计算机应用

• 图形图像处理 • 上一篇    下一篇

基于各向异性扩散方程的Canny边缘检测算法

张洁 檀结庆   

  1. 合肥工业大学 合肥工业大学
  • 收稿日期:2008-02-28 修回日期:2008-04-02 发布日期:2008-08-01 出版日期:2008-08-01
  • 通讯作者: 张洁

Canny edge detection algorithm based on anisotropic diffusion equations

<a href="http://www.joca.cn/EN/article/advancedSearchResult.do?searchSQL=(((Jie ZHANG[Author]) AND 1[Journal]) AND year[Order])" target="_blank">Jie ZHANG</a>   

  • Received:2008-02-28 Revised:2008-04-02 Online:2008-08-01 Published:2008-08-01
  • Contact: Jie ZHANG

摘要: Canny边缘检测算法由于使用高斯滤波对图像进行平滑,往往使得算法的信噪比和定位精度下降,从而产生一些虚假边缘,使角点变圆。针对Canny算法所出现的问题提出了一种改进方法,运用各向异性扩散方程代替高斯滤波,并对扩散后的图像做图像增强。实验结果表明,该算法有效地提高了边缘检测的准确性,得到了比较理想的边缘检测效果。

关键词: 边缘检测, Canny算法, 高斯滤波, 各向异性扩散方程, 图像增强

Abstract: The Canny edge detection algorithm which used Gaussian filter for smoothing the image may lead to lower SNR and higher edge location error, which produced false edge and corner roundness. To solve these problems, an improved edge detection algorithm was proposed by replacing Gaussian filter with anisotropic diffusion equations, and the image enhancement was carried out after diffusion. Experimental results show that the accuracy of edge detection is improved evidently, and a much better edge detection effect is obtained.

Key words: edge detection, Canny algorithm, Gaussian filter, anisotropic diffusion equations, image enhancement