High imaging quality of Fourier single pixel imaging based on generative adversarial networks at low sampling rate

被引:36
作者
Yang, Xu [1 ]
Jiang, Pengfei [1 ]
Jiang, Mingfeng [1 ]
Xu, Lu [1 ]
Wu, Long [1 ]
Yang, Chenghua [2 ]
Zhang, Wei [2 ]
Zhang, Jianlong [3 ]
Zhang, Yong [3 ]
机构
[1] Zhejiang Sci Tech Univ, Sch Informat Sci & Technol, Hangzhou 310018, Peoples R China
[2] Beijing Inst Remote Sensing Equipment, Beijing 110000, Peoples R China
[3] Harbin Inst Technol, Inst Opt Target Simulat & Test Technol, Harbin 150001, Peoples R China
基金
中国国家自然科学基金;
关键词
Single pixel imaging; High quality; Generative adversarial network; Low sampling rate; INVERSE PROBLEMS;
D O I
10.1016/j.optlaseng.2021.106533
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
Single pixel imaging is an innovative imaging scheme using active light to obtain spatial information, which has attracted much attention in the computational imaging field. However, for single pixel imaging, it is a great challenge to find an efficient technique to obtain imaging results with high quality under low sampling conditions. In order to solve this problem, a Fourier single pixel imaging (FSPI) based on a generative adversarial network (GAN) is proposed in this paper. In the proposed GAN model, perceptual loss, pixel and frequency loss are incorporated into the total loss function to better preserve the details of the target. With the help of the GAN model, the FSPI can reconstruct results with high quality at low sampling rate conditions. The numerical simulation and experiment are implemented. Compared with conventional FSPI and FSPI based on a deep convolutional auto-encoder network, the proposed method has a better visual effect and image quality evaluation index. This approach is particularly important to high speed single pixel imaging applications due to its potential for reconstructing the high-quality target image with a low sampling rate.
引用
收藏
页数:13
相关论文
共 46 条
  • [1] Solving ill-posed inverse problems using iterative deep neural networks
    Adler, Jonas
    Oktem, Ozan
    [J]. INVERSE PROBLEMS, 2017, 33 (12)
  • [2] Bengio, 2015, ARXIVABS150204390 CO, P1504
  • [3] Fourier single-pixel imaging using fewer illumination patterns
    Deng, Huaxia
    Gao, Xicheng
    Ma, Mengchao
    Yao, Pengcheng
    Guan, Qingtian
    Zhong, Xiang
    Zhang, Jin
    [J]. APPLIED PHYSICS LETTERS, 2019, 114 (22)
  • [4] Principles and prospects for single-pixel imaging
    Edgar, Matthew P.
    Gibson, Graham M.
    Padgett, Miles J.
    [J]. NATURE PHOTONICS, 2019, 13 (01) : 13 - 20
  • [5] High-resolution far-field ghost imaging via sparsity constraint
    Gong, Wenlin
    Han, Shensheng
    [J]. SCIENTIFIC REPORTS, 2015, 5
  • [6] Route to Intelligent Imaging Reconstruction via Terahertz Nonlinear Ghost Imaging
    Gongora, Juan S. Totero
    Olivieri, Luana
    Peters, Luke
    Tunesi, Jacob
    Cecconi, Vittorio
    Cutrona, Antonio
    Tucker, Robyn
    Kumar, Vivek
    Pasquazi, Alessia
    Peccianti, Marco
    [J]. MICROMACHINES, 2020, 11 (05)
  • [7] Ghost Imaging Based on Deep Learning
    He, Yuchen
    Wang, Gao
    Dong, Guoxiang
    Zhu, Shitao
    Chen, Hui
    Zhang, Anxue
    Xu, Zhuo
    [J]. SCIENTIFIC REPORTS, 2018, 8
  • [8] Deep learning for real-time single-pixel video
    Higham, Catherine F.
    Murray-Smith, Roderick
    Padgett, Miles J.
    Edgar, Matthew P.
    [J]. SCIENTIFIC REPORTS, 2018, 8
  • [9] Single-pixel compressive diffractive imaging
    Horisaki, Ryoichi
    Matsui, Hiroaki
    Egami, Riki
    Tanida, Jun
    [J]. APPLIED OPTICS, 2017, 56 (05) : 1353 - 1357
  • [10] Ioffe S, 2015, PR MACH LEARN RES, V37, P448