Autofocusing using deep learning in off-axis digital holography

Jaesung Lee, Wooyoung Jeong, Kyungchan Son, Wonseok Jeon, Hyunseok Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We have focused on rapid and efficient estimator to find object distance from hologram in order to reconstruct original image. Our approach to find it makes the estimator pre-trained through deep learning. Especially in off-axis holography configuration, our method eliminates the unnecessary factors and reduces information loss occurred by resizing image to plug into Convolution Neural Network (CNN). Training is performed on the generated images at several specific distances under various optical conditions and the accuracy of estimation is validated.

Original languageEnglish
Title of host publicationDigital Holography and Three-Dimensional Imaging, DH 2018
PublisherOSA - The Optical Society
VolumePart F100-DH 2018
ISBN (Electronic)9781557528209
DOIs
Publication statusPublished - 2018 Jan 1
EventDigital Holography and Three-Dimensional Imaging, DH 2018 - Orlando, United States
Duration: 2018 Jun 252018 Jun 28

Other

OtherDigital Holography and Three-Dimensional Imaging, DH 2018
CountryUnited States
CityOrlando
Period18/6/2518/6/28

Fingerprint

Holography
Holograms
Convolution
Neural networks
Deep learning

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Mechanics of Materials

Cite this

Lee, J., Jeong, W., Son, K., Jeon, W., & Yang, H. (2018). Autofocusing using deep learning in off-axis digital holography. In Digital Holography and Three-Dimensional Imaging, DH 2018 (Vol. Part F100-DH 2018). OSA - The Optical Society. https://doi.org/10.1364/DH.2018.DTh1C.4
Lee, Jaesung ; Jeong, Wooyoung ; Son, Kyungchan ; Jeon, Wonseok ; Yang, Hyunseok. / Autofocusing using deep learning in off-axis digital holography. Digital Holography and Three-Dimensional Imaging, DH 2018. Vol. Part F100-DH 2018 OSA - The Optical Society, 2018.
@inproceedings{a3f44aa4c29a4ef78d1bedb1b163ec23,
title = "Autofocusing using deep learning in off-axis digital holography",
abstract = "We have focused on rapid and efficient estimator to find object distance from hologram in order to reconstruct original image. Our approach to find it makes the estimator pre-trained through deep learning. Especially in off-axis holography configuration, our method eliminates the unnecessary factors and reduces information loss occurred by resizing image to plug into Convolution Neural Network (CNN). Training is performed on the generated images at several specific distances under various optical conditions and the accuracy of estimation is validated.",
author = "Jaesung Lee and Wooyoung Jeong and Kyungchan Son and Wonseok Jeon and Hyunseok Yang",
year = "2018",
month = "1",
day = "1",
doi = "10.1364/DH.2018.DTh1C.4",
language = "English",
volume = "Part F100-DH 2018",
booktitle = "Digital Holography and Three-Dimensional Imaging, DH 2018",
publisher = "OSA - The Optical Society",

}

Lee, J, Jeong, W, Son, K, Jeon, W & Yang, H 2018, Autofocusing using deep learning in off-axis digital holography. in Digital Holography and Three-Dimensional Imaging, DH 2018. vol. Part F100-DH 2018, OSA - The Optical Society, Digital Holography and Three-Dimensional Imaging, DH 2018, Orlando, United States, 18/6/25. https://doi.org/10.1364/DH.2018.DTh1C.4

Autofocusing using deep learning in off-axis digital holography. / Lee, Jaesung; Jeong, Wooyoung; Son, Kyungchan; Jeon, Wonseok; Yang, Hyunseok.

Digital Holography and Three-Dimensional Imaging, DH 2018. Vol. Part F100-DH 2018 OSA - The Optical Society, 2018.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Autofocusing using deep learning in off-axis digital holography

AU - Lee, Jaesung

AU - Jeong, Wooyoung

AU - Son, Kyungchan

AU - Jeon, Wonseok

AU - Yang, Hyunseok

PY - 2018/1/1

Y1 - 2018/1/1

N2 - We have focused on rapid and efficient estimator to find object distance from hologram in order to reconstruct original image. Our approach to find it makes the estimator pre-trained through deep learning. Especially in off-axis holography configuration, our method eliminates the unnecessary factors and reduces information loss occurred by resizing image to plug into Convolution Neural Network (CNN). Training is performed on the generated images at several specific distances under various optical conditions and the accuracy of estimation is validated.

AB - We have focused on rapid and efficient estimator to find object distance from hologram in order to reconstruct original image. Our approach to find it makes the estimator pre-trained through deep learning. Especially in off-axis holography configuration, our method eliminates the unnecessary factors and reduces information loss occurred by resizing image to plug into Convolution Neural Network (CNN). Training is performed on the generated images at several specific distances under various optical conditions and the accuracy of estimation is validated.

UR - http://www.scopus.com/inward/record.url?scp=85051262244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051262244&partnerID=8YFLogxK

U2 - 10.1364/DH.2018.DTh1C.4

DO - 10.1364/DH.2018.DTh1C.4

M3 - Conference contribution

VL - Part F100-DH 2018

BT - Digital Holography and Three-Dimensional Imaging, DH 2018

PB - OSA - The Optical Society

ER -

Lee J, Jeong W, Son K, Jeon W, Yang H. Autofocusing using deep learning in off-axis digital holography. In Digital Holography and Three-Dimensional Imaging, DH 2018. Vol. Part F100-DH 2018. OSA - The Optical Society. 2018 https://doi.org/10.1364/DH.2018.DTh1C.4