| Peer-Reviewed

Research Status and Prospects of Generative Adversarial Networks in Seismic Data Denoising

Received: 13 September 2022     Accepted: 29 September 2022     Published: 17 October 2022
Views:       Downloads:
Abstract

Efficient and high-quality processing of seismic data collected by geophone sensors is the core of successful seismic exploration. Seismic denoising is a key step in seismic data processing. Traditional seismic denoising relies on manual empirical parameter selection and comparative analysis, which is time-consuming and limited by subjective errors. Deep learning methods based on convolutional neural networks (CNN) have improved the efficiency of denoising massive amounts of seismic data and reduced the manual errors of traditional methods. However, most CNN methods only consider data loss and are weak in recovering the structure of seismic signals, resulting in severe attenuation of some seismic traces in the recovered effective signals, reducing the continuity of seismic events and the quality of seismic data. Generative adversarial networks (GAN), a popular method for deep learning with unique adversarial ideas and powerful feature extraction capabilities, can overcome the limitations of CNN methods in the field of seismic data denoising. This paper firstly introduces the classification and development process of seismic denoising. Then, starting from the principle of GAN, it introduces the workflow of the original GAN, the objective function in the training process, the existing problems of the original GAN and some mainstream solutions to these problems, and introduces the commonly used model of GAN in the field of earthquake denoising. besides, it summarizes and analyzes the current application and improvement innovation of GAN in the field of seismic denoising, and analyzes the application of GAN in the field of seismic denoising from two aspects of supervised learning and unsupervised learning with examples. Finally, the prospect of GAN for seismic denoising in the future is prospected.

Published in International Journal of Sensors and Sensor Networks (Volume 10, Issue 2)
DOI 10.11648/j.ijssn.20221002.13
Page(s) 33-50
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2022. Published by Science Publishing Group

