小波时空变阈值去噪方法在可控震源资料处理中的应用
|
摘要
鉴于常规的消除地震资料中随机噪声的增强信号或压制噪声方法的局限性,本文基于小波变换阈值去噪和加权叠加的思路,提出一种小波时空变阈值去噪方法:以逐点方差为加权叠加的权系数,通过迭代计算修正权系数得到准确的时空变方差,再进行小波阈值去噪,从而做到多道识别噪声,单道保真压噪。这种方法可消除满足高斯白噪条件和不满足高斯白噪条件的随机噪声。经理论模型和相关未叠加可控震源资料的处理证明,这种小波时空变阈值去噪方法有效而实用。
Due to the limitation of the regular signal enhancement and noise elimination methods which are used to remove the noise in seismic data,in this paper based on the idea of wavelet transform threshold denoising and weighted stacking a wavelet time-space-varying threshold denoising method is presented.With point-to-point variance as weight coefficient for weighted stacking and through iterative calculation to modify weight coefficient,the accurate time-space-varying variance can be obtained,then wavelet threshold denoising is carried out,at last noise can be identified in multi-traces,and the true amplitude can be kept and the noise can be removed in single trace.With this method the random noise can be removed whether or not it meets the Gaussian white noise conditions.Theoretical modeling and related non-stacked vibroseis data processing show that the wavelet time-space-varying threshold denoising method is an effective tool and it is an applicable and practicable denoising method.
Due to the limitation of the regular signal enhancement and noise elimination methods which are used to remove the noise in seismic data,in this paper based on the idea of wavelet transform threshold denoising and weighted stacking a wavelet time-space-varying threshold denoising method is presented.With point-to-point variance as weight coefficient for weighted stacking and through iterative calculation to modify weight coefficient,the accurate time-space-varying variance can be obtained,then wavelet threshold denoising is carried out,at last noise can be identified in multi-traces,and the true amplitude can be kept and the noise can be removed in single trace.With this method the random noise can be removed whether or not it meets the Gaussian white noise conditions.Theoretical modeling and related non-stacked vibroseis data processing show that the wavelet time-space-varying threshold denoising method is an effective tool and it is an applicable and practicable denoising method.
引文
[1]Donoho D L,Johnstone I M.Ideal spatial adaptation via wavelet shrinkage.Biometrika,1994,81:425～455
[2]王坤明.自动选取阈值方法比较研究.抚顺石油学院学报,2002,27(2):70～73
[3]Donoho D L.De-noising by soft-thresholding.IEEE Trans on IT,1995,41:613～627
[4]董永生,羿旭明.基于四种改进阈值的小波去噪方法.数学杂志,2006,26(6):473～477
[5]王云专,郑江峰,孙立志等.基于小波变换的加权叠加.大庆石油学院学报,2005,29(1):9～11
[6]Ramesh Neelamani,Thomas A Dickens and Max De-enbaugh.Stack-and-denoise:A new method to stack seismic datasets.SEG Technical Program Ex-panded Abstracts,2006,25:2827～2831
[2]王坤明.自动选取阈值方法比较研究.抚顺石油学院学报,2002,27(2):70～73
[3]Donoho D L.De-noising by soft-thresholding.IEEE Trans on IT,1995,41:613～627
[4]董永生,羿旭明.基于四种改进阈值的小波去噪方法.数学杂志,2006,26(6):473～477
[5]王云专,郑江峰,孙立志等.基于小波变换的加权叠加.大庆石油学院学报,2005,29(1):9～11
[6]Ramesh Neelamani,Thomas A Dickens and Max De-enbaugh.Stack-and-denoise:A new method to stack seismic datasets.SEG Technical Program Ex-panded Abstracts,2006,25:2827～2831
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心