A Variable Vertex Sparse Hyperbolic Radon Transform Method Considering Seismic Wavelet Stretch Effect

Abstract

The invention discloses a vertex variable sparse hyperbolic Radon transform method considering a seismic wavelet stretching effect. According to the method, the vertex positions and velocities of hyperbolic events are calculated by using the dip angles and the derivatives thereof of seismic events; the number of the hyperbolic events is determined through using a clustering algorithm, and the number of parameters of the estimated events is decreased; based on the calculated vertex positions and velocities of the events, a vertex variable hyperbolic Radon transform operator is constructed by using a method according to which windowing is adopted to segment a seismic channel, and the seismic wavelet stretching effect generated by inverse Radon transform is eliminated; a Radon domain model and the L1 norm minimization constrain of a data fitting term are introduced into an optimization problem, so that the sparsity of the Radon domain model and the adaptability of the Radon domain model to non Gaussian fitting error can be improved; a multiplier alternating direction method is utilized to solve an L1-L1 norm minimization problem; and in a whole iterative process, matrix inversion is only needed to be carried out for once, so that computational complexity is low.

Description

technical field [0001] The invention belongs to the field of seismic signal processing in seismic exploration technology, and in particular relates to a variable vertex sparse hyperbolic Radon transform method considering the stretching effect of seismic wavelets. Background technique [0002] Radon transform is widely used in seismic signal processing, such as multiple wave suppression, seismic data regularization, and wave field separation. Its main idea is to map seismic data from time-offset domain to Radon domain. According to the integral trajectory, Radon transformation can be divided into three categories: linear Radon transformation, parabolic Radon transformation and hyperbolic Radon transformation. [0003] Among them, the hyperbolic Radon transform can be directly applied to the common center point gather or common shot point gather without motion correction, and the hyperbolic trajectory is used to construct the transformation operator, which can better represen...

AI Technical Summary

Problems solved by technology

The traditional variable vertex sparse hyperbolic Radon transform method uses the iterative reweighted least squares algorithm to solve the L1 norm minimization optimization problem, and needs to calculate the matrix inversion once in each iteration step, and the computational complexity is high

Therefore, the traditional variable vertex sparse hyperbolic Radon transform has a large amount of calculation, which limits the practicability of this transform method

In addition, the traditional hyperbolic Radon transform does not consider the existence of seismic wavelets when constructing hyperbolic Radon transform operators for forward and inverse transformations. In the process of reconstructing seismic data using inverse Radon transforms, seismic wavelets will appear. Stretch effect, causing seismic signal distortion

Images