74‐2: Diffracted Image Retrieving with Deep Learning

74‐2: Diffracted Image Retrieving with Deep Learning

We investigated deep image‐to‐image regression models to retrieve diffracted and/or defocused images induced from optical components. For the purpose, a quantitative metric for strength of diffraction/defocus and optical simulation method have been devised. The quality of retrieved images has been i...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:SID International Symposium Digest of technical papers 2020-08, Vol.51 (1), p.1106-1108
Hauptverfasser: Baek, Seungin, Koh, Eunkyung, Lee, Sang-Gu, Park, Byong-Ug, Yoo, Jewon, Kim, Daewook, Choi, Soojin, Son, Hyunjin, Cho, Hyunguk, Kim, Yongjo
Format: Artikel
Sprache:eng
Schlagworte:
Cascaded Fresnel Transform (CFT)
Computer simulation
Deep Learning
Diffraction
Image quality
Image Retrieval
Optical components
Regression models
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We investigated deep image‐to‐image regression models to retrieve diffracted and/or defocused images induced from optical components. For the purpose, a quantitative metric for strength of diffraction/defocus and optical simulation method have been devised. The quality of retrieved images has been improved by optimizing optical components, and by modifying the deep model.
ISSN:0097-966X
2168-0159
DOI:10.1002/sdtp.14068