All Issue

2025 Vol.28, Issue 2 Preview Page

Research Article

Supervised Learning-Based Seismic Inversion in the Laplace Domain Integrated with Source Estimation

송신원 추정을 통합한 라플라스 영역 지도학습 기반 탄성파 역산목

Jun Hyeon Jo1
,
Wansoo Ha1*
조 준현1
,
하 완수1*
1Department of Energy Resources Engineering, Pukyong National University
1부경대학교 에너지자원공학과
*Corresponding Author
31 May 2025. pp. 64-78
PDF XML Citation
Abstract

Deep learning-based seismic inversion techniques have been proposed to overcome the limitations of conventional full-waveform inversion (FWI) and have demonstrated excellent inversion performance on synthetic datasets. However, studies addressing the source estimation challenges in deep learning-based seismic inversion remain scarce. Accurate estimation of the source wavelet is particularly critical for the successful inversion of field seismic data. If the wavelet used deviates significantly from the true source wavelet, the inversion may fail to converge to the true velocity model or incur increased computational costs. Therefore, this study investigates how uncertainties in source wavelet estimation affect the accuracy of the deep learning-based seismic inversion, with a particular focus on supervised learning approaches, which are inherently sensitive to training data. To address this issue, source estimation in the Laplace domain is incorporated as a preprocessing step. Unlike the time or frequency domains, the source wavelet in the Laplace domain retains only the amplitude information, simplifying the estimation process and improving computational efficiency. The source estimation algorithm adopts the same approach as used in conventional Laplace-domain FWI, iteratively updating the source parameters using the Newton method. Furthermore, we propose a novel deep neural network for deep-learning-based seismic inversion in the Laplace-domain and conduct a quantitative comparison with the previous study. The proposed network achieved improved efficiency and superior inversion performance in the Laplace-domain inversion. Numerical examples using two benchmark models demonstrate the importance of source wavelet estimation and the effectiveness of the proposed approach.

Keywords
Deep-Learning
Seismic Inversion
Laplace-Domain
Source Estimation

전통적인 전파형 역산의 한계를 극복하기 위해 제안된 딥러닝 기반 탄성파 역산 기술들은 인공 합성 자료들을 대상으로 우수한 역산 성능을 입증하였다. 그러나, 송신원 추정 문제를 다루는 딥러닝 기반 탄성파 역산 연구는 거의 찾아볼 수 없다. 특히 현장 탐사 자료의 성공적인 역산을 위해서는 정확한 송신 파형을 사용하는 것이 중요하다. 역산에 사용되는 송신원의 파형이 실제 송신원과 크게 다를 경우 역산 결과가 참 속도 모델에 수렴하지 못하거나 계산 비용이 증가할 수 있다. 따라서, 본 연구는 딥러닝 기반 탄성파 역산에서 송신 파형의 불확실성이 역산 정확도에 미치는 영향을 분석하기 위해 수행하였다. 특히, 훈련 자료에 민감한 지도학습 기반 탄성파 역산에 초점을 맞추었으며, 라플라스 영역에서의 송신원 추정을 전처리 단계로 통합하였다. 시간이나 주파수 영역과 달리 라플라스 영역에서 송신 파형은 진폭 정보만을 가지므로 송신원 추정이 보다 단순하고 계산 효율적이다. 송신원 추정 알고리즘은 기존 라플라스 영역 완전 파형 역산에서 사용하는 것과 동일한 알고리즘을 채택하였으며, 뉴턴법을 이용하여 송신원 매개변수를 반복적으로 갱신하였다. 또한, 라플라스 영역 딥러닝 기반 탄성파 역산을 위한 새로운 심층 신경망을 제안하여 선행 연구와 정량적으로 비교 분석하였다. 제안된 신경망을 이용해 라플라스 영역 역산에서 더욱 효율적이고 우수한 역산 성능을 달성하였으며, 두 가지 벤치마크 모델을 이용한 수치 예제를 통해 송신원 추정의 중요성과 제안된 접근법의 효과성을 입증하였다.

키워드
딥러닝
탄성파 역산
라플라스 영역
송신원 추정
References
1

Adler, A., Araya-Polo, M., and Poggio, T., 2021, Deep learning for seismic inverse problems: Toward the acceleration of geophysical analysis workflows, IEEE Signal Processing Magazine, 38(2), 89-119.

10.1109/MSP.2020.3037429
2

Allen-Zhu, Z., Li, Y., and Liang, Y., 2019a, Learning and generalization in overparameterized neural networks, going beyond two layers, Proc. Adv. Neural Inf. Process. Syst., 32, 6155-6166.

3

Allen-Zhu, Z., Li, Y., and Song, Z., 2019b, A convergence theory for deep learning via over-parameterization, Proc. 36th Int. Conf. Mach. Learn. (ICML), 97, 242-252.

