基于不规则三角网的遥感影像自动配准系统设计  

Design of Remote Sensing Image Automatic Registration System Based on Irregular Triangular Network

作　　者：陈振庭 陈富强 许丽娟 CHEN Zhenting;CHEN Fuqiang;XU Lijuan(School of Artificial Intelligence,Guangzhou Huashang College,Guangzhou 511300,China)

机构地区：[1]广州华商学院人工智能学院,广州511300

出　　处：《计算机测量与控制》2025年第4期232-240,共9页Computer Measurement &Control

基　　金：广州华商学院校内导师制科研基金资助项目(2024HSDS12)。

摘　　要：在遥感影像配准中,因多种因素导致影像存在几何畸变、平均正确配准点率低的问题;为解决此问题,利用不规则三角网技术设计遥感影像自动配准系统;通过改装采集器、传感器、处理器,调整电源电路连接方式优化硬件系统保障数据安全可靠;基于采集器光学原理生成影像并经大气校正、影像提高保证质量;采用Harris算子提取特征点构建不规则三角网实现影像自动分割纠正偏移;经粗配准、精配准和误配准剔除三步实现自动配准功能;经系统测试实验,与传统系统相比,优化后的系统对低、高分辨率遥感影像的正确配准点率分别提高3.5%和1.9%,系统运行加速比明显提升。In remote sensing image registration,there are problems such as geometric distortion and low average correct registration point rate due to various factors.To solve this problem,an automatic registration system for remote sensing images is designed using irregular triangulation technology.By modifying the collector,sensor,and processor,adjusting the power circuit connection method,optimizing the hardware system to ensure data security and reliability;Generate images based on the optical principles of the collector and ensure quality through atmospheric correction and image enhancement;Using Harris operator to extract feature points and construct an irregular triangular network to achieve automatic image segmentation and correction of offset;The automatic registration function is achieved through three steps:coarse registration,fine registration,and error registration removal.According to system testing experiments,compared with traditional systems,the optimized system has improved the correct registration rate of low and high-resolution remote sensing images by 3.5%and 1.9%respectively,and the system's acceleration ratio has been significantly improved.

关 键 词：不规则三角网 遥感影像 遥感技术 影像配准 自动配准系统 

分 类 号：TP237[自动化与计算机技术—检测技术与自动化装置]