Remote Sensing Technology and Application, Volume. 40, Issue 3, 734(2025)

Remote Sensing Information Extraction of Yancheng Wetland Rare Bird Reserve based on Multi-Feature Optimization

Senbo YU1, Hong YANG1,2、*, Chunfeng WANG1,2, and Dasong LUO1
Author Affiliations
  • 1Shanghai Ocean University, College Of Oceanography And Ecological Science, Shanghai201306, China
  • 2Shanghai Ocean University,The Centre for Monitoring and Evaluation of the Marine Environment, Shanghai201306, China
  • show less
    Figures & Tables(20)
    Location of the study area
    Vegetation growth cycle in Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’ core area
    Technical flow chart
    Patch segmentation at different scales
    Calculation ROC-LV curve
    Results of the study area’s feature importance ranking
    Distribution of the number of various features in Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’ core area
    Classification results of single-season facies based on random forests in Yancheng Wetland National Nature Reserve’core area
    Accuracy of ground feature classification during vegetation growth and dormancy periods
    Classification results of preferred feature groups in Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’ core area
    Classification accuracy of preferred feature group subset in Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’ core area
    • Table 1. Source information for Sentinel-2 images

      View table
      View in Article

      Table 1. Source information for Sentinel-2 images

      影像获取日期/时间基线/轨道编号卫星影像质量
      2020.02.18/2:37:49N9999/R0892B
      2020.02.18/2:37:49N9999/R0892B
      2020.05.13/2:35:51N0214/R0892A
      2020.05.13/2:35:51N9999/R0892A
      2020.09.05/2:35:49N0214/R0892B
      2020.09.05/2:35:49N0214/R0892B
      2020.12.24/2:41:09N0214/R0892B
      2020.12.24/2:41:09N0214/R0892B
    • Table 2. Spectral bands and resolutions of Sentinel-2A/B sensors

      View table
      View in Article

      Table 2. Spectral bands and resolutions of Sentinel-2A/B sensors

      波段Sentinel-2ASentinel-2B

      分辨率

      /m

      中心波长

      /nm

      波谱宽度

      /nm

      中心波长

      /nm

      波谱宽度

      /nm

      B2-blue496654926510
      B3-green560355593510
      B4-red664306653010
      B5-Red Edge 1703157031520
      B6-Red Edge 2740157391520
      B7-Red Edge 3782207792020
      B8a-Narrow NIR864208642020
      B8-NIR83511583311510
      B11-SWIR116139016109020
      B12-SWIR22202180218518020
    • Table 3. Wetland classification system of Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’ core area

      View table
      View in Article

      Table 3. Wetland classification system of Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’ core area

      一级分类二级分类类别描述
      自然湿地光滩无植被覆盖的淤泥质滩涂
      芦苇沼泽主要以芦苇覆盖的沼泽草甸
      赤碱蓬沼泽主要以赤碱蓬覆盖的沼泽湿地
      互花米草地主要以互花米草覆盖的沼泽湿地
      水体河流及水深<6m的浅海水域
      人工湿地养殖坑塘盐田,鱼虾养殖池塘,水库坑塘
    • Table 4. The dataset of feature variables

      View table
      View in Article

      Table 4. The dataset of feature variables

      特征简称特征全称特征描述变量类型
      BBandB2-B8,B8a,B11,B12光谱特征
      NDVINormalized Difference Vegetation Index(B8a-B4)/(B8a+B4)植被及水体指数
      RVIRatio Vegetation IndexB8a/B4
      DVIDifference Vegetation IndexB8a-B4
      SAVISoil Adjusted Vegetation Index[(B8a-B4)/(B8A+B4+T)]*(1+T)
      NDWINormalized Difference Water Index(B3-B8a)/(B3+B8a)
      MNDWIModified Normalized Difference Water Index(B3-B11)/(B3+B11)
      NDre1Normalized Difference red-edge 1(B6-B5)/(B6+B5)红边指数
      NDre2Normalized Difference red-edge 2(B7-B5)/(B7+B5)
      CIreChlorophyll Index red-edgeB7/B5-1
      MCARIModified Chlorophyll Absorption Ratio Index[(B5-B4)-0.2×(B5-B3)] × (B5-B4)
      GLCM_MMean影像灰度平均值纹理特征
      GLCM-ConContrast灰度值对比度
      GLCM_CorCorrelation灰度值相关性
      GLCM_EEntropy熵(无序性)
    • Table 5. Sentinel-2 Image Feature Combination Scheme

      View table
      View in Article

      Table 5. Sentinel-2 Image Feature Combination Scheme

      方案特征组合时相
      1光谱特征+植被指数+水体指数+红边指数
      植被休眠期
      植被生长期
      2光谱特征+植被指数+水体指数+红边指数+纹理特征
      植被休眠期
      植被生长期
      3植被休眠期优选特征组植被休眠期
      4植被生长期优选特征组植被生长期
    • Table 6. Interpretation marks for terrain information of Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’s core area

      View table
      View in Article

      Table 6. Interpretation marks for terrain information of Jiangsu Yancheng Wetland Rare Bird National Nature Reserve’s core area

      坐标类别实地照片天地图Sentinel-2RF分类结果

      120°67′94″E,

      33°50′96″N

      光滩

      120°65′80″E,

      33°51′71″N

      芦苇沼泽

      120°63′19″E,

      33°54′58″N

      赤碱蓬沼泽

      120°61′80″E,

      33°58′81″N

      互花米草地

      120°67′86″E,

      33°54′53″N

      水体

      120°55′07″E,

      33°56′63″N

      养殖坑塘
    • Table 7. Results of single-season phase classification for scenario 1

      View table
      View in Article

      Table 7. Results of single-season phase classification for scenario 1

      时相KNNRF
      总体精度/%Kappa总体精度/%Kappa
      2月77.000.7294.510.93
      5月81.330.7897.880.97
      9月80.670.7793.010.91
      12月82.670.7894.690.96
    • Table 8. Changes in classification results after adding texture features to single-season phase images

      View table
      View in Article

      Table 8. Changes in classification results after adding texture features to single-season phase images

      时相方案1方案2
      总体精度/%Kappa总体精度/%Kappa
      2月94.510.9294.730.93
      5月97.880.9797.980.97
      9月93.010.9094.510.92
      12月95.920.9497.880.97
    • Table 9. Core area’s classification Results of Vegetation Dormancy Period and Growth Period Images in Different Schemes

      View table
      View in Article

      Table 9. Core area’s classification Results of Vegetation Dormancy Period and Growth Period Images in Different Schemes

      时期方案1方案2方案3
      总体精度/%Kappa总体精度/%Kappa总体精度/%Kappa
      植被生长期98.750.9898.830.9898.930.99
      植被休眠期97.770.9797.790.9897.970.98
    Tools

    Get Citation

    Copy Citation Text

    Senbo YU, Hong YANG, Chunfeng WANG, Dasong LUO. Remote Sensing Information Extraction of Yancheng Wetland Rare Bird Reserve based on Multi-Feature Optimization[J]. Remote Sensing Technology and Application, 2025, 40(3): 734

    Download Citation

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

    Category:

    Received: May. 18, 2023

    Accepted: --

    Published Online: Sep. 28, 2025

    The Author Email: Hong YANG (hyang@shou.edu.cn)

    DOI:10.11873/j.issn.1004-0323.2025.3.0734

    Topics