1887
Volume 22, Issue 5
  • ISSN: 1569-4445
  • E-ISSN: 1873-0604
PDF

Abstract

Abstract

First‐break picking is an essential step in seismic data processing. For reliable results, first arrivals should be picked by an expert. This is a time‐consuming procedure and subjective to a certain degree, leading to different results for different operators. In this study, we have used a U‐Net architecture with residual blocks to perform automatic first‐break picking based on deep learning. Focusing on the effects of weight initialization on first‐break picking, we conduct this research by using the weights of a pre‐trained network that is used for object detection on the ImageNet dataset. The efficiency of the proposed method is tested on two real datasets. For both datasets, we pick manually the first breaks for less than 10 of the seismic shots. The pre‐trained network is fine‐tuned on the picked shots, and the rest of the shots are automatically picked by the neural network. It is shown that this strategy allows to reduce the size of the training set, requiring fine‐tuning with only a few picked shots per survey. Using random weights and more training epochs can lead to a lower training loss, but such a strategy leads to overfitting as the test error is higher than the one of the pre‐trained network. We also assess the possibility of using a general dataset by training a network with data from three different projects that are acquired with different equipment and at different locations. This study shows that if the general dataset is created carefully it can lead to more accurate first‐break picking; otherwise, the general dataset can decrease the accuracy. Focusing on near‐surface geophysics, we perform traveltime tomography and compare the inverted velocity models based on different first‐break picking methodologies. The results of the inversion show that the first breaks obtained by the pre‐trained network lead to a velocity model that is closer to the one obtained from the inversion of expert‐picked first breaks.

© 2024 European Association of Geoscientists & Engineers. © 2024 The Author(s). Near Surface Geophysics published by John Wiley & Sons Ltd on behalf of European Association of Geoscientists and Engineers.
Loading

Article metrics loading...

/content/journals/10.1002/nsg.12316
2024-09-08
2026-02-11

  • From This Site

    /content/journals/10.1002/nsg.12316
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

/deliver/fulltext/nsg/22/5/nsg12316.html?itemId=/content/journals/10.1002/nsg.12316&mimeType=html&fmt=ahah

