Geophysics and Geophysical Exploration (지구물리와물리탐사)

Korean Society of Earth and Exploration Geophysicists (한국지구물리·물리탐사학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of CNN and GAN-based Deep Learning Models for Ground Roll Suppression

그라운드-롤 제거를 위한 CNN과 GAN 기반 딥러닝 모델 비교 분석

  • Sangin Cho (Department of Energy Resources Engineering, Inha University) ;
  • Sukjoon Pyun (Department of Energy Resources Engineering, Inha University)
  • 조상인 (인하대학교 에너지자원공학과) ;
  • 편석준 (인하대학교 에너지자원공학과)
  • Received : 2023.02.27
  • Accepted : 2023.05.07
  • Published : 2023.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

The ground roll is the most common coherent noise in land seismic data and has an amplitude much larger than the reflection event we usually want to obtain. Therefore, ground roll suppression is a crucial step in seismic data processing. Several techniques, such as f-k filtering and curvelet transform, have been developed to suppress the ground roll. However, the existing methods still require improvements in suppression performance and efficiency. Various studies on the suppression of ground roll in seismic data have recently been conducted using deep learning methods developed for image processing. In this paper, we introduce three models (DnCNN (De-noiseCNN), pix2pix, and CycleGAN), based on convolutional neural network (CNN) or conditional generative adversarial network (cGAN), for ground roll suppression and explain them in detail through numerical examples. Common shot gathers from the same field were divided into training and test datasets to compare the algorithms. We trained the models using the training data and evaluated their performances using the test data. When training these models with field data, ground roll removed data are required; therefore, the ground roll is suppressed by f-k filtering and used as the ground-truth data. To evaluate the performance of the deep learning models and compare the training results, we utilized quantitative indicators such as the correlation coefficient and structural similarity index measure (SSIM) based on the similarity to the ground-truth data. The DnCNN model exhibited the best performance, and we confirmed that other models could also be applied to suppress the ground roll.

그라운드-롤(ground roll)은 육상 탄성파 탐사 자료에서 가장 흔하게 나타나는 일관성 잡음(coherent noise)이며 탐사를 통해 얻고자 하는 반사 이벤트 신호보다 훨씬 큰 진폭을 가지고 있다. 따라서 탄성파 자료 처리에서 그라운드-롤 제거는 매우 중요하고 필수적인 과정이다. 그라운드-롤 제거를 위해 주파수-파수 필터링, 커브릿(curvelet) 변환 등 여러 제거 기술이 개발되어 왔으나 제거 성능과 효율성을 개선하기 위한 방법에 대한 수요는 여전히 존재한다. 최근에는 영상처리 분야에서 개발된 딥러닝 기법들을 활용하여 탄성파 자료의 그라운드-롤을 제거하고자 하는 연구도 다양하게 수행되고 있다. 이 논문에서는 그라운드-롤 제거를 위해 CNN (convolutional neural network) 또는 cGAN (conditional generative adversarial network)을 기반으로 하는 세가지 모델(DnCNN (De-noiseCNN), pix2pix, CycleGAN)을 적용한 연구들을 소개하고 수치 예제를 통해 상세히 설명하였다. 알고리듬 비교를 위해 동일한 현장에서 취득한 송신원 모음을 훈련 자료와 테스트 자료로 나누어 모델을 학습하고, 모델 성능을 평가하였다. 이러한 딥러닝 모델은 현장자료를 사용하여 훈련할 때, 그라운드-롤이 제거된 자료가 필요하므로 주파수-파수 필터링으로 그라운드-롤을 제거하여 정답자료로 사용하였다. 딥러닝 모델의 성능 평가 및 훈련 결과 비교는 정답 자료와의 유사성을 기본으로 상관계수와 SSIM (structural similarity index measure)과 같은 정량적 지표를 활용하였다. 결과적으로 DnCNN 모델이 가장 좋은 성능을 보였으며, 다른 모델들도 그라운드-롤 제거에 활용될 수 있음을 확인하였다.

