Abstract

We propose a new learning and inferring model that generates digital holograms using deep neural networks (DNNs). This DNN uses a generative adversarial network, trained to infer a complex two-dimensional fringe pattern from a single object point. The intensity and fringe patterns inferred for each object point were multiplied, and all the fringe patterns were accumulated to generate a perfect hologram. This method can achieve generality by recording holograms for two spaces (16 Space and 32 Space). The reconstruction results of both spaces proved to be almost the same as numerical computer-generated holograms by showing the performance at 44.56 and 35.11 dB, respectively. Through displaying the generated hologram in the optical equipment, we proved that the holograms generated by the proposed DNN can be optically reconstructed.

© 2021 Optical Society of America

Deep learning for hologram generation

Sheng-Chi Liu and Daping Chu
Opt. Express 29(17) 27373-27395 (2021)

HoloPhaseNet: fully automated deep-learning-based hologram reconstruction using a conditional generative adversarial model

Keyvan Jaferzadeh and Thomas Fevens
Biomed. Opt. Express 13(7) 4032-4046 (2022)

Deep neural network for multi-depth hologram generation and its training strategy

Juhyun Lee, Jinsoo Jeong, Jaebum Cho, Dongheon Yoo, Byounghyo Lee, and Byoungho Lee
Opt. Express 28(18) 27137-27154 (2020)

Deep-learning-based binary hologram

Hiroaki Goi, Koshi Komuro, and Takanori Nomura
Appl. Opt. 59(23) 7103-7108 (2020)

HoloSR: deep learning-based super-resolution for real-time high-resolution computer-generated holograms

Siwoo Lee, Seung-Woo Nam, Juhyun Lee, Yoonchan Jeong, and Byoungho Lee
Opt. Express 32(7) 11107-11122 (2024)