• Laser & Optoelectronics Progress
  • Vol. 59, Issue 12, 1228007 (2022)
Lei Zhang1, Xiuzhen Han2、***, Fuzhong Weng3、*, and Zhongfeng Qiu1、**
Author Affiliations
  • 1School of Marine Sciences, Nanjing University of Information Science & Technology, Nanjing 210044, Jiangsu , China
  • 2National Satellite Meteorological Center, Beijing 100081, China
  • 3State Key Laboratory of Hazardous Weather, Chinese Academy of Meteorological Sciences, Beijing 100081, China
  • show less
    DOI: 10.3788/LOP202259.1228007 Cite this Article Set citation alerts
    Lei Zhang, Xiuzhen Han, Fuzhong Weng, Zhongfeng Qiu. Comparison of Water Information Extraction Algorithms Based on Sentinel-2A MSI Data[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1228007 Copy Citation Text show less
    Technology roadmap
    Fig. 1. Technology roadmap
    Schematic diagram of determining optimal segmentation threshold based on Otsu (shown by dotted abscissa)
    Fig. 2. Schematic diagram of determining optimal segmentation threshold based on Otsu (shown by dotted abscissa)
    Comparison of extraction effects of 10 band index in different experimental areas
    Fig. 3. Comparison of extraction effects of 10 band index in different experimental areas
    F1-score of water identification results of all water index in different areas. (a) Mountain; (b) lake; (c) paddy field; (d) town; (e) river
    Fig. 4. F1-score of water identification results of all water index in different areas. (a) Mountain; (b) lake; (c) paddy field; (d) town; (e) river
    False alarm rate and dismissal alarm rate of water identification results of all water index in different areas. (a) Mountain; (b) lake; (c) paddy field; (d) town; (e) river
    Fig. 5. False alarm rate and dismissal alarm rate of water identification results of all water index in different areas. (a) Mountain; (b) lake; (c) paddy field; (d) town; (e) river
    Average accuracy indexes of each model in different regions. (a) F1-Score; (b) false alarm rate and dismissal alarm rate
    Fig. 6. Average accuracy indexes of each model in different regions. (a) F1-Score; (b) false alarm rate and dismissal alarm rate
    Display of recognition effect of each index for land and water boundary. (a) Paddy field area, rectangular box as main research area; (b) false color composite map of research area; (c)-(l) water body recognition results corresponding to each index model
    Fig. 7. Display of recognition effect of each index for land and water boundary. (a) Paddy field area, rectangular box as main research area; (b) false color composite map of research area; (c)-(l) water body recognition results corresponding to each index model
    Identification accuracy of each index in mixed area of land and water. (a) F1-score; (b) false alarm rate; (c) dismissal alarm rate
    Fig. 8. Identification accuracy of each index in mixed area of land and water. (a) F1-score; (b) false alarm rate; (c) dismissal alarm rate
    Variation of reflectance of different land features with Sentinel-2A MSI band
    Fig. 9. Variation of reflectance of different land features with Sentinel-2A MSI band
    True valuePredicted value
    PositiveNegative
    PositiveTrue positive(TP)False negative(FN)
    NegativeFalse positive(FP)True negative(TN)
    Table 1. Confusion matrix
    AreaNDWIMNDWICIWIEWIRNDWIGNDWIAWEInshAWEIshFNDWISWI
    Standard deviation0.050.050.490.080.080.66363.52797.730.4741.81
    Mountain-0.110.110.24-0.44-0.031.23458.06-2108.45-1.89231.58
    Lake-0.080.081.34-0.27-0.11-0.48362.13-1911.47-0.49255.20
    Paddy field0.010.160.68-0.21-0.24-0.101268.90-835.45-1.45256.27
    Town000.06-0.293-0.0390.925782.24-152.43-0.92352.14
    River-0.090.050.03-0.38-0.050.921163.51-363.34-1.22259.38
    Table 2. Threshold and standard deviation of water index model
    Lei Zhang, Xiuzhen Han, Fuzhong Weng, Zhongfeng Qiu. Comparison of Water Information Extraction Algorithms Based on Sentinel-2A MSI Data[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1228007
    Download Citation