Articles | Volume 15, issue 17
https://doi.org/10.5194/gmd-15-6841-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-15-6841-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Development and technical paper
| 
08 Sep 2022
Development and technical paper |
| 08 Sep 2022
| 08 Sep 2022
DeepISMNet: three-dimensional implicit structural modeling with convolutional neural network
Zhengfa Bi
School of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui, P. R. China
School of Earth and Space Sciences, University of Science and Technology of China, Hefei, Anhui, P. R. China
Zhaoliang Li
China Aero Geophysical Survey and Remote Sensing Center for Natural Resources, Beijing, P. R. China
Dekuan Chang
Research Institute of Petroleum Exploration and Development–Northwest (NWGI), PetroChina, Gansu, Lanzhou, P. R. China
Xueshan Yong
Research Institute of Petroleum Exploration and Development–Northwest (NWGI), PetroChina, Gansu, Lanzhou, P. R. China
Related authors
Hui Gao, Xinming Wu, Jinyu Zhang, Xiaoming Sun, and Zhengfa Bi
Geosci. Model Dev., 16, 2495–2513, https://doi.org/10.5194/gmd-16-2495-2023, https://doi.org/10.5194/gmd-16-2495-2023, 2023
Short summary
Short summary
We propose a workflow to automatically generate synthetic seismic data and corresponding stratigraphic labels (e.g., clinoform facies, relative geologic time, and synchronous horizons) by geological and geophysical forward modeling. Trained with only synthetic datasets, our network works well to accurately and efficiently predict clinoform facies in 2D and 3D field seismic data. Such a workflow can be easily extended for other geological and geophysical scenarios in the future.
Zhixiang Guo, Xinming Wu, Yimin Dou, Hui Gao, and Guillaume Caumon
EGUsphere, https://doi.org/10.5194/egusphere-2026-1087, https://doi.org/10.5194/egusphere-2026-1087, 2026
This preprint is open for discussion and under review for Geoscientific Model Development (GMD).
Short summary
Short summary
We present a fast way to generate subsurface structure models from seismic surveys while honoring known horizons and faults. Instead of compressing the data into a hidden representation, our method works directly with the original model values and applies geological constraints during generation. Tests on synthetic and real surveys show more realistic structures and efficient prediction, producing a 512 by 512 model in 1.56 seconds on an NVIDIA H20 graphics processing unit.
Guangyu Wang, Xinming Wu, and Wen Zhang
Earth Syst. Sci. Data, 17, 3447–3471, https://doi.org/10.5194/essd-17-3447-2025, https://doi.org/10.5194/essd-17-3447-2025, 2025
Short summary
Short summary
Seismic paleochannel interpretation is vital for georesource exploration and paleoclimate research yet remains time-consuming. While deep learning offers automation potential, it is limited by the lack of labeled data. We present a workflow to simulate geologically reasonable 3D seismic volumes with diverse paleochannels, generating a large-scale labeled dataset. Field applications demonstrate its effectiveness. The dataset and codes are publicly available to support future research.
Hui Gao, Xinming Wu, Xiaoming Sun, Mingcai Hou, Hang Gao, Guangyu Wang, and Hanlin Sheng
Earth Syst. Sci. Data, 17, 595–609, https://doi.org/10.5194/essd-17-595-2025, https://doi.org/10.5194/essd-17-595-2025, 2025
Short summary
Short summary
We propose three strategies for field seismic data curation, knowledge-guided synthesization, and generative adversarial network (GAN)-based generation to construct a massive-scale, feature-rich, and high-realism benchmark dataset of seismic facies and evaluate its effectiveness in training a deep-learning model for automatic seismic facies classification.
Hai Yang, Shengqing Xiong, Qiankun Liu, Fang Li, Zhiye Jia, Xue Yang, Haofei Yan, and Zhaoliang Li
Solid Earth, 14, 1289–1308, https://doi.org/10.5194/se-14-1289-2023, https://doi.org/10.5194/se-14-1289-2023, 2023
Short summary
Short summary
The Wenchuan (Ms 8.0) and Lushan (Ms 7.0) earthquakes show different geodynamic features and form a 40–60 km area void of aftershocks for both earthquakes. The inverse models suggest that the downward-subducted basement of the Sichuan Basin is irregular in shape and heterogeneous in magnetism and density. The different focal mechanisms of the two earthquakes and the genesis of the seismic gap may be closely related to the differential thrusting mechanism caused by basement heterogeneity.
