A New Approach to Unsupervised Target Classification for Polarimetric SAR Images

Xing RONG  Weijie ZHANG  Jian YANG  Wen HONG  

IEICE TRANSACTIONS on Communications   Vol.E91-B   No.6   pp.2081-2084
Publication Date: 2008/06/01
Online ISSN: 1745-1345
DOI: 10.1093/ietcom/e91-b.6.2081
Print ISSN: 0916-8516
Type of Manuscript: LETTER
Category: Sensing
polarimetric SAR,  unsupervised classification,  Inhomogeneous Markov Random Field,  Graph Cuts,  

Full Text: PDF(341.7KB)>>
Buy this Article

A new unsupervised classification method is proposed for polarimetric SAR images to keep the spatial coherence of pixels and edges of different kinds of targets simultaneously. We consider the label scale variability of images by combining Inhomogeneous Markov Random Field (MRF) and Bayes' theorem. After minimizing an energy function using an expansion algorithm based on Graph Cuts, we can obtain classification results that are discontinuity preserving. Using a NASA/JPL AIRSAR image, we demonstrate the effectiveness of the proposed method.