文章基本信息

标题：Image Registration Using Advanced Topology Preserving Relaxation Labeling
本地全文：下载
作者：Chan-IL Kim ; Ji-Ae Park ; Su Yang 等
期刊名称：Computer Science & Information Technology
电子版ISSN：2231-5403
出版年度：2016
卷号：6
期号：8
页码：31-42
DOI：10.5121/csit.2016.60804
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：This paper presents a relaxation labeling technique with newly defined compatibility measuresfor solving a general non-rigid point matching problem. In the literature, there exists a pointmatching method using relaxation labeling, however, the compatibility coefficients always takea binary value zero or one depending on whether a point and a neighboring point havecorresponding points. Our approach generalizes this relaxation labeling approach. Thecompatibility coefficients take n-discrete values which measures the correlation between edges.We use log-polar diagram to compute correlations. Through simulations, we show that thistopology preserving relaxation method improves the matching performance significantlycompared to other state-of-the-art algorithms such as shape context, thin plate spline-robustpoint matching, robust point matching by preserving local neighborhood structures andcoherent point drift.
关键词：Point pattern matching; graph matching; registration; relaxation labelling