Optics and Precision Engineering, Volume. 31, Issue 3, 417(2023)

Neural network-based computational holographic encryption image reconstruction scheme for chaotic iris phase mask

Tao HU1,2, Xueru SUN1,2, and Weimin JIN1,2、*
Author Affiliations
  • 1Institute of Information Optics, College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua32004, China
  • 2Key Laboratory of Optical Information Detecting and Display Technology in Zhejiang Province, Jinhua31004, China
  • show less
    Figures & Tables(15)
    Flowchart for generating Chaotic Iris Phase Masks(CIPMs)
    Flowchart of computational holographic image encryption
    Flowchart of residual network structure
    Structure of residual network module
    Neural network architecture based on residual network
    Reconstructed images by proposed neural network
    Reconstruction results of images added salt and pepper noise with different ratios
    Reconstruction results of images with Gaussian noise of 0.1%
    Reconstruction results of MNIST images by ResNet neural network
    Reconstruction results of images by CNN and VGG13
    • Table 1. CCs,PSNRs and SSIMs of decrypted images

      View table
      View in Article

      Table 1. CCs,PSNRs and SSIMs of decrypted images

      ImageCCPSNR/dBSSIM
      10.99070.0570.913
      20.98360.9580.687
      30.99053.5350.772
      40.97862.8720.686
      50.98157.5910.792
    • Table 2. CCs,PSNRs and SSIMs of decrypted images with salt and pepper noise of different ratios

      View table
      View in Article

      Table 2. CCs,PSNRs and SSIMs of decrypted images with salt and pepper noise of different ratios

      ImageCCPSNR/dBSSIM
      0.1%0.2%0.3%0.5%1%0.1%0.2%0.3%0.5%1%0.1%0.2%0.3%0.5%1%
      10.9710.9790.9790.9210.89855.6360.7063.6349.7647.840.8720.8780.8760.7210.678
      20.9770.9760.9670.8630.86160.8063.9160.7149.5750.130.6770.6610.6630.5250.518
      30.9870.9750.8740.8870.74453.5051.1444.8845.1240.480.7330.6820.5480.4350.289
      40.9770.9390.9520.8840.83663.1755.2055.8148.7146.650.6850.5930.5790.4410.416
      50.9810.9480.9660.9330.87157.5463.2457.9456.0549.790.7900.7320.6640.6740.561
    • Table 3. CCs,PSNRs and SSIMs of decrypted images with Gaussian noise of 0.1%

      View table
      View in Article

      Table 3. CCs,PSNRs and SSIMs of decrypted images with Gaussian noise of 0.1%

      ImageCCPSNR/dBSSIM
      10.91845.770.525
      20.88154.430.437
      30.83044.260.267
      40.88351.350.470
      50.87947.660.436
    • Table 4. CCs,PSNRs and SSIMs of decrypted images

      View table
      View in Article

      Table 4. CCs,PSNRs and SSIMs of decrypted images

      ImagesCCPSNR/dBSSIM
      10.98972.5510.877
      20.84360.8180.626
      30.92756.2750.666
      40.91360.0200.677
      50.87656.5460.698
    • Table 5. Average CC, PSNR and SSIM of decrypted images by three network structures

      View table
      View in Article

      Table 5. Average CC, PSNR and SSIM of decrypted images by three network structures

      Networks

      types

      CCPSNR/dBSSIM
      CNN0.90755.5750.477
      VGG130.94558.5530.611
      ResNet0.98461.0030.770
    Tools

    Get Citation

    Copy Citation Text

    Tao HU, Xueru SUN, Weimin JIN. Neural network-based computational holographic encryption image reconstruction scheme for chaotic iris phase mask[J]. Optics and Precision Engineering, 2023, 31(3): 417

    Download Citation

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

    Category: Information Sciences

    Received: May. 20, 2022

    Accepted: --

    Published Online: Mar. 7, 2023

    The Author Email: JIN Weimin (jhjinwm@163.com)

    DOI:10.37188/OPE.20233103.0417

    Topics