华南理工大学学报(自然科学版) ›› 2019, Vol. 47 ›› Issue (2): 50-58.doi: 10.12141/j.issn.1000-565X.180278

• 电子、通信与自动控制 • 上一篇    下一篇

基于有向超图的图像匹配算法

朱明1, 2, 3 张健2 梁栋1, 2, 3† 唐俊1, 2 张艳1, 2   

  1. 1. 安徽大学 农业生态大数据分析与应用技术国家地方联合工程研究中心,安徽 合肥 230601; 2. 安徽大学 电子信息工程学院,安徽 合肥 230601; 3. 偏振光成像探测技术安徽省重点实验室,安徽 合肥 230031
  • 收稿日期:2018-06-08 修回日期:2018-09-28 出版日期:2019-02-25 发布日期:2019-01-02
  • 通信作者: 梁栋( 1963-) ,男,教授,博士生导师,主要从事计算机视觉、图像处理、模式识别研究 E-mail:dliang@ahu.edu.cn
  • 作者简介:朱明( 1984-) ,男,博士,讲师,主要从事计算机视觉、图像处理、模式识别研究
  • 基金资助:
    国家自然科学基金资助项目( 61501003, 61772032, 61672032) ;偏振光成像探测技术安徽省重点实验室开放课题 ( 2016-KFJJ-002) 

Image Matching Algorithm Based on Directed Hypergraph
 

 ZHU Ming1, 2, 3 ZHANG Jian2 LIANG Dong1, 2, 3 TANG Jun1, 2 ZHANG Yan1, 2   

  1.  1. National Engineering Research Center for Agro-Ecological Big Data Analysis and Application,Anhui University,Hefei 230601, Anhui,China; 2. School of Electronics and Information Engineering,Anhui University,Hefei 230601,Anhui,China; 3. Key Laboratory of Polarization Imaging Detection Technology in Anhui Province,Hefei 230031,Anhui,China
  • Received:2018-06-08 Revised:2018-09-28 Online:2019-02-25 Published:2019-01-02
  • Contact: 梁栋( 1963-) ,男,教授,博士生导师,主要从事计算机视觉、图像处理、模式识别研究 E-mail:dliang@ahu.edu.cn
  • About author:朱明( 1984-) ,男,博士,讲师,主要从事计算机视觉、图像处理、模式识别研究
  • Supported by:
     Supported by the National Natural Science Foundation of China( 61501003, 61772032, 61672032) 

摘要: 为了提高复杂变换下图模型的匹配精度,提出了一种基于有向超图的图像匹配 算法. 该算法首先分别在两个待匹配的特征点集中构造3 一致超图,计算每条超边所包含 三元组的权值,然后利用这些权值来构造加权邻接张量,最后通过凸凹松弛算法实现图像 匹配. 模拟和真实图像的实验结果表明,文中算法能够获得更高的匹配精度,对于复杂变 换的图像也有很好的匹配效果. 

关键词: 超图, 三元组, 邻接张量, 凸凹松弛, 匹配 

Abstract: An algorithm for images matching based on directed hypergraph was proposed to improve the matching accuracy of graph model under the complex transformation. Firstly,3-uniform hypergraphs in the two feature points set to be matched was constructed by the algorithm,and the weights of the triples contained in each hyperedge was calculated, then these weights were used to construct weighted adjacency tensors. Finally,image matching was achieved by convex-concave relaxation algorithm. The experiments with both simulate and real results show that the proposed algorithm can achieve higher matching accuracy and good matching effect for complex transformed images. 

Key words:

 

中图分类号: