A Comparative Evaluation of Self-Supervised Methods Applied to Rock Images Classification - Université d'Évry Access content directly
Conference Papers Year : 2024

A Comparative Evaluation of Self-Supervised Methods Applied to Rock Images Classification

Abstract

Digital Rock Physics DRP is a discipline that employs advanced computational techniques to analyze and simulate rock properties at the pore-scale level. Recently, Self-Supervised Learning (SSL) has shown promising outcomes in various application domains, but its potential in DRP applications remains largely unexplored. In this study, we propose to assess several self-supervised representation learning methods designed for automatic rock category recognition. Hence, we demonstrate how different SSL approaches can be specifically adapted for DRP, and comparatively evaluated on a new dataset. Our objective is to leverage unlabeled micro-CT (Computed Tomography) image data to train models that capture intricate rock features and obtain representations that enhance the accuracy of classical machine-learning-based rock images classification. Experimental results on a newly proposed rock images dataset indicate that a model initialized using SSL pretraining outperforms its non-self-supervised learning counterpart. Particularly, we find that MoCo-v2 pretraining provides the most benefit with limited labeled training data compared to other models, including supervised model.
Fichier principal
Vignette du fichier
VISAPP_2024_88_CR.pdf (848.33 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Licence

Dates and versions

hal-04470950 , version 1 (21-02-2024)

Licence

Identifiers

Cite

Van Thao Nguyen, Dominique Fourer, Desiré Sidibé, Jean-François Lecomte, Souhail Youssef. A Comparative Evaluation of Self-Supervised Methods Applied to Rock Images Classification. 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2024), Feb 2024, Rome, Italy. pp.393--400, ⟨10.5220/0012319400003660⟩. ⟨hal-04470950⟩
98 View
55 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More