4

Araya-Polo, M., Ferris, S., and Florez, M., 2019, Deep learning-driven velocity model building workflow, Lead. edge., 38(11), 872a1-872a9.

10.1190/tle38110872a1.1
5

Cao, Y. and Gu, Q., 2020, Generalization error bounds of gradient descent for learning over-parameterized deep ReLU networks, Proc. AAAI Conf. Artif. Intell., 34(4), 3349-3356.

10.1609/aaai.v34i04.5736
6

Cohen, G., 2002, High-order Numerical Methods for Transient Wave Equations, Springer, 307-320.

10.1007/978-3-662-04823-8_14
7

Dhara, A. and Sen, M. K., 2022, Physics-guided deep autoencoder to overcome the need for a starting model in full-waveform inversion, Lead. Edge., 41(6), 375-381.

10.1190/tle41060375.1
8

Farris, S., Clapp, R., and Araya-Polo, M., 2023, Learning-based seismic velocity inversion with synthetic and field data, Sensors, 23(19), 8277.

10.3390/s2319827737837108PMC10574958
9

Gauthier, O., Virieux, J., and Tarantola, A., 1986, Two-dimensional nonlinear inversion of seismic waveforms: Numerical results, Geophysics, 51(7), 1387-1403.

10.1190/1.1442188
10

Goodfellow, I., Bengio, Y., and Courville, A., 2016, Deep Learning, MIT Press.

11

Ha, W., Chung, W., Park, E., and Shin, C., 2012, 2-D acoustic Laplace-domain waveform inversion of marine field data, Geophys. J. Int., 190(1), 421-428.

10.1111/j.1365-246X.2012.05487.x
12

Hanin, B. and Rolnick, D., 2018, How to start training: The effect of initialization and architecture, arXiv preprint, arXiv:1803.01719.

13

He, Q. and Wang, Y., 2021, Reparameterized full-waveform inversion using deep neural networks, Geophysics, 86(1), V1-V13.

10.1190/geo2019-0382.1
14

Hendrycks, D. and Gimpel, K., 2016, Gaussian error linear units (gelus), arXiv preprint, arXiv:1606.08415.

15

Hornik, K., Stinchcombe, M., and White, H., 1989, Multilayer feedforward networks are universal approximators, Neural Netw., 2(5), 359-366.

10.1016/0893-6080(89)90020-8
16

Jo, J. H. and Ha, W., 2022, Seismic velocity modeling building using depthwise separable convolutional neural network, J. Korea Soc. Mineral Mining Eng., 59(2), 148-160 (in Korean with English abstract).

10.32390/ksmer.2022.59.2.148
17

Jo, J. H. and Ha, W., 2023a, Deep-learning seismic inversion using laplace-domain wavefields, Geophys. and Geophys. Explor., 26(2), 84-93 (in Korean with English abstract).

18

Jo, J. H. and Ha, W., 2023b, Seismic traveltime tomography using deep learning, IEEE Trans. Geosci. Remote Sens., 61, 5923111.

10.1109/TGRS.2023.3334283
19

Jo, J. H. and Ha, W., 2024, Seismic traveltime tomography using transfer learning, IEEE Trans. Geosci. Remote Sens., 62, 5931214.

10.1109/TGRS.2024.3476682
20

Kaiser, L., Gomez, A. N., and Chollet, F., 2017, Depthwise separable convolutions for neural machine translation, arXiv preprint, arXiv:1706.03059v2.

21

Li, S., Liu, B., Ren, Y., Chen, Y., Yang, S., Wang, Y., and Jiang, P., 2020, Deep-learning inversion of seismic data, IEEE Trans. Geosci. Remote Sens., 58(3), 2135-2149.

10.1109/TGRS.2019.2953473
22

Liu, B., Jiang, P., Wang, Q., Ren, Y., Yang, S., and Cohn, A. G., 2023, Physics-driven self-supervised learning system for seismic velocity inversion, Geophysics, 88(2), R145-R161.

10.1190/geo2021-0302.1
23

Liu, Z. and Bleistein, N., 1995, Migration velocity analysis: Theory and an iterative algorithm, Geophysics, 60(1), 142-153.

10.1190/1.1443741
24

Loffe, 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, 37, 448-456.

25

Nesterov, Y., 1983, A method for unconstrained convex minimization problem with the rate of convergence o(1/k2), Doklady USSR, 269(3), 543-547.

26

Odena, A., Dumoulin, V., and Olah, C., 2016, Deconvolution and checkerboard artifacts, Distill, 1(10), e3.

10.23915/distill.00003
27

Park, Y. and Pyun, S., 2015, Source wavelet estimation using common mid-point gathers, Geosystem Engineering, 18(4), 199-204.

10.1080/12269328.2015.1046562
28

Pratt, R. G. and Worthington, M. H., 1990, Inverse theory applied to multi-source cross-hole tomography. Part 1: acoustic wave-equation method, Geophys. Prospect., 38(3), 287-310.

