Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (2): 10-12.DOI: 10.3778/j.issn.1002-8331.2009.02.003

• 博士论坛 • Previous Articles     Next Articles

Research of feature points matching method in 3D reconstruction across images

WANG Jian,ZHOU Guo-min   

  1. Agricultural Information Institute of CAAS,Beijing 100081,China
  • Received:2008-10-28 Revised:2008-11-19 Online:2009-01-11 Published:2009-01-11
  • Contact: WANG Jian

图像三维重建中的特征点匹配方法研究

王 剑,周国民   

  1. 中国农业科学院 农业信息研究所,北京 100081
  • 通讯作者: 王 剑

Abstract: During the course of 3D reconstruction from a set of images,to some limitations in traditional algorithms a new efficient algorithm of points correspondence matching based on a new affinity measure is developed.Through this method,image points matching and 3D reconstruction can be operated in parallel.In this paper,the algorithm of points matching and reconstruction across images is presented.Experimental results on synthetic and real data demonstrate the effectiveness of the method.

Key words: 3D reconstruction, feature correspondence matching, 3D model

摘要: 针对传统三维重建中点特征匹配算法的局限性,提出了基于相关系数的图像点特征匹配方法,实现了对图像点特征匹配与三维重建同步进行的目的。基于同一物体的两幅图像,给出了其特征匹配和三维重建的算法,并通过实验数据分析了此算法的有效性。

关键词: 三维重建, 特征匹配, 三维模型