计算机与现代化 ›› 2009, Vol. 1 ›› Issue (12): 89-91,9.doi: 10.3969/j.issn.1006-2475.2009.12.024

• 图像处理 • 上一篇    下一篇

基于Canny改进的边缘检测算法

黄宝贵,赵景秀,马春梅
  

  1. 曲阜师范大学计算机科学学院,山东 日照 276826
  • 收稿日期:2009-05-19 修回日期:1900-01-01 出版日期:2009-11-27 发布日期:2009-11-27

An Improved Edge Detection Algorithm Based on Canny Theory

HUANG Bao-gui, ZHAO Jing-xiu, MA Chun-mei
  

  1. Computer Science College, Qufu Normal University, Rizhao 276826, China
  • Received:2009-05-19 Revised:1900-01-01 Online:2009-11-27 Published:2009-11-27

摘要: Canny算子受噪声干扰严重,检测噪声污染的图像边缘的效果较差。轮廓结构元素形态学变换能有效滤除图像噪声,同时能保持图像边缘细节。本算法先用轮廓结构元素形态学变换的方法抑制噪声,求取图像梯度并用非极大值抑制法求其局部模极大值,采用双阈值法二值化图像边缘并进行连接。实验表明,该算法能精确提取噪声图像边缘。

关键词: Canny算子, 边缘检测, 数学形态学

Abstract: Canny operator is disturbed seriously by noise, so it has poor effect on noised image edge detection. The transformation based on contour structuring elements morphology not only can filter noise, but also can preserve detail of image edge. This arithmetic restrains noise by using transformation based on contour structuring elements morphology, calculates the grads and has local modulus maximum by using non-maximum suppression, binary edges by using double-threshold and connects the edges. The experiment shows that the arithmetic can extract noised image edges accurately.

Key words: Canny operator, edge detection, mathematic morphology

中图分类号: