none 10.1190/geo2023-0006.1 Society of Exploration Geophysicists Society of Exploration Geophysicists 186 149526441 5981 2025121211490591000 10.1190 2025-12-12T16:49:20Z 2023-07-12T09:46:50Z 31 Geophysics 0016-8033 1942-2156 11 01 2023 88 6 Dawei Liu Xi’an Jiaotong University 1 , School of Information and Communications Engineering, Xi’an, China and , Department of Physics, Edmonton, Alberta, Canada. 409791715@qq.com University of Alberta 1 , School of Information and Communications Engineering, Xi’an, China and , Department of Physics, Edmonton, Alberta, Canada. 409791715@qq.com http://orcid.org/0000-0001-5553-2379 Wenli Niu Xi’an Jiaotong University 2 , School of Information and Communications Engineering, Xi’an, China. 3101434366@qq.com Xiaokai Wang Xi’an Jiaotong University 2 , School of Information and Communications Engineering, Xi’an, China. xkwang@xjtu.edu.cn http://orcid.org/0000-0001-5396-985X Mauricio D. Sacchi University of Alberta 3 , Department of Physics, Edmonton, Alberta, Canada. msacchi@ualberta.ca http://orcid.org/0000-0001-6654-6661 Wenchao Chen Xi’an Jiaotong University 2 , School of Information and Communications Engineering, Xi’an, China. wencchen@xjtu.edu.cn (corresponding author) http://orcid.org/0000-0002-2686-067X Cheng Wang Daqing Oilfield Company Ltd. 4 , the Exploration and Development Research Institute, Daqing, China. wangcheng88@petrochina.com.cn ABSTRACT Seismic vertical resolution is critical for accurately identifying subsurface structures and reservoir properties. Improving the vertical resolution of vintage seismic data with strongly supervised deep learning is challenging due to scarce or costly labels. To remedy the label-lacking problem, we develop a weakly supervised deep-learning method to improve vintage seismic data with poor resolution by extrapolating from nearby high-resolution seismic data. Our method uses a cycle generative adversarial network with an improved identity loss function. In addition, we contribute a pseudo-3D training data construction strategy that reduces discontinuity artifacts caused by accessing 3D field data with a 2D network. We determine the feasibility of our method on 2D synthetic data and achieve results comparable to the classic time-varying spectrum whitening method on field poststack migration data while effectively recovering more high-frequency information. 09 22 2023 11 01 2023 07 12 2023 V445 V458 National Natural Science Foundation of China http://dx.doi.org/10.13039/501100001809 41974131 41774135 1 10.1190/crossmark-policy library.seg.org true 2023-01-03 2023-07-02 2023-07-05 2023-09-22 National Natural Science Foundation of China http://dx.doi.org/10.13039/501100001809 41974131 41774135 10.1190/geo2023-0006.1 https://pubs.geoscienceworld.org/geophysics/article/88/6/V445/628278/Improving-vertical-resolution-of-vintage-seismic https://pubs.geoscienceworld.org/seg/geophysics/article-pdf/88/6/V445/6009061/geo-2023-0006.1.pdf https://pubs.geoscienceworld.org/seg/geophysics/article-pdf/88/6/V445/6009061/geo-2023-0006.1.pdf https://pubs.geoscienceworld.org/geophysics/article/88/6/v445/628278/improving-vertical-resolution-of-vintage-seismic Interpretation Alaudah 7 3 SE175 2019 10.1190/INT-2018-0249.1 A machine-learning benchmark for facies classification Geophysical Prospecting for Petroleum Bian 2 26 1986 Spectral whitening of seismic data 10.1190/segam2019-3216178.1 Chen D. Gao J. Hou Y. Gao Z. , 2019, High resolution inversion of seismic wavelet and reflectivity using iterative deep neural networks: 89th Annual International Meeting, SEG, Expanded Abstracts, 2538–2542, doi:http://dx.doi.org/10.1190/segam2019-3216178.1. 0016-8033 Geophysics Chen 86 3 R265 2021 10.1190/geo2020-0034.1 Optimization-inspired deep learning high-resolution inversion for seismic data 10.1190/segam2019-3216042.1 Choi Y. Seol S. J. Byun J. Kim Y. , 2019, Vertical resolution enhancement of seismic data with convolutional U-net: 89th Annual International Meeting, SEG, Expanded Abstracts, 2388–2392, doi:http://dx.doi.org/10.1190/segam2019-3216042.1. 0162-8828 IEEE Transactions on Pattern Analysis and Machine Intelligence Dong 38 295 2015 10.1109/TPAMI.2015.2439281 Image super-resolution using deep convolutional networks SIAM Journal on Imaging Sciences Gholami 6 2350 2013 10.1137/130905009 Fast 3D blind seismic deconvolution via constrained total variation and GCV 0001-0782 Communications of the ACM Goodfellow 63 139 2020 10.1145/3422622 Generative adversarial networks 10.1190/image2022-3746892.1 Hamida A. Alfarraj M. Al-Shuhail A. A. Zummo S. A. , 2022, Efficient seismic image super-resolution: Second International Meeting for Applied Geoscience & Energy, SEG, Expanded Abstracts, 1659–1663, doi: http://dx.doi.org/10.1190/image2022-3746892.1. 