• Acta Photonica Sinica
  • Vol. 50, Issue 12, 1210004 (2021)
Chen HE, Hong FANG*, and Ningchao ZHANG
Author Affiliations
  • School of Sciences, Xi′an Technological University, Xi′an710021, China
  • show less
    DOI: 10.3788/gzxb20215012.1210004 Cite this Article
    Chen HE, Hong FANG, Ningchao ZHANG. Single-shot On-axis Digital Holography Reconstruction Method Based on Deep Learning[J]. Acta Photonica Sinica, 2021, 50(12): 1210004 Copy Citation Text show less
    U-Net network structure
    Fig. 1. U-Net network structure
    Five intensity distributions and holograms generated for training
    Fig. 2. Five intensity distributions and holograms generated for training
    The on-axis holograms in training set and the reconstructed results
    Fig. 3. The on-axis holograms in training set and the reconstructed results
    The SSIM between U-Net results and the intensity distributions of the objects
    Fig. 4. The SSIM between U-Net results and the intensity distributions of the objects
    The on-axis holograms of phase objects and the corresponding phase distributions
    Fig. 5. The on-axis holograms of phase objects and the corresponding phase distributions
    The test holograms and the results reconstructed by U-Net
    Fig. 6. The test holograms and the results reconstructed by U-Net
    The difference between real phase distributions and the phase distributions reconstructed by Fig. 6(a)
    Fig. 7. The difference between real phase distributions and the phase distributions reconstructed by Fig. 6(a)
    The on-axis digital holograms generated by the letter data set and the corresponding reconstruction results
    Fig. 8. The on-axis digital holograms generated by the letter data set and the corresponding reconstruction results
    The reconstruction results of on-axis hologram of letter “F” under different noise levers
    Fig. 9. The reconstruction results of on-axis hologram of letter “F” under different noise levers
    Chen HE, Hong FANG, Ningchao ZHANG. Single-shot On-axis Digital Holography Reconstruction Method Based on Deep Learning[J]. Acta Photonica Sinica, 2021, 50(12): 1210004
    Download Citation