摘要
水平叠加虽然在很大程度上压制了噪声,提高了地震剖面的信噪比,但CMP道集上还存在不少不是一次波的规则干扰和随机噪声,不利于叠前资料的岩性反演和叠后资料的波阻抗反演.本文提出了基于改进的正交多项式变换压制地震资料中随机噪声的方法,其优势在于:通过对不同时间信号的奇异值分解,确定有效信号正交多项式系数谱的阶数;再利用小波变换,改善有效信号和噪声在低阶上的混叠.文中给出了具体处理的过程,数据试验和实际资料的处理结果表明该方法不仅能有效地压制噪声,而且还能较好地保护地震数据中AVO变化特征.
To a large extent, Horizontal stacking suppresses the noise and improves the signal to noise ratio of seismic profile, but there remains much organized noise such as non-primary wave and random noise in the CMP gather. In the meantime, it will not be helpful for lithological inversion on pre-stack data and wave impedance inversion on post-stack data. Based on the problems above, a novel method for seismic noise reduction is put forward on the basis of an improved orthogonal polynomials transform in seismic material. Compared with conventional orthogonal polynomials transform, the proposed method has some superiorities： to determine orders of spectrum of orthogonal polynomials coefficients of effective signal by SVD for the signal at different time; to improve the aliasing problem of effective signal and noise at lower order in some way by wavelet transform. The specific processing procedure is given, and results of synthetic and field data examples show that the improved method can effectively suppress seismic noise and preserve the variation characteristics of AVO in seismic data.
出处
《地球物理学进展》
CSCD
北大核心
2013年第4期2085-2093,共9页
Progress in Geophysics
基金
国家重大专项课题(2011ZX05024-001-01)
国家自然科学基金(41140033)联合资助
关键词
去噪
AVO
正交多项式系数谱
SVD技术
小波变换
De-noising, AVO, spectrum of orthogonal polynomials coefficients, singular value decomposition（SVD） technique, wavelet transform
