首页 | 本学科首页   官方微博 | 高级检索  
     

基于多幅同目标图像和HMM的SAR图像目标识别
引用本文:宦若虹,杨汝良.基于多幅同目标图像和HMM的SAR图像目标识别[J].电子与信息学报,2008,30(9):2051-2054.
作者姓名:宦若虹  杨汝良
作者单位:1. 中国科学院电子学研究所,北京 100190;中国科学院研究生院,北京 100190
2. 中国科学院电子学研究所,北京,100190
摘    要:该文提出了一种基于多幅同目标图像和隐马尔可夫模型的合成孔径雷达图像目标识别方法.该方法通过小波域主成分分析提取目标图像特征向量,结合多幅不同方位角下的同目标图像的特征向量生成单幅图像的特征序列,用隐马尔可夫模型对特征序列进行识别.实验结果表明,该方法可明显提高目标的正确识别率,是一种有效的合成孔径雷达图像目标识别方法.

关 键 词:合成孔径雷达  隐马尔可夫模型  多幅同目标图像  识别
收稿时间:2007-3-12
修稿时间:2007-7-19

Synthetic Aperture Radar Image Target Recognition Based on Multi-Images of the Same Target and Hidden Markov Models
Huan Ruo-hong,Yang Ru-liang.Synthetic Aperture Radar Image Target Recognition Based on Multi-Images of the Same Target and Hidden Markov Models[J].Journal of Electronics & Information Technology,2008,30(9):2051-2054.
Authors:Huan Ruo-hong  Yang Ru-liang
Affiliation:(Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China)  (Graduate University of the Chinese Academy of Sciences, Beijing 100190, China)
Abstract:This paper presents a method for synthetic aperture radar images target recognition based on multi images of the same target and Hidden Markov Models (HMM). Feature vectors of target images are extracted with principal component analysis in wavelet domain. Feature sequences of an image are obtained from the feature vectors of multi images of the same target in different azimuths. Target recognition is carried out for feature sequences with the HMM. The experiments results show that the correctness of recognition is enhanced obviously with the proposed method, and it is an effective method for SAR images target recognition.
Keywords:Synthetic Aperture Radar (SAR)  Hidden Markov Model (HMM)  Multi images of the same target  Recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《电子与信息学报》浏览原始摘要信息
点击此处可从《电子与信息学报》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