Random Interference Of The Seismic Data To Suppress The Technology Research

For low signal-to-noise ratio(SNR) data, the primary task of seismic data processing is to suppress random noise and improve the SNR, which is also the difficult problem in seismic data denoising field.The paper starts with description of characteristics and classification of random noise,and then several methods of suppressing random noise are briefly introduced and analysised.This paper main study singular value decomposition(SVD) and wavelet analysis method to suppress random noise.SVD is a coherency-based technique that provides both signal enhancement and noise suppression.When signal in seismic gather with high laterally coherent,SVD filtering is very successful at suppressing random noise on condition of minimal signal damagement.However,traditional SVD can't effectively cope with discontinuous, isolated and dipping events. It filter out some of non-lateral coherent signal and resulting in waste of seismic data while suppressing the random noise in seismic data processing.The paper presents a modified approach that is based on traditional SVD, which can overcome the limitations of traditional SVD. Data within a local window are first extracted. Any coherent events in window are aligned laterally and then SVD is used to enhence lateral events, simultaneously suppress random noise.Finally, the output data are shifted back to their original positions. The results of synthetic and field data processing show that local SVD, compare with global SVD,can effectively cope with discontinuous and dipping events, as well as improve SNR of seismic data.Wavelet analysis is a common tool for analyzing localized variations of signal in time-frequency space.Based on the study of wavelet denoising theory, threshold filter in wavelet domain is used for suppressing random noise.The method apply different threshold approach deal with wavelet coefficients in different wavelet resolution,according to the distribution of noise in signal component. After that,the treated coefficients are used for reconstructing data by reverse wavelet transform.The denoising results of Synthesis and real data are satisfactory. Finally, Combining wavelet denoising with local SVD overcomes the main weaknesses associated with each individual method.The experiment proves that its results better than that of applying one of them alone.