Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (10): 140-143.DOI: 10.3778/j.issn.1002-8331.2010.10.045

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

Corner detection operator based on LoG of edge contour

LIU Ya-wei1,2,LI Jian-wei1,ZHANG Xiao-hong3   

  1. 1.Key Lab of Opto-Electronic Technique of State Education Ministry,Chongqing University,Chongqing 400030,China
    2.School of Science,Chongqing Jiaotong University,Chongqing 400074,China
    3.School of Software Engineering,Chongqing University,Chongqing 400030,China
  • Received:2009-11-02 Revised:2009-12-18 Online:2010-04-01 Published:2010-04-01
  • Contact: LIU Ya-wei

一种基于边缘轮廓线的LoG角点检测

刘亚威1,2,李见为1,张小洪3   

  1. 1.重庆大学 光电子技术及系统教育部重点实验室,重庆 400030
    2.重庆交通大学 理学院,重庆 400074
    3.重庆大学 软件学院,重庆 400030
  • 通讯作者: 刘亚威

Abstract: Corner is a significant local feature of object in digital image,and it can offer the vision feature of object in low vision level.Laplacian of Gaussian is lead into planar curves,and a corner detection operator based on LoG of edge contour is proposed.The geometric feature of LoG is discussed against ?祝-corner and circumference models,and equivalence relation between the curvature and LoG norm of edge contour is also framed.Then,corner response function is defined as LoG norm of edge contour,and corners are constructed as the local extreme points of LoG norm.Experiments also demonstrate that proposed algorithm is efficient and stability compared with other operators.

Key words: edge detection, feature extract, corner detection, LoG operator

摘要: 角点是数字图像中目标的重要局部特征,提供了目标的低层次视觉特性。将Laplace of Gaussian变换引入到平面曲线,提出了一种基于边缘轮廓的LoG角点检测算子。深入分析LoG变换的几何特征,建立了边缘轮廓曲率和LoG范数间的等价度量关系。角点响应函数被定义为边缘点的LoG变换范数,边缘轮廓上LoG范数的局部极值点被认为是角点。并针对?祝-角点模型和圆周曲线模型,验证了这种定义的合理性。实验结果表明,相比经典角点检测算子,该文算法具有高效、稳定的优点。

关键词: 边缘检测, 特征提取, 角点检测, LoG算子

CLC Number: