摘要
该文结合脉冲耦合神经网络(PCNN)和非线性各向异性扩散(NAD)模型,提出了改进的ADPCNN模型,并对新模型进行了理论分析。新模型对NAD模型的扩散次数问题和PCNN模型对像素的修改策略问题提出了解决方法,在医学眼底图像处理中取得了较好的效果。
Based on the model of Pulse Coupled Neural Network (PCNN) and the model of Nonlinear Anisotropic Diffusion (NAD), an improved model named ADPCNN is proposed and analyzed in theory. The new model resolves the problems of diffusion number of NAD model and the modification rule of pixel gray value of PCNN model. The application to medical image processing shows the good effect of the new model.
出处
《电子与信息学报》
EI
CSCD
北大核心
2007年第10期2316-2320,共5页
Journal of Electronics & Information Technology
基金
教育部"新世纪优秀人才支持计划"
教育部科学技术研究重点项(NO106030)资助课题
关键词
脉冲耦合神经网络
非线性各向异性扩散
图像处理
边缘提取
Pulse coupled neural network
Nonlinear anisotropic diffusion
Image processing
Edge detection