2017_一种结合形状上下文的分析的Laplace谱匹配算法_唐俊.pdfVIP

第21 卷第14 期 系 统 仿 真 学 报© Vol. 21 No. 14 2009 年7 月 Journal of System Simulation Jul., 2009 一种结合形状上下文分析的Laplace谱匹配算法 1 1 1 1,2 1 唐 俊 ，王 年 ，梁 栋 ，范益政 ，贾兆红 （1.安徽大学计算智能与信号处理教育部重点实验室，合肥 230039 ；2.安徽大学数学系，合肥 230039 ） 摘 要：提出了一种结合形状上下文分析的Laplace 谱匹配算法。工作主要侧重于如何提高Laplace 谱匹配算法对点的位置随机抖动的鲁棒性。首先，使用Laplace 矩阵的特征向量和特征值以及双随 机矩阵的方法计算初始匹配概率。然后，借助于概率松弛算法，将用形状上下文表示的局部相似 性融入Laplace 谱匹配算法以优化谱匹配的结果。对真实和合成数据的实验表明该方法具有比较高 的精度。 关键词：Laplace 谱；点模式匹配；双随机矩阵；形状上下文；概率松弛 中图分类号：TP391 文献标识码：A 文章编号：1004-731X (2009) 14-4345-06 Laplacian Spectral Correspondence Combined with Shape Context Analysis 1 1 1 1,2 1 TANG Jun , WANG Nian , LIANG Dong , FAN Yi-zheng , JIA Zhao-hong (1. Key Lab of Intelligent Computing Signal Processing Ministry of Education, Anhui University, Hefei 230039, China; 2. Department of Mathematics, Anhui University, Hefei 230039, China) Abstract: A Laplacian spectral method combined with shape context analysis was proposed for point pattern matching. This work mainly focused on the problem of how to render the Laplacian spectral method robust for random position jitter. Firstly, the initial correspondence probabilities were computed by using the eigenvectors and eigenvalues of the Laplacian matrix as well as the method of doubly stochastic matrix. Secondly, local similarity evaluated by shape context was embedded into the Laplacian spectral method to refine the results of