Remote Sensing Technology and Application, Volume. 40, Issue 3, 647(2025)
Fine Classification Mapping of Liaohe Estuary Wetland based on Google Earth Engine and Dense Time Series Information
Figures & Tables(15)
Fig. 1. Diagram of the study area location， field sampling points， and validation sample dataset
Fig. 5. Results of the classification of the different feature combination scenarios
Fig. 8. Classification results of the study area from 2018 to 2022
Fig. 9. Area of various ground feature types in the study area from 2018 to 2022
Table 1.
Classification system of coastal wetland in the Liaohe Estuary View tableView in ArticleTable 1.
Classification system of coastal wetland in the Liaohe Estuary 一级分类系统 二级分类系统 描述 自然湿地 天然水域 全年淹没，水深<6 m（浅水）和河口，包括河口的突然增加（河流湿地） 滩涂 植被覆盖率<30%的淤泥滩 盐地碱蓬 生长在河流两侧的先锋物种 芦苇 芦苇，植被覆盖率>30% 苇塘水系 芦苇田中未被芦苇覆盖的天然水域 人工湿地 池塘 水库、坑池等 水产养殖池 用来养鱼、虾等的池塘 稻田 稻田、荷花池 非湿地 林地
建设用地
自然生长的树木和灌木，人工种植的苗圃工厂、公园、道路、城镇和农村居民点等
Table 2.
Numbers of Sentinel-2 images by year View tableView in ArticleTable 2.
Numbers of Sentinel-2 images by year 成像日期 影像数量 2018.03.01~2019.03.01 101 2019.03.01~2020.03.01 96 2020.03.01~2021.03.01 107 2021.03.01~2022.03.01 91 2022.03.01~2023.03.01 95
Table 3.
Initial feature variables View tableView in ArticleTable 3.
Initial feature variables 数据源 特征类别 详细描述 说明或计算公式 合成类型 Sentinel-2 光谱特征 B1、B2、B3、B4、B5、B6、B7、B8、B8A、B9、B11、B12 光谱波段 全年中值 指数特征 NDVI (B8-B4)/(B8+B4) NDBI (B11-B8)/(B11+B8) NDWI (B3-B8)/(B3+B8) SSVI NDVI×(B2-B3/B3-B4)2 EVI (B8-B4)/(B8+B4+0.16) OSAVI B8-B4 DVI (0.3×B8)+(0.59×B4)+(0.11×B3) 物候特征 SOS、EOS、POS、LOS，各季12个原始波段与NDVI的中值合成 — 密集时序与季节中值合成 纹理特征 GLCM生成的16个特征 NDVI中值合成灰度共生矩阵 全年中值 缨帽变换特征 Greenness、Brightness、Wetness 亮度、绿度、湿度 全年中值 Sentinel-1 雷达特征 VV、VH 后向散射系数 全年中值 SRTM 地形特征 Slope、Elevation 坡度、海拔高度 —
Table 4.
Numbers of regions of interest for the sample dataset by year View tableView in ArticleTable 4.
Numbers of regions of interest for the sample dataset by year 年份 样本感兴趣区域个数 2018 945 2019 947 2020 946 2021 951 2022 953
Table 5.
Experimental protocols for different features View tableView in ArticleTable 5.
Experimental protocols for different features 特征组合方案 光谱特征 指数
特征
物候特征 缨帽变换 雷达特征 纹理特征 地形特征 1 √ √ √ √ √ √ √ 2 √ 无SSVI √ √ √ √ √ 3 √ √ 无 √ √ √ √ 4 √ √ √ 无 √ √ √ 5 √ √ √ √ 无 √ √ 6 √ √ √ √ √ 无 √ 7 √ √ √ √ √ √ 无
Table 6.
Classification accuracy statistics for each feature combination scheme in the study area View tableView in ArticleTable 6.
Classification accuracy statistics for each feature combination scheme in the study area 类别 全特征 无碱蓬指数特征 无物候特征 无缨帽变换特征 无雷达特征 无纹理特征 无地形特征 UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% 天然水域 97.94 98.79 97.94 98.60 97.94 98.79 97.41 98.78 99.54 98.79 97.94 98.79 98.18 98.79 滩涂 100.00 98.64 98.51 98.64 100.00 98.64 98.55 98.62 100.00 98.64 99.08 98.63 100.00 98.64 芦苇 99.02 98.61 98.50 96.89 82.51 87.55 94.02 94.38 97.93 98.60 97.48 98.03 99.02 98.50 盐地碱蓬 99.90 93.27 85.83 86.35 87.22 87.43 97.09 88.96 99.61 93.21 99.61 92.53 99.41 93.67 苇塘水系 99.34 76.30 98.51 70.10 99.34 77.78 98.18 77.18 99.34 57.74 98.42 65.93 99.34 76.55 稻田 100.00 97.78 96.85 97.42 86.94 77.57 93.88 92.37 100.00 97.31 99.36 97.11 100.00 97.78 池塘 98.64 96.49 98.02 96.00 98.64 96.49 98.64 95.22 98.64 86.08 98.64 86.31 98.64 96.49 水产养殖池 83.65 99.84 78.87 92.46 85.31 95.88 83.65 99.00 68.66 99.74 74.14 93.43 83.81 99.84 建设用地 99.54 92.86 98.69 92.80 99.54 94.42 99.54 92.86 100.00 94.42 82.60 91.19 99.54 92.86 林地 100.00 99.83 100.00 99.83 95.83 98.97 93.83 92.75 100.00 99.50 100.00 99.50 100.00 98.20 总体精度/% 95.75 92.64 90.88 93.72 92.18 91.88 95.76 Kappa系数 0.837 6 0.806 3 0.788 2 0.817 1 0.807 4 0.804 3 0.837 7
Tools
Get Citation
Copy Citation Text
Jiaochan HU, Shenyu TANG, Keyu YUAN, Shuai XIE, Kaizhen ZHOU, Haoyang YU. Fine Classification Mapping of Liaohe Estuary Wetland based on Google Earth Engine and Dense Time Series Information[J]. Remote Sensing Technology and Application, 2025, 40(3): 647
Paper Information
Category:
Received: Mar. 21, 2024
Accepted: --
Published Online: Sep. 28, 2025
The Author Email: Haoyang YU (dlmubs@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20