Prior-Driven Self-Supervised Lightweight Method for Seismic Signal Denoising

TL;DR

This paper introduces a lightweight, prior-driven self-supervised neural network for seismic signal denoising that requires minimal parameters and no training data, effectively reducing noise in both synthetic and real data.

Contribution

The paper presents a novel, extremely parameter-efficient neural network that leverages priors for self-supervised learning, eliminating the need for large training datasets in seismic denoising.

Findings

Effectively reduces random noise in seismic signals

Achieves high denoising performance with only 2464 parameters

Validated on synthetic and field data

Abstract

Seismic exploration is currently the most mature approach for studying subsurface structures, yet the presence of noise greatly restricts its imaging accuracy. Previous methods still face significant challenges: traditional computational methods are often computationally complex and their effectiveness is hard to guarantee; deep learning methods rely heavily on datasets, and the complexity of network training makes them difficult to apply in practical field scenarios. In this paper, we proposed a neural network that has only 2464 learnable parameters, which is hundreds or even thousands of times lower than that of the current mainstream deep learning networks. And its parameter constraints rely on priors rather than requiring training data. We proposed two types of priors: the local prior and the global variance prior for self-supervised learning, and put forward low-scale learning to further enhance its performance in noise processing. We validated our method on both synthetic and field data, and the results indicate that our proposed approach effectively attenuates random noise.