Self-Supervised Learning for Seismic Image Segmentation From Few-Labeled Samples

Bruno A. A. Monteiro

doi:10.1109/LGRS.2022.3193567

Self-Supervised Learning for Seismic Image Segmentation From Few-Labeled Samples

2022-01-01

13

PRE

AI

B

Bruno A. A. Monteiro *

H

Hugo Oliveira

J

Jefersson A. dos Santos

DOI：10.1109/LGRS.2022.3193567

Original

Original request for help

Save

Abstract

En 中文

Current deep learning methods for interpreting seismic images require large amounts of labeled data, and due to strategic and economic interests, these data are not plenty available. In this scenario, seismic interpretation can benefit from self-supervised learning (SSL) by relying on prior training without manually annotated labels within the target data domain and subsequent fine-tuning with few shots. To demonstrate the potential of such an approach, we conducted experiments with three classic context-based pretext tasks: rotation, jigsaw, and frame order prediction. Our results for 1, 5, 10, and 20 shots showed significant improvement for mean Intersection-over-Union (mIoU) measurements for semantic segmentation in most scenarios, outperforming the baseline method in 38% in the one-shot scenario for the F3 Netherlands Dataset and 16.4% in the New Zealand Parihaka dataset, and this gap grows even higher after performing ensemble modeling. These experiments suggest that applying SSL methods can also bring great benefits in seismic interpretation when few labeled data are available.

Keywords:

Task analysis

Image segmentation

Semantics

Training

Deep learning

Feature extraction

Data models

Convolutional neural network (CNN)

seismic image

self-supervised learning (SSL)

semantic segmentation

Self-Supervised Learning for Seismic Image Segmentation From Few-Labeled Samples

Abstract

Journal

Organization