Machine-learning Based Noise Attenuation of Field Seismic Data using Noise Data Acquisition

Hyunggu Jun; Chung-Ho Kim; Han-Joon Kim

doi:10.32390/ksmer.2021.58.5.408

All Issue

2021 Vol.58, Issue 5S Preview Page Next Page

Research Paper (Special Issue)

Machine-learning Based Noise Attenuation of Field Seismic Data using Noise Data Acquisition 잡음 자료 취득을 통한 기계학습 기반 현장 탄성파 자료 잡음 억제 연구

31 October 2021. pp. 408-417

PDF XML

Abstract

The seismic data acquired by field seismic exploration contains not only seismic reflections but also various types of noise. Therefore, noise attenuation in seismic data is one of the most important tasks in seismic data processing. In this study, we attempted to attenuate random noise in field seismic data by using machine learning. The proper training data is an important part of machine learning, and the characteristics of training data should be similar to the characteristics of target data. To construct the proper training data for noise attenuation, we acquired noise data from field seismic exploration without a seismic source. The training data was generated by combining the acquired noise data and synthetic seismic data. The trained model was applied to the synthetic and field seismic data, and the results were used to verify the effectiveness of the proposed method.

Keywords

Seismic noise

Machine learning

Seismic data processing

현장 탄성파 탐사를 통해 자료를 취득하는 과정에서 탄성파 자료에는 반사 신호뿐만 아니라 다양한 형태의 잡음이 함께 기록되므로 탄성파 자료처리를 통해 잡음을 억제하는 것은 매우 중요한 부분이다. 본 연구에서는 기계학습 적용을 통해 탄성파 자료의 무작위 잡음을 억제하고자 하였다. 탄성파 자료에 기록된 무작위 잡음은 탄성파 탐사환경에 따라 서로 다른 특성을 가지게 된다. 기계학습을 통한 잡음 억제를 위해서는 적절한 학습자료의 구축이 필수적이며 잡음 제거 대상 탄성파 자료와 학습자료에 포함된 잡음의 특성이 서로 유사해야 한다. 본 연구에서는 학습자료 구축을 위해서 현장 탄성파 탐사 중 송신원 발파 없이 잡음자료를 직접 취득하였다. 취득한 잡음자료와 인공 합성 탄성파 자료를 조합하여 학습자료를 구축하였으며 기계학습 모델을 학습시켰다. 학습된 모델을 인공 합성 자료 및 현장자료에 적용하여 무작위 잡음을 제거하였으며, 제안된 방법을 통해 구축된 학습자료가 유효함을 검증하였다.

키워드

탄성파 잡음

기계학습

탄성파 자료처리

References

Alwon, S., 2018. Generative adversarial networks in seismic data processing, SEG Technical Program Expanded Abstracts 2018, Society of Exploration Geophysicists, p.1991-1995. 10.1190/segam2018-2996002.1

Dondurur, D., 2018. Acquisition and processing of marine seismic data, Elsevier, Amsterdam, Netherlands, 5p. 10.1016/B978-0-12-811490-2.00002-5

Ebadi, M.R., 2017. Coherent and incoherent seismic noise attenuation using parabolic radon transform and its application in environmental geophysics, Modeling Earth Systems and Environment, 3(1), 18p. 10.1007/s40808-017-0273-4

He, K., Zhang, X., Ren, S., and Sun, J., 2016. Deep residual learning for image recognition, Proceedings of the IEEE conference on computer vision and pattern recognition, p.770-778. 10.1109/CVPR.2016.9026180094

Hlebnikov, V., Elboth, T., Vinje, V., and Gelius, L.J., 2021. Noise types and their attenuation in towed marine seismic: A tutorial, Geophysics, 86(2), W1-W19. 10.1190/geo2019-0808.1

Hore, A. and Ziou, D., 2010. Image quality metrics: PSNR vs. SSIM, Proceedings of the 20th International Conference on Pattern Recognition, Istanbul, Turkey, p.2366-2369. 10.1109/ICPR.2010.579

Ioffe, S. and Szegedy, C., 2015. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, Proceedings of the 32nd International Conference on Machine Learning, PMLR 37, p.448-456.

