计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (2): 63-65.DOI: 10.3778/j.issn.1002-8331.2009.02.017

• 研究、探讨 • 上一篇    下一篇

改进的边缘特征点提取算法

李竹林1,王文东1,赵宗涛2,王红珍1   

  1. 1.延安大学 计算机学院,陕西 延安 716000
    2.第二炮兵学院 402室,西安 710025
  • 收稿日期:2007-12-11 修回日期:2008-02-21 出版日期:2009-01-11 发布日期:2009-01-11
  • 通讯作者: 李竹林

Advanced extracting algorithm for edge feature points

LI Zhu-lin1,WANG Wen-dong1,ZHAO Zong-tao2,WANG Hong-zhen1   

  1. 1.Institute of Computer Science,Yan’an University,Yan’an,Shaanxi 716000,China
    2.402 Division of the Second Artillery Engineering College,Xi’an 710025,China
  • Received:2007-12-11 Revised:2008-02-21 Online:2009-01-11 Published:2009-01-11
  • Contact: LI Zhu-lin

摘要: 对一种基于图像几何重心的特征点提取算法进行了改进。改进后的算法可以提取任何边缘线上凹凸处的特征点。其基本思想是,首先提取图像的边缘轮廓线,计算封闭边缘线和非封闭边缘线的几何重心。然后以重心为极点,对边缘点极坐标化,形成幅角-极径曲线,在该曲线上寻找局部极大极小值点。最后根据局部非极大(小)抑制原则对极大(小)值点进行取舍获得特征点。模拟结果表明算法计算简单、稳定性好。该方法提取的特征点能准确地反映目标的形状,对目标的识别、跟踪以及三维表面重建等都具有重要的价值。

关键词: 特征点, 重心, 极径, 幅角

Abstract: This paper modifies a kind of feature extracting algorithm based on image edge.First,image edge outline is extracted.Second,geometric gravity center of edge outline is calculated.Third,argument-polar radius curve of polar coordinate system whose pole is geometric gravity is formed,and maximum and minimum value points are searched.At last,feature points are obtained by selecting maximum(minimum) value points according to local non-maximum(minimum) depression principle.The result indicates that the algorithm is simple and its stability is good.It is important for object recognition and three-dimension reconstruction.

Key words: feature points, geometric gravity center, polar radius, argument angle