10.1109/CVPR.2016.90 He K. Zhang X. Ren S. Sun J. , 2016, Deep residual learning for image recognition: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770–778. 10.1109/CVPRW.2018.00112 Ignatov A. Kobyshev N. Timofte R. Vanhoey K. Van Gool L. , 2018, WESPE: Weakly supervised photo enhancer for digital cameras: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 691–700. 1057-7149 IEEE Transactions on Image Processing Jiang 30 2340 2021 10.1109/TIP.2021.3051462 EnlightenGAN: Deep light enhancement without paired supervision 0016-8033 Geophysics Kaur 85 4 WA173 2020 10.1190/geo2019-0315.1 Improving the resolution of migrated images by approximating the inverse Hessian using deep learning 0016-8033 Geophysics Kazemeini 75 6 V101 2010 10.1190/1.3483900 Enhancing seismic data resolution using the prestack blueing technique: An example from the Ketzin CO2 injection site, Germany 0196-2892 IEEE Transactions on Geoscience and Remote Sensing Ke 61 1 2023 The nth power Fourier spectrum analysis for the generalized seismic wavelets 10.1109/CVPR.2016.181 Kim J. Lee J. K. Lee K. M. , 2016a, Deeply recursive convolutional network for image super-resolution: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1637–1645. 10.1109/CVPR.2016.182 Kim J. Lee J. K. Lee K. M. , 2016b, Accurate image super-resolution using very deep convolutional networks: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1646–1654. Kim T. Cha M. Kim H. Lee J. K. Kim J. , 2017, Learning to discover cross-domain relations with generative adversarial networks: International Conference on Machine Learning, 1857–1865. Kingma 2014 Adam: A method for stochastic optimization 10.1109/CVPR.2017.618 Lai W.-S. Huang J.-B. Ahuja N. Yang M.-H. , 2017, Deep Laplacian pyramid networks for fast and accurate super-resolution: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 624–632. 10.1190/1.1815711 Lancaster S. Whitcombe D. , 2000, Fast-track ‘coloured’ inversion: 80th Annual International Meeting, SEG, Expanded Abstracts, 1572–1575, doi: http://dx.doi.org/10.1190/1.1815711. 10.1109/CVPR.2017.19 Ledig C. Theis L. Huszár F. Caballero J. Cunningham A. Acosta A. Aitken A. Tejani A. Totz J. Wang Z. Shi W. , 2017, Photo-realistic single image super-resolution using a generative adversarial network: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 4681–4690. 0016-8033 Geophysics Levy 46 1235 1981 10.1190/1.1441261 Reconstruction of a sparse spike train from a portion of its spectrum and application to high-resolution deconvolution IEEE Signal Processing Letters Li 25 323 2018 10.1109/LSP.2018.2792050 Emerging from water: Underwater image color correction based on weakly supervised color transfer 0196-2892 IEEE Transactions on Geoscience and Remote Sensing Li 60 1 2021 Deep learning for simultaneous seismic image super-resolution and denoising IEEE Geoscience and Remote Sensing Letters Li 18 2016 2020 10.1109/LGRS.2020.3011130 The denoising of desert seismic data based on cycle-GAN with unpaired data training 10.1109/ICCVW54120.2021.00210 Liang J. Cao J. Sun G. Zhang K. Van Gool L. Timofte R. , 2021, SwinIR: Image restoration using swin transformer: Proceedings of the IEEE/CVF International Conference on Computer Vision, 1833–1844. 0016-8033 Geophysics Liu 88 1 WA253 2022 10.1190/geo2022-0179.1 Improving sparse representation with deep learning: A workflow for separating strong background interference 0196-2892 IEEE Transactions on Geoscience and Remote Sensing Liu 58 1598 2020 10.1109/TGRS.2019.2947149 Poststack seismic data denoising based on 3-D convolutional neural network 0016-8033 Geophysics Liu 87 5 V505 2022 10.1190/geo2021-0654.1 Accelerating seismic scattered noise attenuation in offset-vector tile domain: Application of deep learning Maas A. L. Hannun A. Y. Ng A. Y. , 2013, Rectifier nonlinearities improve neural network acoustic model: ICMLWorkshop on Deep Learning for Audio, Speech and Language Processing. Mao 2016 Multi-class generative adversarial networks with the L2 loss function Mao X. Shen C. Yang Y.