Cite this article
1
Download
50
Views
Journal Browser
Volume | Year
Issue
Search
News and Announcements
View All
ARTICLE

Enhancing seismic resolution based on U-Net deep learning network

Zeyu Li2 Guoquan Wang2 Chenghong Zhu2,3,4 Shuangquan Chen1
Show Less
1 College of Geophysics, China University of Petroleum (Beijing), 102249 Beijing, P.R.China.,
2 State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective Development, 100083 Beijing, P.R. China.,
3 Sinopec Key Laboratory of Seismic Elastic Wave Technology, 100083 Beijing, P.R.China.,
4 Sinopec Petroleum Exploration and Production Research Institute, 100083 Beijing, P.R. China.,
JSE 2023, 32(4), 315–336;
Submitted: 6 June 2023 | Accepted: 10 July 2023 | Published: 1 August 2023
© 2023 by the Authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Abstract

Li, Z.Y, Wang, G.Q., Zhu, C.H. and Chen, S.Q., 2023. Enhancing seismic resolution based reservoirs are being developed for oil and gas exploration. High signal high-to-noise ratio seismic data is required for accurate reservoir resolution seismic data processing is a challenging and im on U-Net deep learning network. Journal of Seismic Exploration, 32: 315-336. Deep and ultra-deep reservoirs, unconventional hydrocarbons, and other complex -resolution with high description. Therefore, portant endeavor. The conventional high-resolution processing techniques, such as inverse Q-filtering technique based on the stratum attenuation model and deconvolution met hods, are nearly mod- el-dependent. Deep learning method based on image processing has been widely used in seismic data processing and inversion in recent years, which is a data-driven method and has p owerful abilities to extract data features. In this paper, we rebuild a deep learning network based on U-net network and proposed a high-resolution seismic data processing method and workflow. In the network, we utilized ResPath structure instead of the straightforward skip connection in the conventional Res-Unet, and employed a combina- tion loss function by using the mean absolute error and multiscal e structural similarity. Moreover, by introducing pre-training strategy into the processing workflow, the pro- posed network can reduce the loss of low- frequency components and perceive low-frequency information. Synthetic and real field seismic data are used to validate the network, the trained model by using pre-training strategy can better maintain the low-frequency components of the original data and recover the high-frequency compo- nents, and the signal-to-noise ratio and resolution of seismic data cantly enhanced. ave both been signifi-

