Abstract

Ghost imaging (GI), which employs speckle patterns and bucket signals to reconstruct target images, can be regarded as a typical inverse problem. Iterative algorithms are commonly considered to solve the inverse problem in GI. However, high computational complexity and difficult hyperparameter selection are the bottlenecks. An improved inversion method for GI based on the neural network architecture TransUNet is proposed in this work, called TransUNet-GI. The main idea of this work is to utilize a neural network to avoid issues caused by conventional iterative algorithms in GI. The inversion process is unrolled and implemented on the framework of TransUNet. The demonstrations in simulation and physical experiment show that TransUNet-GI has more promising performance than other methods.

© 2022 Optica Publishing Group

Semantic ghost imaging based on recurrent-neural-network

Yuchen He, Sihong Duan, Yuan Yuan, Hui Chen, Jianxing Li, and Zhuo Xu
Opt. Express 30(13) 23475-23484 (2022)

Ghost translation: an end-to-end ghost imaging approach based on the transformer network

Wenhan Ren, Xiaoyu Nie, Tao Peng, and Marlan O. Scully
Opt. Express 30(26) 47921-47932 (2022)

Color computational ghost imaging based on a plug-and-play generalized alternating projection

Shoupei Liu, Qi Li, Huazheng Wu, and Xiangfeng Meng
Opt. Express 30(11) 18364-18373 (2022)

End-to-end computational ghost imaging method that suppresses atmospheric turbulence

Leihong Zhang, Yunjie Zhai, Runchu Xu, Kaimin Wang, and Dawei Zhang
Appl. Opt. 62(3) 697-705 (2023)

High-quality and high-diversity conditionally generative ghost imaging based on denoising diffusion probabilistic model

Shuai Mao, Yuchen He, Hui Chen, Huaibin Zheng, Jianbin Liu, Yuan Yuan, Mingnan Le, Bin Li, Juan Chen, and Zhuo Xu
Opt. Express 31(15) 25104-25116 (2023)