Keywords

Acknowledgement

논문은 2023년도 정부(산업통상자원부)의 재원으로 한국에너지기술평가원의 지원을 받아 수행된 연구임(20226A10100030, 고성능 해양 CO2 모니터링 기술개발).

References

  1. Benaim, S., and Wolf, L., 2017, One-sided unsupervised domain mapping, Advances in Neural Information Processing Systems, 30. https://doi.org/10.48550/arXiv.1706.00826
  2. Choi, W. C., Lee, G. H., Cho, S. I., Choi, B. H., and Pyun, S. J., 2020, Fault Detection for Seismic Data Interpretation Based on Machine Learning: Research Trends and Technological Introduction, Geophysics and Geophysical Exploration, 23(2), 97-114 (In Korean with English abstract). https://doi.org/10.7582/GGE.2020.23.2.097
  3. Deighan, A. J., and Watts, D. R., 1997, Ground-roll suppression using the wavelet transform, Geophysics, 62(6), 1896-1903. https://doi.org/10.1190/1.1444290
  4. Embree, P., Burg, J. P., and Backus, M. M., 1963, Wide-band velocity filtering; The Pie-Slice process, Geophysics, 28(6), 948-974. https://doi.org/10.1190/1.1439310
  5. Fomel, S., 2002, Applications of plane-wave destruction filters, Geophysics, 67(6), 1946-1960. https://doi.org/10.1190/1.1527095
  6. Fomel, S., 2009, Adaptive multiple subtraction using regularized nonstationary regression, Geophysics, 74(1), V25-V33. https://doi.org/10.1190/1.3043447
  7. Gallant, E. V., Stewart, R. R., Bertram, M. B., and Lawton, D. C., 1995, Acquisition of the Blackfoot broad-band seismic survey, CREWES, 7(36), 1-9. https://www.crewes.org/Documents/ResearchReports/1995/1995-36.pdf
  8. Geron, A., 2017, Hands-on machine learning with scikit-learn and tensorflow: Concepts, Tools, and Techniques to build intelligent systems, O'Reilly Media. https://www.oreilly.com/library/view/hands-on-machine-learning/9781491962282/
  9. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y., 2014, Generative adversarial networks, arXiv preprint arXiv:1406.2661. https://doi.org/10.48550/arXiv.1406.2661
  10. Guo, R., Maniar, H., Di, H., Moldoveanu, N., Abubakar, A., and Li, M., 2020, Ground roll attenuation with an unsupervised deep learning approach, In SEG Technical Program Expanded Abstracts 2020, Society of Exploration Geophysicists, 3164-3168. https://doi.org/10.1190/segam2020-3425792.1
  11. He, K., Zhang, X., Ren, S., and Sun, J., 2016, Deep residual learning for image recognition, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770-778. https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/He_Deep_Residual_Learning_CVPR_2016_paper.pdf
  12. Ioffe, S., and Szegedy, C., 2015, Batch normalization: Accelerating deep network training by reducing internal covariate shift, In International Conference on Machine Learning, 448-456. https://doi.org/10.48550/arXiv.1502.03167
  13. Isola, P., Zhu, J. Y., Zhou, T., and Efros, A. A., 2017, Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1125-1134. https://doi.org/10.48550/arXiv.1611.07004
  14. Jia, Z., Lu, W., Zhang, M., and Miao, Y., 2018, Separating ground-roll from land seismic record via convolutional neural network, In SEG 2018 Workshop: SEG Maximizing Asset Value Through Artificial Intelligence and Machine Learning, 60-63. DOI: 10.1190/AIML2018-16.1
  15. Jo, Y. J., Bae, K. M., and Park, J. Y., 2020, Research Trends of Generative Adversarial Networks and Image Generation and Translation, Electronics and Telecommunications Trends, 35(4), 91-102 (In Korean with English abstract). https://doi.org/10.22648/ETRI.2020.J.350409
  16. Kaji, S., and Kida, S., 2019, Overview of image-to-image translation by use of deep neural networks: denoising, super-resolution, modality conversion, and reconstruction in medical imaging, Radiological Physics and Technology, 12, 235-248. DOI: 10.1007/s12194-019-00520-y
