Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (29): 208-210.

• 图形、图像、模式识别 • Previous Articles     Next Articles

Adaptive optimal grouping of feature point matching strategy

ZHAO Hongyu1,LI Yanhua1,HOU Zhenjie1,DUO Huaqiong2   

  1. 1.Computer and Information Engineering College,Inner Mongolia Agricultural University,Hohhot 010018,China
    2.Materials Science and Arts Design College,Inner Mongolia Agricultural University,Hohhot 010018,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-10-11 Published:2011-10-11

一种自适应特征点最优分组匹配的策略

赵宏宇1,李燕华1,侯振杰1,多化琼2   

  1. 1.内蒙古农业大学 计算机与信息工程学院,呼和浩特 010018
    2.内蒙古农业大学 材料科学与艺术设计学院,呼和浩特 010018

Abstract: This paper proposes an adaptive optimal grouping of feature point matching strategy.The optimal group of corner points in accordance with eigenvalue is got.The method uses the monogenic phase or NCC algorithm and Ransac estimation algorithm to a formal group matching and excluding external point.Experimental results show that this method can effectively remove the error matching points,reduce the feature point search and match verification time,which also has the characteristics of easy to compute and good practicality.

Key words: matching, optimal grouping, monogenic phase, Ransac

摘要: 提出一种自适应对特征点进行最优分组匹配的方法策略,按照特征值将获得的角点进行最优分组,采用单演相位或NCC算法与Ransac估计算法进行正式的分组匹配和剔除外点。通过实验证明,该算法能够有效去除误匹配点,减少特征点的匹配搜索与匹配验证时间,具有计算简单,实用性强等特点。

关键词: 匹配, 最优分组, 单演相位, Ransac算法