Abstract
For seismic exploration, excellent seismic records are of extreme significance. However, because of extensive mining over a long period, spare natural resources are primarily distributed in sophisticated areas with weak signals and low signal-to-noise ratios (SNRs). Besides, the incoherent noise overlaps with signals in both the time and frequency domains. All these characteristics place a greater demand on denoising methods. In this article, we propose a multiple attention mechanism-based modular convolutional neural network for desert seismic denoising. To extract weak signals from noisy recordings, two types of attention modules are chosen in this network. The channel-wise mechanism in the enhanced attention module is added to the network to extract detailed information from noisy records. Then, the supervised attention module can learn from the input noisy records again, emphasizing the significant features again to further enhance the denoising capability of our network. Besides, different types of dilated convolution layers are added to the network to extend the receptive field to cover minute information as much as possible. Both synthetic and real seismic denoising experiments show that the proposed network framework can suppress the random noise and recover signals even in low SNR situations.
Similar content being viewed by others
Data availability
No data are available.
References
Anne, O., & Gregor, H. (2019). Chapter two—seismic time-lapse interferometry across scales. Advances in Geophysics. https://doi.org/10.1016/bs.agph.2019.06.001
Anwar, S. (2019). Real image denoising with feature attention. in IEEE/CVF International Conference on Computer Vision (ICCV). IEEE, 2019, pp. 3155–3164. https://doi.org/10.1109/ICCV.2019.00325
Bekara, M., & van der Baan, M. (2009). Random and coherent noise attenuation by empirical mode decomposition. Geophysics, 74(5), 89–98. https://doi.org/10.3997/2214-4609.20149662
Canales, L.L. (1984). Random noise reduction. in 54th Annual International Meeting of SEG Technical Program Expanded Abstracts, pp. 525–527. https://doi.org/10.1190/1.1894168
Chen, Y., & Ma, J. (2014). Random noise attenuation by f-x empirical-mode decomposition predictive filtering. Geophysics, 79(3), 81–91. https://doi.org/10.1190/GEO2013-0080.1
Civiero, C., Custodio, S., Rawlinson, N., Strak, V., Silveira, G., Arroucau, P., & Corela, C. (2019). Thermal nature of mantle upwellings below the Ibero-western Maghreb region inferred from teleseismic tomography. Journal of Geophysical Research: Solid Earth, 124(2), 1781–1801. https://doi.org/10.1029/2018JB016531
Comaniciu, D., & Meer, P. (2002). Mean shift: A robust approach toward feature space analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(5), 603–619. https://doi.org/10.1109/34.1000236
Dabov, K., Foi, A., Katkovnik, V., & Egiazarian, K. (2007). Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Transactions on Image Processing, 16(8), 2080–2095. https://doi.org/10.1109/TIP.2007.901238
Emma, L. C., Nicholas, H., Catherine, A. R., Ryan, J. G., & Derek, K. (2022). Imaging the seismic velocity structure of the crust and upper mantle in the northern East African Rift using Rayleigh wave tomography. Geophysical Journal International, 5, 156. https://doi.org/10.31223/X5XK6X
Fabien-Ouellet, G. (2020). Low-frequency generation and denoising with recursive convolutional neural networks. SEG Technical Program Expanded Abstracts, 2020, 870–874. https://doi.org/10.1190/segam2020-3428270.1
Gu, S., Zhang, L., Zuo, W., & Feng, X. (2014). Weighted nuclear norm minimization with application to image denoising. in IEEE Conference on Computer Vision and Pattern Recognition, pp. 2862–2869. https://doi.org/10.1109/CVPR.2014.366
Gupta, H., & Pasquale, V. (2013). Encyclopedia of Solid Earth Geophysics. Springer International Publishing.
Jiang, X., Liu, S., Dai, X., Hu, G., Huang, X., Yao, Y., & Shao, L. (2022). Deep metric learning based on meta-mining strategy with semiglobal information. IEEE Transactions on Neural Networks and Learning Systems. https://doi.org/10.1109/TNNLS.2022.3202571
Jie, H., Li, S. & Gang. S. (2018). Squeeze-and-excitation networks. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 7132–7141. https://doi.org/10.1109/CVPR.2018.00745
Leslie, R. V. (2018). 1.16—microwave sensors. Comprehensive Remote Sensing, 2, 435–474. https://doi.org/10.1016/B978-0-12-409548-9.10328-8
Li, H., Chen, D. & Chang, D. (2019). Ground-roll noise attenuation based on convolutional neural network.in SEG 2019 Workshop: Fractured Reservoir & Unconventional Resources Forum: Prospects and Challenges in the Era of Big Data, Lanzhou, China, 1–3, pp. 69–73. https://doi.org/10.1190/frur2019_18.1
Li, F., Zhang, B., Verma, S., & Marfurt, K. J. (2018). Seismic signal denoising using thresholded variational mode decomposition. Exploration Geophysics, 49(4), 450–461. https://doi.org/10.1071/EG17004
Li, J.-H., Zhang, Y.-J., Qi, R., & 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, 10(8), 3810–3820. https://doi.org/10.1109/JSTARS.2017.2685628
Liu, N., et al. (2022c). NS2NS: Self-learning for seismic image denoising. IEEE Transactions on Geoscience and Remote Sensing, 60, 1–11. https://doi.org/10.1109/TGRS.2022.3217289