  17. Kaur, H., Fomel, S., and Pham, N., 2019, Ground roll attenuation using generative adversarial network, 81st Annual International Conference and Exhibition, EAGE, Extended Abstracts, 1-5. DOI: https://doi.org/10.3997/2214-4609.201900762
  18. Li, H., Yang, W., and Yong, X., 2018, Deep learning for ground-roll noise attenuation, In SEG Technical Program Expanded Abstracts 2018, Society of Exploration Geophysicists, 1981-1985. https://doi.org/10.1190/segam2018-2981295.1
  19. Liu, X., 1999, Ground roll suppression using the Karhunen-Loeve transform, Geophysics, 64(2), 564-566. https://doi.org/10.1190/1.1444562
  20. Liu, Y., and Fomel, S., 2013, Seismic data analysis using local time-frequency decomposition, Geophysical Prospecting, 61(3), 516-525. DOI: 10.1111/j.1365-2478.2012.01062.x
  21. Mirza, M., and Osindero, S., 2014, Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784. https://doi.org/10.48550/arXiv.1411.1784
  22. Naghizadeh, M., and Sacchi, M., 2018, Ground-roll attenuation using curvelet downscaling, Geophysics, 83(3), V185-V195. https://doi.org/10.1190/geo2017-0562.1
  23. Pham, N., and Li, W., 2022, Physics-constrained deep learning for ground roll attenuation, Geophysics, 87(1), V15-V27. https://doi.org/10.1190/geo2020-0691.1
  24. Ronneberger, O., Fischer, P., and Brox, T., 2015, U-net: Convolutional networks for biomedical image segmentation, In Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015: 18th International Conference, Springer International Publishing, 234-241. https://link.springer.com/chapter/10.1007/978-3-319-24574-4_28
  25. Russell, B., Hampson, D., and Chun, J., 1990a, Noise elimination and the Radon transform, part 1, The Leading Edge, 9(10), 18-23. https://doi.org/10.1190/1.1439677
  26. Russell, B., Hampson, D., and Chun, J., 1990b, Noise elimination and the Radon transform, part 2, The Leading Edge, 9(11), 31-37. https://doi.org/10.1190/1.1439700
  27. Serdyukov, A. S., 2022, Ground-roll extraction using the Karhunen-Loeve transform in the time-frequency domain, Geophysics, 87(2), A19-A24. https://doi.org/10.1190/geo2021-0453.1
  28. Taigman, Y., Polyak, A., and Wolf, L., 2016, Unsupervised cross-domain image generation, arXiv preprint arXiv:1611.02200. https://doi.org/10.48550/arXiv.1611.02200
  29. Treitel, S., Shanks, J. L., and Frasier, C. W., 1967, Some aspects of fan filtering, Geophysics, 32(5), 789-800. https://doi.org/10.1190/1.1439889
  30. Yarham, C., Boeniger, U., and Herrmann, F., 2006, Curvelet-based ground roll removal, 76th Annual International Meeting, SEG, Expanded Abstracts, 2777-2782. DOI: 10.1190/1.2370101
  31. Yilmaz, O., 2001, Seismic data analysis: Processing, inversion, and interpretation of seismic data, Society of exploration geophysicists. https://doi.org/10.1190/1.9781560801580
  32. Yuan, Y., Si, X., and Zheng, Y., 2020, Ground-roll attenuation using generative adversarial networks, Geophysics, 85(4), WA255-WA267. https://doi.org/10.1190/geo2019-0414.1
  33. Yuan, Y., Zhou, Z., Niu, B., Wang, H., and Xiang, A., 2005, A method for improving the signal to noise ratio in seismic data, Oil Geophysical Prospecting, 40, 168-171.
  34. Zhang, K., Zuo, W., Chen, Y., Meng, D., and Zhang, L., 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
  35. Zhu, J. Y., Park, T., Isola, P., and Efros, A. A., 2017, Unpaired image-to-image translation using cycle-consistent adversarial networks, In Proceedings of the IEEE International Conference on Computer Vision, 2223-2232. https://doi.org/10.48550/arXiv.1703.10593