Research On Optimization Of Seismic Data Reconstruction Algorithm Based On Compressed Sensing

Prestack seismic record reconstruction is a key step in seismic record processing.The quality of the reconstruction will directly affect the display of underground structures.Compressed sensing theory breaks through the limitations of the traditional sampling theorem and provides a new solution for the missing reconstruction of seismic data.In the framework of compressed sensing theory,we chose curvelet transform to get sparse coefficient.Then,the four commonly used reconstruction algorithms are studied in depth,and subjective evaluation and objective evaluation are combined to compare the reconstruction effects of seismic data with different degrees of missing.During the research process,it was found that the threshold iteration strategy is an important parameter that affects the reconstruction quality and running time.In view of the shortcomings of the existing iteration strategies,this paper proposes a nonlinear iteration strategy with a parameter q.Four commonly used algorithms are optimized,which improves the reconstruction quality of the algorithm and shortens the running time.Because the field data often contains both effective signals and noise,the noise immunity analysis of the optimized algorithm is performed.The experimental results show that the IHTA and POCS have strong noise immunity.Finally,the two optimization algorithms with better noise immunity are used for the missing reconstruction of the field data,which can also achieve good recovery results,which also shows that the optimization algorithm is practical.