Research On Seismic Data Signal-to-Noise Ratio Improvement Method Based On Improved Wavelet Transform

Generally,the collected seismic data is affected by various factors and is often mixed with a lot of noise.If the noise interference in the seismic signal can be effectively eliminated,this will provide a great guarantee for the subsequent interpretation work.Therefore,this paper starts research from the direction of eliminating the random noise of seismic signals and improving the signal-to-noise ratio of seismic data.The main research contents are as follows:Firstly,this paper introduces the content and process of wavelet transform from theoretical knowledge to practical application,and deeply analyzes the advantages and disadvantages of commonly used wavelet threshold selection algorithms.Then,the wavelet transform threshold denoising algorithm is improved to remove the random noise of seismic data.Aiming at the problem that the commonly used wavelet threshold denoising algorithm often calculates the threshold value by pre-estimating the noise square,there is a large error in the calculation of the threshold.The GCV threshold selection function that does not require the participation of the noise variance is selected as the basis function for threshold selection,and combined with the simulated annealing algorithm The wavelet threshold is optimized with Drosophila optimization algorithm to obtain the optimal threshold.The improved algorithm solves the problem of local optimal threshold,and can dynamically adjust the search step size of the algorithm according to the current iteration number and select a more appropriate starting point of the new iteration on the premise of ensuring the complexity of the algorithm,so as to obtain the optimal threshold of the wavelet transform.Finally,the wavelet transform based on NLM algorithm is improved to remove random noise from seismic data.In this paper,in view of the spatial redundancy of seismic random noise and the problem that the traditional NLM algorithm only uses fixed filter parameters to cause unsatisfactory denoising effects,the analysis of the noise variance calculation when one-dimensional wavelet entropy is applied to the NLM algorithm is only After the shortcomings obtained by averaging the one-dimensional single-channel signal,the two-dimensional wavelet entropy is combined with the NLM algorithm,and the filter parameters are adjusted by the noise variance calculated by the two-dimensional wavelet entropy,and the parameter error is optimized by the control factor.Improve the denoising effect.The improved algorithm is applied to simulated seismic records and actual seismic data,and the effectiveness of the proposed method is proved by comparing the data before and after processing.