Nuclear norm and generalized total variation joint constraint seismic random noise suppression method

Abstract

The invention provides a nuclear norm and generalized total variation joint constraint seismic random noise suppression method. The method comprises the steps of (1) inputting a three-dimensional seismic data volume; (2) constructing a target function; (3) decomposing the target function into two sub-problems by adopting a split Bergman iterative strategy; (4) solving the first sub-problem by adopting a singular value soft thresholding algorithm, and solving an auxiliary variable matrix; (5) solving a second sub-problem by adopting a gradient projection algorithm, and solving a seismic data matrix; (6) judging whether Bergman iteration is stopped or not, and if so, outputting a seismic data matrix; and (7) completing calculation of the whole seismic data volume to obtain a final denoised seismic data volume. According to the method, generalized total variation constraints in image processing, nuclear norm constraints in matrix analysis and split Bergman iteration of an optimization theory are organically combined, the method is suitable for seismic data random noise suppression processing and the like, and the signal-to-noise ratio and fidelity of seismic data are effectively improved.

Description

technical field [0001] The invention belongs to the technical field of petroleum geophysical exploration, and the specific technology is a seismic random noise suppression method constrained jointly by nuclear norm and generalized total variation, which can effectively suppress random noise in seismic data, improve the signal-to-noise ratio of seismic data, and maintain The local smoothness of seismic data improves the fidelity of seismic data. Background technique [0002] With the development of petroleum exploration to today, exploration and development are gradually shifting to complex, subtle, deep and unconventional oil and gas reservoirs, and the requirements for seismic data quality are getting higher and higher. Affected by the complex acquisition environment, the seismic data collected in the field often contain strong random noise, and even the effective information is completely submerged in severe cases. Seismic random noise suppression is an important problem ...

AI Technical Summary

Problems solved by technology

However, in many cases, the sparseness of seismic data may not be able to effectively represent the internal structural characteristics of seismic data, especially the waveform similarity between adjacent seismic traces

However, noise suppression methods based on the similarity of adjacent seismic trace waveforms, such as noise suppression with nuclear norm constraints, seldom take into account the local smoothness of seismic data, so that random noise suppresses seismic data at the edge of geological bodies. "staircase chatter" phenomenon

Images