基于小波变换的地震资料局域自适应去噪研究

英文篇名：Locally adaptive seismic data denoising based on wavelet transform
中文刊名：勘探地球物理进展
英文刊名：Progress in Exploration Geophysics
作者：刘鑫 ; 李可恩 ; 贺振华 ; 黄德济
英文作者：Liu Xin ; Li Keen ; He Zhenhua ; Huang Deji. College of Information Engineering ; Chengdu University of Technology ; Chengdu 610059 ; China
中文关键词：小波变换 ; 双变量收缩 ; 统计模型 ; 地震资料
英文关键词：wavelet transform ; bivariate shrinkage ; statistical modeling ; seismic data
出版日期：2007-08-25
机构：成都理工大学信息工程学院,成都理工大学信息工程学院,成都理工大学信息工程学院,成都理工大学信息工程学院四川成都610059,四川成都610059,四川成都610059,四川成都610059
年：2007
期：04
出版单位：勘探地球物理进展

摘要

常规小波阈值去噪方法未能充分利用地震信号相关性的特点进行去噪,为此在多层小波变换中引入了双变量概率分布模型。基于贝叶斯估计理论,得到了相应的双变量收缩函数;基于层内局域方差估计,得到了一种局域自适应去噪算法。在实验中,将该算法分别应用于实值离散小波变换域和复数小波变换域,并和隐马尔科夫模型的去噪方法进行了比较。数值模拟实验和实际地震资料处理结果表明:复数小波变换的局域自适应收缩算法去噪效果最好。
Conventional denoising methods that use threshold filter in wavelet domain do not utilize the correlations of seismic data to remove noises. This paper presents a local adaptive denoising algorithm. The new algorithm assumes that there is a statistical dependence among wavelet coefficients. Firstly, a bivariate probability distribution model was introduced to model the statistical behaviors of wavelet coefficients, and then the corresponding nonlinear threshold function (bivariate shrinkage function) was derived from the model using the Bayesian estimation theory. Finally, using local variance estimation, a locally adaptive image-denoising algorithm was contrived. Numerical simulation and real seismic data processing examples are given to illustrate the effectiveness of the denoising algorithm. The new algorithm is also applicable to the complex wavelet domain.

引文

1 Crouse M S,Nowak R D,Baraniuk R G.Wavelet-based statistical signal processing using hidden Markov models[J].IEEE Trans on Signal Processing,1998,46(4):886～902
    2 Cai Zhaohui,Cheng T H,Lu Chao,et al.Efficientwavelet-based image denoising algorithm[J].ElectronLett,2001,37(11):683～685
    3 Chang S,Yu B,Vetterli M.Adaptive wavelet threshol-ding for image denoising compression[J].IEEE Transon Image Processing,2000,9(9):1 532～1 546
    4 Levent S,Selesnick I W.A bivariate shrinkage functions for wavelet-based denoising[J].IEEE Trans on Sig-nal Processing,2002,50(11):2 744～2 756
    5 Levent S,Selesnick I W.Bivariate shrinkage with lo-cal variance estimation[J].IEEE Signal ProcessingLetters,2002,9(12):438～441
    6 Levent S,Selesnick I W.Bivariate shrinkage function for wavelet based denoising exploiting interscale de-pendency[J].IEEE Trans on Signal processing,2002,12(50):2 744～2 756
    7 Kingsbury N G.The dual-tree complex wavelet trans-form:a new efficient tool for image restoration and en-hancemen[A].In:Proceedings of EUSIPCO'98[C],Island of Rhodes,Greek:EVRASIP,1998.319～322
    8 Kingsbury N G.A dual-tree complex wavelet trans-form with improved orthogonality and symmetry prop-erties[A].In:Proc.IEEE Int Conf Image Processing[C],Vancouver,Canada,2000.2:375～378
    9易翔,王蔚然.一种概率自适应图像去噪模型[J].电子学报,2005,33(1):63～66
    10刘西宁,刘司红,杜贤等.复杂山地低噪比地震资料处理方法[J].勘探地球物理进展,2004,27(1):41～44
    11宁俊瑞.图像数字处理技术在叠后去噪中的应用[J].石油物探,2006,45(1):70～73
    12陈香朋,曹思远.第二代小波变换及其在地震信号去噪中的应用[J].石油物探,2004,43(6):547～550