一种有效的SAR图像目标识别方法  

Efficient Method of SAR Image Target Recognition

作　　者：薛笑荣[1] 章慧蓉[2] 赵荣椿[3] 

机构地区：[1]北京工业大学机械工程与应用电子技术学院,北京100022 [2]北京工业大学建筑工程学院,北京100022 [3]西北工业大学计算机学院,西安710072

出　　处：《北京工业大学学报》2007年第12期1262-1266,共5页Journal of Beijing University of Technology

基　　金：'十五'国防课题研究项目(413160103);北京工业大学博士科研启动基金(52001111200701)

摘　　要：根据不变矩特征提取和支撑矢量机分类的优势,提出了一种有效的SAR图像目标识别方法.首先对样本SAR图像进行预处理。然后提取目标区域的不变矩特征并计算灰度均值,将其组成特征向量训练SVM分类器,最后用训练好的SVM分类器对要识别的SAR图像进行目标识别.采用该方法对一些含有桥梁和坦克的SAR图像进行目标识别实验,取得了较好的识别结果.According to the advantages of invariant moments extraction and support vector machine （SVM） classification, an efficient method of SAR image target recognition is proposed. Firstly, training sets of SAR images are preprocessed. Then in the target fields of SAR image, seven Hu＇s invariant moments are extract- ed, and average gray level values are also calculated. Following that, by using the vectors composed of normalized invariant moments and average gray level values, SVM is trained. Finally, the testing sets of SAR images are recognized by trained SVM. In the experiments of recognizing bridges and tanks, better results have been obtained with the new method.

关 键 词：SAR图像 不变矩 支撑矢量机 目标识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]