一种改进的遥感图像地标边缘匹配方法

摘要

The distance transformation algorithm for edge matching is an effective landmark matching method. The disad-vantages of the traditional distance transformation algorithms are that searching processes is redundant, and the computation efficiency is poor. The new distance transformation algorithm, which cuts the redundant searching processes by adopting the steps of expanding edge-distance and setting approximate distance search threshold, is presented. Simulation results show that, the proposed algorithm is efficient for computation and has the same accuracy of matching as the traditional one.%基于距离变换的边缘匹配算法是一种有效的地标匹配算法，针对传统算法中重复冗余的计算和搜索效率低的问题，采用边缘距离扩展和设置相似门限的方法，对该算法进行了计算和搜索的优化。仿真实验表明，改进算法在保证匹配精度的基础上，大大降低了算法的计算量，提高了算法的搜索效率。