Seismic Fault SAM: Adapting SAM with Lightweight Modules and 2.5D Strategy for Fault Detection

TL;DR

This paper introduces Seismic Fault SAM, a novel approach that adapts the Segment Anything Model with lightweight modules and a 2.5D strategy, significantly improving seismic fault detection performance with limited data.

Contribution

It pioneers applying the SAM model to seismic fault interpretation by designing lightweight adapters, employing a 2.5D input strategy, and integrating geological constraints for better generalization.

Findings

Surpasses existing 3D models on seismic fault detection metrics.

Achieves state-of-the-art performance on the Thebe dataset.

Demonstrates effective extension to other seismic tasks with limited labeled data.

Abstract

Seismic fault detection holds significant geographical and practical application value, aiding experts in subsurface structure interpretation and resource exploration. Despite some progress made by automated methods based on deep learning, research in the seismic domain faces significant challenges, particularly because it is difficult to obtain high-quality, large-scale, open-source, and diverse datasets, which hinders the development of general foundation models. Therefore, this paper proposes Seismic Fault SAM, which, for the first time, applies the general pre-training foundation model-Segment Anything Model (SAM)-to seismic fault interpretation. This method aligns the universal knowledge learned from a vast amount of images with the seismic domain tasks through an Adapter design. Specifically, our innovative points include designing lightweight Adapter modules, freezing most of the pre-training weights, and only updating a small number of parameters to allow the model to converge quickly and effectively learn fault features; combining 2.5D input strategy to capture 3D spatial patterns with 2D models; integrating geological constraints into the model through prior-based data augmentation techniques to enhance the model's generalization capability. Experimental results on the largest publicly available seismic dataset, Thebe, show that our method surpasses existing 3D models on both OIS and ODS metrics, achieving state-of-the-art performance and providing an effective extension scheme for other seismic domain downstream tasks that lack labeled data.