ENVIRONMENTAL APPLICATIONS: EFFECTIVE ANALYSIS OFLAND CLASSIFICATION FOR CHANGE DETECTION OF SATELLITE IMAGES IN CHENNAI REGION, INDIAN. Prabhakaran, S.S. Ramakrishnan and N.R. Shanker
Remote sensing image needs segmentation for various end-user applications such as urban planning encroachment, and change detection. Image segmentation segments the satellite image into many regions.Satellite image segmentation plays a vital role in change detection. The change detection algorithms for satellite images provide different results for different resolution image. Still the change detection algorithm is a challenging task in terms of accuracy, delineation, optimization and ground truth verification. We apply a novel cat optimization method to different resolution image from earth observation, Landsat and sentinel satellite imagery. The cat optimization algorithm segment the image from best to worst range of segmentation for a single image and differentiates the region in the image for delineation of vegetation,urban, hilly region and water bodies. From the experimental result, the ground truth verification of cat optimization algorithm performs 85% more than the existing hybrid algorithm.
Enter your contact information below to receive full paper.