DuRIN: A Deep-unfolded Sparse Seismic Reflectivity Inversion Network

TL;DR

This paper introduces DuRIN, a neural network based on deep unfolding for sparse seismic reflectivity inversion, improving subsurface interface detection from seismic data.

Contribution

It proposes a novel deep-unfolded neural network with a weighted minimax-concave penalty for seismic reflectivity inversion, combining model-based and data-driven methods.

Findings

Outperforms benchmark techniques on synthetic and real seismic data.

Effective in recovering subsurface interface locations and reflection coefficients.

Validated on 1-D, 2-D, and real seismic datasets.

Abstract

We consider the reflection seismology problem of recovering the locations of interfaces and the amplitudes of reflection coefficients from seismic data, which are vital for estimating the subsurface structure. The reflectivity inversion problem is typically solved using greedy algorithms and iterative techniques. Sparse Bayesian learning framework, and more recently, deep learning techniques have shown the potential of data-driven approaches to solve the problem. In this paper, we propose a weighted minimax-concave penalty-regularized reflectivity inversion formulation and solve it through a model-based neural network. The network is referred to as deep-unfolded reflectivity inversion network (DuRIN). We demonstrate the efficacy of the proposed approach over the benchmark techniques by testing on synthetic 1-D seismic traces and 2-D wedge models and validation with the simulated 2-D Marmousi2 model and real data from the Penobscot 3D survey off the coast of Nova Scotia, Canada.