Laser & Optoelectronics Progress, Volume. 59, Issue 12, 1233001(2022)

Infrared Polarized Face Recognition Based on RGB Color Space

Fangbin Wang1,2,3、*, Xu Jin1,2, Darong Zhu1,2,3, Ziliang Hu1,2, Sheng Tang1,2, and Jingfa Lei1,2,3
Author Affiliations
  • 1School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei 230601, Anhui , China
  • 2Key Laboratory of Construction Machinery Fault Diagnosis and Early Warning Technology, Anhui Jianzhu University, Hefei 230601, Anhui , China
  • 3Key Laboratory of Intelligent Manufacturing of Construction Machinery, Hefei 230601, Anhui , China
  • show less
    Figures & Tables(9)
    Flow chart of RGB color space fusion
    Flow chart of HOG-SVM face recognition
    Experimental equipment and scene. (a) Equipment; (b) scene
    Different polarization azimuth images and Stokes parameter images obtained by analysis
    RGB image
    Quantitative analysis. (a) Original images; (b) structural similarity images; (c) DoG edge detection images
    HOG feature map
    • Table 1. SSIM and entropy of different images

      View table

      Table 1. SSIM and entropy of different images

      ParameterIQUDOPRGB imageVisible image
      Entropy /(bit·pixel-16.2145.8026.1063.5976.7777.334
      SSIM0.7120.7170.7010.4780.7431
    • Table 2. Recognition rate of infrared polarized face image

      View table

      Table 2. Recognition rate of infrared polarized face image

      ParameterIQUDOPRGB image
      Recognition rate /%51.640.837.948.975.6
    Tools

    Get Citation

    Copy Citation Text

    Fangbin Wang, Xu Jin, Darong Zhu, Ziliang Hu, Sheng Tang, Jingfa Lei. Infrared Polarized Face Recognition Based on RGB Color Space[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1233001

    Download Citation

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

    Category: Vision, Color, and Visual Optics

    Received: Jul. 28, 2021

    Accepted: Oct. 11, 2021

    Published Online: Jun. 9, 2022

    The Author Email: Fangbin Wang (wangfb@ahjzu.edu.cn)

    DOI:10.3788/LOP202259.1233001

    Topics