J4 ›› 2010, Vol. 37 ›› Issue (6): 1071-1076.doi: 10.3969/j.issn.1001-2400.2010.06.016

• 研究论文 • 上一篇    下一篇



  1. (西安电子科技大学 智能感知与图像理解教育部重点实验室和智能信息处理研究所,陕西 西安  710071)
  • 收稿日期:2009-11-13 出版日期:2010-12-20 发布日期:2011-01-22
  • 通讯作者: 刘汉强
  • 作者简介:刘汉强(1981-),男,西安电子科技大学博士研究生,E-mail: maxliuhq@hotmail.com.
  • 基金资助:


SAR image multilevel segmentation based on local similarity measure

LIU Han-qiang;JIAO Li-cheng;ZHAO Feng   

  1. (Ministry of Education Key Lab. of Intelligent Perception and Image Understanding,
     Research Inst. of Intelligent Information Processing, Xidian Univ., Xi'an  710071, China)
  • Received:2009-11-13 Online:2010-12-20 Published:2011-01-22
  • Contact: LIU Han-qiang



关键词: 图像分割, 合成孔径雷达图像, 相似性测度, 谱聚类, 权核k均值


Aiming at the high computational complexity of spectral clustering algorithms and their unsuitablity to synthetic aperture radar image segmentation, a novel SAR image multilevel segmentation based on local similarity measure is proposed which utilizes the equivalence of spectral clustering and weighted kernel k-means. First, the wavelet texture features of every pixel are extracted from the image and the sparse adjacent matrix among pixels is constructed by computing the local scale parameter of each pixel, and then multilevel merging, basic clustering and multilevel refining operations are used to cluster the image pixels, and finally the image segmentation result is obtained. Experimental results on artificial texture images and SAR images show that the proposed method can avoid the sensitivity of traditional spectral clustering to the scale parameters and obtain a better segmentation performance.

Key words: image segmentation, synthetic aperture radar (SAR) image, similarity measure, spectral clustering, weighted kernel k means