Keywords
U-net
deep learning
pre-training strategy
seismic resolution
data-driven
References
  1. Braga, I.L.S. and Moraes, F.S., 2013. High-resolution gathers by inverse Q-filtering inthe wavelet domain. Geophysics, 78(2), 53-61.
  2. Canning, A., Mouliére-Reiser, D., Weiss, Y., Malkin, A., Phillip, E., Grinberg, N., Teitel,
  3. A., Reznikov, M. and Yehezkel, V., 2017. Neural networks approach to spectral en-hancement. Expanded Abstr., 87th Ann. Internat. SEG Mtg., Houston: 4283-4286.
  4. Chen, S. and Wang, Y., 2018. Seismic resolution enhancement by frequency-dependentwavelet scaling. IEEE Geosci. Remote Sens. Lett., 15: 654-658.
  5. Chen, S., Wei, Q., Liu, L. and Li, X., 2018. Data-driven attenuation compensation via ashaping regularization scheme. IEEE Geosci. Remote Sens. Lett., 15: 1667-1671.
  6. Choi, Y., Seol, S.J. and Jo, Y., 2021. Deep learning spectral enhancement consideringfeatures of seismic field data. Geophysics, 86(5): V389-V408.
  7. Deng, M., Jia, R. and Tian, Y., 2020. Super-resolution reconstruction of seismic profileimages based on deep learning. Comput. Engineer. Design (in Chinese), 41:2332-2337.
  8. Dong, C., He, K., Loy, C.C. and Tang, X,. 2015. Image super-resolution using deepconvolutional networks. IEEE Transact. Pattern Analys. Mach. Intellig., 38:295-307.
  9. Gao. H., Liu, G. and Wu, X., 2021. ChannelSeg3D: Channel simulation and deep learn-ing for channel interpretation in 3D seismic images. Geophysics, 86(4): IM73-IM83.
  10. Halpert, A.D., 2018. Deep learning-enabled seismic image enhancement. Expanded Ab-str., 88th Ann. Internat. SEG Mtg., Anaheim: 2081-2085.
  11. He, K., Ren, S., Zhang, X. and Sun, J., 2016. Deep Residual Learning for Image Recog-nition. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR): 770-778.
  12. Ibtehaz, N. and Rahman, M.S., 2019. MultiResUNet: Rethinking the U-Net Architecturefor Multimodal Biomedical Image Segmentation. Neural Netw.. 121: 74-87.loffe, S. and Szegedy, C., 2015. Batch normalization: accelerating deep network trainingby reducing internal covariate shift. Proc. 32nd Internat. Conf. Machine Learn. - Vol.37 (ICML'15). JMLR.org, 448-456.
  13. Jervis, M., Liu, M. and Smith, R., 2021. Deep learning network optimization and hy-perparameter tuning for seismic lithofacies classification. The Leading Edge, 40:514-523.
  14. Kazemeini, S.H., Juhlin, C., Yang, C., et al., 2010. Enhancing seismic data resolutionusing the prestack bluing technique: An example from the Ketzin CO2 injection siteGermany. Geophysics, 75(6): V101-V110.
  15. LeCun, Y., Bengio, Y. and Hinton, G., 2015. Deep learning. Nature, 521: 436-444.
  16. Ledig, C., Huszar, F., Theis, L., et al., 2017. Photo-realistic single image su-per-resolution using a generative adversarial network. IEEE Conf. Comput. Vis. Patt.Recogn. (CVPR), 105-114.
  17. Li, J., Wu, X. and Hu, Z., 2020. Deep learning for simultaneous seismic image su-per-resolution and denoising. Expanded Abstr., 90th Ann. Internat. SEG Mtg., Hou-ston: 1661-1665.
  18. Li, Q., 1994. The road to accurate exploration: Engineering analysis of high-resolutionseismic exploration system. Petroleum Industry Press, Houston, TX.
  19. Long, J., Darrell. T. and Shelhamer, E., 2015. Fully convolutional networks for semanticsegmentation. IEEE Transact. Patt. Analys. Machine Intellig., 39: 640-651.
  20. Margrave, G.F., Hrnley, D.C. and Lamoureux, M.P., 2011. Gabor deconvolution: esti-mating reflectivity by nonstationary deconvolution of seismic data. Geophysics,76(3): W15-W30.
  21. Oliveira, D.A.B., Ferreira, R.S., Silva, R. and Brazil, E.V., 2019. Improving seismic dataresolution with deep generative networks. IEEE Geosci. Remote Sens. Lett.: 1-5.
  22. Picetti, F., Lipari, V., Bestagini, P. and Tubaro, S., 2018. A generative adversarial net-work for seismic imaging applications. Expanded Abstr., 88 th Ann. Internat. SEGMtg., Anaheim: 2231-2235.
  23. Robinson, E.A. and Treitel, S., 1967. Principles of digital Wiener filtering. Geophys.Prosp., 15: 311-332 .
  24. Ronneberger, O., Brox, T. and Fischer, P., 2015. U-Net: Convolutional Networks for
  25. Biomedical Image Segmentation. Internat. Conf. Medical Image Comput.-AssistedIntervention. Springer Verlag, Berlin, 234-241.
  26. Sun, Y., Yu, W. and Huang, Y., 2021. High resolution end-to-end seismic processingtechnology based on U-NET network. Progr. Geophys. (in Chinese), 36:1297-1305.
  27. Wu, X., Liang L., Shi, Y. and Fomel, S., 2019. FaultSeg3D: Using synthetic data sets totrain an end-to-end convolutional neural network for 3D seismic fault segmentation.Geophysics, 84(3): IM35-IM45.
  28. Xiao, X., Lian, S., Luo, Z. and Li, S., 2018. Weighted Res-Unet for High-Quality Retina
  29. Vessel Segmentation. 9th Internat. Conf. Informat. Technol. Medicine Education(ITME): 327-331.
  30. Yan, Z., Fang, G., Xu, H.. Liu, J., Shi, J., Pan, J. and Wang, J., 2018. The application of
  31. Hilbert spectral whitening method to high resolution processing of marine seismicdata. Marine Geol. Quatern. Geol., 38: 212-220.
  32. Yuan, Z., Huang, H. and Jiang, Y., 2019. An enhanced fault-detection method based onadaptive spectral decomposition and super-resolution deep learning. Interpretation.7(3): 1-63.
  33. Yu, S., 1996. Broadband Ricker wavelet. Oil Geophys. Prosp. (in Chinese), 5:605-615+750.
  34. Yu, S., Ma, J. and Wang, W., 2019. Deep learning for denoising. Geophysics, 84(6):V333-V350.
  35. Zhao, H., Gallo, O., Frosio, I. and Kautz, J., 2016. Loss functions for image restorationwith neural networks. IEEE Transact. Computat. Imag., 3: 47-57.
  36. Zhang, J., Zhang, B., Zhang, Z., Liang, H. and Ge, D. 2015. Low-frequency data analysisand expansion. Appl. Geophys., 12: 212-220.
Share
Back to top
Journal of Seismic Exploration, Electronic ISSN: 0963-0651 Print ISSN: 0963-0651, Published by AccScience Publishing