10.1111/j.1365-2478.1990.tb01846.x
29

Pyun, S. and Park, Y., 2022, A review of seismic full waveform inversion based on deep learning, Geophys. and Geophys. Explor., 25(4), 227-241 (in Korean with English abstract).

30

Ronneberger, O., Fischer, P., and Brox, T., 2015, U-net: Convolutional networks for biomedical image segmentation, International Conference on Medical Image Computing and Computer-Assisted Intervention, 234-241.

10.1007/978-3-319-24574-4_28
31

Sashank, J. R., Satyen, K., and Sanjiv, K., 2019, On the convergence of adam and beyond, arXiv preprint, arXiv:1904.09237.

32

Shin, C. and Cha, Y., 2008, Waveform inversion in the Laplace domain, Geophys. J. Int., 173(3), 922-931.

10.1111/j.1365-246X.2008.03768.x
33

Shin, C., Pyun, S., and Bednar, J., 2007, Comparison of waveform inversion, Part 1: Conventional wavefield vs logarithmic wavefield, Geophysical Prospecting, 55(4), 449-464.

10.1111/j.1365-2478.2007.00617.x
34

Stoughton, D., Stefani, J., and Michell, S., 2001, 2D elastic model for wavefield investigations of subsalt objectives, deep water Gulf of Mexico, SEG Technical Program, Expanded Abstracts, 1269-1272.

10.1190/1.1816325
35

Tarantola, A., 1984, Inversion of seismic reflection data in the acoustic approximation, Geophysics, 49(8), 1259-1266.

10.1190/1.1441754
36

Versteeg, R., 1994, The marmousi experience: Velocity model determination on a synthetic complex data set, Lead. Edge, 13(9), 927-936.

10.1190/1.1437051
37

Virieux, J. and Operto, S., 2009, An overview of full-waveform inversion in exploration geophysics, Geophysics, 74(6), WCC1-WCC26.

10.1190/1.3238367
38

Wang, W. and Ma, J., 2020, Velocity model building in a crosswell acquisition geometry with image-trained artificial neural networks, Geophysics, 85(2), U31-U46.

10.1190/geo2018-0591.1
39

Wang, Z., Bovik, A. C., Sheikh, H. R., and Simoncelli, E. P., 2004, Image quality assessment: from error visibility to structural similarity, IEEE Transactions on Image Processing, 13(4), 600-612.

10.1109/TIP.2003.81986115376593
40

Wu, B., Meng, D., Wang, L., Liu, N., and Wang, Y., 2020, Seismic impedance inversion using fully convolutional residual network and transfer learning, IEEE Geoscience and Remote Sensing Letters, 17(12), 2140-2144.

10.1109/LGRS.2019.2963106
41

Wu, Y. and McMechan, G. A., 2019, Parametric convolutional neural network-domain full-waveform inversion, Geophysics, 84(6), R881-R896.

10.1190/geo2018-0224.1
42

Yang, F. and Ma, J., 2019, Deep-learning inversion: A next-generation seismic velocity model building method, Geophysics, 84(4), R583-R599.

10.1190/geo2018-0249.1
43

Yilmaz, Ö., 2001, Seismic Data Analysis: Processing, Inversion and Interpretation of Seismic Data, Vol. I, Soc. Expl. Geophys., 288-323.

10.1190/1.9781560801580
44

Zhang, J., ten Brink, U. S., and Toksöz, M. N., 1998, Nonlinear refraction and reflection travel time tomography, J. Geophys. Res. Solid Earth, 103(B12), 29743-29757.

10.1029/98JB01981
45

Zhang, Z. and Lin, Y., 2020, Data-driven seismic waveform inversion: A study on the robustness and generalization, IEEE Trans. Geosci. Remote Sens., 58(10), 6900-6913.

10.1109/TGRS.2020.2977635
46

Zhu, W., Xu, K., Darve, E., Biondi, B., and Beroza, G. C., 2022, Integrating deep neural networks with full-waveform inversion: Reparameterization, regularization, and uncertainty quantification, Geophysics, 87(1), R93-R109.

10.1190/geo2020-0933.1
Information
  • Publisher :Korean Society of Earth and Exploration Geophysicists
  • Publisher(Ko) :한국지구물리물리탐사학회
  • Journal Title :Geophysics and Geophysical Exploration
  • Journal Title(Ko) :지구물리와 물리탐사
  • Volume : 28
  • No :2
  • Pages :64-78
  • Received Date : 2025-04-25
  • Revised Date : 2025-05-09
  • Accepted Date : 2025-05-19
  • DOI :https://doi.org/10.7582/GGE.2025.28.2.064
Journal Informaiton Geophysics and Geophysical Exploration Geophysics and Geophysical Exploration
Online Submission
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • KCI 문헌 유사도 검사
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close