Jun, H., Jou, H.T., Kim, C.H., Lee, S.H., and Kim, H.J., 2020. Random noise attenuation of sparker seismic oceanography data with machine learning, Ocean Sci., 16, p.1367-1383. 10.5194/os-16-1367-2020

Kaur, H., Fomel, S., and Pham, N., 2019. Ground roll attenuation using generative adversarial network, 81st EAGE Conference and Exhibition, EAGE, Extended Abstracts, p.1-5. 10.3997/2214-4609.201900762

Keras API reference, 2021.07.26, https://keras.io/ko/models/sequential.

Kim, Y., Hardisty, R., and Marfurt, K., 2019. Seismic random noise attenuation in f-x domain using complex-valued residual convolutional neural network, SEG Technical Program Expanded Abstracts 2019, Society of Exploration Geophysicists. p.2579-2583. 10.1190/segam2019-3216543.1

Kragh, E. and Christie, P., 2002. Seismic repeatability, normalized rms, and predictability, The Leading Edge, 21, p.640-647. 10.1190/1.1497316

Krizhevsky, A., Sutskever, I., and Hinton, G.E., 2012. Imagenet classification with deep convolutional neural networks, Advances in neural information processing systems, p.1097- 1105.

Li, H., Yang, W., and Yong, X., 2018. Deep learning for ground- roll noise attenuation, SEG Technical Program Expanded Abstracts 2018, Society of Exploration Geophysicists, p.1981-1985. 10.1190/segam2018-2981295.1

Liu, D., Wang, W., Chen, W., Wang, X., Zhou, Y., and Shi, Z., 2018. Random noise suppression in seismic data: What can deep learning do?, SEG Technical Program Expanded Abstracts 2018, Society of Exploration Geophysicists, p.2016-2020. 10.1190/segam2018-2998114.1

Nam, H., Lim, B., Kweon, I., and Kim, J., 2020. Removal of Seabed Multiples in Seismic Reflection Data using Machine Learning, Geophysics and Geophysical Exploration, 23(3), p.168-177.

Nasser, M., Ronen, S., and Stammeijer, J., 2016. Introduction to this special section: 4D seismic, The Leading Edge, 35, p.828-830. 10.1190/tle35100828.1

Saad, O. and Chen, Y., 2020. Deep denoising autoencoder for seismic random noise attenuation, Geophysics, 85(4), V367-V376. 10.1190/geo2019-0468.1

Si, X. and Yuan, Y., 2018. Random noise attenuation based on residual learning of deep convolutional neural network, SEG Technical Program Expanded Abstracts 2018, Society of Exploration Geophysicists. p.1986-1990. 10.1190/segam2018-2985176.1

Si, X., 2020. Ground roll attenuation with conditional generative adversarial networks, SEG Technical Program Expanded Abstracts 2020, Society of Exploration Geophysicists. p.1511-1515. 10.1190/segam2020-3424945.1

Waage, M., Bünz, S., Landrø, M., Plaza-Faverola, A., and Waghorn, K.A., 2019. Repeatability of high-resolution 3D seismic data, Geophysics, 84(1), B75-B94. 10.1190/geo2018-0099.1

Yilmaz, Ö., 2001. Seismic data analysis: processing, inversion and interpretation of seismic data (2nd Ed.), Vol. I, Society of Exploration Geophysicists, Tulsa, Oklahoma, 2027p. 10.1190/1.9781560801580

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 Trans. Image Process, 26, p.3142-3155. 10.1109/TIP.2017.266220628166495

Zhao, X., Lu, P., Zhang, Y., Chen, J., and Li, X., 2019. Swell-noise attenuation: A deep learning approach, The Leading Edge, 38(12), p.934-942. 10.1190/tle38120934.1

Zheng, Y., Yuan, Y., and Si, X., 2020. The improved DnCNN linear noise attenuation, SEG 2019 Workshop: Mathematical Geophysics: Traditional vs Learning, Beijing, China, Global Meeting Abstracts, p.56-59. 10.1190/iwmg2019_14.1

Information

Publisher :The Korean Society of Mineral and Energy Resources Engineers
Publisher(Ko) :한국자원공학회
Journal Title :Journal of the Korean Society of Mineral and Energy Resources Engineers
Journal Title(Ko) :한국자원공학회지
Volume : 58
No :5
Pages :408-417
Received Date : 2021-08-09
Revised Date : 2021-09-14
Accepted Date : 2021-10-26
DOI :https://doi.org/10.32390/ksmer.2021.58.5.408

Journal of the Korean Society of Mineral and Energy Resources Engineers ISSN:2288-0291(Print) 2288-2790(Online) 한국자원공학회지

All Issue