Economic and Environmental Geology (자원환경지질)

The Korean Society of Economic and Environmental Geology (대한자원환경지질학회)
권호 표지
DOI QR코드

DOI QR Code

Denoising Laplace-domain Seismic Wavefields using Deep Learning

  • Lydie Uwibambe (Department of Energy Resources Engineering, Pukyong National University) ;
  • Jun Hyeon Jo (Department of Energy Resources Engineering, Pukyong National University) ;
  • Wansoo Ha (Department of Energy Resources Engineering, Pukyong National University)
  • Received : 2024.08.29
  • Accepted : 2024.10.11
  • Published : 2024.10.29
Download PDF
⟨ Previous Next ⟩

Abstract

Random noise in seismic data can significantly impair hydrocarbon exploration by degrading the quality of subsurface imaging. We propose a deep learning approach to attenuate random noise in Laplace-domain seismic wavefields. Our method employs a modified U-Net architecture, trained on diverse synthetic P-wave velocity models simulating the Gulf of Mexico subsurface. We rigorously evaluated the network's denoising performance using both the synthetic Pluto velocity model and real Gulf of Mexico field data. We assessed the effectiveness of our approach through Laplace-domain full waveform inversion. The results consistently show that our U-Net approach outperforms traditional singular value decomposition methods in noise attenuation across various scenarios. Numerical examples demonstrate that our method effectively attenuates random noise and significantly enhances the accuracy of subsequent seismic imaging processes.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2021R1F1A1064432).

