1887

Abstract

Summary

In seismic interpretation, identifying and mapping geological features is tedious, time-consuming and prone to errors, especially in 3D. While deep learning algorithms have helped automate this process, the diverse morphologies of geological objects remain challenging. Instead of building separate models for each type of geological object, we propose an alternative approach that identifies objects based on how they stand out from the background seismic response. To do so, we have adapted a general purpose state-of-the-art segmentation model specifically for geological object extraction.

Case studies from the Gulf of Mexico and Campos Basin demonstrate the effectiveness of the proposed approach for salt diapirs and submarine fan systems extraction. The method proves equally effective for other geological features such as channels, injectites, and canyons, provided they stand out from the background seismic response. By combining innovative deep learning segmentation with interactive interpretation tools, we have developed a robust yet intuitive workflow for 3D geological body extraction that leverages both geological expertise and the automation capabilities of artificial intelligence.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202539052
2025-03-24
2025-11-16

  • From This Site

    /content/papers/10.3997/2214-4609.202539052
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Evano, L. and Cubizolle, F. [2023]. Manual Active Learning for Salt Interpretation: An Empirical Study to Avoid Forgetting During Incremental Trainings. European Association ofGeoscientists & Engineers. https://doi.org/10.3997/2214-4609.202332023.
     [Google Scholar]
  2. Lowell, D., Lipton, Z.C. and Wallace, B.C. [2019]. Practical Obstacles to Deploying Active Learning. arXiv preprint. https://arxiv.org/abs/1807.04801.
     [Google Scholar]
  3. Milosavljevic, A. [2020]. Identification of Salt Deposits on Seismic Images Using Deep Learning Method for Semantic Segmentation. ISPRS International Journal of Geo-Information, 9(1), 24. https://doi.org/10.3390/ijgi9010024.
     [Google Scholar]
  4. Osco, L.P., Wu, Q., de Lemos, EX., Gonçalves, W.N., Ramos, A.P.M., Li, J. and Junior, J.M. [2023]. The Segment Anything Model (SAM) for Remote Sensing Applications: From Zero to One Shot. arXiv preprint, https://arxiv.org/abs/2306.16623.
     [Google Scholar]
  5. Pauget, F., Lacaze, S., and Valding, T. [2009]. A global approach in seismic interpretation based on cost function minimization. SEG Technical Program Expanded Abstracts, 2592–2596. https://doi.org/10.1190/l.3255384.
     [Google Scholar]
  6. Pratama, H. and Latiff, A.H.A. [2022]. Automated Geological Features Detection in 3D Seismic Data Using Semi-Supervised Learning. Applied Sciences, 12(13), 6723. https://doi.org/10.3390/app12136723.
     [Google Scholar]
  7. Ravi, N., Gabeur, V., Hu, Y.-T., Hu, R, Ryali, C., Ma, T., Khedr, H., Rädle, R., Rolland, C., Gustafson, L., Mintun, E., Pan, J., Alwala, K.V., Carion, N., Wu, C.-Y., Girshick, R., Dollar, P., and Feichtenhofer, C. [2024]. SAM 2: Segment Anything in Images and Videos. arXiv preprint. https://arxiv.org/abs/2408.00714.
     [Google Scholar]
  8. Zhang, Y., Shen, Z. and Jiao, R. [2024b]. Segment Anything Model for Medical Image Segmentation: Current Applications and Future Directions. arXiv preprint, https://arxiv.org/abs/2401.03495.
     [Google Scholar]
  9. Zhang, Z., Chen, R. and Ma, J. [2024a]. Improving Seismic Fault Recognition with Self-Supervised Pre-Training: A Study of 3D Transformer-Based with Multi-Scale Decoding and Fusion. Remote Sensing, 16(5), 922. https://doi.org/10.3390/rs16050922.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202539052
Loading
Interactive Geobody Extraction: Combining Expert Knowledge with Deep Learning Segmentation
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.202539052
/content/papers/10.3997/2214-4609.202539052
/content/papers/10.3997/2214-4609.202539052
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