• Acta Optica Sinica
  • Vol. 41, Issue 7, 0710002 (2021)
Jiaqi Yin1、2、3, Shiyong Wang1、2、*, and Fanming Li1、2、**
Author Affiliations
  • 1Key Laboratory of Infrared System Detection and Imaging Technology, Chinese Academy of Sciences, Shanghai 200083, China
  • 2Shanghai Institute of Technical Physics of the Chinese Academy of Sciences, Shanghai 200083, China
  • 3University of Chinese Academy of Sciences, Beijing 100049, China
  • show less
    DOI: 10.3788/AOS202141.0710002 Cite this Article Set citation alerts
    Jiaqi Yin, Shiyong Wang, Fanming Li. Division-of-Focal-Plane Polarization Image Denoising Algorithm Based on Improved Principal Component Analysis[J]. Acta Optica Sinica, 2021, 41(7): 0710002 Copy Citation Text show less
    Schematic diagram of the micro-polarizer array
    Fig. 1. Schematic diagram of the micro-polarizer array
    Flow chart of our algorithm
    Fig. 2. Flow chart of our algorithm
    Denoising model of the DoFP polarization image
    Fig. 3. Denoising model of the DoFP polarization image
    DoFP polarization image obtained by simulation. (a) DoFP; (b) 0°; (c) 45°; (d) 90°; (e) 135°
    Fig. 4. DoFP polarization image obtained by simulation. (a) DoFP; (b) 0°; (c) 45°; (d) 90°; (e) 135°
    Test images. (a) Fabrics; (b) leaves; (c) macbeth classic; (d) macbeth enhancement; (e) painting; (f) potery
    Fig. 5. Test images. (a) Fabrics; (b) leaves; (c) macbeth classic; (d) macbeth enhancement; (e) painting; (f) potery
    Potery images before and after denoising. (a) Original image; (b) demosaicing image; (c) noisy image; (d) denoised image
    Fig. 6. Potery images before and after denoising. (a) Original image; (b) demosaicing image; (c) noisy image; (d) denoised image
    Denoising results under different σ. (a) PSNR; (b) SSIM
    Fig. 7. Denoising results under different σ. (a) PSNR; (b) SSIM
    Denoising results of different algorithms. (a) Original image; (b) demosaicing image; (c) noisy image; (d) PCA algorithm; (e) K-SVD algorithm; (f) BM3D algorithm; (g) our algorithm
    Fig. 8. Denoising results of different algorithms. (a) Original image; (b) demosaicing image; (c) noisy image; (d) PCA algorithm; (e) K-SVD algorithm; (f) BM3D algorithm; (g) our algorithm
    Denoising results of different algorithms for image macbeth enhancement. (a) PSNR; (b) SSIM
    Fig. 9. Denoising results of different algorithms for image macbeth enhancement. (a) PSNR; (b) SSIM
    Denoising result of real images by different algorithms. (a) Original image; (b) non-uniformity corrected image; (c) PCA algorithm; (d) K-SVD algorithm; (e) BM3D algorithm; (f) our algorithm
    Fig. 10. Denoising result of real images by different algorithms. (a) Original image; (b) non-uniformity corrected image; (c) PCA algorithm; (d) K-SVD algorithm; (e) BM3D algorithm; (f) our algorithm
    ImageS0S1S2DoLP
    Demosaicing imagePSNR /dB46.523038.616641.695931.3399
    SSIM0.99560.95120.96450.7770
    Noisy imagePSNR /dB35.043626.151726.22775.9415
    SSIM0.81040.23880.21960.1175
    Denoised imagePSNR /dB41.407437.101839.570825.9634
    SSIM0.98090.90600.92150.5934
    Table 1. Stokes vector images and DoLP images before and after denoising
    AlgorithmPCAK-SVDBM3DOurs
    Time /s5.924734.24110.38660.8869
    Table 2. Average running time of different algorithms
    AlgorithmFabricsLeavesMacbethclassicMacbethenhancementPaintingPotery
    Noisy imagePSNR /dB28.137928.123228.131528.144328.141228.1259
    SSIM0.67870.47590.61430.77730.76060.5223
    PCAPSNR /dB34.569337.538538.419038.119433.857037.0131
    SSIM0.90950.90110.94810.97340.93900.9124
    K-SVDPSNR /dB36.149839.027737.173737.756833.666038.0791
    SSIM0.94030.93650.90680.96080.93340.9438
    BM3DPSNR /dB33.147437.766339.379437.640432.747537.3816
    SSIM0.89370.92860.96110.96680.93160.9456
    OursPSNR /dB36.961641.559243.743442.595735.216440.3545
    SSIM0.95430.97780.99000.99400.96260.9826
    Table 3. Denoising results of different algorithms
    AlgorithmS0S1S2DoLP
    DemosaicingPSNR /dB49.405541.960043.429133.9477
    SSIM0.99750.96000.98160.8999
    Noisy imagePSNR /dB35.173626.230126.268311.8090
    SSIM0.82350.29480.37270.1553
    PCAPSNR /dB41.897935.696435.683318.5476
    SSIM0.97230.81110.85400.4134
    K-SVDPSNR /dB41.685234.818333.676912.9843
    SSIM0.97440.78150.81420.4154
    BM3DPSNR /dB40.954936.500135.729320.7019
    SSIM0.96890.87200.89020.5247
    OursPSNR /dB43.084439.969839.613927.0429
    SSIM0.98470.93050.95270.7623
    Table 4. Denoising results of different algorithms on macbeth enhancement images
    Jiaqi Yin, Shiyong Wang, Fanming Li. Division-of-Focal-Plane Polarization Image Denoising Algorithm Based on Improved Principal Component Analysis[J]. Acta Optica Sinica, 2021, 41(7): 0710002
    Download Citation