Abstract
The gravity inversion is to restore genetic density distribution of the underground target to be explored for explaining the internal structure and distribution of the Earth. In this paper, we propose a new 3D gravity inversion method based on 3D U-Net++. Compared with two-dimensional gravity inversion, three-dimensional (3D) gravity inversion can more precisely describe the density distribution of underground space. However, conventional 3D gravity inversion method input is two-dimensional, the input and output of the network proposed in our method are three-dimensional. In the training stage, we design a large number of diversified simulation model-data pairs by using the random walk method to improve the generalization ability of the network. In the test phase, we verify the network performance by using the model-data pairs generated by the simulation. To further illustrate the effectiveness of the algorithm, we apply the method to the inversion of the San Nicolas mining area, and the inversion results are basically consistent with the borehole measurement results. Moreover, the results of the 3D U-Net++ inversion and the 3D U-Net inversion are compared. The density models of the 3D U-Net++ inversion have higher resolution, more concentrated inversion results, and a clearer boundary of the density model.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Banerjee, B., and Das Gupta, S. P., 1977, Gravitational attraction of a rectangular parallelepiped: Geophysics, 42(5), 1053–1055.
Boulanger, O., and Chouteau, M., 2001, Constraints in 3D gravity inversion: Geophysical Prospecting, 49(2), 265–280.
Çiçek, Ö., Abdulkadir, A., Lienkamp, S. S., et al., 2016, 3D U-Net: learning dense volumetric segmentation from sparse annotation: International Conference on Medical Image Computing and Computer-assisted Intervention. Springer, Cham, 424–432.
Geng, M., Huang, D., Yang, Q., et al., 2014, 3D inversion of airborne gravity-gradiometry data using cokriging: Geophysics, 79(4), G37–G47.
Guo, W. B., Zhu, Z. Q., and Lu, G. Y., 2012, 3-D gravity inversion for physical properties using BP network: Progress in Geophysics, 27(2), 409–416.
Haáz, I. B., 1953, Relations between the potential of the attraction of the mass contained in a finite rectangular prism and its first and second derivatives: Geophysical Transactions II, 7, 57–66.
Hu, Z., Liu, S., Hu, X., et al., 2021, Inversion of magnetic data using deep neural networks: Physics of the Earth and Planetary Interiors, 311, 106653.
Johnson, B. J., Montante-Martinez, A., Canela-Barboza, M., et al., 2000, Geology of the san Nicolas deposit, Zacatecas, Mexico: VMS Deposits of Latin America: Geological Association of Canada, Mineral Deposits Division Special Publication, 2, 71–85.
Krizhevsky, A., Sutskever, I., and Hinton, G. E., 2012, Imagenet classification with deep convolutional neural networks: Advances in neural information processing systems, 25, 1097–1105.
Li, Y., and Oldenburg, D. W., 1998, 3-D inversion of gravity data: Geophysics, 63(1), 109–119.
Liu, S., Hu, X., and Liu, T., 2014, A stochastic inversion method for potential field data: ant colony optimization: Pure and Applied Geophysics, 171(7), 1531–1555.
Liu, S., Liang, M., and Hu, X., 2018, Particle swarm optimization inversion of magnetic data: field examples from iron ore deposits in China: Geophysics, 83(4), J43–J59.
Long, J., Shelhamer, E., and Darrell, T., 2015, Fully convolutional networks for semantic segmentation: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 3431–3440.
Montesinos, F. G., Arnoso, J., and Vieira, R., 2005, Using a genetic algorithm for 3-D inversion of gravity data in Fuerteventura (Canary Islands): International Journal of Earth Sciences, 94(2), 301–316.
Nagy, D., 1966, The gravitational attraction of a right rectangular prism: Geophysics, 31(2), 362–371.
Phillips, N., Oldenburg, D., Chen, J., et al., 2001, Cost effectiveness of geophysical inversions in mineral exploration: Applications at San Nicolas: The Leading Edge, 20(12), 1351–1360.
Rezaie, M., 2019, 3D non-smooth inversion of gravity data by zero order minimum entropy stabilizing functional: Physics of the Earth and Planetary Interiors, 294, 106275.
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. Springer, Cham, 234–241.
Roy, R., Benedicto, A., Grare, A., et al., 2017, Three-dimensional gravity modelling applied to the exploration of uranium unconformity-related basement-hosted deposits: the Contact prospect case study, Kiggavik, northeast Thelon region (Nunavut, Canada): Canadian Journal of Earth Sciences, 54(8), 869–882.
Sudre, C. H., Li, W., Vercauteren, T., et al., 2017, Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations: Springer, Cham Press, 240–248.
Toushmalani, R., 2013, Gravity inversion of a fault by Particle swarm optimization (PSO): SpringerPlus, 2(1), 1–7.
Vatankhah, S., Ardestani, V. E., and Renaut, R. A., 2015, Application of the χ2 principle and unbiased predictive risk estimator for determining the regularization parameter in 3-D focusing gravity inversion: Geophysical Journal International, 200(1), 265–277.
Yang, Q., Hu, X., Liu, S., et al., 2021, 3-D gravity inversion based on deep convolution neural networks: IEEE Geoscience and Remote Sensing Letters, 19, 1–5.
Yu, J., Jiang, Y., Wang, Z., et al., 2016, Unitbox: An advanced object detection network: Proceedings of the 24th ACM International Conference on Multimedia, 516–520.
Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., et al., 2018, Unet++: A nested u-net architecture for medical image segmentation: Springer, Cham Press, 3–11.
Zhang, J., and Li, J. B., 2011, Gravity inversion and thermal modeling about the crust-mantle structure of Southwest basin in the South China Sea: Chinese Journal of Geophysics, 54(12): 3026–3037.
Acknowledgment
We are sincerely grateful to the Geophysical Inversion Facility of the University of British Colombia (UBC-GIF) for providing the field data (https://doi.org/10.5281/zenodo.4089070).
Author information
Authors and Affiliations
Corresponding author
Additional information
This work is supported by the Key Laboratory of Geological Survey and Evaluation of Ministry of Education (China University of Geosciences) (No. GLAB2020ZR13).
Wang Yu-Feng, he graduated from the Beijing Information Science and Technology University with a bachelor’s degree in information management and information system in 2016. He is currently a postgraduate student in China University of Geosciences. His research interests include applications of gravity inversion based on deep learning. Email: charmingfrank@163.com
Corresponding author: Li Hong-Wei, He was received a Ph.D. in institute of mathematical sciences from Beijing University, China (in 1996). He has been a professor at the School of Mathematics and Physics, China University of Geosciences, Wuhan, China. His research interests include statistical signal processing, blind signal processing, multidimensional signal processing, deep learning and time series analysis. Email: hwli@cug.edu.cn
Rights and permissions
About this article
Cite this article
Yu-Feng, W., Yu-Jie, Z., Li-Hua, F. et al. Three-dimensional gravity inversion based on 3D U-Net++. Appl. Geophys. 18, 451–460 (2021). https://doi.org/10.1007/s11770-021-0909-z
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11770-021-0909-z