A Multi-view Image Registration Algorithm based on SIFT Feature

摘要：

This page proposes a multi-view image registration algorithm based on SIFT feature by analyzing the principle and the characteristics of multi-view images registration. The improved algorithm has all of the characteristics as the classical SIFT algorithm does, and it is also able to deal with a pair of images correctly, between which there may be existing rotation, scaling, affine change and view transformation. Whats more, By detecting Harris corners in the neighborhood of SIFT feature point, the improved algorithm reduces the number of feature points distinctly and raises the rate of algorithm on one hand, and strengthens the significance of feature point set and raises the accuracy of feature point matching on the other hand.

关键词： Multi-view Image Registration SIFT Algorithm Harris Corner Feature Point Matching

作者: Aoshuang Dong Hui Ying

作者单位: Software College,Northeastern University Shenyang,China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 438-441

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）