Hui Gao, Xinming Wu, Jinyu Zhang, Xiaoming Sun, and Zhengfa Bi
Geosci. Model Dev., 16, 2495–2513, https://doi.org/10.5194/gmd-16-2495-2023, https://doi.org/10.5194/gmd-16-2495-2023, 2023
Short summary
Short summary
We propose a workflow to automatically generate synthetic seismic data and corresponding stratigraphic labels (e.g., clinoform facies, relative geologic time, and synchronous horizons) by geological and geophysical forward modeling. Trained with only synthetic datasets, our network works well to accurately and efficiently predict clinoform facies in 2D and 3D field seismic data. Such a workflow can be easily extended for other geological and geophysical scenarios in the future.
Cited articles
Alon, U. and Yahav, E.: On the bottleneck of graph neural networks and its
practical implications, arXiv [preprint], https://doi.org/10.48550/arXiv.2006.05205, 9 June 2020. a
Bi, Z., Wu, X., Geng, Z., and Li, H.: Deep relative geologic time: a deep
learning method for simultaneously interpreting 3-D seismic horizons and
faults, J. Geophys. Res.-Sol. Ea., 126, e2021JB021882, https://doi.org/10.1029/2021JB021882,
2021. a
Bi, Z., Wu, X., Li, Z., Chang, D., and Yong, X.: Training and validation datasets for “Three-Dimensional Implicit Structural Modeling Using Convolutional Neural Network”, Zenodo [data set], https://doi.org/10.5281/zenodo.6480165, 2022a. a
Bi, Z., Wu, X., Li, Z., Chang, D., and Yong, X.: zfbi/DeepISMNet: DeepISMNet: Three-Dimensional Implicit Structural Modeling with Convolutional Neural Network, Zenodo [code], https://doi.org/10.5281/zenodo.6684269, 2022b. a
Carr, J. C., Beatson, R. K., Cherrie, J. B., Mitchell, T. J., Fright, W. R.,
McCallum, B. C., and Evans, T. R.: Reconstruction and representation of 3D
objects with radial basis functions, in: Proceedings of the 28th annual
conference on Computer graphics and interactive techniques, Los Angeles, California, United States of America, 12–17 August 2001, 67–76, https://doi.org/10.1145/383259.383266, 2001. a
Caumon, G., Collon-Drouaillet, P., Veslud, C. L. C. D., Viseur, S., and Sausse,
J.: Surface-based 3D modeling of geological structures, Math.
Geosci., 41, 927–945, https://doi.org/10.1007/s11004-009-9244-2, 2009. a, b
Caumon, G., Gray, G., Antoine, C., and Titeux, M.-O.: Three-dimensional
implicit stratigraphic model building from remote sensing data on tetrahedral
meshes: theory and application to a regional model of La Popa Basin, NE
Mexico, IEEE T. Geosci. Remote, 51, 1613–1621,
2012. a
Chaodong, F., Peng, Y., and Bo, X.: Rapid geological modeling by using implicit
3D potential field interpolation method, in: 2010 International Conference On
Computer Design and Applications, vol. 5, Qinhuangdao, Hebei, China, 25–27 June 2010, V5–50, IEEE, https://doi.org/10.1109/ICCDA.2010.5540850, 2010. a
Chen, Y., Jiang, H., Li, C., Jia, X., and Ghamisi, P.: Deep feature extraction
and classification of hyperspectral images based on convolutional neural
networks, IEEE T. Geosci. Remote, 54, 6232–6251,
2016. a
Chiles, J.-P., Aug, C., Guillen, A., and Lees, T.: Modelling the geometry of
geological units and its uncertainty in 3D from structural data: the
potential-field method, in: Proceedings of international symposium on orebody
modelling and strategic mine planning, Perth, Australia, vol. 22, p. 24,
Citeseer, https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.583.213&rep=rep1&type=pdf (last access: 29 August 2022), 2004. a
Chollet, F.: Xception: Deep learning with depthwise separable convolutions, in:
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, United States of America, 21–26 July 2017, 1251–1258, https://doi.org/10.48550/arXiv.1610.02357, 2017. a
Collon, P., Steckiewicz-Laurent, W., Pellerin, J., Laurent, G., Caumon, G.,
Reichart, G., and Vaute, L.: 3D geomodelling combining implicit surfaces and
Voronoi-based remeshing: A case study in the Lorraine Coal Basin (France),
Comput. Geosci., 77, 29–43, 2015. a
Cook, R. L.: Stochastic sampling in computer graphics, ACM T.
