Improving NIR single-pixel imaging: using deep image prior and GANs
- 27 January 2025
- journal article
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 42 (2) , 201-210
Abstract
We introduce a hybrid approach that combines deep image prior (DIP) with generative adversarial networks (GANs) to improve the resolution of single-pixel imaging (SPI). SPI excels in challenging conditions such as low light or limited spectral camera availability, particularly in the near-infrared (NIR) range from 850 to 1550 nm. By employing an unsupervised image super-resolution technique based on DIP, we reduce the need for extensive direct SPI image datasets. This innovation simplifies enhancing image quality in specific NIR bands. We provide numerical and experimental evidence to support our method and detail the enhancements in UNet and GAN architectures across four neural network configurations.Keywords
Funding Information
- Secretaría de Ciencia, Humanidades, Tecnología e Innovación (251992)
This publication has 36 references indexed in Scilit:
- HIONet: Deep priors based deep unfolded network for phase retrievalDigital Signal Processing, 2022
- Direct generation of 2D arrays of random numbers for high-fidelity optical ghost diffraction and information transmission through scattering mediaOptics and Lasers in Engineering, 2022
- Far-field super-resolution ghost imaging with a deep neural network constraintLight: Science & Applications, 2022
- U2-Net: Going deeper with nested U-structure for salient object detectionPattern Recognition, 2020
- A novel super-resolution CT image reconstruction via semi-supervised generative adversarial networkNeural Computing and Applications, 2020
- Diverse adversarial network for image super-resolutionSignal Processing: Image Communication, 2019
- Deep-learning-based ghost imagingScientific Reports, 2017
- Deep learning with convolutional neural networks for EEG decoding and visualizationHuman Brain Mapping, 2017
- Single pixel sensing for THz laser beam profiler based on Hadamard TransformOptics & Laser Technology, 2016
- Single-pixel imaging by means of Fourier spectrum acquisitionNature Communications, 2015