High-precision seismic data reconstruction method based on two-dimensional non-uniform curvelet transform

Abstract

The invention discloses a high-precision seismic data reconstruction method based on two-dimensional non-uniform curvelet transform. The high-precision seismic data reconstruction method is characterized in that non-uniform Fourier transform is adopted on the basis of multi-scale multidirectional two-dimensional curvelet forward transform, forward and inverse transform operators between a curvelet coefficient and seismic missing data under spatial non-uniform sampling are established, a spectral projected gradient method is adopted for inversion calculation to obtain a uniform curvelet coefficient under the minimal L1 norm constraint, and the uniform curvelet coefficient undergoes standard two-dimensional curvelet inverse transform processing, thereby finally implementing the high-precision seismic data reconstruction method based on two-dimensional non-uniform curvelet transform. The high-precision seismic data reconstruction method overcomes the defect that the traditional two-dimensional curvelet transform method cannot reconstruct a seismic missing trace under non-uniform sampling, does not loss energy in the reconstruction process, has an anti-aliasing function, significantly improves fidelity and a signal-to-noise ratio of reconstructed signals, protects weak effective wave signals, thus makes co-phase axes of reflected waves more continuous, and is of reference significance in other fields.

Description

Technical field [0001] The invention relates to a seismic data reconstruction method of irregular missing traces under spatial non-uniform sampling, in particular to a high-precision seismic data reconstruction method based on two-dimensional non-uniform curvelet transform. technical background [0002] In the process of field data acquisition, due to the acquisition equipment, field terrain conditions, and economic cost constraints, seismic data is often under-sampling irregularly along the spatial direction, resulting in irregular and incomplete seismic data collected and spatial falsehoods. Frequency affects the effect of other subsequent processing methods and reduces the resolution of seismic exploration. In order to overcome this situation, the irregularly missing seismic data in the field must be reconstructed by pre-stack data so that the missing traces can be effectively restored. However, the precondition of most current data reconstruction methods is the reconstruction...

AI Technical Summary

Problems solved by technology

[0004] However, the previous reconstruction method based on two-dimensional curvelet transform requires spatially uniform sampling, because the conventional curvelet transform must first be applied to Fourier transform in the calculation process, and the prerequisite for Fourier transform is spatial uniformity. As a result, the previous two-dimensional curvelet transform can only deal with the missing reconstruction of seismic traces under spatially uniform sampling, but cannot directly reconstruct seismic data under spatially non-uniform sampling, which limits the further application of this method

Images