E2.4 Deep Neural Networks for optical form measurements
- Event
- SMSI 2020
-
(did not take place because of Covid-19 virus pandemic) - Band
- SMSI 2020 - System of Units and Metrological Infrastructure
- Chapter
- E2 Future Topics in Metrology
- Author(s)
- L. Hoffmann, C. Elster - Physikalisch-Technische Bundesanstalt, Braunschweig and Berlin, (Germany)
- Pages
- 368 - 369
- DOI
- 10.5162/SMSI2020/E2.4
- ISBN
- 978-3-9819376-2-6
- Price
- free
Abstract
Deep neural networks have been successfully applied in many different fields like computational imag-ing, medical healthcare, signal processing or autonomous driving. We demonstrate in a proof-of-princi-ple study that also optical form measurement can benefit from deep learning. A data-driven machine learning approach is considered for solving an inverse problem in the accurate measurements of optical surfaces. The approach is developed and tested using virtual measurements with known ground truth.