Journal of Geo-information Science, Volume. 22, Issue 10, 2010(2020)

Water Body Extraction of High Resolution Remote Sensing Image based on Improved U-Net Network

Hongshu HE1...2, Xiaoxia HUANG1,*, Hongga LI1, Lingjia NI1,2, Xinge WANG3, Chong CHEN3 and Ze LIU3 |Show fewer author(s)
Author Affiliations
  • 1Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
  • 2University of Chinese Academy of Sciences, Beijing 100049, China
  • 3Urban and Rural Planning Management Center of the Ministry of Housing and Urban-Rural Development,Beijing 100835, China
  • show less
    Figures & Tables(16)
    Technical routes for extracting the water body
    U-Net architecture
    Improved U-Net network low-dimensional information enhancement
    Full connection condition random field post-processing model
    Location of Qingdao study area
    GF-2 image processing flowchart
    Training curve of improved U-Net
    • Table 1. VGG16 network structure configuration

      View table
      View in Article

      Table 1. VGG16 network structure configuration

      感受野步长填充输出大小
      InputRGBimage:3@256×256
      Conv+ReLU3×31164@256×256
      Conv+ReLU3×31164@256×256
      Max-pooling64@128×128
      Conv+ReLU3×311128@128×128
      Conv+ReLU3×311128@128×128
      Max-pooling128@64×64
      Conv+ReLU3×311256@64×64
      Conv+ReLU3×311256@64×64
      Conv+ReLU3×311256@64×64
      Max-pooling256@32×32
      Conv+ReLU3×311512@32×32
      Conv+ReLU3×311512@32×32
      Conv+ReLU3×311512@32×32
      Max-pooling512@16×16
      Conv+ReLU3×311512@16×16
      Conv+ReLU3×311512@16×16
      Conv+ReLU3×311512@16×16
      Max-pooling512@8×8
    • Table 2. RemotesensingimageinformationintheQingdaostudyarea

      View table
      View in Article

      Table 2. RemotesensingimageinformationintheQingdaostudyarea

      影像编号中心经度/°E中心纬度/°N成像时间影像大小/像素×像素
      L1A0003593712120.536.72018-11-1227 620×35 273
      L1A0003593719120.436.32018-11-1227 620×35 113
      L1A0003593868120.636.32018-11-1227 620×35 191
    • Table 3. Basic system platform configuration

      View table
      View in Article

      Table 3. Basic system platform configuration

      项目系统CPU内存硬盘显卡
      内容Ubuntu16.04Intel E5-16308 GB500 GBNVIDIA GTX970
    • Table 4. Important software configuration

      View table
      View in Article

      Table 4. Important software configuration

      项目GPU-DriverCUDAPythonKerasTensorflow-gpu
      内容3848.03.62.2.41.4.0
    • Table 5. Comparison of water extraction results by different methods in 5 typical areas of the study area

      View table
      View in Article

      Table 5. Comparison of water extraction results by different methods in 5 typical areas of the study area

    • Table 6. Confusion matrix for accuracy evaluation

      View table
      View in Article

      Table 6. Confusion matrix for accuracy evaluation

      实际正类实际负类
      预测正类TPFP
      预测负类FNTN
    • Table 7. Accuracy comparison of water extraction results

      View table
      View in Article

      Table 7. Accuracy comparison of water extraction results

      方法IoU/%精准率/%Kappa系数
      SegNet77.682.50.79
      经典U-net82.390.40.88
      改进后的U-Net网络88.194.80.93
    • Table 8. Remote sensing image information in the application area

      View table
      View in Article

      Table 8. Remote sensing image information in the application area

      影像编号中心经度/°E中心纬度/°N成像时间影像大小/像素×像素
      L1A0003553729120.136.32018-10-28276 20×292 00
      L1A0003351642101.536.82018-07-26276 20×292 00
    • Table 9. Comparison of water extraction results in 5 typical areas of the application area

      View table
      View in Article

      Table 9. Comparison of water extraction results in 5 typical areas of the application area

      区域1区域2区域3区域4区域5
      青岛原始影像
      水体信息
      西宁原始影像
      水体信息
    Tools

    Get Citation

    Copy Citation Text

    Hongshu HE, Xiaoxia HUANG, Hongga LI, Lingjia NI, Xinge WANG, Chong CHEN, Ze LIU. Water Body Extraction of High Resolution Remote Sensing Image based on Improved U-Net Network[J]. Journal of Geo-information Science, 2020, 22(10): 2010

    Download Citation

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

    Received: Oct. 24, 2019

    Accepted: --

    Published Online: Apr. 23, 2021

    The Author Email: HUANG Xiaoxia (huangxx@aircas.ac.cn)

    DOI:10.12082/dqxxkx.2020.190622

    Topics