-B. , 2016a, Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections: Advances in Neural Information Processing Systems, 29. The Leading Edge Martin 25 156 2006 10.1190/1.2172306 Marmousi2: An elastic upgrade for Marmousi Journal of the ACM Ormsby 8 440 1961 10.1145/321075.321087 Design of numerical filters with applications to missile data processing 0196-2892 IEEE Transactions on Geoscience and Remote Sensing Saad 60 1 2022 Self-attention deep image prior network for unsupervised 3-D seismic data enhancement 0956-540X Geophysical Journal International Sacchi 129 651 1997 10.1111/j.1365-246X.1997.tb04500.x Reweighting strategies in seismic deconvolution 10.1109/CVPR.2016.207 Shi W. Caballero J. Huszár F. Totz J. Aitken A. P. Bishop R. Rueckert D. Wang Z. , 2016, Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1874–1883. 0016-8033 Geophysics Sui 85 6 V481 2020 10.1190/geo2019-0423.1 Blind sparse-spike deconvolution with thin layers and structure 0016-8033 Geophysics Taylor 44 39 1979 10.1190/1.1440921 Deconvolution with the ℓ1 norm 10.1109/ICCV.2017.514 Tong T. Li G. Liu X. Gao Q. , 2017, Image super-resolution using dense skip connections: Proceedings of the IEEE International Conference on Computer Vision, 4799–4807. 0016-8033 Geophysics Velis 73 1 R1 2008 10.1190/1.2790584 Stochastic sparse-spike deconvolution 0196-2892 IEEE Transactions on Geoscience and Remote Sensing Wang 59 7062 2020 10.1109/TGRS.2020.3030692 Generative adversarial network for desert seismic data denoising 10.1109/ICCVW54120.2021.00217 Wang X. Xie L. Dong C. Shan Y. , 2021, Real-ESRGAN: Training real-world blind super-resolution with pure synthetic data: IEEE/CVF International Conference on Computer Vision Workshops, 1905–1914. 10.1007/978-3-030-11021-5_5 Wang X. Yu K. Wu S. Gu J. Liu Y. Dong C. Qiao Y. Loy C. C. , 2018, ESRGAN: Enhanced super-resolution generative adversarial networks: European Conference on Computer Vision, 63–79. 0016-8033 Geophysics Wang 71 3 V51 2006 10.1190/1.2192912 Inverse Q-filter for seismic resolution enhancement 0162-8828 IEEE Transactions on Pattern Analysis and Machine Intelligence Wang 43 3365 2020 10.1109/TPAMI.2020.2982166 Deep learning for image super-resolution: A survey Wang Z. Simoncelli E. P. Bovik A. C. , 2003, Multiscale structural similarity for image quality assessment: 37th Asilomar Conference on Signals, Systems & Computers, 1398–1402. 1057-7149 IEEE Transactions on Image Processing Wei 30 4788 2021 10.1109/TIP.2021.3074804 DerainCycleGAN: Rain attentive CycleGAN for single image deraining and rainmaking 0016-7142 Geoexploration Wiggins 16 21 1978 10.1016/0016-7142(78)90005-4 Minimum entropy deconvolution 0016-8033 Geophysics Wu 84 3 IM35 2019 10.1190/geo2018-0646.1 FaultSeg3D: Using synthetic data sets to train an end-to-end convolutional neural network for 3D seismic fault segmentation 0016-8033 Geophysics Xue 84 2 V121 2019 10.1190/geo2018-0177.1 Inverse Q filtering via synchrosqueezed wavelet transform 10.1109/ICCV.2017.310 Yi Z. Zhang H. Tan P. Gong M. , 2017, DualGAN: Unsupervised dual learning for image-to-image translation: Proceedings of the IEEE International Conference on Computer Vision, 2849–2857. 8755-1209 Reviews of Geophysics Yu 59 2021 10.1029/2021RG000742 Deep learning for geophysics: Current and future trends 10.1109/CVPRW.2018.00113 Yuan Y. Liu S. Zhang J. Zhang Y. Dong C. Lin L. , 2018, Unsupervised image super-resolution using cycle-in-cycle generative adversarial networks: IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 701–710. 10.1190/segam2019-3216229.1 Zhang H. Wang W. Wang X. Chen W. Zhou Y. Wang C. Zhao Z. , 2019, An implementation of the seismic resolution enhancing network based on GAN: 89th Annual International Meeting, SEG, Expanded Abstracts, 2478–2482, doi: http://dx.doi.org/10.1190/segam2019-3216229.1. 0196-2892 IEEE Transactions on Geoscience and Remote Sensing Zhang 60 1 2022 A novel multichannel seismic deconvolution method via structure-oriented regularization IEEE Transactions on Computational Imaging Zhao 3 47 2016 10.1109/TCI.2016.2644865 Loss functions for image restoration with neural networks National Science Review Zhou 5 44 2018 10.1093/nsr/nwx106 A brief introduction to weakly supervised learning 10.1109/ICCV.2017.244 Zhu J.-Y. Park T. Isola P. Efros A. A. , 2017, Unpaired image-to-image translation using cycle-consistent adversarial networks: Proceedings of the IEEE International Conference on Computer Vision, 2223–2232.