Abstract

Deep-learning-augmented single-pixel imaging (SPI) provides an efficient solution for target compressive sensing. However, the conventional supervised strategy suffers from laborious training and poor generalization. In this Letter, we report a self-supervised learning method for SPI reconstruction. It introduces dual-domain constraints to integrate the SPI physics model into a neural network. Specifically, in addition to the traditional measurement constraint, an extra transformation constraint is employed to ensure target plane consistency. The transformation constraint uses the invariance of reversible transformation to impose an implicit prior, which avoids the non-uniqueness of measurement constraint. A series of experiments validate that the reported technique realizes self-supervised reconstruction in various complex scenes without any paired data, ground truth, or pre-trained prior. It can tackle the underdetermined degradation and noise, with ∼3.7-dB improvement on the PSNR index compared with the existing method.

© 2023 Optica Publishing Group

Global-optimal semi-supervised learning for single-pixel image-free sensing

Xinrui Zhan, Hui Lu, Rong Yan, and Liheng Bian
Opt. Lett. 49(3) 682-685 (2024)

Learning-based adaptive under-sampling for Fourier single-pixel imaging

Wenxin Huang, Fei Wang, Xiangyu Zhang, Ying Jin, and Guohai Situ
Opt. Lett. 48(11) 2985-2988 (2023)

Local-enhanced transformer for single-pixel imaging

Ye Tian, Ying Fu, and Jun Zhang
Opt. Lett. 48(10) 2635-2638 (2023)

Masked autoencoder for highly compressed single-pixel imaging

Haiyan Liu, Xuyang Chang, Jun Yan, Pengyu Guo, Dong Xu, and Liheng Bian
Opt. Lett. 48(16) 4392-4395 (2023)

Pol2Pol: self-supervised polarimetric image denoising

Hedong Liu, Xiaobo Li, Zhenzhou Cheng, Tiegen Liu, Jingsheng Zhai, and Haofeng Hu
Opt. Lett. 48(18) 4821-4824 (2023)