Improving the quality control of seismic data through active learning

TL;DR

This paper introduces an active learning approach to improve seismic data quality control, reducing the need for extensive human labeling and enhancing efficiency in seismic signal processing.

Contribution

The paper presents a novel active learning method based on local error estimates, specifically designed for seismic data quality control, with demonstrated superior performance over existing strategies.

Findings

The proposed method outperforms alternative active learning strategies.

It effectively reduces labeling effort in seismic data QC.

Empirical results show improved accuracy on synthetic and real datasets.

Abstract

In image denoising problems, the increasing density of available images makes an exhaustive visual inspection impossible and therefore automated methods based on machine-learning must be deployed for this purpose. This is particulary the case in seismic signal processing. Engineers/geophysicists have to deal with millions of seismic time series. Finding the sub-surface properties useful for the oil industry may take up to a year and is very costly in terms of computing/human resources. In particular, the data must go through different steps of noise attenuation. Each denoise step is then ideally followed by a quality control (QC) stage performed by means of human expertise. To learn a quality control classifier in a supervised manner, labeled training data must be available, but collecting the labels from human experts is extremely time-consuming. We therefore propose a novel active learning methodology to sequentially select the most relevant data, which are then given back to a human expert for labeling. Beyond the application in geophysics, the technique we promote in this paper, based on estimates of the local error and its uncertainty, is generic. Its performance is supported by strong empirical evidence, as illustrated by the numerical experiments presented in this article, where it is compared to alternative active learning strategies both on synthetic and real seismic datasets.