一种自适应模糊聚类神经网络滤波器设计  

The design of Filtering based on adapted fuzzy cluster neural network

作　　者：欧阳春娟[1] 欧阳迎春[2] 杨群生[3] 

机构地区：[1]井冈山学院,江西343009 [2]江西师范大学,江西343009 [3]广东工业大学,江西343009

出　　处：《微计算机信息》2007年第02X期233-234,115,共3页Control & Automation

基　　金：测绘遥感信息工程国家重点实验室基金项目(WKL(03)0101)

摘　　要：为了获得滤除噪声和细节保留两方面更好的平衡,提出了一种自适应模糊聚类神经网络。采用模糊C—均值聚类算法对网络进行模糊化,利用改进的LMS算法对网络进行训练。仿真表明,与模糊BP神经网络及改进的BP神经网络相比,AFCN是一种性能较好的智能神经网络。A AFCN （adapted fuzzy cluster neural network） is introduced to keep up trade-off between noise attenuation and detail preservation in image processing. The FCNN is fuzzied by FCM algorithm and improved LMS algorithm is applied to tune the weight of FCNN. The result of simulation indicates that the AFCN has better performance than fuzzy BP network and improved BP network in image filtering. The AFCN is a sort of intelligent neural network which has good performance.

关 键 词：模糊聚类 神经网络 LMS算法 图像滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]