Remote Sensing Technology and Application, Volume. 40, Issue 2, 461(2025)
Methods of Improving Land Cover Classification based on Terrain Factors and Time-series NDVI
Figures & Tables(14)
Fig. 1. MCD12Q1 in the study area and 1∶100 000 land use product in the study area
Fig. 5. Technology roadmap of the two schemes by integrating terrain factors and time-series NDVI
Fig. 7. Improved result of time-series NDVI method（MCD-NEW- time-series NDVI）
Fig. 8. The improved result of the two methodsⅠ and the improved result of the two methodsⅡ
Table 1.
Correspondence of unified classification system View tableView in ArticleTable 1.
Correspondence of unified classification system 统一新类别 1:10万土地利用数据 MCD12Q1 1 耕地 水田、旱地 农用地、农用地与自然植被拼接体 2 林地 有林地、灌木林、疏林地、其他林地 常绿针叶林、常绿阔叶林、落叶针叶林、落叶阔叶林、混交林、稠密灌丛、稀疏灌丛 3 草地 高覆盖度草地、中覆盖度草地、低覆盖度草地 木本热带稀树草原、热带稀树草原、草地 4 人工地表 城镇用地、农村居民点、其它建设用地 城市和建筑区 5 其它地类 河渠、湖泊、水库坑塘、永久性冰川雪地、滩涂、滩地、沙地、戈壁、盐碱地、沼泽地、裸土地、裸岩石质地、其他未利用土地、海洋 水、永久湿地、雪和冰、稀疏植被
Table 2.
Regression model equation of the plain area View tableView in ArticleTable 2.
Regression model equation of the plain area 类型 回归方程（常数为0） 修正R2 回归方程（常数非0） 修正R2 耕地 y=0.007x3-1.139x4+0.004x1+2.14x5 0.907 y=0.815+0.003x1+2.037x5-1.072x4 0.171 林地 y=0.223x4-0.282x2 0.802 y=0.089-0.262x2+0.211x4-0.001x3 0.799 草地 y=0.026x4-0.00004677x3-0.043x5 0.702 y=-0.008+0.023x2-0.025x5 0.660 人工地表 y=0.001x3 0.197 * 0.082 其他地类 y=0.001x3-0.001x1 0.370 y=0.331-0.001x1-0.002x3+0.013x4 0.145
Table 3.
Regression model equation of the non plain area View tableView in ArticleTable 3.
Regression model equation of the non plain area 类型 回归方程（常数为0） 修正R2 回归方程（常数非0） 修正R2 耕地 y=0.009x3-0.058x2 0.885 y=1.11-0.052x2 0.639 林地 y=0.061x5-0.002x3 0.831 y=-0.2+0.06x5 0.614 草地 y=0.162x2-0.106x4 0.725 y=-0.048+0.194x2-0.127x4 0.260 人工地表 y=0.000458x3-0.0000363x1 0.398 y=0.059-0.0000343x1 0.069 其他地类 y=0.000133x3 0.581 * -0.022
Table 4.
Accuracy analysis of the plain and the non plain parts of land cover products View tableView in ArticleTable 4.
Accuracy analysis of the plain and the non plain parts of land cover products 产品 平原部分精度 非平原部分精度 总体精度 MCD12Q1 73.72% 40.07% 56.50% MCD-NEW-地形因子 73.94% 62.22% 67.46%
Table 5.
Accuracy analysis of improved product with time-series NDVI View tableView in ArticleTable 5.
Accuracy analysis of improved product with time-series NDVI 产品 总体精度/% Kappa系数 MCD12Q1 56.50 0.38 MCD-NEW-时序NDVI 74.97 0.64
Table 6.
Precision analysis of confusion matrix of land cover products View tableView in ArticleTable 6.
Precision analysis of confusion matrix of land cover products 产品 总体精度 Kappa系数 MCD12Q1 56.50% 0.38 MCD-NEW-地形因子 67.46% 0.50 MCD-NEW-时序NDVI 74.97% 0.64 MCD-NEW-地形因子与时序结合Ⅰ 68.34% 0.52 MCD-NEW-地形因子与时序结合Ⅱ 82.78% 0.74
Tools
Get Citation
Copy Citation Text
Yuexin CHEN, Shunbao LIAO, Yanping WANG, Feng LI. Methods of Improving Land Cover Classification based on Terrain Factors and Time-series NDVI[J]. Remote Sensing Technology and Application, 2025, 40(2): 461
Paper Information
Category:
Received: May. 16, 2022
Accepted: --
Published Online: May. 23, 2025
The Author Email: Shunbao LIAO (liaoshunbao@cidp.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20