References

  1. Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., Isard, M., Kudlur, M., Levenberg, J., Monga, R., Moore, S., Murray, D.G., Steiner, B., Tucker, P., Vasudevan, V., Warden, P., Wicke, M., Yu, Y., and Zheng, X. (2016) TensorFlow: A system for large-scale machine learning. 12th USENIX symposium on operating systems design and implementation (OSDI 16), p.265-283. doi: 10.48550/arXiv.1605.08695
  2. Almadani, M., Waheed, U.B., Masood, M., and Chen, Y. (2021) Dictionary learning with convolutional structure for seismic data denoising and interpolation. Geophysics, 86(5), p.V361-V374. doi: 10.1190/geo2019-0689.1
  3. Chen, Y., and Ma, J. (2014) Random noise attenuation by f-x empirical-mode decomposition predictive filtering. Geophysics, v.79(3), p.V81-V91. doi: 10.1190/geo2013-0080.1
  4. Gan, S., Chen, Y., Zu, S., Qu, S., and Zhong, W. (2015) Structure-oriented singular value decomposition for random noise attenuation of seismic data. Journal of Geophysics and Engineering, v.12(2), p.262-272. doi: 10.1088/1742-2132/12/2/262
  5. Ha, W., and Shin, C. (2013) Why do Laplace-domain waveform inversions yield long-wavelength results? Geophysics, v.78(4), p.R167-R173. doi: 10.1190/geo2012-0365.1
  6. Ha, W., and Shin, C. (2021a) Seismic random noise attenuation in the Laplace domain using singular value decomposition. IEEE Access, v.9, p.62029-62037. doi: 10.1109/ACCESS.2021.3074648
  7. Ha, W., and Shin, C. (2021b) Handling negative values for the logarithmic objective function in acoustic Laplace-domain full-waveform inversion using real variables. IEEE Transactions on Geoscience and Remote Sensing, v.59(7), p.6218-6224. doi: 10.1109/TGRS.2020.3019510
  8. Kingma, D.P., and Ba, J. (2015) Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980. doi: 10.48550/arXiv.1412.6980
  9. Kopsinis, Y., and McLaughlin, S. (2009) Development of EMD-based denoising methods inspired by wavelet thresholding. IEEE Transactions on Signal Processing, v.57(4), p.1351-1362. doi: 10.1109/TSP.2009.2013885
  10. Li, J.-H., Zhang, Y.-J., Qi, R., and Liu, Q.H. (2017) Wavelet-based higher order correlative stacking for seismic data denoising in the curvelet domain. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, v.10(8), p.3810-3820. doi: 10.1109/JSTARS.2017.2685628
  11. Liu, B., Yue, J., Zuo, Z., Xu, X., Fu, C., Yang, S., and Jiang, P. (2022) Unsupervised deep learning for random noise attenuation of seismic data. IEEE Geoscience and Remote Sensing Letters, v.19, p.1-5. doi: 10.1109/LGRS.2021.3057631
  12. Liu, Z., Chen, Y., and Ma, J. (2018) Ground roll attenuation by synchrosqueezed curvelet transform. Journal of Applied Geophysics, v.151, p.246-262. doi: 10.1016/j.jappgeo.2018.02.016
  13. Meng, F., Fan, Q., and Li, Y. (2022) Self-supervised learning for seismic data reconstruction and denoising. IEEE Geoscience and Remote Sensing Letters, v.19, p.1-5. doi: 10.1109/LGRS.2021.3068132
  14. Nihei, K.T., and Li, X. (2007) Frequency response modelling of seismic waves using finite difference time domain with phase sensitive detection (TD-PSD). Geophysical Journal International, v.169(3), p.1069-1078. doi: 10.1111/j.1365-246X.2006.03262.x
  15. Park, B., Ha, W., and Shin, C. (2020) A comparison of the preconditioning effects of different parameterization methods for monoparameter full waveform inversions in the Laplace domain. Journal of Applied Geophysics, v.172, 103883. doi: 10.1016/j.jappgeo.2019.103883
  16. Reddi, S.J., Kale, S., and Kumar, S. (2018) On the convergence of Adam and beyond. International Conference on Learning Representations, 1-23. doi: 10.48550/arXiv.1904.09237
  17. Ronneberger, O., Fischer, P., and Brox, T. (2015) U-Net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 234-241). Springer, Cham. doi: 10.1007/978-3-319-24574-4_28
  18. Saad, O.M., and Chen, Y. (2020) Deep denoising autoencoder for seismic random noise attenuation. Geophysics, v.85(4), p.V367-V376. doi: 10.1190/geo2019-0468.1
  19. Shin, C., and Cha, Y.H. (2008) Waveform inversion in the Laplace domain. Geophysical Journal International, v.173(3), p.922-931. doi: 10.1111/j.1365-246X.2008.03768.x
  20. Shin, C., and Ha, W. (2008) A comparison between the behavior of objective functions for waveform inversion in the frequency and Laplace domains. Geophysics, v.73(5), p.VE119-VE133. doi.org/10.1190/1.2953978
  21. Shin, C., Ko, S., Kim, W., Min, D.-J., Yang, D., Marfurt, K.J., Shin, S., Yoon, K., and Yoon, C.H. (2003) Traveltime calculations from frequency-domain downward-continuation algorithms. Geophysics, v.68(4), p.1380-1388. doi: 10.1190/1.1598131
  22. Stoughton, D., Stefani, J., and Michell, S. (2001) 2D elastic model for wavefield investigations of subsalt objectives, deep water Gulf of Mexico. SEG Expanded Abstracts, v.20, p.1269-1272. doi: 10.1190/1.1816325
  23. Xue, Y.-J., Cao, J.-X., and Wang, X.-J. (2019) Inverse Q filtering via synchrosqueezed wavelet transform. Geophysics, v.84(2), p.V121-V132. doi: 10.1190/geo2018-0177.1
  24. Yang, L., Wang, S., Chen, X., Saad, O.M., Chen, W., Oboue, Y.A.S.I., and Chen, Y. (2021) Unsupervised 3-D random noise attenuation using deep skip autoencoder. IEEE Transactions on Geoscience and Remote Sensing, v.60, p.1-16. doi: 10.1109/TGRS.2021.3100455
  25. Yilmaz, O. (2001) Seismic Data Analysis: Processing, Inversion, and Interpretation of Seismic Data. Society of Exploration Geophysicists. doi: 10.1190/1.9781560801580
  26. Zhang, H., Yang, H., Li, H., Huang, G., and Ding, Z. (2018) Random noise attenuation of non-uniformly sampled 3D seismic data along two spatial coordinates using non-equispaced curvelet transform. Journal of Applied Geophysics, v.151, p.221-233. doi: 10.1016/j.jappgeo.2018.02.018
  27. Zhang, M., Liu, Y., Zhang, H., and Chen, Y. (2020) Incoherent noise suppression of seismic data based on robust low-rank approximation. IEEE Transactions on Geoscience and Remote Sensing, v.58(12), p.8874-8887. doi: 10.1109/TGRS.2020.2991438
  28. Zhong, T., Cheng, M., Dong, X., Li, Y., and Wu, N. (2022) Seismic random noise suppression by using deep residual U-Net. Journal of Petroleum Science and Engineering, v.209, 109901. doi: 10.1016/j.petrol.2021.109901
  29. Zhu, L., Liu, E., and McClellan, J.H. (2015) Seismic data denoising through multiscale and sparsity-promoting dictionary learning. Geophysics, v.80(6), p.WD45-WD57. doi: 10.1190/geo2015-0047.1
  30. Zhu, W., Mousavi, S.M., and Beroza, G.C. (2019) Seismic signal denoising and decomposition using deep neural networks. IEEE Transactions on Geoscience and Remote Sensing, v.57(11), p.9476-9488. doi: 10.1109/TGRS.2019.2926772