SAR image segmentation for land cover change detection

Authors:

Abstract
Synthetic aperture radar is used for land cover change detection which can either be mounted on a drone or an aircraft, spacecraft. It can be used for land cover change detection by comparing two images which are taken at different intervals of time. For this we are using differencing methods. The operators used in differencing methods are log ratio and mean ratio. For obtaining a better difference in image. The image fusion technique is applied using complementary information obtained from log ratio and mean ratio differenced image then the image is segmented using k means clustering algorithms in which k centroids are placed as one for each cluster such that they are at maximum distance away from each other. The image obtained is compared with the ground truth, which has been already implemented. Now if the image is normal there is no change, and if the image is abnormal then the change has occurred during that time interval.

This publication has 22 references indexed in Scilit: