1887
Volume 73, Issue 2
  • E-ISSN: 1365-2478

Abstract

Abstract

Deep learning‐based methods have performed well in seismic waveform inversion tasks in recent years, while the need for velocity models as labels has somewhat limited their application. Unsupervised learning allows us to train the neural network without labels. When inverting seismic velocity models from observed data, labels are often unavailable for real data. To address this problem and improve network generalization, we introduce a multi‐scale strategy to enhance the performance of unsupervised learning. The first ‘multi‐scale’ is derived from the conventional full waveform inversion strategy, in which the low‐, middle‐ and high‐frequency inversion results are successively predicted during the network training. Another ‘multi‐scale’ is to introduce multi‐scale similarity as an additional data loss term to improve the inversion results. With 12,000 samples from the overthrust model, our method obtains comparable results with the supervised learning method and outperforms unsupervised methods that rely only on the mean square error as a loss function. We compare the performance of the proposed method with multi‐scale full waveform inversion on the Marmousi model, and the proposed method achieves better results at low‐ and middle‐frequencies, and, as a result, it provides good initial models for further full waveform inversion updates.

© 2025 European Association of Geoscientists & Engineers. © 2025 European Association of Geoscientists & Engineers.
Loading

Article metrics loading...

/content/journals/10.1111/1365-2478.13665
2025-01-26
2025-11-12

  • From This Site

    /content/journals/10.1111/1365-2478.13665
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Alkhalifah, T. & Choi, Y. (2014) From tomography to full‐waveform inversion with a single objective function. Geophysics, 79, R55–R61.
     [Google Scholar]
  2. Alkhalifah, T. & Ovcharenko, O. (2022) Direct domain adaptation through reciprocal linear transformations. Frontiers in Artificial Intelligence, 5, 927676.
     [Google Scholar]
  3. Alkhalifah, T., Song, C., Waheed, U.b. & Hao, Q. (2021) Wavefield solutions from machine learned functions constrained by the Helmholtz equation. Artificial Intelligence in Geosciences, 2, 11–19.
     [Google Scholar]
  4. Alkhalifah, T., Wang, H. & Ovcharenko, O. (2022) Mlreal: bridging the gap between training on synthetic data and real data applications in machine learning. Artificial Intelligence in Geosciences, 3, 101–114.
     [Google Scholar]
  5. Araya‐Polo, M., Dahlke, T., Frogner, C., Zhang, C., Poggio, T. & Hohl, D. (2017) Automated fault detection without seismic processing. The Leading Edge, 36, 208–214.
     [Google Scholar]
  6. Askan, A. & Bielak, J. (2008) Full anelastic waveform tomography including model uncertainty. Bulletin of the Seismological Society of America, 98, 2975–2989.
     [Google Scholar]
  7. Boonyasiriwat, C., Valasek, P., Routh, P., Cao, W., Schuster, G.T. & Macy, B. (2009) An efficient multiscale method for time‐domain waveform tomography. Geophysics, 74, WCC59–WCC68.
     [Google Scholar]
  8. Bunks, C., Saleck, F.M., Zaleski, S. & Chavent, G. (1995) Multiscale seismic waveform inversion. Geophysics, 60, 1457–1473.
     [Google Scholar]
  9. Chen, L., Yang, S., Li, K., Zhang, P., Shao, W. & Xu, X. (2023) Seismic ahead‐prospecting based on deep learning of retrieving seismic wavefield. Underground Space. https://doi.org/10.1016/j.undsp.2023.02.001
  10. Deng, C., Feng, Y., Feng, S., Jin, P., Zhang, X., Zeng, Q. & Lin, Y. (2021) OpenFWI: benchmark seismic datasets for machine learning‐based full waveform inversion. CoRR, abs/2111.02926. https://dl.acm.org/doi/10.5555/3600270.3600705
  11. Donno, D., Chauris, H. & Calandra, H. (2013) Estimating the background velocity model with the normalized integration method. https://doi.org/10.3997/2214‐4609.2013041
  12. Feng, R., Mejer Hansen, T., Grana, D. & Balling, N. (2020) An unsupervised deep‐learning method for porosity estimation based on poststack seismic data. Geophysics, 85, M97–M105.
     [Google Scholar]
  13. Feng, S., Lin, Y. & Wohlberg, B. (2022) Multiscale data‐driven seismic full‐waveform inversion with field data study. IEEE Transactions on Geoscience and Remote Sensing, 60, 1–14.
     [Google Scholar]
  14. Fichtner, A., Trampert, J., Cupillard, P., Saygin, E., Taymaz, T., Capdeville, Y. & Villaseñor, A. (2013) Multiscale full waveform inversion. Geophysical Journal International, 194, 534–556.
     [Google Scholar]
  15. Gao, Z., Pan, Z., Gao, J. & Wu, R. (2019) Frequency controllable envelope operator and its application in multiscale full‐waveform inversion. IEEE Transactions on Geoscience and Remote Sensing, 57, 683–699.
     [Google Scholar]
  16. Guo, R., Yao, H.M., Li, M., Ng, M. K.P., Jiang, L. & Abubakar, A. (2021) Joint inversion of audio‐magnetotelluric and seismic travel time data with deep learning constraint. IEEE Transactions on Geoscience and Remote Sensing, 59, 7982–7995.
     [Google Scholar]
  17. He, K., Zhang, X., Ren, S. & Sun, J. (2015) Delving deep into rectifiers: Surpassing human‐level performance on Imagenet classification. In Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV'15). Washington, DC: IEEE Computer Society, pp. 1026–1034.
  18. Hu, W., Chen, J., Liu, J. & Abubakar, A. (2018) Retrieving low wavenumber information in FWI: an overview of the cycle‐skipping phenomenon and solutions. IEEE Signal Processing Magazine, 35, 132–141.
     [Google Scholar]
  19. Jiang, P., Wang, Q., Ren, Y., Yang, S. & Li, N. (2024) Full waveform inversion based on inversion network reparameterized velocity. Geophysical Prospecting, 72(1), 52–67.
     [Google Scholar]
  20. Jin, P., Zhang, X., Chen, Y., Huang, S.X., Liu, Z. & Lin, Y. (2022) Unsupervised learning of full‐waveform inversion: connecting CNN and partial differential equation in a loop. CoRR, https://arxiv.org/abs/2110.07584
  21. Johnson, C.W., Ben‐Zion, Y., Meng, H. & Vernon, F. (2020) Identifying different classes of seismic noise signals using unsupervised learning. Geophysical Research Letters, 47, e2020GL088353.
     [Google Scholar]
  22. Kazei, V., Ovcharenko, O., Plotnitskii, P., Peter, D., Zhang, X. & Alkhalifah, T. (2021) Mapping full seismic waveforms to vertical velocity profiles by deep learning. Geophysics, 86, R711–R721.
     [Google Scholar]
  23. Kazei, V., Tessmer, E. & Alkhalifah, T. (2016) Scattering angle‐based filtering via extension in velocity. In SEG technical program expanded abstracts 2016, Houston, TX: Society of Exploration Geophysicists, pp. 1157–1162.
     [Google Scholar]
  24. Krizhevsky, A., Sutskever, I. & Hinton, G.E. (2012) Imagenet classification with deep convolutional neural networks. In Pereira, F., Burges, C., Bottou, L. & Weinberger, K. (Eds.) Advances in neural information processing systems, volume 25. Red Hook, NY: Curran Associates. 1097–1105.
     [Google Scholar]
  25. Lailly, P. & Bednar, J. (1983) The seismic inverse problem as a sequence of before stack migrations. In Conference on inverse scattering: theory and application. Philadelphia, PA: SIAM, pp. 206–220.
     [Google Scholar]
  26. Li, S., Liu, B., Ren, Y., Chen, Y., Yang, S., Wang, Y. & Jiang, P. (2020) Deep‐learning inversion of seismic data. IEEE Transactions on Geoscience and Remote Sensing, 58, 2135–2149.
     [Google Scholar]
  27. Li, Y., Alali, A. & Alkhalifah, T. (2022) High‐wavenumber extrapolation for band‐limited FWI results using deep learning. In 83rd EAGE Annual Conference & Exhibition, June 2022, volume 2022. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. 1–5.
     [Google Scholar]
  28. Li, Y. & Alkhalifah, T. (2021) Extended full waveform inversion with matching filter. Geophysical Prospecting, 69, 1441–1454.
     [Google Scholar]
  29. Li, Y., Choi, Y., Alkhalifah, T., Li, Z. & Zhang, K. (2018) Full‐waveform inversion using a nonlinearly smoothed wavefield. Geophysics, 83, R117–R127.
     [Google Scholar]
  30. Liu, B., Guo, Q., Li, S., Liu, B., Ren, Y., Pang, Y., Guo, X., Liu, L. & Jiang, P. (2020) Deep learning inversion of electrical resistivity data. IEEE Transactions on Geoscience and Remote Sensing, 58, 5715–5728.
     [Google Scholar]
  31. Liu, B., Jiang, P., Wang, Q., Ren, Y., Yang, S. & Cohn, A.G. (2023) Physics‐driven self‐supervised learning system for seismic velocity inversion. Geophysics, 88, R145–R161.
     [Google Scholar]
  32. Liu, B., Ren, Y., Liu, H., Xu, H., Wang, Z., Cohn, A.G. & Jiang, P. (2021) GPRInvNet: Deep learning‐based ground‐penetrating radar data inversion for tunnel linings. IEEE Transactions on Geoscience and Remote Sensing, 59, 8305–8325.
     [Google Scholar]
  33. Liu, B., Yang, S., Ren, Y., Xu, X., Jiang, P. & Chen, Y. (2021) Deep‐learning seismic full‐waveform inversion for realistic structural models. Geophysics, 86, R31–R44.
     [Google Scholar]
  34. Liu, B., Yue, J., Zuo, Z., Xu, X., Fu, C., Yang, S. & Jiang, P. (2022) Unsupervised deep learning for random noise attenuation of seismic data. IEEE Geoscience and Remote Sensing Letters, 19, 1–5.
     [Google Scholar]
  35. Liu, J., Chauris, H. & Calandra, H. (2011) The normalized integration method ‐ an alternative to full waveform inversion?
  36. Liu, Y., Teng, J., Xu, T., Wang, Y., Liu, Q. & Badal, J. (2016) Robust time‐domain full waveform inversion with normalized zero‐lag cross‐correlation objective function. Geophysical Journal International, 209, 106–122.
     [Google Scholar]
  37. Mathew, A., Amudha, P. & Sivakumari, S. (2021) Deep learning techniques: An overview. In Hassanien, A.E., Bhatnagar, R. & Darwish, A. (Eds.) Advanced machine learning technologies and applications. Singapore: Springer Singapore, pp. 599–608.
     [Google Scholar]
  38. Moores, E.M. & Day, H.W. (1984) Overthrust model for the Sierra Nevada. Geology, 12, 416–419.
     [Google Scholar]
  39. Ovcharenko, O., Kazei, V., Alkhalifah, T. & Peter, D. (2022) Multi‐task learning for low‐frequency extrapolation and elastic model building from seismic data. IEEE Transactions on Geoscience and Remote Sensing, 60, 1–17.
     [Google Scholar]
  40. Pladys, A., Brossier, R. & Métivier, L. (2017) FWI alternative misfit functions ‐ what properties should they satisfy? In 79th EAGE Conference and Exhibition 2017, volume 2017, No. 1. European Association of Geoscientists & Engineers, pp. 1–5.
     [Google Scholar]
  41. Ren, Y., Liu, B., Yang, S., Li, D. & Jiang, P. (2021) Seismic data inversion with acquisition adaptive convolutional neural network for geologic forward prospecting in tunnels. Geophysics, 86, R659–R670.
     [Google Scholar]
  42. Ren, Y., Nie, L., Yang, S., Jiang, P. & Chen, Y. (2021) Building complex seismic velocity models for deep learning inversion. IEEE Access, 9, 63767–63778.
     [Google Scholar]
  43. Ren, Y., Xu, X., Yang, S., Nie, L. & Chen, Y. (2020) A physics‐based neural‐network way to perform seismic full waveform inversion. IEEE Access, 8, 112266–112277.
     [Google Scholar]
  44. Ronneberger, O., Fischer, P. & Brox, T. (2015) U‐net: convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer‐Assisted Intervention‐MICCAI 2015. Cham, Switzerland: Springer. https://link.springer.com/chapter/10.1007/978‐3‐319‐24574‐4_28
  45. Sheng‐Chang, C. & Guo‐Xin, C. (2016) Full waveform inversion of the second‐order time integral wavefield. Chinese Journal of Geophysics, 59, 676–690.
     [Google Scholar]
  46. Shin, C. & Cha, Y.H. (2008) Waveform inversion in the Laplace domain. Geophysical Journal International, 173, 922–931.
     [Google Scholar]
  47. Sun, B. & Alkhalifah, T. (2019) The application of an optimal transport to a preconditioned data matching function for robust waveform inversion. Geophysics, 84, R923–R945.
     [Google Scholar]
  48. Sun, B. & Alkhalifah, T. (2022) ML‐misfit: a neural network formulation of the misfit function for full‐waveform inversion. Frontiers in Earth Science, 10, 1011825.
     [Google Scholar]
  49. Sun, J., Innanen, K., Zhang, T. & Trad, D. (2023) Implicit seismic full waveform inversion with deep neural representation. Journal of Geophysical Research: Solid Earth, 128, e2022JB025964.
     [Google Scholar]
  50. Sun, J., Innanen, K.A. & Huang, C. (2021) Physics‐guided deep learning for seismic inversion with hybrid training and uncertainty analysis. Geophysics, 86, R303–R317.
     [Google Scholar]
  51. Symes, W.W. (2008) Migration velocity analysis and waveform inversion. Geophysical Prospecting, 56, 765–790.
     [Google Scholar]
  52. Tarantola, A. (1984) Inversion of seismic reflection data in the acoustic approximation. Geophysics, 49, 1259–1266.
     [Google Scholar]
  53. Wang, Z., Simoncelli, E.P. & Bovik, A.C. (2003) Multiscale structural similarity for image quality assessment. In The Thirty‐Seventh Asilomar Conference on Signals, Systems Computers, 2003, Pacific Grove, CA, USA, 2003, Vol. 2. Piscataway, NJ: IEEE, pp. 1398‐1402.
     [Google Scholar]
  54. Wu, R., Luo, J. & Wu, B. (2014) Seismic envelope inversion and modulation signal model. Geophysics, 79, WA13–WA24.
     [Google Scholar]
  55. Wu, Y. & Lin, Y. (2020) InversionNet: an efficient and accurate data‐driven full waveform inversion. IEEE Transactions on Computational Imaging, 6, 419–433.
     [Google Scholar]
  56. Wu, Z. & Alkhalifah, T. (2017) Efficient scattering‐angle enrichment for a nonlinear inversion of the background and perturbations components of a velocity model. Geophysical Journal International, 210, 1981–1992.
     [Google Scholar]
  57. Xu, Z.J., Zhang, Y. & Luo, T. (2022) Overview frequency principle/spectral bias in deep learning. Communications on Applied Mathematics and Computation, https://doi.org/10.1007/s42967‐024‐00398‐7
  58. Yang, F. & Ma, J. (2019) Deep‐learning inversion: A next‐generation seismic velocity model building method. Geophysics, 84, R583–R599.
     [Google Scholar]
  59. Young, T., Hazarika, D., Poria, S. & Cambria, E. (2018) Recent trends in deep learning based natural language processing. IEEE Computational Intelligence Magazine, 13, 55–75.
     [Google Scholar]
  60. Yuan, Y.O. & Simons, F.J. (2014) Multiscale adjoint waveform‐difference tomography using wavelets. Geophysics, 79, WA79–WA95.
     [Google Scholar]
  61. Zhang, C. & van der Baan, M. (2020) Microseismic denoising and reconstruction by unsupervised machine learning. IEEE Geoscience and Remote Sensing Letters, 17, 1114–1118.
     [Google Scholar]
  62. Zhang, Z. & Alkhalifah, T. (2022) Regularized elastic full‐waveform inversion using deep learning. In S.Bhattacharya & H.Di (Eds.) Advances in subsurface data analytics. Amsterdam: Elsevier, pp. 219–250.
     [Google Scholar]
/content/journals/10.1111/1365-2478.13665
Loading
Unsupervised learning inversion of seismic velocity models based on a multi‐scale strategy
Geophysical Prospecting 73, 471 (2025); https://doi.org/10.1111/1365-2478.13665
/content/journals/10.1111/1365-2478.13665
/content/journals/10.1111/1365-2478.13665
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): Full waveform; Inverse; Numerical study; Seismics

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