Reconstruction Of Shallow Seismic Data Based On Compressive Sensing Theory

As an important geophysical exploration method,seismic exploration has been widely used in oil and gas resources exploration,coalfield and engineering geological exploration,regional geological research and crustal research.Among them,shallow seismic exploration has a wide range of applications in engineering construction,hydrology and environmental geological exploration and so on.It is a high-precision,low-cost engineering geophysical exploration method.The acquisition of seismic data seriously affects the accuracy of imaging results.In an ideal situation,the collected seismic data should be regular,dense,and satisfy the sampling theorem.Due to the influence of terrain environment,surface obstacles,instrument hardware and other factors during the actual acquisition process,the collected data is always irregular.The missing track,that is,the seismic record presents irregular sampling or sparse sampling in the spatial direction.The lack of irregularity of seismic data often causes the generation of spatial aliasing,reduces the signal-to-noise ratio of seismic data processing results,and affects the imaging effect of seismic data.At the same time,shallow earthquakes often require high detection accuracy,and the loss of the track will eventually reduce the success rate of seismic exploration.In recent years,the research and application of seismic data reconstruction methods based on compressed sensing theory have many research results in petroleum seismic exploration.In the shallow seismic exploration,there are also missing problems,but due to the signal-to-noise ratio of shallow seismic data is not high and the cost of acquisition and processing of shallow seismic data.For the reasons of cost,etc.,seismic data reconstruction methods are not used,and traditional processing methods will affect the imaging effect of seismic data.The traditional signal processing method is limited by Nyquist-Shannon sampling theorem.When processing data with irregular low sampling rate,the effect is very poor.This paper is free from the limitation of Nyquist-Shannon's theorem,based on the theory of compressed sensing,the application of seismic data reconstruction method in shallow seismic exploration can not only achieve good processing results,improve detection accuracy and success rate,but also reduce seismic data sampling rate and save seismic exploration cost.In this paper,the curvelet transform is used to construct the sparse transform base,and a sparse inversion model is constructed by constructing an approximation function of 0-norm.In this paper,seismic data is simulated in various ways and reconstruction experiments are carried out.Experiments show that the proposed method can reconstruct the random missing seismic data very well.Compressed sensing sampling and data reconstruction were carried out on seismic survey data of active faults in a certain urban area and regional geological surveys in a certain area.Seismic data processing was performed on the original data and reconstruction data respectively.The offset imaging results were verified based on the potential value of compressed sensing reconstruction methods in shallow seismic exploration is of great significance for improving the accuracy and success rate of shallow seismic exploration.