Display options
Cite
Anderson TI, Vega B, McKinzie J, et al. 2D-to-3D image translation of complex nanoporous volumes using generative networks. Sci Rep. 2021;11(1):20768doi: 10.1038/s41598-021-00080-5.
Anderson, T. I., Vega, B., McKinzie, J., Aryana, S. A., & Kovscek, A. R. (2021). 2D-to-3D image translation of complex nanoporous volumes using generative networks. Scientific reports, 11(1), 20768. https://doi.org/10.1038/s41598-021-00080-5
Anderson, Timothy I, et al. "2D-to-3D image translation of complex nanoporous volumes using generative networks." Scientific reports vol. 11,1 (2021): 20768. doi: https://doi.org/10.1038/s41598-021-00080-5
Anderson TI, Vega B, McKinzie J, Aryana SA, Kovscek AR. 2D-to-3D image translation of complex nanoporous volumes using generative networks. Sci Rep. 2021 Oct 21;11(1):20768. doi: 10.1038/s41598-021-00080-5. PMID: 34675247; PMCID: PMC8531351.
Copy Download .nbib Format: NLM
Share it on

Sci Rep. 2021 Oct 21;11(1):20768. doi: 10.1038/s41598-021-00080-5.

2D-to-3D image translation of complex nanoporous volumes using generative networks.

Scientific reports

Timothy I Anderson, Bolivia Vega, Jesse McKinzie, Saman A Aryana, Anthony R Kovscek

Affiliations

  1. Department of Electrical Engineering, Stanford University, Stanford, CA, 94305, USA.
  2. Department of Energy Resources Engineering, Stanford University, Stanford, CA, 94305, USA.
  3. Department of Chemical Engineering, University of Wyoming, Laramie, WY, 82071, USA.
  4. Department of Energy Resources Engineering, Stanford University, Stanford, CA, 94305, USA. [email protected].

PMID: 34675247 PMCID: PMC8531351 DOI: 10.1038/s41598-021-00080-5

Abstract

Image-based characterization offers a powerful approach to studying geological porous media at the nanoscale and images are critical to understanding reactive transport mechanisms in reservoirs relevant to energy and sustainability technologies such as carbon sequestration, subsurface hydrogen storage, and natural gas recovery. Nanoimaging presents a trade off, however, between higher-contrast sample-destructive and lower-contrast sample-preserving imaging modalities. Furthermore, high-contrast imaging modalities often acquire only 2D images, while 3D volumes are needed to characterize fully a source rock sample. In this work, we present deep learning image translation models to predict high-contrast focused ion beam-scanning electron microscopy (FIB-SEM) image volumes from transmission X-ray microscopy (TXM) images when only 2D paired training data is available. We introduce a regularization method for improving 3D volume generation from 2D-to-2D deep learning image models and apply this approach to translate 3D TXM volumes to FIB-SEM fidelity. We then segment a predicted FIB-SEM volume into a flow simulation domain and calculate the sample apparent permeability using a lattice Boltzmann method (LBM) technique. Results show that our image translation approach produces simulation domains suitable for flow visualization and allows for accurate characterization of petrophysical properties from non-destructive imaging data.

© 2021. The Author(s).

References

  1. Phys Rev E Stat Nonlin Soft Matter Phys. 2004 Dec;70(6 Pt 2):066135 - PubMed
  2. Phys Rev E Stat Nonlin Soft Matter Phys. 2006 Oct;74(4 Pt 2):046703 - PubMed
  3. Nat Methods. 2019 Dec;16(12):1226-1232 - PubMed
  4. Neural Netw. 2019 Oct;118:310-320 - PubMed
  5. IEEE Trans Pattern Anal Mach Intell. 2016 Feb;38(2):295-307 - PubMed
  6. Phys Rev E. 2017 Oct;96(4-1):043309 - PubMed
  7. Deep Learn Data Label Med Appl (2016). 2016;2016:170-178 - PubMed
  8. AJNR Am J Neuroradiol. 2018 Oct;39(10):1776-1784 - PubMed
  9. J Synchrotron Radiat. 2012 Mar;19(Pt 2):281-7 - PubMed
  10. Mach Learn Med Imaging. 2018 Sep;11046:55-63 - PubMed
  11. Tomography. 2018 Sep;4(3):138-147 - PubMed
  12. IEEE Trans Image Process. 2012 Aug;21(8):3467-78 - PubMed
  13. J Nucl Med. 2016 Jan;57(1):136-43 - PubMed

Publication Types

Grant support