1887

Abstract

Summary

The analysis and description of petrographic thin sections is a critical, highly specialized and time-consuming activity. Recently, a lot of effort has been made to digitalize this process: image acquisition equipment has been integrated within microscopes, and a vast collection of thin section images is being constantly acquired. This broadens the opportunity of applying machine learning techniques to facilitate the current workflow.

However, supervised methodologies rely heavily on labeled data. Labeling thin sections is expensive and, considering the large passive of yet unanalyzed sections, mostly impractical. Unsupervised and self-supervised approaches could allow the generation of insights without the imposition of workload overhead. If associated with section descriptions and metadata - which can be used as weak labels, or as multimodal training data - they may prove powerful tools for accelerating processes.

This work, then, proposes an initial attempt at developing a self-distillation based self-supervised approach capable of creating high-quality latent representation spaces of petrographic thin sections - playfully referred to as the equivalent to a Thin Section DNA. We aim to create a foundation over which it is possible to develop, without the need for extensive labeling work, applications ranging from similarity retrieval of sections to multimodal automatic description generation.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202539051
2025-03-24
2026-02-15

  • From This Site

    /content/papers/10.3997/2214-4609.202539051
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Buono, A., Fullmer, S., Luck, K., Peterson, K., king, H., More, P., and LeBlanc, S. [2019]. Quantitative Digital Petrography: Full Thin Section Quantification of Pore Space and Grains. Presented at the SPE Middle East Oil and Gas Show and Conference, https://doi.org/10.2118/194899-ms
     [Google Scholar]
  2. Caron, M., Touvron, H., Misra, I., Jegou, H., Mairal, J., Bojanowski, P., and Joulin, A. [2021]. Emerging Properties in Self-Supervised Vision Transformers. https://doi.org/10.48550/ARXIV.2104.14294
     [Google Scholar]
  3. Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C, Gustafson, L., … Girshick, R. [2023]. Segment Anything. https://doi.org/10.48550/ARXIV.2304.02643
     [Google Scholar]
  4. Knight, P. A. [2008]. The Sinkhorn-Knopp Algorithm: Convergence and Applications. https://doi.org/10.1137/060659624
     [Google Scholar]
  5. Korenchendler, A., Sobreira, M., and Pesce, P. [2023]. An unsupervised approach for geoscientific data similarity analysis: well logs and thin sections applications, https://doi.org/10.3997/2214–4609.202310811
     [Google Scholar]
  6. Liu, H., Ren, Y.-L., Li, X., Hu, Y.-X., Wu, J.-P., Li, B., … Fang, W.-K. [2022]. Rock thin-section analysis and identification based on artificial intelligent technique. https://doi.org/10.1016/j.petsci. 2022.03.011
     [Google Scholar]
  7. Oquab, M., Darcet, T., Moutakanni, T., Vo, H., Szafraniec, M., Khalidov, V., … Bojanowski, P. [2023]. DINOv2: Learning Robust Visual Features without Supervision. https://doi.org/10.48550/ARXIV.2304.07193
     [Google Scholar]
  8. Pattnaik, S., Chen, S., Helba, A., and Ma, S. [2020]. Automatic Carbonate Rock Facies Identification with Deep Learning. Presented at the SPE Annual Technical Conference and Exhibition. https://doi.org/10.2118/201673-ms
     [Google Scholar]
  9. Rubo, R. A., de Carvalho Carneiro, C, Michelon, M. F., and Gioria, R. dos S. [2019]. Digital petrography: Mineralogy and porosity identification using machine learning algorithms in petrographic thin section images. https://doi.org/10.1016/j.petrol.2019.106382
     [Google Scholar]
  10. Sablayrolles, A., Douze, M., Schmid, C., and Jegou, H. [2018]. Spreading vectors for similarity search. https://doi.org/10.48550/ARXIV.1806.03198
     [Google Scholar]
  11. Sylvester, Z. [2024]. Segment Every Grain. https://github.com/zsylvester/segmenteverygrain
     [Google Scholar]
  12. Tang, D. G., Milliken, K. L., and Spikes, K. T. [2020]. Machine learning for point counting and segmentation of arenite in thin section. https://doi.org/10.1016/j.marpetgeo.2020.104518
     [Google Scholar]
  13. Xu, Y., Dai, Z., and Luo, Y. [2020]. Research on Application of Image Enhancement Technology in Automatic Recognition of Rock Thin Section. https://doi.org/10.1088/1755–1315/605/1/012024
     [Google Scholar]
  14. Zhou, J., Wei, C., Wang, H., Shen, W., Xie, C., Yuille, A., and Kong, T. [2021]. iBOT: Image BERT Pre-Training with Online Tokenizer. https://doi.org/10.48550/ARXIV.2111.07832
     [Google Scholar]
/content/papers/10.3997/2214-4609.202539051
Loading
Thin Section DNA: A Self-supervised Approach for Petrographic Thin Section Analysis
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.202539051
/content/papers/10.3997/2214-4609.202539051
/content/papers/10.3997/2214-4609.202539051
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error