首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于非均匀五株采样滤波器组的图像阈值除噪
引用本文:王佳宁.基于非均匀五株采样滤波器组的图像阈值除噪[J].科学技术与工程,2011(26):6341-6344.
作者姓名:王佳宁
作者单位:陕西教育学院物理与电子技术系,西安,710061
摘    要:针对传统二维小波缺少多方向特性和传统多方向滤波器组不能进行多尺度分解的缺陷,提出了一种具有多尺度多方向特性的非均匀五株采样滤波器组。该滤波器组每层分解具有一个低频子带和十二个高频子带,并可通过对低频子带的重复分解得到图像的多尺度信息,实现了多方向分解和多尺度分析的统一完成。将提出的非均匀五株采样滤波器组结合统一阈值方法应用于图像除噪。经试验验证,与小波统一阈值除噪相比,所提出的方法对除噪后的图像无论从视觉效果和峰值信噪比(PSNR)上都有较显著提高。

关 键 词:小波变换  多方向滤波器组  五株采样  统一阈值  图像除噪
收稿时间:2011/6/19 0:00:00
修稿时间:2011/6/19 0:00:00

Image Denoising Based on Nonuniform Quincunx Directional Filter Banks via Wavelet shrinkage
wangjianing.Image Denoising Based on Nonuniform Quincunx Directional Filter Banks via Wavelet shrinkage[J].Science Technology and Engineering,2011(26):6341-6344.
Authors:wangjianing
Institution:WANG Jia-ning(Shaanxi Institute of Education,Xi’an 710061,P.R.China)
Abstract:Traditional 2D-wavelet transform lacks multidirectional information, Directional filter banks lacks multiresolution property. In this paper, an image denoising method based on nonuniform quincunx directional filter banks via Wavelet shrinkage has been proposed. The nonuniform quincunx directional filter banks has twelve highpass subbands and one lowpass subband in each level, and the multiresolution information can be obtained by repeating decomposition in lowpass subband. By wavelet shrinkage method to realize image denoising. The experiment results show that the performance of the proposed method is obviously superior in both vision and Pulse Signal to Noise Ratio(PSNR).
Keywords:Wavelet thansform  Directional filter banks  Quincunx Sampling  Wavelet shrinkage  Image denoising
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《科学技术与工程》浏览原始摘要信息
点击此处可从《科学技术与工程》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号