Keywords

Deep Learning, GAN, Seismic Denoising

References
[1] Zhao, B.; Yong, X. Progress and development direction of Petro China intelligent seismic processing and interpretation technology. China Petroleum Exploration. 2021, 26, 12-23.
[2] Yang, W.; Wei, X. Development plan for intelligent geophysical prospecting technology of applied geophysical+AI. Oil Forum. 2019, 38, 40-47.
[3] Zhu, L.; Liu, E.; McClellan, J. H. Seismic Data Denoising through Multiscale and Sparsity-Promoting Dictionary Learning. GEOPHYSICS 2015, 80 (6), WD45–WD57.
[4] Zhang Junhua, Lv Ning, Tian Lianyu, et al. A comprehensive review of seismic data denoising methods and techniques. Advances in Geophysics, 2006, 2.
[5] Han, G.; Jie, Z. Simultaneous Denoising and Interpolation of Seismic Data via the Deep Learning Method. 33 (1), 15.
[6] Yang, L.; Chen, W.; Wang, H.; Chen, Y. Deep Learning Seismic Random Noise Attenuation via Improved Residual Convolutional Neural Network. IEEE Trans. Geosci. Remote Sensing 2021, 59 (9), 7968–7981.
[7] Chen Tian, Yi Yuanyuan. Random Noise Suppression for Seismic Data Based on Deep Convolutional Neural Networks. Chinese Journal of Seismology, 2021, 43 (4): 474-482.
[8] Yu, S.; Ma, J.; Wang, W. Deep Learning for Denoising. GEOPHYSICS 2019, 84 (6), V333–V350.
[9] Zhang Chaoming, Wen Xiaotao, Zhang Xiaoqi, Lan Yunlin, He Yilong. Seismic data denoising method based on DNCNN and constrained convolution. Advances in Geophysics, 2022, 1-18.
[10] Richardson, A.; Feller, C. Seismic Data Denoising and Deblending Using Deep Learning. arXiv July 2, 2019.
[11] Wang, E.; Nealon, J. Applying Machine Learning to 3D Seismic Image Denoising and Enhancement. Interpretation 2019, 7 (3), SE131–SE139.
[12] Zhang, F.; Liu, D.; Wang, X.; Chen, W.; Wang, W. Random Noise Attenuation Method for Seismic Data Based on Deep Residual Networks. In International Geophysical Conference, Beijing, China, 24-27 April 2018; Society of Exploration Geophysicists and Chinese Petroleum Society: Beijing, China, 2018; pp 1774–1777.
[13] Wang, F.; Chen, S. Residual Learning of Deep Convolutional Neural Network for Seismic Random Noise Attenuation. IEEE Geosci. Remote Sensing Lett. 2019, 16 (8), 1314–1318.
[14] Jin, Y.; Wu, X.; Chen, J.; Han, Z.; Hu, W. Seismic Data Denoising by Deep-Residual Networks. In SEG Technical Program Expanded Abstracts 2018; Society of Exploration Geophysicists: Anaheim, California, 2018; pp 4593–4597.
[15] Han, W.; Zhou, Y. Deep learning convolutional neural networks for random noise attenuation in seismic data. Geophysical Prospecting for Petroleum. 2018, 57, 862-869.
[16] Li, S.; Chen, W. Pre-stack Random Noise Deep Residual Network Suppression Method. Oil Geophysical Prospecting. 2020, 55, 493-503.
[17] Yang, L.; Chen, W.; Liu, W.; Zha, B.; Zhu, L. Random Noise Attenuation Based on Residual Convolutional Neural Network in Seismic Datasets. IEEE Access 2020, 8, 30271–30286.
[18] Wang, Y.; Lu, W. Seismic random noise suppression based on data augmentation and CNN [J]. CHINESE J GEOPHYS-CH, 2019, 62 (01): 421-433.
[19] Dong, X.; Zhong, T.; Li, Y. New Suppression Technology for Low-Frequency Noise in Desert Region: The Improved Robust Principal Component Analysis Based on Prediction of Neural Network. IEEE Trans. Geosci. Remote Sensing 2020, 58 (7), 4680–4690.
[20] Zhao, Y.; Li, Y.; Dong, X.; Yang, B. Low-Frequency Noise Suppression Method Based on Improved DNCNN in Desert Seismic Data. IEEE Geosci. Remote Sensing Lett. 2019, 16 (5), 811–815.
[21] Liu, D.; Wang, W.; Chen, W.; Wang, X.; Zhou, Y.; Shi, Z. Random-Noise Suppression in Seismic Data: What Can Deep Learning Do? In SEG Technical Program Expanded Abstracts 2018; Society of Exploration Geophysicists: Anaheim, California, 2018; pp 2016–2020.
[22] Liu, D.; Wang, W.; Wang, X.; Wang, C.; Pei, J.; Chen, W. Poststack Seismic Data Denoising Based on 3-D Convolutional Neural Network. IEEE Trans. Geosci. Remote Sensing 2020, 58 (3), 1598–1629.