Liu, N., Wang, J., Gao, J., Chang, S., & Lou, Y. (2022b). Similarity-informed self-learning and its application on seismic image denoising. IEEE Transactions on Geoscience and Remote Sensing, 60, 1–13. https://doi.org/10.1109/TGRS.2022.3210217
Liu, N., Wu, L., Wang, J., Wu, H., Gao, J., & Wang, D. (2022a). Seismic data reconstruction via wavelet-based residual deep learning. IEEE Transactions on Geoscience and Remote Sensing, 60, 1–13. https://doi.org/10.1109/TGRS.2022.3152984
Liu, W., & Duan, Z. (2020). Seismic signal denoising using f–x variational mode decomposition. IEEE Geoscience and Remote Sensing Letters, 17(8), 1313–1317. https://doi.org/10.1109/LGRS.2019.2948631
Liu, Y., Fomel, S., & Liu, C. (2015). Signal and noise separation in prestack seismic data using velocity-dependent seislet transform. Geophysics, 80(6), 117–128. https://doi.org/10.1190/geo2014-0234.1
Luo, J., Yan, Z., Wu, R. S., Chen, G., & Huang, X. (2022). Angle domain illumination compensated full waveform inversion. IEEE Transactions on Geoscience and Remote Sensing, Society of Exploration Geophysicists, 60, 1–11. https://doi.org/10.1109/TGRS.2022.3204553
Moore, R., Ezekiel, S., & Blasch, E. (2014). Denoising one-dimensional signals with curvelets and contourlets, Proc. in IEEE Nat. Aerosp. Electron. Conf., pp. 189–194. https://doi.org/10.1109/NAECON.2014.7045801
Mousavi, S. M., & Langston, C. A. (2016). Hybrid seismic denoising using higher-order statistics and improved wavelet block thresholding. Bulletin of the Seismological Society of America, 106(4), 1380–1393. https://doi.org/10.1785/0120150345
Robinson, E. A., & Treitel, S. (1967). Principles of digital Wiener FILTERING. Geophysical Prospecting, 15, 3. https://doi.org/10.1111/j.1365-2478.1967.tb01793.x
Roman, S., Frank, N., & Alan, G. G. (2001). Reducing source-generated noise in shallow seismic data using linear and hyperbolic τ–p transformations. Geophysics, 66(5), 1612–1621. https://doi.org/10.1190/1.1487106
Stephen, R.M., Glenn, T., Jeffrey, J., de Silvio, A, & David, F. (2015). Chapter 63—Seismic and Infrasonic Monitoring, The Encyclopedia of Volcanoes (Second Edition), 1071–1099, https://doi.org/10.1016/B978-0-12-385938-9.00063-8
Tibi, R., Hammond, P., Brogan, R., Young, C. J., & Koper, K. (2021). Deep learning denoising applied to regional distance seismic data in Utah. Bulletin of the Seismological Society of America, 111(2), 775–790. https://doi.org/10.1785/0120200292
Vrolijk, W. J., & Blacquiere, G. (2021). Source deghosting of coarsely sampled common-receiver data using a convolutional neural network. Geophysics, 86(3), 185–196. https://doi.org/10.1190/geo2020-0186.1
Wang, C., Zhu, Z., & Gu, H. (2020). Low-rank seismic denoising with optimal rank selection for hankel matrices. Geophysical Prospecting, 68(3), 892–909. https://doi.org/10.1111/1365-2478.12899
Webb, S. C. (2002). 19—Seismic noise on land and on the sea floor. International Geophysics, 5, 305–318. https://doi.org/10.1016/S0074-6142(02)80222-4
Yang, C., Zhou, Y., He, H., He, J., & Chi, Y. (2021). Pyramid residual neural network with attention for seismic data denoising. In 2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI), pp. 387–390. https://doi.org/10.1109/ICCEAI52939.2021.00077
Yatong, Z., Shuhua, L., Dong, Z., & Yangkang, C. (2020). Seismic noise attenuation using an online subspace tracking algorithm. Geophysical Journal International, 222(3), 1765–1788. https://doi.org/10.1093/gji/ggaa187
Zamir, S.W., Arora, A., & Khan, S. (2021). Multi-stage progressive image restoration. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 14821–14831. https://doi.org/10.1109/CVPR46437.2021.01458
Zhang, C., & Baan, M. (2019). Microseismic and seismic noise attenuation by supervised deep learning with noisy natural images. SEG Technical Program Expanded Abstracts, 2019, 4485–4489. https://doi.org/10.1190/segam2019-3213158.1
Zhang, K., Zuo, W., & Chen, Y. (2017). Beyond a gaussian denoiser: Residual learning of deep cnn for image denoising. IEEE Transactions on Image Processing, 26(7), 3142–3155. https://doi.org/10.1109/TIP.2017.2662206
Zhao, Y., Li, Y., & Yan, J. (2021). Desert seismic data denoising based on Gaussian conditional random field with sparsity measurement. IEEE Geoscience and Remote Sensing Letters, 18(10), 1851–1855. https://doi.org/10.1109/LGRS.2020.3008231
Funding
This research is supported by the Natural Science Foundation of the Department of Science and Technology of Jilin Province under grant no. 20200201046JC and the National Natural Science Foundation of China under grant 41730422.
Author information
Authors and Affiliations
Contributions
Author 1: Conceptualization, Methodology, Software, Investigation, Formal Analysis; Author 2: Data Curation, Writing - Original Draft; Author 3: Resources, Supervision; All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Conflict of interest
The authors declare no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, J., Qu, R. & Lu, C. Multiple Attention Mechanisms-Based Convolutional Neural Network for Desert Seismic Denoising. Pure Appl. Geophys. 180, 2135–2155 (2023). https://doi.org/10.1007/s00024-023-03255-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00024-023-03255-5