Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (18): 160-162.

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

Edge detection of gray-scale images based on self-adaptive CNN

ZHANG Ying,WANG Tai-yong,HUANG Guo-long,LENG Yong-gang   

  1. School of Mechanical Engineering,Tianjin University,Tianjin 300072,China
  • Received:2007-12-24 Revised:2008-03-10 Online:2008-06-21 Published:2008-06-21
  • Contact: ZHANG Ying

基于参数自适应CNN的灰度图像边缘检测

张 莹,王太勇,黄国龙,冷永刚   

  1. 天津大学 机械工程学院,天津 300072

  • 通讯作者: 张 莹

Abstract: Edge is an important character of the image.The stability of the network and the parameter choice are the key problems when Cellular Neural Network(CNN) is used to pick up the image edge.The stable condition of CNN is discussed,and the self-adaptive template parameters of the network are fixed reasonably.Based on Matlab 7.0,the simulation applied to the edge detection has presented its validity in the image processing,as well as the noise control.

摘要: 边缘是图像的重要特征。在应用细胞神经网络提取图像边缘时,网络的稳定性和参数的选择是关键。文中推导了细胞神经网络的稳定条件,并提出了网络参数的自适应设计思路。基于Matlab 7.0平台,通过编写仿真程序,检测灰度图像边缘,得到良好效果。实验证明,该法还能有效抑制噪声的干扰。