基于ransac的sift匹配阔值自适应估计-计算机科学.pdfVIP

第 44 卷第6A 期 Vo l. 44 No. 6A 计算机科学 2017 年 6 月 CO岛1PUTER SCIENCE June 2017 基于 RANSAC 的 SIFT 匹配阔值自适应估计 刘川熙赵汝进刘恩海洪裕珍 (中国科学院光电技术研究所 成都 610209) 摘 要 针对基于欧氏距离比值作为图像尺度不变特征变换(SIFT)特征匹配相似性度量时，距离比闵值难以设置最 优，且固定距离比闵值易引起误匹自己或漏匹配等问题，引入随机抽样一致性(RANSAC)算法。该算法对 SIIT 匹配算 法中的距离比阂值进行自适应优化，确定最佳的阂值，再利用双向匹配的方法剔除误匹配点。实验结果表明，针对不 同的实验图像，所提算法都能自适应地求解出一个最优的比例阅佳，使得匹配点数最多，同时具有较高的匹配正确率， 经过双向匹配的策略优化后效果更好。 关键词 尺度不变特征变换(SIFT) ，随机抽样一致性 (RANSAC) ，自适应，匹配阅值，双向 中图法分类号 TP91 1. 73 文献标识码 A Estimate Threshold of SIFT Matching Adaptively ßased on RANSAC LIU Chuan-xi ZHAO Ru-jin LIU En-hai HONG Yu-zhen (Institute of Optics and Electronics ,Chinese Academy of Sciences ,Chengdu 610209 ,China) Abstract When matching images with scale invariant feature transform(SIIT) ,the Euclidean distance between feature vectors is used as the similarity measurement. But it was difficult to get the best distance ratio. Moreover ,when the ratio was a constant ,there would be some problems of error matching or matching leakage. Deal with the problem ,the Ran dom Sample Consensus (RANSAC) algorithm was introduced. Optimize the ratio in the process adaptively ,and we can get the best threshold. SIIT-based image matching algorithm was analyzed ,and a bi-direction matching was used to im- prove the accuracy of image matching and ensure the correctness of matching at maximum level. Finally ,the experiment results show that the proposed methods can obtain an optimal threshold for different images. It can get the most ma tching points and a better matching rate ,and by bi-direction matching ,better results