Acta Optica Sinica, Volume. 44, Issue 24, 2428001(2024)

A Method for Hyperspectral Reflectance Expansion from Multispectral Reflectance Model of Stable Sites and Uncertainty Analysis

Xin Lu1,2, Yongguang Zhao1、**, Lingling Ma1、*, Ning Wang1, Qijin Han3, Zui Tao1, Jianghong Zhao4, Wan Li1, Renfei Wang1, and Xuyuan Zhang1
Author Affiliations
  • 1National Engineering Laboratory for Satellite Remote Sensing Applications, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
  • 2School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
  • 3China Centre for Resources Satellite Data and Application, Beijing 100094, China
  • 4School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing 100044, China
  • show less
    Figures & Tables(18)
    Flowchart of hyperspectral reflectance expansion for TOA reflectance model for stable sites
    Average profile and standard deviation for AHSI of East of Dazaohuo site. (a) Average profile of AHSI TOA reflectance; (b) standard deviation of AHSI TOA reflectance
    GF1/PMS images of stable sites. (a) East of Dazaohuo site (Nov. 20, 2019); (b) West of Xiao Qaidam Lake site (Aug. 14, 2019)
    Relative spectral response function and TOA hyperspectral reflectance of East of Dazaohuo. (a) Sentinel2/MSI; (b) Landsat8/OLI
    Sentinel2/MSI model predicted and observed values and their relative differences. (a) Observed values and model predictions; (b) relative differences
    Landsat8/OLI model predicted and observed values and their relative differences. (a) Observed values and model predictions; (b) relative differences
    Error bar plots for both models. (a) Dataset of Sentinel2/MSI; (b) dataset of Landsat8/OLI
    Effect of atmospheric conditions on spectral scaling factors. (a) Spectral scaling factors under the same atmospheric condition; (b) spectral scaling factors under different atmospheric conditions
    Effect of solar zenith angles on spectral scaling factors. (a) Spectral scaling factors under the same solar zenith angle; (b) spectral scaling factors under different solar zenith angles
    Effect of view zenith angles on spectral scaling factors. (a) Spectral scaling factors under the same view zenith angle; (b) spectral scaling factors under different view zenith angles
    • Table 1. Dataset for model construction and validation

      View table

      Table 1. Dataset for model construction and validation

      Data useRemote sensing sensorSiteTime frameNumber of images
      Model constructionGF5B/AHSIEast of Dazaohuo

      2021-11-20; 2022-04-01; 2022-05-22; 2022-07-12

      2022-09-08; 2022-12-26; 2023-01-02; 2023-02-15

      8
      West of Xiao Qaidam Lake2021-11-05; 2022-07-122
      Model validationSentinel2/MSIEast of DazaohuoNo cloud data in 2019, 2020, 202133
      West of Xiao Qaidam Lake47
      Landsat8/OLIEast of Dazaohuo37
      West of Xiao Qaidam Lake60
    • Table 2. Relative differences between East of Dazaohuo model predicted values and observed values of Sentinel2/MSI

      View table

      Table 2. Relative differences between East of Dazaohuo model predicted values and observed values of Sentinel2/MSI

      Precision indexB1B2B3B4B5B6B7B8B8AB11B12
      Emre /%-3.991.61-0.43-0.390.25-1.41-0.78-2.82-0.810.830.94
      Ermse /%0.760.420.340.390.530.630.481.020.470.560.68
    • Table 3. Relative differences between East of Dazaohuo model predicted values and observed values of Landsat8/OLI

      View table

      Table 3. Relative differences between East of Dazaohuo model predicted values and observed values of Landsat8/OLI

      Precision indexB1B2B3B4B5B6B7B8
      Emre /%1.022.930.851.490.611.503.460.43
      Ermse /%0.290.580.360.540.440.781.410.34
    • Table 4. Comparison of two models based on Sentinel2/MSI dataset

      View table

      Table 4. Comparison of two models based on Sentinel2/MSI dataset

      BandMultispectral modelHyperspectral model
      EmreErmseEmreErmse
      B2-1.02%0.36%-1.61%0.42%
      B30.37%0.33%0.43%0.34%
      B40.63%0.42%0.39%0.39%
      B8A0.63%0.46%0.81%0.47%
      B12-0.12%0.61%-0.94%0.68%
    • Table 5. Comparison of two models based on Landsat8/OLI dataset

      View table

      Table 5. Comparison of two models based on Landsat8/OLI dataset

      BandMultispectral modelHyperspectral model
      Emre /%Ermse /%Emre /%Ermse /%
      B2-1.450.36-2.930.58
      B3-1.380.43-0.850.36
      B4-0.760.42-1.490.54
      B5-0.660.44-0.610.44
      B7-2.711.16-3.461.41
    • Table 6. Relative differences between West of Xiao Qaidam Lake model predicted values and observed values of Sentinel2/MSI

      View table

      Table 6. Relative differences between West of Xiao Qaidam Lake model predicted values and observed values of Sentinel2/MSI

      Precision indexB1B2B3B4B5B6B7B8B8AB11B12
      Emre /%3.01-1.820.390.45-1.370.310.861.881.71-1.562.12
      Ermse /%0.650.470.350.430.710.500.520.870.730.901.05
    • Table 7. Relative differences between West of Xiao Qaidam Lake model predicted values and observed values of Landsat8/OLI

      View table

      Table 7. Relative differences between West of Xiao Qaidam Lake model predicted values and observed values of Landsat8/OLI

      Precision indexB1B2B3B4B5B6B7B8
      Emre /%-1.78-2.74-0.21-0.780.37-1.600.990.21
      Ermse /%0.470.650.410.550.490.900.720.42
    • Table 8. Uncertainty of calibration results of TOA hyperspectral reflectance model for stable sites

      View table

      Table 8. Uncertainty of calibration results of TOA hyperspectral reflectance model for stable sites

      Uncertainty contributorRelative standard uncertainty /%
      B1B2B3B4B5B6B7B8
      Hyperspectral modelVZA0.290.160.070.080.070.080.400.09
      VAA0.030.010.010.010.010.010.050.01
      SZA0.050.010.010.010.010.010.110.01
      SAA0.030.010.010.010.010.010.050.01
      MODIS measurement2.002.002.002.002.002.002.002.00
      Fitting error2.262.242.041.952.001.982.262.15
      Reference spectrum3.003.003.003.003.003.003.003.00
      Spectral scaling factor3.000.500.500.500.500.503.000.50
      Center wavelength offset0.560.960.130.070.190.140.760.39
      Surface homogeneity0.910.790.630.570.630.590.650.64
      Overall uncertainty5.324.454.224.174.214.195.324.29
    Tools

    Get Citation

    Copy Citation Text

    Xin Lu, Yongguang Zhao, Lingling Ma, Ning Wang, Qijin Han, Zui Tao, Jianghong Zhao, Wan Li, Renfei Wang, Xuyuan Zhang. A Method for Hyperspectral Reflectance Expansion from Multispectral Reflectance Model of Stable Sites and Uncertainty Analysis[J]. Acta Optica Sinica, 2024, 44(24): 2428001

    Download Citation

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

    Category: Remote Sensing and Sensors

    Received: Feb. 5, 2024

    Accepted: Apr. 14, 2024

    Published Online: Dec. 12, 2024

    The Author Email: Zhao Yongguang (zhaoyg@aircas.ac.cn), Ma Lingling (mall@aircas.ac.cn)

    DOI:10.3788/AOS240618

    CSTR:32393.14.AOS240618

    Topics