GEO-LEOedocs LogoGEO-LEOedocs Logo
  • GEO-LEO
    • Deutsch
    • English
  • GEO-LEO
  • English 
    • Deutsch
    • English
  • Login
View Item 
  •   Home
  • Alle Publikationen
  • Geochemie, Mineralogie, Petrologie
  • View Item
  •   Home
  • Alle Publikationen
  • Geochemie, Mineralogie, Petrologie
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Tomographic reconstruction with a generative adversarial network

Yang, Xiaogang
Kahnt, Maik
Brückner, Dennis
Schropp, Andreas
Fam, Yakub
Becher, Johannes
Grunwaldt, Jan-Dierk
Sheppard, Thomas L.
Schroer, Christian G.
DOI: https://doi.org/10.1107/S1600577520000831
Persistent URL: http://resolver.sub.uni-goettingen.de/purl?gldocs-11858/9393
Yang, Xiaogang; Kahnt, Maik; Brückner, Dennis; Schropp, Andreas; Fam, Yakub; Becher, Johannes; Grunwaldt, Jan-Dierk; Sheppard, Thomas L.; Schroer, Christian G., 2020: Tomographic reconstruction with a generative adversarial network. In: Journal of Synchrotron Radiation, Band 27, 2: 486 - 493, DOI: 10.1107/S1600577520000831.
 
Thumbnail
View/Open
JSY2_JSY2IL5048.pdf (1.264Mb)
Metadata Export:
Endnote
BibTex
RIS
  • Abstract
This paper presents a deep learning algorithm for tomographic reconstruction (GANrec). The algorithm uses a generative adversarial network (GAN) to solve the inverse of the Radon transform directly. It works for independent sinograms without additional training steps. The GAN has been developed to fit the input sinogram with the model sinogram generated from the predicted reconstruction. Good quality reconstructions can be obtained during the minimization of the fitting errors. The reconstruction is a self-training procedure based on the physics model, instead of on training data. The algorithm showed significant improvements in the reconstruction accuracy, especially for missing-wedge tomography acquired at less than 180° rotational range. It was also validated by reconstructing a missing-wedge X-ray ptychographic tomography (PXCT) data set of a macroporous zeolite particle, for which only 51 projections over 70° could be collected. The GANrec recovered the 3D pore structure with reasonable quality for further analysis. This reconstruction concept can work universally for most of the ill-posed inverse problems if the forward model is well defined, such as phase retrieval of in-line phase-contrast imaging.
Statistik:
View Statistics
Collection
  • Geochemie, Mineralogie, Petrologie [335]
Subjects:
missing-wedge tomography
reconstruction algorithms
generative adversarial network (GAN)
ptychography
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

ImpressumPrivacy (Opt-Out)About usDeposit LicenseSupport: fid-geo-digi@sub.uni-goettingen.de
DFGSUBFID GEOFID Montan
 

 

Submit here
Submission hints
Search hints

All of Geo-Leo e-docsCommunities & CollectionsBy Issue DateContributorsSubjectsPeriodicalsTitlesThis CollectionBy Issue DateContributorsSubjectsPeriodicalsTitles

Statistics

View Usage Statistics

ImpressumPrivacy (Opt-Out)About usDeposit LicenseSupport: fid-geo-digi@sub.uni-goettingen.de
DFGSUBFID GEOFID Montan