Title :
SAR image change detection using modified statistical region merging
Author :
Han Zhang; Weiping Ni; Weidong Yan; Hui Bian; Junzheng Wu; Sha Li
Author_Institution :
Northwest Institute of Nuclear Technology, Xi´an China
Abstract :
A novel algorithm using modified statistical region merging (MSRM) for Synthetic Aperture Radar (SAR) image change detection is proposed in this paper. The statistical region merging algorithm is modified to cope with the speckles of SAR images. A new statistical predicate variable is introduced to control the merging process to end up with three classes in the final map. Comparative experiments show that the MSRM algorithm has the superiority in the change detection accuracy as well as the computational simplicity.
Keywords :
"Merging","Synthetic aperture radar","Image segmentation","Change detection algorithms","Clustering algorithms","Speckle","Probability"
Conference_Titel :
Computer Science and Network Technology (ICCSNT), 2015 4th International Conference on
DOI :
10.1109/ICCSNT.2015.7490976