首页 | 本学科首页   官方微博 | 高级检索  
     

结合多特征和SVM的SAR图像分割
引用本文:钟微宇,沈 汀.结合多特征和SVM的SAR图像分割[J].计算机应用研究,2013,30(9):2846-2851.
作者姓名:钟微宇  沈 汀
作者单位:1. 中国科学院对地观测与数字地球科学中心, 北京 100094; 中国科学院大学, 北京 100049
2. 中国科学院对地观测与数字地球科学中心,北京,100094
摘    要:为实现灰度共生矩阵(GLCM)多尺度、多方向的纹理特征提取, 提出了一种结合非下采样轮廓变换(NSCT)和GLCM的纹理特征提取方法。先用NSCT对合成孔径雷达(SAR)图像进行多尺度、多方向分解; 再对得到的子带图像使用GLCM提取灰度共生量; 然后对提取的灰度共生量进行相关性分析, 去除冗余特征量, 并将其与灰度特征构成多特征矢量; 最后, 充分利用支持向量机(SVM)在小样本数据库和泛化能力方面的优势, 由SVM完成多特征矢量的划分, 实现SAR图像分割。实验结果表明, 基于NSCT域的GLCM纹理提取方法和多特征融合用于SAR图像分割, 可以提高分割准确率, 获得较好的边缘保持效果。

关 键 词:合成孔径雷达  图像分割  非下采样轮廓变换  灰度共生矩阵  支持向量机  特征选择  多特征融合

Multiple features and SVM combined SAR image segmentation
ZHONG Wei-yu,SHEN Ting.Multiple features and SVM combined SAR image segmentation[J].Application Research of Computers,2013,30(9):2846-2851.
Authors:ZHONG Wei-yu  SHEN Ting
Affiliation:1. Center for Earth Observation & Digital Earth, Chinese Academy of Sciences, Beijing 100094, China; 2. University of Chinese Academy of Sciences, Beijing 100049, China
Abstract:In order to implement multi-scale and multi-directional texture extraction, this paper proposed a texture feature extraction algorithm, which combined the nonsubsampled contourlet transform(NSCT) and gray level co-occurrence matric(GLCM). Firstly, it translated the SAR image to be segmented via NSCT. Then, it computed the gray co-occurrence features via GLCM for the decomposed sub-bands, and selected the features extracted by correlation analysis to remove redundant features. Meanwhile, it extracted gray features to constitute a multi-feature vector with the gray co-occurrence features. Finally, making full use of advantages of resolving the small-sample statistics and generalizing ability of support vector machines(SVM), it used SVM to divide the multi-feature vector to segment the SAR image. Experimental results show that the proposed method for SAR image segmentation can improve segmentation precision, and obtain better edge preservation results.
Keywords:SAR  image segmentation  NSCT  GLCM  SVM  feature selection  multiple features fusion
本文献已被 万方数据 等数据库收录!
点击此处可从《计算机应用研究》浏览原始摘要信息
点击此处可从《计算机应用研究》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号