Deep neural networks for 1D impedance inversion

Vladimir Puzyrev; Anton Egorov; Anastasia Pirogova; Chris Elders; Claus Otto

doi:10.1080/22020586.2019.12073187

2nd Australasian Exploration Geoscience Conference: Data to Discovery

ISSN: 2202-0586
E-ISSN:

oa Deep neural networks for 1D impedance inversion
Authors Vladimir Puzyrev^a, Anton Egorov^b, Anastasia Pirogova^c, Chris Elders^d and Claus Otto^e
View Affiliations Hide Affiliations

^a Curtin University, Kent St., Bentley, [email protected] ^b Curtin University and Lomonosov Moscow State University, [email protected] ^c Curtin University and Lomonosov Moscow State University, [email protected] ^d Curtin University, Kent St., Bentley, [email protected] ^e Curtin University, Kent St., Bentley, [email protected]
Publisher: Australian Society of Exploration Geophysicists
Source: ASEG Extended Abstracts, Volume 2019, Issue 2nd Australasian Exploration Geoscience Conference: Data to Discovery, Dec 2019, p. 1 - 4
DOI: https://doi.org/10.1080/22020586.2019.12073187
- Published online: 01 Dec 2019

Abstract

Summary

We investigate the applicability of different types of deep neural networks for the estimation of subsurface properties from seismic data. The pre-trained networks can predict velocity models from new data in a few milliseconds, which makes this data-driven approach especially important for multidimensional inversion, where conventional methods inversion methods suffer from large computational cost. At the same time, realistic one-dimensional models such as the 160-layer velocity model used as an example in this study require large synthetic datasets for training, which are not always possible to obtain. Hence, we also study the impact of extending the training data by adding random noise to the modelled examples. We observe that enlarging training datasets by adding synthetic noise to existing samples improves the quality of inversion without a significant increase in computational complexity.

Article metrics loading...

/content/journals/10.1080/22020586.2019.12073187

2019-12-01

2026-01-17

From This Site

/content/journals/10.1080/22020586.2019.12073187

dcterms_title,dcterms_subject,pub_keyword

-contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution

10

5

Full text loading...

References

Araya-Polo, M., Dahlke, T., Frogner, C., Zhang, C., Poggio, T. and Hohl, D., 2017, Automated fault detection without seismic processing: The Leading Edge, 36(3), 208-214.
Das, V., Pollack, A., Wollner, U., and Mukerji, T., 2018, Convolutional neural network for seismic impedance inversion: SEG Technical Program Expanded Abstracts, 2071-2075.
Kingma, D.P., and Ba, J., 2014, Adam: A method for stochastic optimization: arXiv preprint arXiv:1412.6980.
Krizhevsky, A., Sutskever, I., and Hinton, G.E., 2012, Imagenet classification with deep convolutional neural networks: Advances in neural information processing systems 1097-1105.
Ma, Y., Loures, L., and Margrave, G. F., 2004, Seismic modeling with the reflectivity method: CREWES Research Report, 16-1.
Puzyrev, V., 2018, Deep learning electromagnetic inversion with convolutional neural networks: arXiv preprint arXiv:1812.10247.
Puzyrev, V., Egorov, A., Pirogova, A., Elders, C., and Otto, C., 2019, Seismic inversion with deep neural networks: a feasibility analysis: 81st EAGE Conference and Exhibition.
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R., 2014, Dropout: a simple way to prevent neural networks from overfitting: The Journal of Machine Learning Research, 15(1), 1929-1958.
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., and Rabinovich, A., 2015, Going deeper with convolutions: Proceedings of the IEEE conference on computer vision and pattern recognition, 1-9.
Wu, Y., Lin, Y., and Zhou, Z., 2018, InversionNet: Accurate and efficient seismic waveform inversion with convolutional neural networks: SEG Technical Program Expanded Abstracts, 2096-2100.
Wu, X., Liang, L., Shi, Y. and Fomel, S., 2019, FaultSeg3D: using synthetic datasets to train an end-to-end convolutional neural network for 3D seismic fault segmentation: Geophysics, 84(3), 1-36.
Zhang, G., Wang, Z., and Chen, Y., 2018, Deep learning for seismic lithology prediction: Geophysical Journal International, 215(2), 1368-1387.

/content/journals/10.1080/22020586.2019.12073187

Article Type: Research Article

Keyword(s): deep learning; impedance inversion; neural networks; seismic wave

Most Cited This Month Most Cited RSS feed

- Inversion of data from electrical imaging surveys in water-covered areas
  
  Authors M. H. Loke and J. W. Lane
- Inversion of Magnetic Data from Remanent and Induced Sources
  
  Authors Robert G. Elllis, Barrry de Wet and Ian N. Macleod
- A comparison of smooth and blocky inversion methods in 2-D electrical imaging surveys
  
  Authors M. H. Loke, Ian Acworth and Torleif Dahlin
- Designing matched bandpass and azimuthal filters for the separation of potential-field anomalies by source region and source type
  
  Authors Jeffrey D. Phillips
- Bad Colour Maps Hide Big Features and Create False Anomalies
  
  Authors Peter Kovesi
- Practical 3D EM inversion – the P223F software suite
  
  Authors Art Raiche, Fred Sugeng and Glenn Wilson
- Estimating Noise Levels in AEM Data
  
  Authors Andy Green and Richard Lane
- Target delineation using Full Tensor Gravity Gradiometry data
  
  Authors Colm A. Murphy and James Brewster
- Transdimensional Monte Carlo Inversion of AEM Data
  
  Authors Ross C Brodie and Malcolm Sambridge
- The geotech VTEM time domain helicopter em system
  
  Authors Ken Witherly, Richard Irvine and Edward Morrison
More Less

oa Deep neural networks for 1D impedance inversion

Abstract

Most Read This Month

Most Cited This Month Most Cited RSS feed