基于CenSurE特征的SAR景象匹配技术研究-信号与信息处理专业论文.docxVIP

万方数据 万方数据 Abstract SURF algorithm has a good matching performance in SAR scene matching applications. Then we introduce the CenSurE algorithm, and make a deep analysis to the characteristic process of the CenSurE algorithm. In order to simplify algorithm, we introduced a new descriptor constructing in a sub-ring region , and then we design a improved CenSurE-based scene matching technical solution. The experimental results show that, CenSurE has a good invariant performance to image deformation capability and it has good real-time performance in the SAR scene matching. Finally, we made a detailed experimental and performance evaluation on SAR scene matching with the three algorithms, SIFT, SURF and improved CenSurE respectively. The experimental results show that the improved CenSurE algorithm has a higher performance than SIFT and SURF on the real-time requirement, and the matching precision is also better than SIFT and SURF in a small rotation angle, the improved CenSurE algorithm can better meet the SAR scene matching aided navigation project, it has a valued engineering application. Key Words: SAR; scene matching; SIFT; SURF; CenSurE; local features III 目录 目录 目 录 摘 要..............................................................................................................................1 Abstract ......................................................................................................................... II 目 录............................................................................................................................IV 第一章 绪论..................................................................................................................1 1.1 课题研究背景及意义.....................................................................................1 1.2 SAR 景象匹配技术的发展及研究现状 4 1.3 SAR 景象匹配导航技术的关键问题 6 1.4 论文主要工作安排.........................................................................................8 第二章 基于局部特征的 SAR 景象匹配技术研究..................