结合高阶图模型与蚁群优化的图像匹配方法  

Second-order graph model ant and colony optimization based image matching

作　　者：杨思燕[1] 曹文灿 李世平 

机构地区：[1]陕西广播电视大学计算机与信息管理系,陕西西安710119 [2]西安市西光中学,陕西西安710043

出　　处：《西安电子科技大学学报》2017年第1期159-164,共6页Journal of Xidian University

基　　金：国家自然科学基金资助项目(61272280);大数据环境下计算机类课程MOOC研究资助项目(15G-04-A04);大数据下的计算机类课程资源建设实践研究资助项目(GJ1529)

摘　　要：图像匹配是计算机视觉领域中的一个重要的问题.针对基于图结构模型的图像匹配方法,研究了图模型框架的建立方法以及二阶约束和高阶约束下的图匹配算法框架.为了克服传统的求驻点的优化方法易陷入局部最优解的不足,采用蚁群算法优化目标函数,提出一种基于蚁群算法的高阶图匹配方法.该算法使用张量值计算启发因子提供先验知识,然后根据启发因子和信息素计算转移概率,最后利用搜索到的解对信息素进行局部更新和全局更新.实验表明,该算法能获得比较高的匹配精度,并且在形变噪声、外点和视角变化等因素的干扰下仍具有很强的鲁棒性.Image matching is a fundamental problem in the computer vision field. This paper focuses on image matching based on the graph structure model. The methods of the graph model establishment in the second-order or high-order constraint are studied. In order to overcome the defects of traditional optimal algorithms which fall easily into the local optimal solution, this paper adopts the ant colony optimization algorithm to optimize the match score function and proposes an high-order graph matching algorithm based on ant colony optimization. It first applies the tensor matching algorithm to initialize the pheromone matrix to provide a good start point, adopts the affinity tensor to provide the priori knowledge for computing the heuristic factor, then calculates the transition probability using the pheromone and heuristic factor, and finally updates the pheromone in two ways by the solutions which have been searched. The two updating rules of pheromone are local and global. Experimental results show that this algorithm can get a higher matching accuracy and has a stronger robustness against deformation noises and outliers compared with others.

关 键 词：图像匹配 蚁群算法 高阶图匹配 优化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]