系统工程与电子技术 ›› 2019, Vol. 41 ›› Issue (8): 1735-1741.doi: 10.3969/j.issn.1001-506X.2019.08.09

• 传感器与信号处理 • 上一篇    下一篇

基于地形起伏和SVM的SAR景象MSA选取

苏娟1, 王延钊1,2, 伍薇1   

  1. 1. 火箭军工程大学核工程学院, 陕西 西安 710025;
    2. 火箭军士官学校测试控制系, 山东 潍坊 262500
  • 出版日期:2019-07-25 发布日期:2019-07-25

SAR scene MSA selection based on topographic relief and SVM

SU Juan1, WANG Yanzhao1,2, WU Wei1   

  1. 1. College of Nuclear Engineering, Rocket Force University of Engineering, Xi’an 710025, China; 
    2. Department of Measurement and Control, Rocket Force Sergeancy School, Wei Fang 262500, China
  • Online:2019-07-25 Published:2019-07-25

PDF (PC)

摘要：

为提高合成孔径雷达(synthetic aperture radar, SAR)景象匹配区(matchingsuitable areas,MSA)的匹配概率和匹配精度,提出基于支持向量机(support vector machine, SVM)的SAR景象MSA选取方法。首先结合SAR图像特性,选取基于数字高程模型、基于灰度和基于特征的3级特征参数;然后基于综合特征量,利用SVM训练样本数据,得到MSA决策函数。该方法充分考虑了SAR图像信息对MSA选取结果的影响及特征参数间的相关性,能够有效规划出高性能的匹配区。

关键词: 适配区选取, 合成孔径雷达, 数字高程模型, 支持向量机

Abstract:

In order to improve the matching probability and matching accuracy of synthetic aperture radar (SAR) scene matching-suitable area (MSA), a SAR scene MSA selection method based on support vector machine(SVM) is proposed. Firstly, digital elevation modelbased, intensitybased and featurebased measure parameters are selected based on the characteristics of SAR images. Then based on the integrated feature vectors, SVM is employed to training sample data, and decision function which could separate suitable and unsuitable-matching area class is obtained. This method fully considers the impact of SAR image information on the results of the selecting MSA and the correlation between the characteristic parameters. Experimental results show that this method can effectively single out high performance MSA.

Key words: matching-suitable area (MSA) selection, synthetic aperture radar (SAR), digital elevation model, support vector machine (SVM)

京ICP备11035014号-5
版权所有 © 2019 北京航天情报与信息研究所《系统工程与电子技术》编辑部
通讯地址：北京市142信箱32分箱《系统工程与电子技术》编辑部(100854)
电　话：(010)68388406/68386014    E-mail：xtgcydzjs@126.com(系统工程与电子技术)/jseeoffice@126.com(JSEE)
本系统由北京玛格泰克科技发展有限公司设计开发