J4

• 计算机科学 • 上一篇    下一篇

一种改进的基于Canny算子的图像边缘提取算法

张震1,2, 马驷良1, 张忠波1, 刘辉1, 宫跃欣1, 孙秋成1   

  1. 1. 吉林大学 数学研究所, 长春 130012; 2. 大庆油田 物业集团, 黑龙江省 大庆 163453
  • 收稿日期:2006-06-30 修回日期:1900-01-01 出版日期:2007-03-26 发布日期:2007-03-26
  • 通讯作者: 马驷良

Improved Image Edge Extraction Algorithm Based on Canny Operator

ZHANG Zhen1,2, MA Siliang1, ZHANG Zhongbo1, LIU Hui1, GONG Yuexin1, SUN Qiucheng1   

  1. 1. Institute of Mathematics, Jilin University, Changchun 130012, China;2. Estate Group, Daqing Oilfield, Daqing 163453, Heilongjiang Province, China
  • Received:2006-06-30 Revised:1900-01-01 Online:2007-03-26 Published:2007-03-26
  • Contact: MA Siliang

摘要: 对Canny边缘提取算法进行改进, 采用二维高斯函数的一阶偏导数构造滤波器计算梯度幅值, 利用一种四阈值边缘检测定位方法定位边缘, 最后引入数学形态学方法对边缘细化处理. 改进算法对噪声抑制效果明显, 能够删除伪边缘, 检出边缘更加精细, 实验结果表明了改进算法的可行性和有效性.

关键词: Canny算子, 边缘提取, 图像处理

Abstract: This paper introduces a new approach of improvedCanny operator for edge detection. The filter operator constructed with the first order partial derivative of twodimensional Gaussian function is employed to compute gradient. Then the edge was located via the edge detection and localization method with four threshold values. And finally, edge thin operation was conducted by introducing a morphological operator. The improved algorithm can enhance the capability of suppressing noise, delete false edges and obtain exact edges. Experimental results indicate the feasibility and validity of the improved algorithm.

Key words: Canny operator, edge extraction, image processing

中图分类号: 

  • TP391.41