• Photonics Research
  • Vol. 11, Issue 6, 1038 (2023)
Xuyu Zhang1,2,†, Shengfu Cheng3,4,†, Jingjing Gao2,5, Yu Gan2,5..., Chunyuan Song2,5, Dawei Zhang1,8, Songlin Zhuang1, Shensheng Han2,5,6, Puxiang Lai3,4,7,9 and Honglin Liu2,4,5,*|Show fewer author(s)
Author Affiliations
  • 1School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
  • 2Key Laboratory for Quantum Optics, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, China
  • 3Department of Biomedical Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, China
  • 4Hong Kong Polytechnic University Shenzhen Research Institute, Shenzhen 518000, China
  • 5Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
  • 6Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China
  • 7Photonics Research Institute, The Hong Kong Polytechnic University, Hong Kong SAR, China
  • 8e-mail: dwzhang@usst.edu.cn
  • 9e-mail: puxiang.lai@polyu.edu.hk
  • show less
    DOI: 10.1364/PRJ.490125 Cite this Article Set citation alerts
    Xuyu Zhang, Shengfu Cheng, Jingjing Gao, Yu Gan, Chunyuan Song, Dawei Zhang, Songlin Zhuang, Shensheng Han, Puxiang Lai, Honglin Liu, "Physical origin and boundary of scalable imaging through scattering media: a deep learning-based exploration," Photonics Res. 11, 1038 (2023) Copy Citation Text show less
    Schematic of the experimental setup of imaging through a diffuser with the coordinate system labeled. The insets (a) and (b) show the settings of imaging regions to acquire the training and test data in Tests I and II, respectively. In Test I, the training data were obtained from region A (red circle) only with the test data from regions 1–5 (green circles). In Test II, the training data were obtained from regions A–E (red circles) with the test data from regions 1–5 (green circles).
    Fig. 1. Schematic of the experimental setup of imaging through a diffuser with the coordinate system labeled. The insets (a) and (b) show the settings of imaging regions to acquire the training and test data in Tests I and II, respectively. In Test I, the training data were obtained from region A (red circle) only with the test data from regions 1–5 (green circles). In Test II, the training data were obtained from regions AE (red circles) with the test data from regions 1–5 (green circles).
    Experimental results. (a) Image reconstruction through a homemade diffuser in Tests I and II. (b) Curves of the averaged PCC with error bar for 10 reconstructed images at each of the test regions 1–5. Note a nonuniform abscissa is adopted to better reflect the whole trend, given the nonuniform distributed displacements. (c) The CCC curve measured in experiment with a FWHM of ∼34 μm.
    Fig. 2. Experimental results. (a) Image reconstruction through a homemade diffuser in Tests I and II. (b) Curves of the averaged PCC with error bar for 10 reconstructed images at each of the test regions 1–5. Note a nonuniform abscissa is adopted to better reflect the whole trend, given the nonuniform distributed displacements. (c) The CCC curve measured in experiment with a FWHM of 34  μm.
    Results of Simulation I where no ballistic light is involved. (a) The phase map of the simulated diffuser in which the color bar denotes the range of phase value in radian. (b) The characterized CCC curve of the simulated diffuser, which has an FWHM of ∼36 μm and a base level of zero. (c) The speckle patterns and predicted images in both Tests I and II at regions 1, 3, 5, 7, and 9 with the ground truth on the left. (d) Curves of averaged PCC with error bar for two tests in Simulation I in which experimental PCC results are also included for comparison. The right subplot shows the zoom-in area of the dash rectangle.
    Fig. 3. Results of Simulation I where no ballistic light is involved. (a) The phase map of the simulated diffuser in which the color bar denotes the range of phase value in radian. (b) The characterized CCC curve of the simulated diffuser, which has an FWHM of 36  μm and a base level of zero. (c) The speckle patterns and predicted images in both Tests I and II at regions 1, 3, 5, 7, and 9 with the ground truth on the left. (d) Curves of averaged PCC with error bar for two tests in Simulation I in which experimental PCC results are also included for comparison. The right subplot shows the zoom-in area of the dash rectangle.
    Results of Simulation II that involves different weights of ballistic light (η). (a) Phase distributions of the simulated diffusers corresponding to different value of η. (b) The image reconstruction results on test regions 1–6 of varying Δx when the network is trained under different η. Note that rows I–VI correspond to η=0.1,0.3,0.5,0.7,0.9,and 1, respectively. (c) The curves of average PCC as a function of displacement for different η. (d) The CCC curves of output field under different η.
    Fig. 4. Results of Simulation II that involves different weights of ballistic light (η). (a) Phase distributions of the simulated diffusers corresponding to different value of η. (b) The image reconstruction results on test regions 1–6 of varying Δx when the network is trained under different η. Note that rows I–VI correspond to η=0.1,0.3,0.5,0.7,0.9,and1, respectively. (c) The curves of average PCC as a function of displacement for different η. (d) The CCC curves of output field under different η.
    Improved model generalization by increasing distance z3 in experiment. (a) The CCC curves measured experimentally for z3=5,10,20 cm, respectively. (b) The curves of average PCC for network testing on a series of regions in experimental Tests I and II under the case of different z3.
    Fig. 5. Improved model generalization by increasing distance z3 in experiment. (a) The CCC curves measured experimentally for z3=5,10,20  cm, respectively. (b) The curves of average PCC for network testing on a series of regions in experimental Tests I and II under the case of different z3.
    Xuyu Zhang, Shengfu Cheng, Jingjing Gao, Yu Gan, Chunyuan Song, Dawei Zhang, Songlin Zhuang, Shensheng Han, Puxiang Lai, Honglin Liu, "Physical origin and boundary of scalable imaging through scattering media: a deep learning-based exploration," Photonics Res. 11, 1038 (2023)
    Download Citation