利用雷达散射截面积序列的特定航天器工作平台分类识别方法  

An Identification Method for Specific Spacecraft Platform Using Radar Cross Section Sequence

作　　者：牛威[1] 苏威[2] 郝培杰[2] 李少敏[2] 

机构地区：[1]中国西安卫星测控中心宇航动力学国家重点实验室,西安710043 [2]中国西安卫星测控中心,西安710043

出　　处：《西安交通大学学报》2011年第12期53-58,共6页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金资助项目(61179010)

摘　　要：针对三维运动特性及雷达散射截面积(RCS)的姿态敏感性等因素影响航天器在轨运行识别的问题,结合特定工作平台RCS测量序列的典型特征,提出将每个观测弧段的RCS序列视为图像来提取边缘、骨架和密度等特征,再利用基于Hausdorff距离的图像匹配方法与特定工作平台上的航天器模板进行匹配识别,从而有针对性地实现了航天器的分类识别.实测结果表明,所提方法的分类正确率可高达85%,识别效果明显优于传统的统计和小波变换的方法.A method for identifying specific spacecraft is proposed to improve the problem that statistic characteristics or various translation characteristics of RCS（radar cross section） sequence are inconsistent and can not be used to correctly identify spacecrafts which is caused by the influence of the three-dimensional movement characteristic of the spacecraft and the attitude sensitivity of RCS.The proposed method bases upon analyses on the RCS sequence,and identifies specific platforms using narrow-band radar RCS sequence.The method extracts edge,skeleton,density and other characteristics of arcs by regarding the RCS sequence of each arc as an image,and uses an image matching method based on Hausdorff distance to classify and identify spacecrafts,so that spacecrafts can be effectively identified on specific platforms.Test results show that the proposed method achieves a 85% accurate rate for the specific spacecraft platform identification,and is much better than traditional statistics and wavelet decomposition methods.

关 键 词：图像匹配 航天器模板 匹配识别 

分 类 号：TN95[电子电信—信号与信息处理]