多特征辅助的SAR-GMTI雷达扩展动目标检测方法  

A Multi-features-Aided Method for Extended Target Detection of SAR-GMTI

作　　者：杨志伟[1] 张蕊[1] 许华健 廖桂生[1] 杨剑[1] 

机构地区：[1]西安电子科技大学雷达信号处理重点实验室,陕西西安710071

出　　处：《电子学报》2016年第12期2916-2923,共8页Acta Electronica Sinica

基　　金：国家自然科学基金(No.61501471);国防预研基金(No.9140xxx005)

摘　　要：在远程监视模式下,高分辨率SAR-GMTI雷达系统面临目标扩散、信噪比低等挑战,传统基于点目标检测方法性能恶化明显.本文提出一种目标形状与阴影辅助的高分辨率SAR-GMTI雷达扩展动目标检测方法.该方法首先对杂波抑制后的残差图聚类并生成目标形状信息;然后依据目标阴影形成的几何模型计算与目标形状相匹配的阴影模板,并利用目标径向速度估计值生成阴影位置匹配条件;最后综合利用阴影形状、位置及幅度信息,剔除不符合匹配条件的虚假目标,降低虚警.仿真验证了所提方法对低信噪比扩展目标检测性能的改善.This paper deals with the target-detection issue that a high-resolution synthetic aperture radar （SAR）- ground moving target indication （GMTI） system under remote surveillance mode faces many challenges, i. e. , target-diffu- sing and low signal-to-noise ratio, and based on the fact that extended targets with conventional target detection methods based on the poin target cause decrease in detection performance, a Shape-Shadow-Aided Moving Target Detector（ S2A- MTD ） is proposed for extended target detection of airborne multichannel SAR-GMTI. First, the shaped information of a tar- get is obtained when the residual image with adaptive clutter suppression clustered. Then, shadow-template matched with shaped information of a target is evaluated, which is based on the model of the geometric relationship on a moving extended target and its shadow. Meanwhile, according to estimative radial velocity, position-matched condition with shadow is genera- ted. Finally, taking all the information in shape, position and amplitude of shadow into consideration is to eliminate false tar- gets that not satisfy with matching conditions, so as to reduce false alarms, the simulation results show the validity that this method improves the detection performance of the extended target under low signal-to-noise ratio.

关 键 词：合成孔径雷达 地面运动目标检测 扩展目标 目标形状与阴影辅助 模板匹配 

分 类 号：TN957[电子电信—信号与信息处理]