A convolutional neural network for fast upsampling of undersampled tomograms in X-ray CT time-series using a representative highly sampled tomogram

Bellos, D.; Basham, M.; Pridmore, T.; French, A.P.

doi:10.1107/S1600577519003448

Journal of Synchrotron Radiation Journal of
Synchrotron Radiation

IUCr IT WDC

home archive editors for authors for readers submit subscribe open access

view article

Figure 12
The selected area marked by the red window in Fig. 11(A)

enlarged (area [373:573, 343:543] of [1:1280,1:1280], pixel-value [0:0.5]). (A) From the original dataset without noise and 3601 projections. (B) From the original dataset with noise and 181 projections. (C) Using UDNN-128 for upscaling (181 projections). (D) Using the Bertram's directional interpolation (181 projections). (E) Using cubic interpolation (181 projections). (F) From the original dataset with noise and 91 projections. Note the with the proposed method (panel C) the presence of noise is decreased and that there are fewer rotational artefacts.

JOURNAL OF
SYNCHROTRON
RADIATION

ISSN: 1600-5775

Volume 26| Part 3| May 2019| Pages 839-853

https://doi.org/10.1107/S1600577519003448

Open

access

Follow J. Synchrotron Rad.

Search IUCr Journals		doi		Advanced search
Author		volume	page