| 引用本文: | 项海兵， 吴涛， 张玉营， 盛佳佳. 机载SAR图像解译系统设计研究[J]. 雷达科学与技术, 2020, 18(3): 345-350.[点击复制] |
| XIANG Haibing, WU Tao, ZHANG Yuying, SHENG Jiajia. Research and Design of Airborne SAR Image Interpretation System[J]. Radar Science and Technology, 2020, 18(3): 345-350.[点击复制] |
|
| 摘要: |
| 针对机载SAR系统应用的海量SAR图像数据的存储管理、态势显示和快速解译等需求，本文提出了一种基于客户端/服务器端（C/S）架构的机载SAR图像解译系统的解决方案。本文介绍了该方案的系统软件架构和硬件架构，重点阐述多源异构海量数据管理、基于OSGEarth的地图和图像叠加显示、基于金字塔的大图像显示、基于深度学习的目标自动检测等关键技术；描述了数据管理、SAR图像显示和SAR图像目标检测识别等功能，其中海量数据的有序管理和查询功能，有效解决海量数据存储管理问题；SAR图像数据的地图叠加显示功能，为图像解译奠定基础；基于深度学习的智能检测识别等功能，满足SAR图像实时快速解译需求。 |
| 关键词: 机载SAR 图像解译 深度学习 金字塔 |
| DOI：DOI:10.3969/j.issn.1672-2337.2020.03.018 |
| 分类号:TN958;TP753 |
| 基金项目:空军“十三五”预研项目 |
|
| Research and Design of Airborne SAR Image Interpretation System |
|
XIANG Haibing, WU Tao, ZHANG Yuying, SHENG Jiajia
|
|
1.The 38th Research Institute of China Electronics Technology Group Corporation, Hefei 230088, China;2.Key Laboratory of Aperture Array and Space Application, Hefei 230088, China
|
| Abstract: |
| Aiming at the requirement of storage management, situation display and fast interpretation of massive SAR image data in airborne SAR system, this paper proposes a solution of airborne SAR image interpretation system based on client/server architecture. The system software architecture and hardware architecture of the scheme are introduced, focusing on the key technologies such as multi-source heterogeneous mass data management, OSGEarth-based map and image overlay display, pyramid-based large image display, and deep learning-based automatic target detection. The functions of data management, SAR image display, and SAR image target detection and recognition are described, among which, the orderly management and query functions of mass data can effectively solve the problem of mass data storage and management. The superposition display function of SAR image data lays a foundation for image interpretation. Intelligent detection and recognition based on deep learning can meet the demand of real-time and fast interpretation of SAR images. |
| Key words: airborne synthetic aperture radar image interpretation deep learning pyramid |
