Seismic Data Denoising Method Based On Compressed Sensing Theory

Due to the complex working environment of seismic exploration,the collected seismic data often contains noise.The existence of noise brings interference to the utilization of effective signals of seismic data.Sometimes most of the effective information is submerged and it is difficult to identify,which seriously affects the subsequent data.Processing and interpreting work,so how to effectively suppress random noise in seismic data has become an important task in the field of seismic data processing.This paper is aimed at the noise problem existing in seismic data.From the perspective of compressed sensing,this paper deeply studies a seismic data denoising method.The main research contents are as follows:(1)The principle and advantages and disadvantages of traditional noise reduction methods are analyzed.The method of random noise in seismic data is presented.The related theory and the principle of noise reduction are elaborated in detail,which provides a theoretical basis for subsequent simulation experiments.(2)The idea of noise reduction method in this paper is introduced and the model is constructed.The specific noise reduction process of this method is given.In the data reconstruction part of this method,the reconstruction principle,advantages and disadvantages and research status of the reconstruction algorithm are analyzed.The orthogonal matching pursuit(OMP)algorithm involved is improved in the selection of the atomic part.(3)The related parameters involved in the method are discussed,and multiple sets of simulation experiments are designed.The appropriate values of relevant parameters are determined through experiments.(4)The simulation experiment design is carried out,and the noise reduction method and wavelet denoising method given in this paper,as well as the DCT-random Gauss-ROMP combined from the selection of sparse basis,the selection of measurement matrix and the selection of reconstruction algorithm are selected.,DCT-random Gaussian-OMP,FFT-random Gaussian-OMP,DCT-circulant matrix-OMP,etc.,respectively,simulated and contrast experiments.Introduced signal-to-noise ratio(SNR)and peak signal-to-noise ratio(PSNR)as evaluation indicators to verify the noise reduction The noise reduction effect of the method and the simulation results show that for the simulation experimental data given in this paper,the noise reduction method of this method is compared with other comparison methods to reduce noise.The seismic data has high signal-to-noise ratio(SNR)and peak signal-to-noise ratio(PSNR),which indicates that the proposed method has better suppression effect on random non-stationary noise in seismic data,and has achieved ideal noise reduction effect,which verifies the effectiveness of the proposed method.