您的位置:首页  > 论文页面

基于模糊神经网络的非线性滤波算法

发表时间:2009-10-31  浏览量:1556  下载量:711
全部作者: 赵雪青,王晅,白春丽
作者单位: 陕西师范大学物理学与信息技术学院;西安电子科技大学微电子学院
摘 要: 针对现有图像椒盐噪声滤除算法缺乏对小于滤波窗口的图像细节与边缘信息的保护问题,在分析现有图像滤波算法的基础上,提出一种基于模糊神经网络自适应神经模糊推理系统(self-adaptive neural-fuzzy inference system,ANFIS)的非线性滤波算法。算法将滤波过程分为噪声检测和去除噪声两个阶段。在噪声检测过程中,用自适应中值原理对图像中的噪声点进行初步检测,然后通过局部模糊隶属度函数对检测出的噪声点进行二次判断,有效提高了噪声的准确度。在噪声恢复阶段,利用ANFIS通过自适应神经模糊推理系统对非线性系统的结构和参数进行辨识与自学习,采用混合学习算法,对前向参数和结论参数分别辨识,在提高精度的同时可加快训练收敛的速度。此算法充分利用了图像的局部特征,实验结果表明:对椒盐噪声具有很好的细节保护与噪声滤除能力,其效果明显优于现有的其他滤波算法。
关 键 词: 计算机应用;二次噪声检测;非线性滤波;自适应神经模糊推理系统;椒盐噪声
Title: Nonlinear filtering algorithm using adaptive neural-fuzzy inference systems
Author: ZHAO Xueqing, WANG Xuan, BAI Chunli
Organization: College of Physics and Information Technology, Shaanxi Normal University;School of Micro Electron, Xidian University
Abstract: The major drawback of existing image filtering algorithms for removing salt-and-pepper noise is lack of the ability of preserving the image details and edges which are smaller than the size of filtering windows. A nonlinear filtering algorithm using self-adaptive neural-fuzzy inference systems (ANFIS) is proposed. The proposed filter has a two-stage scheme: detecting noise and removing noise. In order to improve the accuracy of noise detection, noise candidates identified with the noise detection algorithm of the adaptive median filter are judged again by local fuzzy membership function, and the structures and parameters of nonlinear system are identified and learned by ANFIS. This network uses hybrid learning algorithm to identify former parameters and conclusion parameters. The algorithm improves the precision as well as quickens the training speed. The proposed filtering algorithm takes advantages of local characteristics of images, and achieves a better performance than the other filtering algorithms in noise suppression and detail preservation.
Key words: computer application; double noise detector; nonlinear filtering; adaptive neural-fuzzy inference systems; salt-and-pepper noise
发表期数: 2009年10月第20期
引用格式: 赵雪青,王晅,白春丽. 基于模糊神经网络的非线性滤波算法[J]. 中国科技论文在线精品论文,2009,2(20):2176-2184.
 
0 评论数 0
暂无评论
友情链接