[23] Zhang, M.; Liu, Y.; Chen, Y. Unsupervised Seismic Random Noise Attenuation Based on Deep Convolutional Neural Network. IEEE Access 2019, 7, 179810–179822.
[24] Wang, H.; Li, Y.; Dong, X. Generative Adversarial Network for Desert Seismic Data Denoising. IEEE Trans. Geosci. Remote Sensing 2021, 59 (8), 7062–7075.
[25] Goodfellow, I. J.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; Bengio, Y. Generative Adversarial Networks. arXiv June 10, 2014.
[26] Alwon, S. Generative Adversarial Networks in Seismic Data Processing. In SEG Technical Program Expanded Abstracts 2018; Society of Exploration Geophysicists: Anaheim, California, 2018; pp 1991–1995.
[27] Picetti, F.; Lipari, V.; Bestagini, P.; Tubaro, S. A Generative-Adversarial Network For Seismic-Imaging Applications. In SEG Technical Program Expanded Abstracts 2018; Society of Exploration Geophysicists: Anaheim, California, 2018; pp 2231–2235.
[28] Wang, S.; Li, Y.; Wu, N.; Zhao, Y.; Yao, H. Attribute-Based Double Constraint Denoising Network for Seismic Data. IEEE Trans. Geosci. Remote Sensing 2021, 59 (6), 5304–5316.
[29] Ma, H.; Sun, Y.; Wu, N.; Li, Y. Relative Attributes-Based Generative Adversarial Network for Desert Seismic Noise Suppression. IEEE Geosci. Remote Sensing Lett. 2022, 19, 1–5.
[30] Zheng, J.; Wu, Z. Generative adversarial network-based denoising method for micro seismic data. Research & Explorationin Laboratory, 2021, 40 (05): 18-21.
[31] Li, Y.; Luo, X.; Wu, N.; Dong, X. The Application of Semisupervised Attentional Generative Adversarial Networks in Desert Seismic Data Denoising. IEEE Geosci. Remote Sensing Lett. 2022, 19, 1–5.
[32] Yu, R.; Zhang, Y. Random noise removal from Rayleigh wave signals based on deep convolutional generative adversarial networks. Prog Geophys, 2020, 35 (06): 2276-2283.
[33] Liu, Y.; Wei, H. Random noise removal from seismic data based on convolutional recurrent generative adversarial networks. Journal of Jilin University (Information Science Edition), 2022, 1-9.
[34] Li, W.; Wang, J. Residual Learning of Cycle-GAN for Seismic Data Denoising. 2021, 9, 13.
[35] Ma, W.; Pan, Z.; Guo, J.; Lei, B. Achieving Super-Resolution Remote Sensing Images via the Wavelet Transform Combined With the Recursive Res-Net. IEEE Trans. Geosci. Remote Sensing 2019, 57 (6), 3512–3527.
[36] Rudin, L. I.; Osher, S.; Fatemi, E. Nonlinear Total Variation Based Noise Removal Algorithms. Physica D: Nonlinear Phenomena 1992, 60 (1–4), 259–268.
[37] Bonar, D.; Sacchi, M. Denoising Seismic Data Using the Nonlocal Means Algorithm. GEOPHYSICS 2012, 77 (1), A5–A8.
[38] Gulunay, N. FXDECON and Complex Wiener Prediction Filter. In SEG Technical Program Expanded Abstracts 1986; Society of Exploration Geophysicists, 1986; pp 279–281.
[39] Hennenfent, G.; Herrmann, F. J. Seismic Denoising with Nonuniformly Sampled Curvelets. Comput. Sci. Eng. 2006, 8 (3), 16–25.
[40] Dabov, K.; Foi, A.; Katkovnik, V.; Egiazarian, K. Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering. IEEE Trans. on Image Process. 2007, 16 (8), 2080–2095.
[41] Dong, W.; Shi, G.; Li, X. Nonlocal Image Restoration With Bilateral Variance Estimation: A Low-Rank Approach. IEEE Trans. on Image Process. 2013, 22 (2), 700–711.
[42] Mertens, J.-F.; Zamir, S. The Value of Two-Person Zero-Sum Repeated Games with Lack of Information on Both Sides. Int J Game Theory 1971, 1 (1), 39–64.
[43] Fuglede, B.; Topsoe, F. Jensen-Shannon Divergence and Hilbert Space Embedding. In International Symposium onInformation Theory, 2004. ISIT 2004. Proceedings.; IEEE: Chicago, Illinois, USA, 2004; pp 30–30.
[44] Raiber, F.; Kurland, O. Kullback-Leibler Divergence Revisited. In Proceedings of the ACM SIGIR International Conference on Theory of Information Retrieval; ACM: Amsterdam The Netherlands, 2017; pp 117–124.
[45] Arjovsky, M.; Chintala, S.; Bottou, L. Wasserstein GAN. arXiv December 6, 2017.
[46] Gulrajani, I.; Ahmed, F.; Arjovsky, M.; Dumoulin, V.; Courville, A. Improved Training of Wasserstein GANs. arXiv December 25, 2017.
