遥感影像配准中控制点的自动提取  

Automatic selection of control point based on image matching

作　　者：李国胜[1] 张继贤[1] 宋伟东[1] 皇甫洁[1] 

机构地区：[1]石油大学资源与信息学院,山东东营257061

出　　处：《辽宁工程技术大学学报（自然科学版）》2005年第1期41-44,共4页Journal of Liaoning Technical University (Natural Science)

摘　　要：针对遥感影像的处理应用中,传统的手工选取控制点速度慢、精度低和强度大等缺点的问题,提出了一种基于影像灰度和影像特征混合自动提取控制点的新方法。在进行控制点的自动提取时,该方法速度快、错误匹配率低、精度高。方法内容包括特征点提取、构造影像金字塔以及基于金字塔数据结构的互相关匹配。The enormous increase in the volume of remotely sensed data is acquired by an ever growing number of earth observation satellites means that automation now plays a more vital role than ever before in application of remote sensing. However, traditional method manually selected control point is laborious, tedious and complex task. To speed the selection of control-point, a mixed algorithm on automatic selection of control point is proposed. The main contents include extraction of feature points, generation of pyramid image structure and similarity extent computation based on a pyramid image structure.

关 键 词：自动提取 遥感影像 控制点 特征点提取 配准 数据结构 互相关 造影 影像特征 处理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP751[自动化与计算机技术—计算机科学与技术]