
Single-pixel imaging using a recurrent neural network combined with convolutional layers
Author(s) -
Ikuo Hoshi,
Tomoyoshi Shimobaba,
Takashi Kakue,
Tomoyoshi Ito
Publication year - 2020
Publication title -
optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.394
H-Index - 271
ISSN - 1094-4087
DOI - 10.1364/oe.410191
Subject(s) - undersampling , pixel , computer science , artificial intelligence , image quality , robustness (evolution) , miniaturization , convolutional neural network , optics , computer vision , image (mathematics) , materials science , physics , biochemistry , chemistry , gene , nanotechnology
Single-pixel imaging allows for high-speed imaging, miniaturization of optical systems, and imaging over a broad wavelength range, which is difficult by conventional imaging sensors, such as pixel arrays. However, a challenge in single-pixel imaging is low image quality in the presence of undersampling. Deep learning is an effective method for solving this challenge; however, a large amount of memory is required for the internal parameters. In this study, we propose single-pixel imaging based on a recurrent neural network. The proposed approach succeeds in reducing the internal parameters, reconstructing images with higher quality, and showing robustness to noise.