[47] Radford, A.; Metz, L.; Chintala, S. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. arXiv January 7, 2016.
[48] Zhang, K.; Zuo, W.; Chen, Y.; Meng, D.; Zhang, L. Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising. IEEE Trans. on Image Process. 2017, 26 (7), 3142–3155.
[49] Mirza, M.; Osindero, S. Conditional Generative Adversarial Nets. arXiv November 6, 2014.
[50] Zhu, J.-Y.; Park, T.; Isola, P.; Efros, A. A. Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks. In 2017 IEEE International Conference on Computer Vision (ICCV); IEEE: Venice, 2017; pp 2242–2251.
[51] Xia, Y.; He, D.; Qin, T.; Wang, L.; Yu, N.; Liu, T.-Y.; Ma, W.-Y. Dual Learning for Machine Translation. arXiv November 1, 2016.
[52] Si, X.; Yuan, Y.; Ping, F.; Zheng, Y.; Feng, L. Ground Roll Attenuation Based on Conditional and Cycle Generative Adversarial Networks. In SEG 2019 Workshop: Mathematical Geophysics: Traditional vs Learning, Beijing, China, 5–7 November 2019; Society of Exploration Geophysicists: Beijing, China, 2020; pp 95–98.
[53] Yuan, Y.; Si, X.; Zheng, Y. Ground-Roll Attenuation Using Generative Adversarial Networks. GEOPHYSICS 2020, 85 (4), WA255–WA267.
[54] Si, X. Ground Roll Attenuation with Conditional Generative Adversarial Networks. In SEG Technical Program Expanded Abstracts 2020; Society of Exploration Geophysicists: Virtual, 2020; pp 1511–1515.
[55] Dong, X.; Li, Y. Denoising the Optical Fiber Seismic Data by Using Convolutional Adversarial Network Based on Loss Balance. IEEE Trans. Geosci. Remote Sensing 2021, 59 (12), 10544–10554.
[56] Kaur, H.; Fomel, S.; Pham, N. Seismic Ground-roll Noise Attenuation Using Deep Learning. Geophysical Prospecting 2020, 68 (7), 2064–2077.
[57] Li, Y.; Wang, H.; Dong, X. The Denoising of Desert Seismic Data Based on Cycle-GAN With Unpaired Data Training. IEEE Geosci. Remote Sensing Lett. 2021, 18 (11), 2016–2020.
[58] Wu, X.; Zhang, H. Random noise suppression method for seismic data based on cyclic consistent Generative adversarial network. Oil Geophysical Prospecting. 2021, 56, 958-968.
[59] Zhu, W.; Mousavi, S. M.; Beroza, G. C. Seismic Signal Denoising and Decomposition Using Deep Neural Networks. IEEE Trans. Geosci. Remote Sensing 2019, 57 (11), 9476–9488.
[60] Pan, Z.; Yu, W.; Yi, X.; Khan, A.; Yuan, F.; Zheng, Y. Recent Progress on Generative Adversarial Networks (GANs): A Survey. IEEE Access 2019, 7, 36322–36333.
[61] Alnujaim, I.; Kim, Y. Augmentation of Doppler Radar Data Using Generative Adversarial Network for Human Motion Analysis. Healthc Inform Res 2019, 25 (4), 344.
[62] Leinonen, J.; Guillaume, A.; Yuan, T. Reconstruction of Cloud Vertical Structure With a Generative Adversarial Network. Geophysical Research Letters 2019, 46 (12), 7035–7044.
[63] Wolterink, J. M.; Leiner, T.; Viergever, M. A.; Isgum, I. Generative Adversarial Networks for Noise Reduction in Low-Dose CT. IEEE Trans. Med. Imaging 2017, 36 (12), 2536–2545.
[64] Sandfort, V.; Yan, K.; Pickhardt, P. J.; Summers, R. M. Data Augmentation Using Generative Adversarial Networks (CycleGAN) to Improve Generalizability in CT Segmentation Tasks. Sci Rep 2019, 9 (1), 16884.
[65] Mao, X.; Shen, C.; Yang, Y.-B. Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections. 9.
[66] Dumoulin, V.; Visin, F. A Guide to Convolution Arithmetic for Deep Learning. arXiv January 11, 2018.
[67] Eckle, K.; Schmidt-Hieber, J. A Comparison of Deep Networks with ReLU Activation Function and Linear Spline-Type Methods. Neural Networks 2019, 110, 232–242.
[68] Ma, H.; Yao, H.; Li, Y.; Wang, H. Deep Residual Encoder–Decoder Networks for Desert Seismic Noise Suppression. IEEE Geosci. Remote Sensing Lett. 2020, 17 (3), 529–533.
[69] Liu, H.; Zhang, J. A study on the impact of introducing feature loss on CycleGAN. Computer Engineering and Applications, 2020, 56 (22): 217-223.
[70] Li, C.; Liu, H.; Chen, C.; Pu, Y.; Chen, L.; Henao, R.; Carin, L. ALICE: Towards Understanding Adversarial Learning for Joint Distribution Matching. 9.
Cite This Article
  • APA Style

