1887

Abstract

Summary

Delimiting salt inclusions from migrated images during the velocity model building flow is a time-consuming activity that depends on highly human-curated analysis and is subject to interpretation errors or limitations of the images and methods available. We present a supervised Deep-Learning (DL) approach to predict the complete salt geometries included in 3D velocity models. The input for the DL algorithm is the subsurface offset gathers migrated with the sediment velocity but without salt inclusions, an approach which has been previously validated for 2D velocity models. To solve the 3D salt segmentation, we used a U-Net with Convolutional Long Short-Term Memory (C-LSTM) layers, where the 3D data is represented as a stack of 2D slices to the C-LSTM, avoiding the memory issues related to 3D convolutional layers. The training process tuned the DL algorithm to successfully learn the shape of complex 3D salt body masks with high accuracy and also performed well when applied to a synthetic benchmark data set that was not previously introduced in network training.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202310124
2023-06-05
2026-02-09

  • From This Site

    /content/papers/10.3997/2214-4609.202310124
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. AlAli, A. and Anifowose, F. [2022]. Seismic velocity modeling in the digital transformation era: a review of the role of machine learning. J Petrol Explor Prod Technol12, 21–34.
     [Google Scholar]
  2. AlAli, A., Kazei, V., Kalita, M. and Alkhalifah, T. [2022]. Deep learning unflooding for robust subsalt waveform inversion. Geophysical Prospecting.
     [Google Scholar]
  3. Aminzadeh, F., Burkhard, N., Long, J. Kunz, T., Duclos, P. [1996]. Three dimensional seg/eaeg models - an update: The Leading Edge, 15, 131–134.
     [Google Scholar]
  4. Arbelle, A. and Raviv, T. R. [2019]. Microscopy Cell Segmentation Via Convolutional LSTM Networks, 2019 IEEE 16th International Symposium on Biomedical Imaging, 1008–1012.
     [Google Scholar]
  5. Chen, J., Yang, L., Zhang, Y., Alber, M., Chen, D. Z. [2016]. Combining fully convolutional and recurrent neural networks for 3d biomedical image segmentation, Advances in Neural Information Processing Systems, 3036–3044.
     [Google Scholar]
  6. Dellinger, J., Brenders, A J., Sandschaper, J.R., Regone, C., Etgen, J., Ahmed, I., Lee, K. J. [2017]. The garden banks model experience, The Leading Edge, 36, 151–158.
     [Google Scholar]
  7. Klatt, M., Faria, E.L., Muller, A.P.O., Coelho, J.M., González, J.L., de Albuquerque, M.P., Bom, C.R., Correia, M.D. [2022]. Deep learning strategy for salt model building, Geophysics87, IM221–IM233.
     [Google Scholar]
  8. Muller, A.P.O., Costa, J.C., Bom, C.R., Faria, E.L., Klatt, M., Teixeira, G., de Albuquerque, M.P. [2022]. Complete identification of complex salt geometries from inaccurate migrated subsurface offset gathers using deep learning, Geophysics87, R453–R463.
     [Google Scholar]
  9. Sava, P. C., and Fomel, S. [2003]. Angle-domain common-image gathers by wavefield continuation methods, Geophysics68, 1065–1074
     [Google Scholar]
  10. Shi, Y., Wu, X., Fomel, S. [2019]. SaltSeg: Automatic 3D salt segmentation using a deep convolutional neural network, Interpretation7, SE113–SE122
     [Google Scholar]
  11. Solovyev, R., Kalinin, A.A., Gabruseva, T. [2022]. 3D convolutional neural networks for stalled brain capillary detection, Computers in Biology and Medicine141, 105089.
     [Google Scholar]
  12. Yu, S. and Ma, J. [2021]. Deep learning for geophysics: Current and future trends, Reviews of Geophysics59, e2021RG000742.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202310124
Loading
Deep-salt: 3D salt segmentation from inaccurate migrated subsurface offset gathers using convolutional LSTM layers
Conference Proceedings 2023, 1 (2023); https://doi.org/10.3997/2214-4609.202310124
/content/papers/10.3997/2214-4609.202310124
/content/papers/10.3997/2214-4609.202310124
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