Lensless computational imaging through deep learning

Deep learning has been proven to yield reliably generalizable solutions to numerous classification and decision tasks. Here, we demonstrate for the first time to our knowledge that deep neural networks (DNNs) can be trained to solve end-to-end inverse problems in computational imaging. We experiment...

Full description

Saved in:
Bibliographic Details
Published inOptica Vol. 4; no. 9; p. 1117
Main Authors Sinha, Ayan, Lee, Justin, Li, Shuai, Barbastathis, George
Format Journal Article
LanguageEnglish
Published United States Optical Society of America 20.09.2017
Subjects
MATHEMATICS AND COMPUTING
Online AccessGet full text

Cover

More Information
Summary:Deep learning has been proven to yield reliably generalizable solutions to numerous classification and decision tasks. Here, we demonstrate for the first time to our knowledge that deep neural networks (DNNs) can be trained to solve end-to-end inverse problems in computational imaging. We experimentally built and tested a lensless imaging system where a DNN was trained to recover phase objects given their propagated intensity diffraction patterns.
Bibliography:USDOE Office of Science (SC)
FG02-97ER25308
ISSN:2334-2536
2334-2536
DOI:10.1364/OPTICA.4.001117