Journal of Systems Engineering and Electronics ›› 2011, Vol. 33 ›› Issue (5): 1036-.doi: 10.3969/j.issn.1001-506X.2011.05.15

• 传感器与信号处理 • 上一篇    下一篇

一种极化SAR图像模糊分类方法

张涛1,2, 孙建涛1,2, 杨汝良1   

  1. 1. 中国科学院电子学研究所, 北京 100190; 2. 中国科学院研究生院, 北京 100049
  • 出版日期:2011-05-25 发布日期:2010-01-03

Fuzzy classification of polarimetric SAR images

ZHANG Tao1,2, SUN Jian-tao1,2, YANG Ru-liang1   

  1. 1. Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;
    2. Graduate School, Chinese Academy of Sciences, Beijing 100049, China
  • Online:2011-05-25 Published:2010-01-03

摘要:

针对极化合成孔径雷达图像模糊非监督分类问题,给出了一种改进的极化合成孔径雷达图像模糊分类方法。该方法通过引入极化总功率参数——span,改进极化合成孔径雷达图像模糊H/α分类方法,进行极化合成孔径雷达数据模糊H/α/span非监督分类。利用机载极化合成孔径雷达数据进行实验。实验结果表明,改进的方法提高了分类性能,聚类中心更为合理。

Abstract:

For the problem of unsupervised classification of polarimetric SAR images based on fuzzy theory, an improved fuzzy classification method of polarimetric SAR images is presented. The fuzzy H/α  classification method is improved by introducing the total polarimetric power parameter —span. The polarimetric SAR image is classified by the fuzzy H/α/span method. A classified experiment for airborne polarimetric SAR data is carried out by this method. Experimental results indicate that this method can improve the performance of classification and the cluster centers are more reasonable.