Graphic, 5, 51–72, 1986. a
de Kemp, E., Jessell, M., Aillères, L., Schetselaar, E., Hillier, M.,
Lindsay, M., and Brodaric, B.: Earth model construction in challenging
geologic terrain: Designing workflows and algorithms that makes sense, in:
Proceedings of the Sixth Decennial International Conference on Mineral Exploration, vol. 17, Toronto, Canada, 22 to 25 October 2017, 419–439, 2017. a
Donmez, P.: Introduction to Machine Learning, by Ethem Alpaydin, 2010. a
Fornberg, B.: Generation of finite difference formulas on arbitrarily spaced
grids, Math. Comput., 51, 699–706, 1988. a
Fossen, H.: Structural geology, Cambridge University Press, ISBN 978-1-107-05764-7, 2016. a
Grose, L., Laurent, G., Aillères, L., Armit, R., Jessell, M., and
Cousin-Dechenaud, T.: Inversion of Structural Geology Data for Fold Geometry,
J. Geophys. Res.-Sol. Ea., 123, 6318–6333,
https://doi.org/10.1029/2017JB015177, 2018. a
Grose, L., Ailleres, L., Laurent, G., Caumon, G., Jessell, M., and Armit, R.: Modelling of faults in LoopStructural 1.0, Geosci. Model Dev., 14, 6197–6213, https://doi.org/10.5194/gmd-14-6197-2021, 2021. a
Hennenfent, G. and Herrmann, F. J.: Simply denoise: Wavefield reconstruction
via jittered undersampling, Geophysics, 73, V19–V28, 2008. a
Hillier, M. J., Schetselaar, E. M., de Kemp, E. A., and Perron, G.:
Three-dimensional modelling of geological surfaces using generalized
interpolation with radial basis functions, Math. Geosci., 46,
931–953, 2014. a
Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W.,
Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., and Adam, H.: Searching for mobilenetv3, in:
Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, South Korea, 27 October–2 November 2019,
1314–1324, https://openaccess.thecvf.com/content_ICCV_2019/papers/Howard_Searching_for_MobileNetV3_ICCV_2019_paper.pdf (last access: 29 August 2022), 2019. a
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T.,
Andreetto, M., and Adam, H.: Mobilenets: Efficient convolutional neural
networks for mobile vision applications, arXiv [preprint], https://doi.org/10.48550/arXiv.1704.04861, 17 April
2017. a, b
Hu, J., Shen, L., and Sun, G.: Squeeze-and-excitation networks, in: Proceedings
of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 18–23 June 2018,
7132–7141, https://doi.org/10.48550/arXiv.1709.01507, 2018. a
Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., and
Keutzer, K.: SqueezeNet: AlexNet-level accuracy with 50x fewer parameters
and <0.5 MB model size, arXiv [preprint], https://doi.org/10.48550/arXiv.1602.07360, 24 February 2016. a
Jessell, M.: Three-dimensional geological modelling of potential-field data,
Comput. Geosci., 27, 455–465, 2001. a
Jessell, M., Guo, J., Li, Y., Lindsay, M., Scalzo, R., Giraud, J., Pirot, G., Cripps, E., and Ogarko, V.: Into the Noddyverse: a massive data store of 3D geological models for machine learning and inversion applications, Earth Syst. Sci. Data, 14, 381–392, https://doi.org/10.5194/essd-14-381-2022, 2022. a, b
Kingma, D. P. and Ba, J.: Adam: A method for stochastic optimization, arXiv
[preprint], https://doi.org/10.48550/arXiv.1412.6980, 22 December 2014. a
Kirkwood, C., Economou, T., Pugeault, N., and Odbert, H.: Bayesian Deep
Learning for Spatial Interpolation in the Presence of Auxiliary Information,
Math. Geosci., 54, 507–531, https://doi.org/10.1007/s11004-021-09988-0,
2022. a
Lajaunie, C., Courrioux, G., and Manuel, L.: Foliation fields and 3D
cartography in geology: principles of a method based on potential
interpolation, Math. Geol., 29, 571–584, 1997. a
Laurent, G., Aillères, L., Caumon, G., and Grose, L.: Folding and
poly-deformation modelling in implicit modelling approach, 34th Gocad Meet.
