针对合成孔径雷达深度展开网络成像方法研究  

Research on Depth Unfolding Network Imaging Method for Synthetic Aperture Radar

作　　者：李文静[1] LI Wenjing(Beijing University of Chemical Technology,Beijing,100029 China)

机构地区：[1]北京化工大学,北京100029

出　　处：《科技资讯》2025年第1期39-42,共4页Science & Technology Information

基　　金：北京化工大学科学技术发展研究院——中央高校基本科研业务费自由探索项目“12060098070/24年自由探索项目”(项目编号:ZY2427)。

摘　　要：在灾害监控和遥感制图等多个领域,合成孔径雷达(Synthetic Aperture Radar,SAR)成像技术因其不受天气和光照条件限制的优势而被广泛应用。近年来,深度学习技术与SAR成像的结合,尤其是SAR深度展开网络成像技术,已经成为该领域的研究热点。它不仅能够高效地整合大量的历史SAR数据,还能够融合传统成像方法中的丰富先验信息,使得即使面对严重欠采样的情况,也能够实现高分辨率、宽视场的SAR图像精确重建,从而显著减少了对数据采集、储存和传输的资源需求。从深度展开网络构建原理出发,对深度展开网络构建方法进行分析,并提出基于先验分布的SAR展开网络成像方法与基于图像特征的SAR展开网络成像方法,希望能够为我国SAR成像提供一定参考。Synthetic Aperture Radar(SAR)imaging technology is widely used in various fields such as disaster monitoring and remote sensing mapping due to its advantage of not being limited by weather and lighting conditions.In recent years,the combination of deep learning technology and SAR imaging,especially SAR depth unfolding network imaging technology,has become a research hotspot in this field.It not only efficiently integrates a large amount of historical SAR data,but also integrates rich prior information from traditional imaging methods,enabling accurate reconstruction of high-resolution and wide field of view SAR images even in the face of severe undersampling,thereby significantly reducing the resource requirements for data acquisition,storage,and transmission.Starting from the principle of depth unfolding network construction,this article analyzes the construction methods of depth unfolding networks and proposes SAR unfolding network imaging methods based on prior distribution and image features,hoping to provide some reference for SAR imaging in China.

关 键 词：合成孔径雷达 深度展开 网络成像 欠采样成像 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]