互联网 qkzz.net
全刊杂志网:首页 > 女性 > 文章正文
刊社推荐

基于细胞神经网络的图像阈值化方法


  摘 要:图像阈值化是一种经典、简单而又非常有效的图像分割方法,并已得到了广泛的研究。在分析灰度图像直方图分布的基础上提出了一种基于细胞神经网络(CNN)结合直方图分析的图像阈值化方法,并给出了阈值化CNN所需阈值的自动搜索算法。实验结果表明,相对于其他两种经典的阈值化方法,该方法的阈值化分割结果较好。

  关键词:图像阈值化; 细胞神经网络; 直方图; 模板参数

  中图分类号:TP391.4文献标志码:A

  文章编号:1001-3695(2010)07-2780-03

  doi:10.3969/j.issn.1001-3695.2010.07.108

  Image thresholding approach based on CNN

  KANG Jia-yin1a,2, ZHANG Wen-juan1b

  (1.a.School of Electronics Engineering, b.School of Computer Engineering, Huaihai Institute of Technology, Lianyungang Jiangsu 222005, China; 2. School of Communication & Control Engineering, Jiangnan University, Wuxi Jiangsu 214122, China)

  Abstract:Image thresholding is one of the classical, simple, and very effective techniques for image segmentation, and has been widely studied. This paper proposed an approach for image thresholding based on cellular neural network (CNN) associated with histogram analysis, and gave an automatic searching algorithm of threshold needed by threshold CNN. Experimental results show that the proposed approach obtains better segmenting results than other two classical methods.

  Key words:image thresholding; cellular neural network; histogram; template parameters

......
很抱歉,暂无全文,若需要阅读全文或喜欢本刊物请联系《计算机应用研究》杂志社购买。
欢迎作者提供全文,请点击编辑
分享:
 

了解更多资讯,请关注“木兰百花园”
分享:
 
精彩图文


关键字
支持中国杂志产业发展,请购买、订阅纸质杂志,欢迎杂志社提供过刊、样刊及电子版。
关于我们 | 网站声明 | 刊社管理 | 网站地图 | 联系方式 | 中图分类法 | RSS 2.0订阅 | IP查询
全刊杂志赏析网 2017