Infrared and Laser Engineering, Volume. 51, Issue 6, 20220249(2022)

Polarized multispectral image classification of typical ground objects based on neural network (Invited)

Ying Zhang... Heshen Li, Hao Wang, Junhua Sun*, Xi Zhang, Huilan Liu and Yanhong Lv |Show fewer author(s)
Author Affiliations
  • School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing 100191, China
  • show less
    Figures & Tables(8)
    (a) Structure diagram of polarized multispectral imaging system; (b) Physical system image
    Algorithm flow chart for registration
    Network structure diagram
    (a) Samples selected from a set of images; (b) I, Q, U components of a sample; (c) Degree of linear polarization of a sample; (d) Polarization angle of a sample
    (a) Training accuracy iterative image of network; (b) Training loss iterative image of network
    (a) Experimental scene; (b) One original image; (c) Classification results of this set of images by neural network
    • Table 1. Main device parameters of polarized multispectral imaging system

      View table
      View in Article

      Table 1. Main device parameters of polarized multispectral imaging system

      DevicePerformanceParameter
      Specifications of CCD sensorTypeMER2-502-79 U3 M/C
      PortUSB3.0
      Resolution2 448(H) × 2 048(V)
      Frame rate79.1 fps
      Sensor2/3", Sony IMX250 Global shutter CMOS
      Pixel size3.45 μm×3.45 μm
      Specifications of lensLens mountPort C
      Specifications of polarizerTypeGSP-25
      Extinction ratio>1000:1
      Specifications of filterHalf bandwidth10 nm
      Peak transmittance50%
      Mechanical specificationsWeight1.5 kg
      Size18(L) cm×10(W) cm×13(H) cm
    • Table 2. Comparison of different classification methods

      View table
      View in Article

      Table 2. Comparison of different classification methods

      Classification methodTraining set accuracyValidation set accuracyAccuracy rankingKappa coefficientKappa coefficient ranking
      Network97.3%94.2%10.8981
      MLC95.3%91.7%30.8513
      MDC76.5%76.5%50.5845
      SVM-RBF71.9%73.6%60.4566
      SVM-linear88.9%85.5%40.7254
    Tools

    Get Citation

    Copy Citation Text

    Ying Zhang, Heshen Li, Hao Wang, Junhua Sun, Xi Zhang, Huilan Liu, Yanhong Lv. Polarized multispectral image classification of typical ground objects based on neural network (Invited)[J]. Infrared and Laser Engineering, 2022, 51(6): 20220249

    Download Citation

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

    Category: Invited paper

    Received: Apr. 10, 2022

    Accepted: --

    Published Online: Dec. 20, 2022

    The Author Email:

    DOI:10.3788/IRLA20220249

    Topics