Laser & Optoelectronics Progress, Volume. 61, Issue 22, 2211001(2024)

Optimization and Efficient Computation of Mask Extension for Holographic Display Viewing Angle Method

Jiachen Kang* and Shaowen Hao**
Author Affiliations
  • School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541000, Guangxi , China
  • show less
    Figures & Tables(23)
    Diffraction pattern at 100 mm. (a) Result of traditional angular spectrum diffraction; (b) result of band-limited angular spectrum diffraction
    Optimization scheme framework
    Principle diagram of mask extension for holographic display viewing angle of spatial light modulator. (a) Phase mask extension for holographic display viewing angle optical system structure; (b) schematic of resolution difference matching
    Band-limited angular spectrum diffraction GS optimization algorithm based on phase mask extension for holographic display viewing angle method
    Reconstruction using time-averaging method
    Layered hologram computation with additional random phase mask for three-dimensional holography
    Improved GS iterative optimization process
    Parallel computation flow
    Structure of transfer function lookup table for band-limited angular spectrum
    Hologram and its reconstruction results. (a) Original image; (b) hologram computed with traditional angular spectrum diffraction; (c) reconstruction result of traditional angular spectrum diffraction hologram; (d) hologram computed with band-limited angular spectrum diffraction; (e) reconstruction result of band-limited angular spectrum diffraction hologram; (f) hologram computed with band-limited angular spectrum diffraction GS iteration for 20 times; (g) reconstruction result of band-limited angular spectrum diffraction GS iteration for 20 times
    Comparison of results of reconstruction using traditional angular spectrum diffraction GS and band-limited angular spectrum diffraction GS with time-averaging method respectively. (a) Original image; (b)‒(d) reconstruction images using traditional angular spectrum diffraction GS optimization for iteration 20 times combined with time-averaging method; (e)‒(g) reconstruction images using band-limited angular spectrum diffraction GS optimization for iteration 20 times combined with time-averaging method
    Objective evaluation metrics for reconstruction results. (a) Speckle contrast; (b) SSIM; (c) PSNR
    Final reconstruction results. (a) Optimized final result from Ref. [10]; (b) optimized result of method in this paper (reconstruction of 20 holograms)
    Intensity and depth images of three-dimensional object. (a) Intensity image; (b) depth image
    Holograms with or without GS optimization and reconstruction images with holograms participating in focusing and head reconstruction. (a) Three-dimensional phase hologram without GS optimization; (b) three-dimensional phase hologram with GS optimization iteration for 40 times; (c) reconstruction image with 1 hologram without GS optimization participating in focusing and head reconstruction; (d) reconstruction image with 1 hologram with GS optimization iteration for 40 times participating in focusing and head reconstruction; (e) reconstruction image with 20 holograms without GS optimization participating in focusing and head reconstruction; (f) reconstruction image with 20 holograms with GS optimization iteration for 40 times participating in focusing and head reconstruction
    Local enlargement of three-dimensional reconstructed images. (a) Without GS optimization; (b) GS iteration for 40 times
    Before and after improvement, holograms and reconstruction results computed with band-limited angular spectrum diffraction GS iteration optimization. (a) Original image; (b) hologram computed with traditional angular spectrum diffraction; (c) reconstruction result of traditional angular spectrum diffraction hologram; (d) hologram computed with band-limited angular spectrum diffraction GS iteration for 20 times; (e) reconstruction result of band-limited angular spectrum diffraction GS iteration for 20 times; (f) improved hologram computed with band-limited angular spectrum diffraction GS iteration for 20 times; (g) reconstruction result of improved hologram computed with band-limited angular spectrum diffraction GS iteration for 20 times
    Holograms and reconstruction images computed with parallel computation with multiple streams and lookup table. (a) Hologram computed sequentially with 40 iterations; (b) hologram computed in parallel with 40 iterations; (c) reconstruction image of a single hologram computed sequentially; (d) reconstruction image of a single hologram computed in parallel; (e) reconstruction image of 20 holograms computed sequentially; (f) reconstruction image of 20 holograms computed in parallel; (g) local enlargement of reconstruction image with 20 holograms computed sequentially; (h) local enlargement of reconstruction image with 20 holograms computed in parallel
    • Table 1. Objective evaluation

      View table

      Table 1. Objective evaluation

      Optimization resultSpeckle contrastPSNRSSIM
      Method from Ref. [10]0.410010.68080.2778
      Method in this paper0.206518.50460.7338
    • Table 2. Objective evaluation metrics for reconstruction image quality

      View table

      Table 2. Objective evaluation metrics for reconstruction image quality

      Calculation methodPSNRSSIM
      Before improvement10.54800.2933
      After improvement10.12590.3066
    • Table 3. Computer configuration for algorithm implementation

      View table

      Table 3. Computer configuration for algorithm implementation

      Processor and memoryDevice model
      CPU12th Gen Intel (R) Core (TM) i5-12500H
      GPU 1NVIDIA GeForce RTX 3050Ti Laptop GPU
      GPU 2NVIDIA GeForce RTX 4080 Laptop GPU
      Computer memory16 GB
      GPU 1graphics memory4 GB
      GPU 2graphics memory12 GB
    • Table 4. Time for parallel computation with multiple streams only

      View table

      Table 4. Time for parallel computation with multiple streams only

      Number of calculationsCPUsequentialGPU 1-10 streamsGPU 2-10 streams
      20 holograms28210.602874.132363.02
      1 hologram1410.53143.70118.15
    • Table 5. Time for parallel computation with multiple streams and lookup table

      View table

      Table 5. Time for parallel computation with multiple streams and lookup table

      Number of calculationsCPU sequentialGPU 1-10 streamsGPU 2-10 streams
      20 holograms28210.601898.20329.41
      1 hologram1410.5394.9116.47
    Tools

    Get Citation

    Copy Citation Text

    Jiachen Kang, Shaowen Hao. Optimization and Efficient Computation of Mask Extension for Holographic Display Viewing Angle Method[J]. Laser & Optoelectronics Progress, 2024, 61(22): 2211001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Imaging Systems

    Received: Feb. 21, 2024

    Accepted: Mar. 12, 2024

    Published Online: Nov. 13, 2024

    The Author Email: Jiachen Kang (kangjc233@163.com), Shaowen Hao (hsw6826@163.com)

    DOI:10.3788/LOP240719

    CSTR:32186.14.LOP240719

    Topics