    Yuan Xu, Qing Wang, Shiguang Guo. (2022). Research Status and Prospects of Generative Adversarial Networks in Seismic Data Denoising. International Journal of Sensors and Sensor Networks, 10(2), 33-50. https://doi.org/10.11648/j.ijssn.20221002.13

    Copy | Download

    ACS Style

    Yuan Xu; Qing Wang; Shiguang Guo. Research Status and Prospects of Generative Adversarial Networks in Seismic Data Denoising. Int. J. Sens. Sens. Netw. 2022, 10(2), 33-50. doi: 10.11648/j.ijssn.20221002.13

    Copy | Download

    AMA Style

    Yuan Xu, Qing Wang, Shiguang Guo. Research Status and Prospects of Generative Adversarial Networks in Seismic Data Denoising. Int J Sens Sens Netw. 2022;10(2):33-50. doi: 10.11648/j.ijssn.20221002.13

    Copy | Download

  • @article{10.11648/j.ijssn.20221002.13,
      author = {Yuan Xu and Qing Wang and Shiguang Guo},
      title = {Research Status and Prospects of Generative Adversarial Networks in Seismic Data Denoising},
      journal = {International Journal of Sensors and Sensor Networks},
      volume = {10},
      number = {2},
      pages = {33-50},
      doi = {10.11648/j.ijssn.20221002.13},
      url = {https://doi.org/10.11648/j.ijssn.20221002.13},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijssn.20221002.13},
      abstract = {Efficient and high-quality processing of seismic data collected by geophone sensors is the core of successful seismic exploration. Seismic denoising is a key step in seismic data processing. Traditional seismic denoising relies on manual empirical parameter selection and comparative analysis, which is time-consuming and limited by subjective errors. Deep learning methods based on convolutional neural networks (CNN) have improved the efficiency of denoising massive amounts of seismic data and reduced the manual errors of traditional methods. However, most CNN methods only consider data loss and are weak in recovering the structure of seismic signals, resulting in severe attenuation of some seismic traces in the recovered effective signals, reducing the continuity of seismic events and the quality of seismic data. Generative adversarial networks (GAN), a popular method for deep learning with unique adversarial ideas and powerful feature extraction capabilities, can overcome the limitations of CNN methods in the field of seismic data denoising. This paper firstly introduces the classification and development process of seismic denoising. Then, starting from the principle of GAN, it introduces the workflow of the original GAN, the objective function in the training process, the existing problems of the original GAN and some mainstream solutions to these problems, and introduces the commonly used model of GAN in the field of earthquake denoising. besides, it summarizes and analyzes the current application and improvement innovation of GAN in the field of seismic denoising, and analyzes the application of GAN in the field of seismic denoising from two aspects of supervised learning and unsupervised learning with examples. Finally, the prospect of GAN for seismic denoising in the future is prospected.},
     year = {2022}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Research Status and Prospects of Generative Adversarial Networks in Seismic Data Denoising
    AU  - Yuan Xu
    AU  - Qing Wang
    AU  - Shiguang Guo
    Y1  - 2022/10/17
    PY  - 2022
    N1  - https://doi.org/10.11648/j.ijssn.20221002.13
    DO  - 10.11648/j.ijssn.20221002.13
    T2  - International Journal of Sensors and Sensor Networks
    JF  - International Journal of Sensors and Sensor Networks
    JO  - International Journal of Sensors and Sensor Networks
    SP  - 33
    EP  - 50
    PB  - Science Publishing Group
    SN  - 2329-1788
    UR  - https://doi.org/10.11648/j.ijssn.20221002.13
    AB  - Efficient and high-quality processing of seismic data collected by geophone sensors is the core of successful seismic exploration. Seismic denoising is a key step in seismic data processing. Traditional seismic denoising relies on manual empirical parameter selection and comparative analysis, which is time-consuming and limited by subjective errors. Deep learning methods based on convolutional neural networks (CNN) have improved the efficiency of denoising massive amounts of seismic data and reduced the manual errors of traditional methods. However, most CNN methods only consider data loss and are weak in recovering the structure of seismic signals, resulting in severe attenuation of some seismic traces in the recovered effective signals, reducing the continuity of seismic events and the quality of seismic data. Generative adversarial networks (GAN), a popular method for deep learning with unique adversarial ideas and powerful feature extraction capabilities, can overcome the limitations of CNN methods in the field of seismic data denoising. This paper firstly introduces the classification and development process of seismic denoising. Then, starting from the principle of GAN, it introduces the workflow of the original GAN, the objective function in the training process, the existing problems of the original GAN and some mainstream solutions to these problems, and introduces the commonly used model of GAN in the field of earthquake denoising. besides, it summarizes and analyzes the current application and improvement innovation of GAN in the field of seismic denoising, and analyzes the application of GAN in the field of seismic denoising from two aspects of supervised learning and unsupervised learning with examples. Finally, the prospect of GAN for seismic denoising in the future is prospected.
    VL  - 10
    IS  - 2
    ER  - 

    Copy | Download

Author Information
  • Key Laboratory of Information and Communication Systems, Ministry of Information Industry, Beijing Information Science and Technology University, Beijing, China

  • Key Laboratory of Information and Communication Systems, Ministry of Information Industry, Beijing Information Science and Technology University, Beijing, China

  • Key Laboratory of Information and Communication Systems, Ministry of Information Industry, Beijing Information Science and Technology University, Beijing, China

  • Sections