SAR) having great use, due to their independence of atmospheric. They have become valuable sources of information in change detection. Change detection in SAR images is the process of the analysis of two co registered SAR images that are acquired over the same geographical area at different times. Such analysis is unsupervised when it aims to discriminate between two opposite classes with no prior knowledge about the scene.