基于时频分析的成像小弱运动目标检测方法  

Dim moving target detection method based on time-frequency analysis

作　　者：李正周[1] 田蕾[1] 郑微[1] 张月华[1] 金钢[2,3] 

机构地区：[1]重庆大学通信工程学院,重庆400044 [2]中国空气动力研究与发展中心,四川绵阳621000 [3]中国科学院光电技术研究所,成都610209

出　　处：《强激光与粒子束》2011年第9期2292-2296,共5页High Power Laser and Particle Beams

基　　金：中央高校基本科研业务费项目(CDJZR10160004)

摘　　要：针对传统空时域滤波器难以剔除杂波(尤其是杂波边缘)的缺点,提出了一种基于时频分析的弱小运动目标检测方法。理论分析表明,目标经过处的时频是一个小波包,波包的幅度与目标的幅度一致,而宽度则与目标速度成反比关系;杂波边缘经过处的时频则是一个上坡或者下坡。采用两级滤波方法检测运动目标,即首先采用恒虚警率准则过滤噪声,然后再利用目标出现处存在的波包,分别统计主瓣与旁瓣能量及其能量比,去除杂波,检测出运动目标。云背景下弱小目标检测试验结果证明了该方法的有效性。For it is hard to remove clutters,especially the clutter edge,with conventional high-pass filters in spatial or temporal domain,a dim moving target detection method based on the time-frequency characters of target,noise and clutter is presented.Theoretical analysis shows that the magnitude at the target location in time-frequency domain is a small wave packet,whose magnitude is consistent with the target amplitude,and width is inversely proportional to target speed;the magnitude at clutter edge in time-frequency domain is an uphill or downhill.This target detection method uses two-stage filtering to detect dim target.At first,a threshold based on false alarm ratio criteria in time-frequency domain is adopted to remove noise,and then the energy ratio between main lobe and side lobe in the wave packet is counted to remove clutter and detect target.The experiment of dim target detection under cloud background is conducted and the result shows that the method is effective.

关 键 词：小弱运动目标 杂波 时频分析 短时加窗傅里叶变换 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]