首页 | 官方网站   微博 | 高级检索  
     

基于Gabor滤波器和ICA技术的SAR目标识别算法
引用本文:胡风明,王满玉,范学花.基于Gabor滤波器和ICA技术的SAR目标识别算法[J].现代雷达,2011,33(6).
作者姓名:胡风明  王满玉  范学花
作者单位:1. 空军装备研究院装备总体论证研究所,北京,100076
2. 中国科学院大气物理研究所,北京,100029
摘    要:提出了一种基于Gabor滤波器和独立分量分析(ICA)技术对合成孔径雷达(SAR)目标识别的算法.该方法提取预处理后SAR图像的低频子带图像,利用Gabor滤波器组对该低频子带图像在不同方向和尺度上滤波,再用主成分分析(PCA)+ICA方法对Gabor滤波后图像提取有效特征向量作为目标识别特征,最后用支持向量机(SVM)对该特征进行分类完成目标识别.使用MSTAR数据库中3类SAR目标数据对该方法进行目标识别的仿真实验,平均识别率最高可达96.56%.通过与其他识别方法对比实验,验证了文中方法的有效性.

关 键 词:合成孔径雷达目标识别  低频子带图像  Gabor滤波器  独立分量分析  支持向量机

A Method of SAR Target Recognition Based on Gabor Filter and ICA
HU Feng-ming,WANG Man-yu,FAN Xue-hua.A Method of SAR Target Recognition Based on Gabor Filter and ICA[J].Modern Radar,2011,33(6).
Authors:HU Feng-ming  WANG Man-yu  FAN Xue-hua
Affiliation:HU Feng-ming1,WANG Man-yu1,FAN Xue-hua2(1.Research Institute of General Development and Evaluation of Equipment,Academe of Air Force,Beijing 100076,China)(2.Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029,China)
Abstract:A method for SAR target recognition using Gabor filter and ICA is presented in the paper.The low-frequency sub-band image extracted from the pre-processed SAR image is filtered by Gabor filter on different directions and scales.Then the effective feature vectors derived using PCA and ICA from Gabor filtered images are regarded as the target recognition feature,which could be used to recognize the targets with SVM.The proposed method is tested on MSTAR dataset for 3-type SAR target recognition and the highes...
Keywords:SAR target recognition  low-frequency sub-band image  gabor filter  ICA  SVM  
本文献已被 CNKI 万方数据 等数据库收录!
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-23

京公网安备 11010802026262号