SAR image segmentation using co-occurrence matrix and slope magnitude

Authors:
H. B. Kekre;Saylee Gharge
Affiliations:
MPSTME, NMIMS University, Vile-Parle(w), Mumbai, India;MPSTME, NMIMS University, V.E.S.I.T, Mumbai, India
Venue:
Proceedings of the International Conference on Advances in Computing, Communication and Control
Year:
2009

Citing 2
Cited 1

A Computational Approach to Edge Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Edge and Curve Detection for Visual Scene Analysis

IEEE Transactions on Computers

Kekre's Fast Codebook Generation algorithm for tumor detection in mammography images

Proceedings of the International Conference and Workshop on Emerging Trends in Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Edge detector is fundamental issue in image analysis. Due the presence of speckle, which can be modeled as a strong, multiplicative noise, edge detection in synthetic aperture radar (SAR) images is extremely difficult. A common approach is to identify edges as local maxima of the gradient magnitude in the gradient direction. We here proposed a new method as an edge detector for SAR images. It computes actual magnitude of slope in horizontal as well as in vertical direction by using any edge operator and then the resultant of these gradient of slope obtained and image of slope magnitude is constructed. On this image of slope magnitude canny's edge operator is used for getting segmented image. The results are compared with cooccurrence matrix method using energy and correlation criteria. Experimental results obtained for SAR images are presented.