References

  1. Alali, A., Kazei, V., Kalita, M. & Alkhalifah, T. (2024) Deep learning unflooding for robust subsaltwaveform inversion. Geophysical Prospecting, 72, 7–19.
     [Google Scholar]
  2. Azwin, I., Saad, R. & Nordiana, M. (2013) Applying the seismic refraction tomography for site characterization. APCBEE Procedia, 5, 227–231.
     [Google Scholar]
  3. Badrinarayanan, V., Kendall, A. & Cipolla, R. (2017) SegNet: a deep convolutional encoder‐decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12), 2481–2495.
     [Google Scholar]
  4. Bozinovski, S. & Fulgosi, A. (1976) The influence of pattern similarity and transfer learning upon training of a base perceptron B2. In Proceedings of Symposium Informatica. Bled, volume 3. pp. 121–126.
     [Google Scholar]
  5. Coppens, F. (1985) First arrival picking on common‐offset trace collections for automatic estimation of static corrections. Geophysical Prospecting, 33(8), 1212–1231.
     [Google Scholar]
  6. Cunha, A., Pochet, A., Lopes, H. & Gattass, M. (2020) Seismic fault detection in real data using transfer learning from a convolutional neural network pre‐trained with synthetic seismic data. Computers & Geosciences, 135, 104344.
     [Google Scholar]
  7. Deng, J., Dong, W., Socher, R., Li, L.J., Li, K. & Fei‐Fei, L. (2009) ImageNet: A large‐scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, pp. 248–255.
     [Google Scholar]
  8. Dip, A.C., Giroux, B. & Gloaguen, E. (2021) Microseismic monitoring of rockbursts with the ensemble Kalman filter. Near Surface Geophysics, 19(4), 429–445.
     [Google Scholar]
  9. Fabien‐Ouellet, G. & Fortier, R. (2014) Using all seismic arrivals in shallow seismic investigations. Journal of Applied Geophysics, 103, 31–42.
     [Google Scholar]
  10. Fabien‐Ouellet, G. & Sarkar, R. (2020) Seismic velocity estimation: a deep recurrent neural‐network approach. Geophysics, 85(1), U21–U29.
     [Google Scholar]
  11. Glorot, X. & Bengio, Y. (2010) Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the thirteenth international conference on artificial intelligence and statistics. JMLR Workshop and Conference Proceedings. Cambridge MA: JMLR, pp. 249–256.
     [Google Scholar]
  12. Goodfellow, I., Bengio, Y. & Courville, A. (2016) Deep learning. Cambridge MA: MIT Press.
     [Google Scholar]
  13. Hatherly, P. (1982) A computer method for determining seismic first arrival times. Geophysics, 47(10), 1431–1436.
     [Google Scholar]
  14. He, K., Zhang, X., Ren, S. & Sun, J. (2016) Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. Piscataway, NJ: IEEE, pp. 770–778.
     [Google Scholar]
  15. Hossain, M.S., Betts, J.M. & Paplinski, A.P. (2021) Dual focal loss to address class imbalance in semantic segmentation. Neurocomputing, 462, 69–87.
     [Google Scholar]
  16. Hu, L., Zheng, X., Duan, Y., Yan, X., Hu, Y. & Zhang, X. (2019) First‐arrival picking with a U‐net convolutional network. Geophysics, 84(6), U45–U57.
     [Google Scholar]
  17. Kingma, D.P. & Ba, J. (2014) Adam: A method for stochastic optimization. Preprint. arXiv:1412.6980.
  18. Leite, E.P. & Vidal, A.C. (2011) 3D porosity prediction from seismic inversion and neural networks. Computers & Geosciences, 37(8), 1174–1180.
     [Google Scholar]
  19. Li, H., Xu, Z., Taylor, G., Studer, C. & Goldstein, T. (2018) Visualizing the loss landscape of neural nets. Advances in Neural Information Processing Systems, 31. Red Hook, NY: Curran Associates, 6391–6401.
  20. Lim, J.S. (2005) Reservoir properties determination using fuzzy logic and neural networks from well data in offshore Korea. Journal of Petroleum Science and Engineering, 49(3‐4), 182–192.
     [Google Scholar]
  21. Ma, Y., Cao, S., Rector, J.W. & Zhang, Z. (2020) Automated arrival‐time picking using a pixel‐level network. Geophysics, 85(5), V415–V423.
     [Google Scholar]
  22. Mardan, A. & Fabien‐Ouellet, G. (2024) Physics‐informed attention‐based neural networks for creating initial model for full‐waveform inversion. In 85th EAGE Annual Conference & Exhibition. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. 1–5.
     [Google Scholar]
  23. Mardan, A., Giroux, B. & Fabien‐Ouellet, G. (2023) Weighted‐average time‐lapse seismic full‐waveform inversion. Geophysics, 88(1), R25–R38.
     [Google Scholar]
  24. Mardan, A., Javaherian, A. & Mirzakhanian, M. (2017) Channel characterization using support vector machine. In 79th EAGE Conference and Exhibition 2017‐Workshops. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. cp–519.