Proc., 1–18, 2014. a
Li, Z., Pan, M., Han, D., Liu, W., Hu, S., Liu, P., and Yan, M.:
Three-Dimensional Structural Modeling Technique, Earth Sci., 41,
2136–2146, 2016. a
Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B., and Belongie,
S.: Feature pyramid networks for object detection, in: Proceedings of the
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 21–26 July 2017, 2117–2125, https://doi.org/10.48550/arXiv.1612.03144,
2017. a
Lindsay, M. D., Aillères, L., Jessell, M. W., de Kemp, E. A., and Betts,
P. G.: Locating and quantifying geological uncertainty in three-dimensional
models: Analysis of the Gippsland Basin, southeastern Australia,
Tectonophysics, 546, 10–27, 2012. a
Maggiori, E., Tarabalka, Y., Charpiat, G., and Alliez, P.: Convolutional neural
networks for large-scale remote-sensing image classification, IEEE
T. Geosci. Remote, 55, 645–657, 2016. a
Mallet, J.: Three-dimensional graphic display of disconnected bodies,
Math. Geol., 20, 977–990, 1988. a
Mallet, J.-L.: Discrete smooth interpolation in geometric modelling,
Computer-Aided Design, 24, 178–191, 1992. a
Mallet, J.-L.: Discrete modeling for natural objects, Math. Geol., 29,
199–219, 1997. a
Mallet, J.-L.: Elements of Mathematical Sedimentary Geology: the GeoChron Model, EAGE publications, ISBN 9789073834811, 2014. a
McInerney, P., Goldberg, A., Calcagno, P., Courrioux, G., Guillen, A., and
Seikel, R.: Improved 3D geology modelling using an implicit function
interpolator and forward modelling of potential field data, in: Proceedings
of exploration, vol. 7, 919–922, https://www.911metallurgist.com/blog/wp-content/uploads/2015/10/Improved-3D-Geology-Modelling-using-an-Implicit-Function-Interpolator-and-Forward-Modelling-of-Potential-Field-Data.pdf, 2007. a
Perol, T., Gharbi, M., and Denolle, M.: Convolutional neural network for
earthquake detection and location, Science Advances, 4, e1700578, https://doi.org/10.1126/sciadv.1700578, 2018. a
Phillips, J. D., Hansen, R. O., and Blakely, R. J.: The use of curvature in
potential-field interpretation, Explor. Geophys., 38, 111–119, 2007. a
Pirot, G., Joshi, R., Giraud, J., Lindsay, M. D., and Jessell, M. W.: loopUI-0.1: indicators to support needs and practices in 3D geological modelling uncertainty quantification, Geosci. Model Dev., 15, 4689–4708, https://doi.org/10.5194/gmd-15-4689-2022, 2022. a
Renaudeau, J., Malvesin, E., Maerten, F., and Caumon, G.: Implicit structural
modeling by minimization of the bending energy with moving least squares
functions, Math. Geosci., 51, 693–724, 2019. a
Ronneberger, O., Fischer, P., and Brox, T.: U-net: Convolutional networks for
biomedical image segmentation, in: International Conference on Medical image
computing and computer-assisted intervention, 234–241, Springer, https://3dvar.com/Ronneberger2015U.pdf (last access: 29 August 2022), 2015. a
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.-C.: Mobilenetv2:
Inverted residuals and linear bottlenecks, in: Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 18–23 June 2018, 4510–4520, https://doi.org/10.48550/arXiv.1801.04381, 2018. a
Shewchuk, J.: What is a good linear finite element? interpolation,
conditioning, anisotropy, and quality measures (preprint), University of
California at Berkeley, 73, 137, 2002. a
Shi, Y., Wu, X., and Fomel, S.: SaltSeg: Automatic 3D salt segmentation using a
deep convolutional neural network, Interpretation, 7, SE113–SE122, 2019. a
Souche, L., Iskenova, G., Lepage, F., and Desmarest, D.: Construction of
structurally and stratigraphically consistent structural models using the
volume-based modelling technology: Applications to an Australian dataset, in:
International Petroleum Technology Conference, Kuala Lumpur, Malaysia, December 2014, https://doi.org/10.2523/IPTC-18216-MS, 2014. a
Viard, T., Caumon, G., and Levy, B.: Adjacent versus coincident representations
of geospatial uncertainty: Which promote better decisions?, Comput.
