说明：  提出了一种基于线特征和SIFT点特征的多源遥感影像配准方法。该方法首先匹配待配准影像和参 考影像中的线特征,利用匹配直线构建虚拟角点 其次,针对传统SIFT算法匹配多源遥感影像特征点存在的 不足,采用线特征约束点特征的方法进行SIFT同名点对的提取 最后结合虚拟角点对及SIFT同名点对构建三角网进行小面元微分纠正。
(A line-based features and SIFT features multi-point sources of remote sensing image registration. This method first match when matching images and reference images in the line features, using the matching Zhixian construct the virtual corner Secondly, the traditional SIFT algorithm for multi-source remote sensing image matching feature points and its shortcomings, bound with thread features characteristic of the method for point SIFT extraction of the same name points Finally, a virtual corner and SIFT on the same name point to the building triangulation using small panel differential corrected.)