• Journal of Infrared and Millimeter Waves
  • Vol. 41, Issue 1, 2021015 (2022)
Yi-Ming LIU1, Lei ZHANG2, Mei ZHOU1、3、4, Jian LIANG5, Yan WANG1, Li SUN1、3, and Qing-Li LI1、3、4、*
Author Affiliations
  • 1Shanghai Key Laboratory of Multidimensional Information Processing,East China Normal University,Shanghai 200241,China
  • 2Beijing Tracking and communication Technology Institute,Beijing 100094,China
  • 3Engineering Center of SHMEC for Space Information and GNSS,Shanghai 200241,China
  • 4Engineering Research Center of Nanophotonics & Advanced Instrument,Ministry of Education,East China Normal University,Shanghai 200241,China
  • 5Nantong Academy of Intelligent Sensing,Nantong 226000,China
  • show less
    DOI: 10.11972/j.issn.1001-9014.2022.01.029 Cite this Article
    Yi-Ming LIU, Lei ZHANG, Mei ZHOU, Jian LIANG, Yan WANG, Li SUN, Qing-Li LI. A neural networks based method for suspended sediment concentration retrieval from GF-5 hyperspectral images[J]. Journal of Infrared and Millimeter Waves, 2022, 41(1): 2021015 Copy Citation Text show less
    Locations of 14 SSC field measurements on March 27(blue),May 24(brown)and 31 October(black)2019 near the Yangtze estuarine and coastal waters. The stars and diamonds represent the field measurements collected by the buoy stations and ships,respectively
    Fig. 1. Locations of 14 SSC field measurements on March 27(blue),May 24(brown)and 31 October(black)2019 near the Yangtze estuarine and coastal waters. The stars and diamonds represent the field measurements collected by the buoy stations and ships,respectively
    Flow diagram for the entire SSC retrieval process.
    Fig. 2. Flow diagram for the entire SSC retrieval process.
    Line chart of total in situ SSC data. The number 1~7,8~10,11~14 samples were measured on 31 October,24 May and 27 March 2019,separately. A separation line(purple)is plotted to highlight the water samples 1~7 used for the final retrieval. The blue to yellow colors of dots intuitively show the low to high SSC levels. The lines drew in blue and orange represent the origin SSC values of all 3 days and sorted SSC values of 31 October 2019,respectively
    Fig. 3. Line chart of total in situ SSC data. The number 1~7,8~10,11~14 samples were measured on 31 October,24 May and 27 March 2019,separately. A separation line(purple)is plotted to highlight the water samples 1~7 used for the final retrieval. The blue to yellow colors of dots intuitively show the low to high SSC levels. The lines drew in blue and orange represent the origin SSC values of all 3 days and sorted SSC values of 31 October 2019,respectively
    Spectra of the surface reflectance in the research region on 27 March (a) 24 May (b) and 31 October (c) 2019. The dotted,dashed and solid lines represent the low,middle and high SSC values,respectively (d) some surface reflectance spectra extracted from different typical ground objects on 31 October 2019
    Fig. 4. Spectra of the surface reflectance in the research region on 27 March (a) 24 May (b) and 31 October (c) 2019. The dotted,dashed and solid lines represent the low,middle and high SSC values,respectively (d) some surface reflectance spectra extracted from different typical ground objects on 31 October 2019
    The 7 examples of preprocessed surface reflectance spectra for different SSCs measured on 31 October 2019
    Fig. 5. The 7 examples of preprocessed surface reflectance spectra for different SSCs measured on 31 October 2019
    The relationships between the regularization hyperparameter λ,RMSE,MAPE and R2 for D’Sa (a) Nechad (b) Ruhl (c) and Loisel (d) models in the application for baseline model calibration
    Fig. 6. The relationships between the regularization hyperparameter λ,RMSE,MAPE and R2 for D’Sa (a) Nechad (b) Ruhl (c) and Loisel (d) models in the application for baseline model calibration
    The scatter diagrams (left) between the predicted values and field measurement values and the NNC calibration curves (right) for D’Sa (a),Nechad (b),Ruhl(c) and Loisel(d) models in the application for baseline model calibration
    Fig. 7. The scatter diagrams (left) between the predicted values and field measurement values and the NNC calibration curves (right) for D’Sa (a),Nechad (b),Ruhl(c) and Loisel(d) models in the application for baseline model calibration
    The scatter diagrams (left) between the predicted values and field measurement values and the NNC calibration curves (right) for D’Sa (a),Nechad (b),Ruhl (c) and Loisel (d) models in the application for temporal calibration
    Fig. 8. The scatter diagrams (left) between the predicted values and field measurement values and the NNC calibration curves (right) for D’Sa (a),Nechad (b),Ruhl (c) and Loisel (d) models in the application for temporal calibration
    SSC retrieval results of the baseline model (a) and NNC double calibration (b) using the D’Sa model in the application of temporal calibration based on the GF-5 images in the Yangtze estuarine and coastal waters on 31 October 2019. For result comparison,the magnified images of the region of interest(ROI)labelled in the red area are provided in the top left of each picture. The green star and pink diamond denote the samples with 0.14 and 0.63 g/L SSC values,respectively
    Fig. 9. SSC retrieval results of the baseline model (a) and NNC double calibration (b) using the D’Sa model in the application of temporal calibration based on the GF-5 images in the Yangtze estuarine and coastal waters on 31 October 2019. For result comparison,the magnified images of the region of interest(ROI)labelled in the red area are provided in the top left of each picture. The green star and pink diamond denote the samples with 0.14 and 0.63 g/L SSC values,respectively
    ParametersCapability
    Spatial Coverage60 km
    Spectral Range400~ 2500 nm
    Spectral ResolutionVNIR:5 nm,SWIR:10 nm
    Spatial Resolution30 m
    Signal to Noise100~200
    Table 1. Main parameters for GF-5 AHSI
    Modeling MethodIndependent Variables(nm)BaselineNNC
    RMSE(g/L)MAPER2RMSE(g/L)MAPER2
    D’Sa668,5490.14950.78210.68050.14360.75800.6926
    Nechad7580.15870.80490.67290.15670.76570.6772
    Ruhl7450.21041.11420.60390.19390.98490.6336
    Loisel557,489,6680.49412.58120.29140.39932.19950.3992
    Table 2. Comparison between baseline and NNC results in the application for baseline model calibration.
    Modeling MethodIndependent Variables(nm)BaselineNNC
    RMSE(g/L)MAPER2RMSE(g/L)MAPER2
    D’Sa668,5490.12180.86570.66880.13520.78170.7155
    Nechad7620.31660.70160.40830.15880.76830.6670
    Ruhl7620.29930.58670.39780.18040.99470.6456
    Loisel557,489,6680.41600.69720.36850.36153.5580.3037
    Table 3. Comparison between baseline and NNC results in the application for temporal calibration
    Yi-Ming LIU, Lei ZHANG, Mei ZHOU, Jian LIANG, Yan WANG, Li SUN, Qing-Li LI. A neural networks based method for suspended sediment concentration retrieval from GF-5 hyperspectral images[J]. Journal of Infrared and Millimeter Waves, 2022, 41(1): 2021015
    Download Citation