一种新的自适应阈值SSDA算法  

New Adaptive Threshold SSDA Algorithm

作　　者：杜德生[1] 叶建平[1] 

机构地区：[1]哈尔滨理工大学,黑龙江哈尔滨150080

出　　处：《现代电子技术》2010年第6期135-136,139,共3页Modern Electronics Technique

摘　　要：在传统的序惯相似度检测算法和自适应阈值的序惯相似度检测算法的基础上,运用粗匹配中取三点最小值分别作为精匹配中的初始门限值,然后进行三次精匹配,比较三次匹配结果取最小值,转变为最终的图像匹配点,实现一种新的自适应阈值的序惯相似度检测算法。实验结果表明,这种改进自适应阈值的序惯相似度检测算法,在略微降低算法的执行速度的同时,能够很好地保持自适应阈值的序惯相似度检测算法的图像匹配准确性,具有很好的实时性。In the traditional sequential similarity detection algorithm and adaptive threshold sequential similarity detection algorithm, based on the use of coarse match, taking three matches in a fine match, respectively, the minimum initial threshold, and then refining for three matches to compare the results of three matches from the minimum, changing the images for the fi- nal match point,a new adaptive threshold of sequential similarity detection algorithm is realized,experimental results show that the improved adaptive sequential threshold similarity detection algorithm, the algorithm in a slightly lower speed at the same time,able to maintain the adaptive threshold of sequential similarity detection algorithm and to match accuracy of the images, it has very good real - time.

关 键 词：序惯相似度检测算法 自适应阈值 图像匹配 实时性 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]