Geosci., 37, 511–520, 2011. a
Wang, Z., Simoncelli, E. P., and Bovik, A. C.: Multi-scale structural similarity
for image quality assessment, in: The Thrity-Seventh Asilomar Conference on
Signals, Systems & Computers, 2003, vol. 2, 1398–1402, IEEE, https://utw10503.utweb.utexas.edu/publications/2003/zw_asil2003_msssim.pdf (last access: 29 August 2022), 2003. a, b, c
Wellmann, F. and Caumon, G.: 3-D Structural geological models: Concepts,
methods, and uncertainties, Adv. Geophys., 59, 1–121, https://doi.org/10.1016/bs.agph.2018.09.001, 2018. a
Wu, X., Liang, L., Shi, Y., and Fomel, S.: FaultSeg3D: using synthetic
datasets to train an end-to-end convolutional neural network for 3D seismic
fault segmentation, Geophysics, 84, IM35–IM45, 2019. a
Wu, X., Geng, Z., Shi, Y., Pham, N., Fomel, S., and Caumon, G.: Building
realistic structure models to train convolutional neural networks for seismic
structural interpretation, Geophysics, 85, WA27–WA39, 2020. a
Wu, Y., Lin, Y., Zhou, Z., Bolton, D. C., Liu, J., and Johnson, P.: DeepDetect:
A cascaded region-based densely connected network for seismic event
detection, IEEE T. Geosci. Remote, 57, 62–75,
2018. a
Yeh, R. A., Chen, C., Yian Lim, T., Schwing, A. G., Hasegawa-Johnson, M., and
Do, M. N.: Semantic image inpainting with deep generative models, in:
Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 21–26 July 2017, https://doi.org/10.48550/arXiv.1607.07539, 5485–5493, 2017. a
Yu, C., Wang, J., Peng, C., Gao, C., Yu, G., and Sang, N.: Learning a
discriminative feature network for semantic segmentation, in: Proceedings of
the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 18–23 June 2018,
1857–1866, https://doi.org/10.48550/arXiv.1804.09337, 2018. a
Yu, J., Lin, Z., Yang, J., Shen, X., Lu, X., and Huang, T. S.: Generative image
inpainting with contextual attention, in: Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, 18–23 June 2018, 5505–5514, https://doi.org/10.48550/arXiv.1801.07892,
2018. a
Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., and Liang, J.: Unet++: A nested
u-net architecture for medical image segmentation, in: Deep learning in
medical image analysis and multimodal learning for clinical decision support, Springer, 3–11, https://doi.org/10.1007/978-3-030-00889-5_1, 2018. a
Short summary
We present an implicit modeling method based on deep learning to produce a geologically valid and structurally compatible model from unevenly sampled structural data. Trained with automatically generated synthetic data with realistic features, our network can efficiently model geological structures without the need to solve large systems of mathematical equations, opening new opportunities for further leveraging deep learning to improve modeling capacity in many Earth science applications.
We present an implicit modeling method based on deep learning to produce a geologically valid...
