Abstract: A new matching method called Progressive Sample Consensus（PROSAC） is proposed.The algorithm exploits the linear ordering defined on the set of correspondences by a similarity function used in establishing tentative correspondences.Under the mild assumption that the similarity measure predicts correctness of a match better than random guessing.The authors introduce the main concept of our method including the growth function and stopping criterion.The power of the method is demonstrated on image matching problems.

摘要： 介绍了一种新的特征匹配算法-渐进式抽样一致算法，该算法首先按匹配度大小将初步匹配集合中的匹配对按匹配度作降序排列，然后根据规定的增长函数从前至后作抽样。详细介绍了算法的设计思想、增长函数的设计策略、抽样中指标准。将该算法成功的应用到图像配准问题中，实验结果表明该算法较传统算法在抽样终止标准相同条件下时间复杂度有较大改善。