SAR image segmentation based on Kullback-Leibler distance of edgeworth

Authors:
Lei Hu;Yan Ji;Yang Li;Feng Gao
Affiliations:
Digital Media Laboratory, School of Computer Science and Engineering, Beihang University, Beijing, China and School of Computer and Information Engineering, Jiangxi Normal University, Nanchang, Ch ...;Digital Media Laboratory, School of Computer Science and Engineering, Beihang University, Beijing, China;Digital Media Laboratory, School of Computer Science and Engineering, Beihang University, Beijing, China;Digital Media Laboratory, School of Computer Science and Engineering, Beihang University, Beijing, China
Venue:
PCM'10 Proceedings of the 11th Pacific Rim conference on Advances in multimedia information processing: Part I
Year:
2010

Citing 4
Cited 0

A segmentation algorithm for SAR images based on the anisotropic heat diffusion equation

Pattern Recognition
SAR image segmentation based on mixture context and wavelet hidden-class-label Markov random field

Computers & Mathematics with Applications
Multiscale segmentation and anomaly enhancement of SAR imagery

IEEE Transactions on Image Processing
Minimum description length synthetic aperture radar image segmentation

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new segmentation method based on Kullback-Leibler distance (KLD) of Edgeworth is proposed to accurately segment synthetic aperture radar (SAR) images into homogeneous regions and reduce the over-segmentation phenomenon. The proposed method uses a coarse-to-fine scheme. In the coarse phase, the SAR image is divided into fragments based on KLD, in which the Edgeworth expansion is employed to represent SAR data. In the fine phase, the divided fragments with the same shape or texture are merged in order to achieve an integrated segmentation. Experiments are performed based on highresolution satellite SAR images and the experimental results demonstrate the efficiency of the proposed method.