摘要
提出一种基于分布式压缩感知(DCS)的重轨干涉SAR形变检测方法.DCS理论利用多次观测信号集的联合稀疏特性和相关性,对信号集进行联合重建.本文将DCS理论引入微波成像形变检测中,并对地基SAR复数据进行处理,利用相位数据检测场景形变,比较压缩感知(CS)算法和DCS算法在降采样条件下的重建结果.CS算法和DCS算法都具有保相性,在幅度和相位图像中可以很好地消除副瓣,成像效果比Omega-k算法好.基于DCS的稀疏微波联合观测系统可以利用多幅场景间回波数据的联合稀疏特性,进一步降低数据采集,实现准确重建和检测.
A model for repeat-pass interferometric synthetic aperture radar （InSAR） is proposed to detect deformation based on distributed compressed sensing （DCS）. DCS is applied to recover the signals from independent observations of multiple sensors, which are sparse in a transform domain and coherent with each other. We conduct a series of ground-based SAR experiments to compare the recovery performance using different imaging algorithms, such as compressed sensing （CS） and DCS. We find that images recovered by CS and DCS can preserve the phase information of the complex data and have better focus and lower sidelobes than those by Omega-k. Furthermore, by taking advantage of the joint sparsity among multiple scene echoes, DCS joint observation system can further reduce independent observations than CS and achieve scene reconstruction and deformation detection.
出处
《中国科学院大学学报（中英文）》
CSCD
北大核心
2016年第1期107-114,共8页
Journal of University of Chinese Academy of Sciences
关键词
形变检测
重轨干涉测量
干涉合成孔径雷达
分布式压缩感知
deformation detection
repeat track interferometry （RTI）
interferometric synthetic aperture radar （InSAR）
distributed compressed sensing （DCS）
