首页 | 本学科首页   官方微博 | 高级检索  
     

基于低秩结构的宽角SAR稀疏成像
引用本文:赵曜,曹翔宇,全相印,崔莉,许洪玮.基于低秩结构的宽角SAR稀疏成像[J].信号处理,2022,38(12):2532-2541.
作者姓名:赵曜  曹翔宇  全相印  崔莉  许洪玮
作者单位:1.广东工业大学信息工程学院, 广东 广州 510006
基金项目:广东省自然科学基金2021A1515012009国家自然科学基金61907008
摘    要:在合成孔径雷达(Synthetic Aperture Radar,SAR)成像中,当方位向合成孔径较大时,观测区域内目标的电磁特征会表现为各向异性,导致基于各向同性假设的传统SAR成像方法不再适用。为此,宽角SAR成像方法通过将宽孔径划分为多个子孔径,利用每个子孔径对应的回波数据单独成像,实现对目标雷达图像的多角度重构。由于目标的散射特性在相邻子孔径中通常不会发生较大改变,每个子孔径的强散射中心分布高度相似,使得宽角SAR的成像结果具有低秩结构,即相邻子孔径对应的目标成像结果的支撑集相近。为了使用这种相关性,将Karhunen Loeve (KL)变换引入到宽角SAR成像过程中,再利用目标强散射中心分布的稀疏特性,建立基于低秩结构的宽角SAR稀疏成像模型。采用增广拉格朗日法(Augmented Lagrangian,AL)和交替方向乘子(Alternating Direction Method of Multipliers,ADMM)对上述成像模型进行迭代求解,从而获得宽角SAR成像结果。相比于传统的宽角SAR成像方法,本文所述方法不仅能提高目标后向散射系数的重建精度,还能有效抑制旁瓣效应与信号噪声对成像质量的影响,对目标电磁特征的多角度恢复具有更好的效果。 

关 键 词:??宽角合成孔径雷达    低秩结构    稀疏特性    Karhunen  Loeve变换
收稿时间:2022-06-16

WASAR Sparse Imaging Method Based on Low-Rank Structures
Affiliation:1.Guangdong University of Technology School of Information Engineering, Guangzhou, Guangdong 510006, China2.China Academy of Launch Vehicle Technology, Beijing 100076, China3.Beijing Institute of Remote Sensing, Beijing 100192, China
Abstract:? ?When the synthetic aperture in azimuth exceeds a certain angle range, the electromagnetic characters of observed targets generally represent anisotropic in synthetic aperture radar (SAR) imaging. And it leads to the failure of the conventional SAR imaging methods, which are derived from the hypothesis of isotropic targets. By dividing the wide aperture into some subapertures, the wide angle synthetic aperture radar (WASAR) can use the echo data in subapertures to achieve radar imaging of the targets from different angles. Since the scattering properties of radar targets are similar in the adjacent subapertures, which results in the similarity of the scattering-center distribution corresponding to every subaperture, the subaperture imaging results of WASAR have the similar support. And the specific property is called low-rank structures. In order to take advantage of the property, Karhunen Loeve (KL) transform is introduced into the imaging of WASAR. Meanwhile, we construct the WASAR sparse imaging model based on low-rank structures by the sparsity of the distribution of strong scattering centers. Then, augmented Lagrangian method and alternating direction method of multipliers are utilized to solve the model iteratively, so that the imaging results of WASAR can be obtained. Compared with the conventional WASAR imaging methods, the proposed algorithm in this paper not only improves the reconstruction precision of the backscattering coefficients of targets, but also suppresses the negative influence of sidelobe and noise on the image quality. The experimental results have also shown that it is able to recover the electromagnetic characters of targets in multiple angles better. 
Keywords:
点击此处可从《信号处理》浏览原始摘要信息
点击此处可从《信号处理》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号