Attention-enhanced computational ghost imaging

被引：0
作者：
Chen, Yifan [1 ]
Tian, Tong [2 ,3 ]
Lu, Xin [1 ]
Li, Chen [1 ,4 ]
Zhu, Ruolan [1 ]
Sun, Zhe [1 ]
Li, Xuelong [1 ]
机构：
[1] Northwestern Polytech Univ, Sch Artificial Intelligence, OPt & Elect iOPEN, Xian 710072, Peoples R China
[2] Friedrich Schiller Univ, Inst Opt & Quantum Elect, Abbe Ctr Photon, D-07743 Jena, Germany
[3] Helmholtz Inst Jena, D-07743 Jena, Germany
[4] Northwestern Polytech Univ, Sch Elect & Informat, Xian 710072, Peoples R China
来源：
SCIENCE CHINA-INFORMATION SCIENCES | 2025年 / 68卷 / 06期
关键词：
computational ghost imaging; attention mechanism; deep learning; self-supervised; speckle pattern; SCATTERING MEDIA;
D O I：
10.1007/s11432-024-4434-5
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
In this study, we propose an attention-enhanced computational ghost imaging method (AEGI). AEGI integrates the attention mechanism into the framework of computational ghost imaging. In the process of image reconstruction, the attention mechanism identifies and captures object-relevant information from the extracted features, constrained by the discrepancy between the 1D intensity of the predicted and actual object image. This process can adjust the weight of the feature by exploring the relationship between the feature and adjacent features, thus enhancing the object signal while suppressing background noise in the final image. In addition, we have conducted some experiments in unfamiliar space and underwater environments to verify the effectiveness of AEGI. The results show that AEGI can reconstruct object images with high quality, which greatly enhances the practical application capabilities of computational ghost imaging.
引用
收藏
页数：10
相关论文
共 34 条
[1]   Self-supervised learning for single-pixel imaging via dual-domain constraints [J].
Chang, Xuyang ;
Wu, Ze ;
LI, Daoyu ;
Zhan, Xinrui ;
Yan, Rong ;
Bian, Liheng .
OPTICS LETTERS, 2023, 48 (07) :1566-1569
[2]   Computational ghost imaging in turbulent water based on self-supervised information extraction network [J].
Chen, Yifan ;
Sun, Zhe ;
Li, Chen ;
Li, Xuelong .
OPTICS AND LASER TECHNOLOGY, 2023, 167
[3]   High-turbidity underwater active single-pixel imaging based on generative adversarial networks with double Attention U-Net under low sampling rate [J].
Feng, Wei ;
Zhou, Shiqi ;
Li, Shuyang ;
Yi, Yongcong ;
Zhai, Zhongsheng .
OPTICS COMMUNICATIONS, 2023, 538
[4]   Differential Ghost Imaging [J].
Ferri, F. ;
Magatti, D. ;
Lugiato, L. A. ;
Gatti, A. .
PHYSICAL REVIEW LETTERS, 2010, 104 (25)
[5]   Dual Attention Network for Scene Segmentation [J].
Fu, Jun ;
Liu, Jing ;
Tian, Haijie ;
Li, Yong ;
Bao, Yongjun ;
Fang, Zhiwei ;
Lu, Hanqing .
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, :3141-3149
[6]   Improving the Imaging Quality of Ghost Imaging Lidar via Sparsity Constraint by Time-Resolved Technique [J].
Gong, Wenlin ;
Yu, Hong ;
Zhao, Chengqiang ;
Bo, Zunwang ;
Chen, Mingliang ;
Xu, Wendong .
REMOTE SENSING, 2016, 8 (12)
[7]   Three-dimensional ghost imaging lidar via sparsity constraint [J].
Gong, Wenlin ;
Zhao, Chengqiang ;
Yu, Hong ;
Chen, Mingliang ;
Xu, Wendong ;
Han, Shensheng .
SCIENTIFIC REPORTS, 2016, 6
[8]   Denoising ghost imaging under a small sampling rate via deep learning for tracking and imaging moving objects [J].
Hu, Hong-Kang ;
Sun, Shuai ;
Lin, Hui-Zu ;
Jiang, Liang ;
Liu, Wei-Tao .
OPTICS EXPRESS, 2020, 28 (25) :37284-37293
[9]   Compressive ghost imaging through scattering media with deep learning [J].
Li, Fengqiang ;
Zhao, Ming ;
Tian, Zhiming ;
Willomitzer, Florian ;
Cossairt, Oliver .
OPTICS EXPRESS, 2020, 28 (12) :17395-17408
[10]   URNet: High-quality single-pixel imaging with untrained reconstruction network [J].
Li, Jiaosheng ;
Wu, Bo ;
Liu, Tianyun ;
Zhang, Qinnan .
OPTICS AND LASERS IN ENGINEERING, 2023, 166
1234