Articles | Volume 9, issue 2
https://doi.org/10.5194/jsss-9-301-2020
https://doi.org/10.5194/jsss-9-301-2020
Regular research article
 | 
24 Sep 2020
Regular research article |  | 24 Sep 2020

Deep neural networks for computational optical form measurements

Lara Hoffmann and Clemens Elster

Viewed

Total article views: 1,573 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
1,118 436 19 1,573 33 17
  • HTML: 1,118
  • PDF: 436
  • XML: 19
  • Total: 1,573
  • BibTeX: 33
  • EndNote: 17
Views and downloads (calculated since 24 Sep 2020)
Cumulative views and downloads (calculated since 24 Sep 2020)

Viewed (geographical distribution)

Total article views: 1,418 (including HTML, PDF, and XML) Thereof 1,414 with geography defined and 4 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 28 Mar 2024
Download
Short summary
Deep learning has become a state-of-the-art method in machine learning, with a broad range of successful applications. Our goal is to explore the benefits of deep learning techniques for computational optical form measurements. The research is based on solving a nonlinear inverse problem aimed at the reconstruction of optical topographies from given processed interferograms. A U-Net network structure is chosen and tested on a simulated database. The obtained results are promising.