     [Google Scholar]
  25. Marsden, D. (1993) Static corrections–a review. The Leading Edge, 12(2), 115–120.
     [Google Scholar]
  26. McCormack, M.D., Zaucha, D.E. & Dushek, D.W. (1993) First‐break refraction event picking and seismic data trace editing using neural networks. Geophysics, 58(1), 67–78.
     [Google Scholar]
  27. Moser, T. (1991) Shortest path calculation of seismic rays. Geophysics, 56(1), 59–67.
     [Google Scholar]
  28. Murat, M.E. & Rudman, A.J. (1992) Automated first arrival picking: a neural network approach. Geophysical Prospecting, 40(6), 587–604.
     [Google Scholar]
  29. Nasr, M., Giroux, B. & Dupuis, J.C. (2022) Python package for 3D joint hypocenter‐velocity inversion on tetrahedral meshes: parallel implementation and practical considerations. Computational Geosciences, 1–25.
  30. Park, M.J. & Sacchi, M.D. (2020) Automatic velocity analysis using convolutional neural network and transfer learning. Geophysics, 85(1), V33–V43.
     [Google Scholar]
  31. Peraldi, R. & Clement, A. (1972) Digital processing of refraction data study of first arrivals. Geophysical Prospecting, 20(3), 529–548.
     [Google Scholar]
  32. Ronneberger, O., Fischer, P. & Brox, T. (2015) U‐Net: convolutional networks for biomedical image segmentation. In Proceedings of the Medical Image Computing and Computer‐Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5‐9, 2015 Part III 18. Berlin: Springer, pp. 234–241.
     [Google Scholar]
  33. Röth, G. & Tarantola, A. (1994) Neural networks and inversion of seismic data. Journal of Geophysical Research: Solid Earth, 99(B4), 6753–6768.
     [Google Scholar]
  34. Rücker, C., Günther, T. & Wagner, F.M. (2017) pyGIMLi: An open‐source library for modelling and inversion in geophysics. Computers & Geosciences, 109, 106–123.
     [Google Scholar]
  35. Sabbione, J.I. & Velis, D. (2010) Automatic first‐breaks picking: new strategies and algorithms. Geophysics, 75(4), V67–V76.
     [Google Scholar]
  36. Simon, J., Fabien‐Ouellet, G., Gloaguen, E. & Khurjekar, I. (2023) Hierarchical transfer learning for deep learning velocity model building. Geophysics, 88(1), R79–R93.
     [Google Scholar]
  37. Veezhinathan, J. & Wagner, D. (1990) A neural network approach to first break picking. In 1990 IJCNN International Joint Conference on Neural Networks. Piscataway, NJ: IEEE, vol. 1, pp. 235–240.
     [Google Scholar]
  38. Walia, R. & Hannay, D. (1999) Source and receiver geometry corrections for deep towed multichannel seismic data. Geophysical Research Letters, 26(13), 1993–1996.
     [Google Scholar]
  39. Wu, H., Zhang, B., Li, F. & Liu, N. (2019) Semiautomatic first‐arrival picking of microseismic events by using the pixel‐wise convolutional image segmentation method. Geophysics, 84(3), V143–V155.
     [Google Scholar]
  40. Yuan, S., Liu, J., Wang, S., Wang, T. & Shi, P. (2018) Seismic waveform classification and first‐break picking using convolution neural networks. IEEE Geoscience and Remote Sensing Letters, 15(2), 272–276.
     [Google Scholar]
  41. Yuan, S.Y., Zhao, Y., Xie, T., Qi, J. & Wang, S.X. (2022) SegNet‐based first‐break picking via seismic waveform classification directly from shot gathers with sparsely distributed traces. Petroleum Science, 19(1), 162–179.
     [Google Scholar]
  42. Zhao, X. & Mendel, J.M. (1988) Minimum‐variance deconvolution using artificial neural networks. In SEG Technical Program Expanded Abstracts 1988. Houston, TX: Society of Exploration Geophysicists, pp. 738–741.
     [Google Scholar]
  43. Zhu, W. & Beroza, G.C. (2019) PhaseNet: a deep‐neural‐network‐based seismic arrival‐time picking method. Geophysical Journal International, 216(1), 261–273.
     [Google Scholar]
  44. Zwartjes, P. & Yoo, J. (2022) First break picking with deep learning–evaluation of network architectures. Geophysical Prospecting, 70(2), 318–342.
     [Google Scholar]
/content/journals/10.1002/nsg.12316
Loading
A fine‐tuning workflow for automatic first‐break picking with deep learning
Near Surface Geophysics 22, 539 (2024); https://doi.org/10.1002/nsg.12316
/content/journals/10.1002/nsg.12316
/content/journals/10.1002/nsg.12316
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): environmental; hydrogeology; near‐surface; tomography; traveltime

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