Improving the reliability of deep learning computational ghost imaging with prediction uncertainty based on neighborhood feature maps
Impact
![Google Scholar](/xmlui/themes/Mirage2/images/uji/logo_google.png)
![Microsoft Academico](/xmlui/themes/Mirage2/images/uji/logo_microsoft.png)
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/43662
comunitat-uji-handle3:10234/43643
comunitat-uji-handle4:
INVESTIGACIONThis resource is restricted
https://doi.org/10.1364/AO.511817 |
Metadata
Title
Improving the reliability of deep learning computational ghost imaging with prediction uncertainty based on neighborhood feature mapsAuthor (s)
Date
2024-05-10Publisher
Optica Publishing GroupISSN
1559-128X; 2155-3165Bibliographic citation
Shoma Kataoka, Yasuhiro Mizutani, Tsutomu Uenohara, Erick Ipus, Koichi Nitta, Osamu Matoba, Yasuhiro Takaya, and Enrique Tajahuerce, "Improving the reliability of deep learning computational ghost imaging with prediction uncertainty based on neighborhood feature maps," Appl. Opt. 63, 3736-3744 (2024)Type
info:eu-repo/semantics/articlePublisher version
https://opg.optica.org/ao/fulltext.cfm?uri=ao-63-14-3736&id=549566Version
info:eu-repo/semantics/publishedVersionAbstract
Defect inspection is required in various fields, and many researchers have attempted deep-learning algorithms for inspections. Deep-learning algorithms have advantages in terms of accuracy and measurement time; however, ... [+]
Defect inspection is required in various fields, and many researchers have attempted deep-learning algorithms for inspections. Deep-learning algorithms have advantages in terms of accuracy and measurement time; however, the reliability of deep-learning outputs is problematic in precision measurements. This study demonstrates that iterative estimation using neighboring feature maps can evaluate the uncertainty of the outputs and shows that unconfident error predictions have higher uncertainties. In ghost imaging using deep learning, the experimental results show that removing outputs with higher uncertainties improves the accuracy by approximately 15.7%. [-]
Is part of
Applied Optics, 2024, vol. 63, no 14Funder Name
Japan Society for the Promotion of Science | Ministerio de Ciencia e Innovación | Generalitat Valenciana
Funder ID
http://dx.doi.org/10.13039/501100011033
Project code
JP19H02154 | JP20H05886 | JP22H01499 | JP23KJ1439 | JPJSBP120229927 | MCIN/PEICTI2021-2023/PID2022-142907OB-I00 | Prometeo/2020/029
Project title or grant
Caracterización y control espaciotemporal de haces de luz y su aplicación a las ciencias de la vida
Rights
© Optica Publishing Group
http://rightsstatements.org/vocab/CNE/1.0/
info:eu-repo/semantics/restrictedAccess
http://rightsstatements.org/vocab/CNE/1.0/
info:eu-repo/semantics/restrictedAccess
This item appears in the folowing collection(s)
- INIT_Articles [751]