• Spectroscopy and Spectral Analysis
  • Vol. 41, Issue 8, 2565 (2021)
Schematic diagram of study area
Fig. 1. Schematic diagram of study area
The ED3Modified values vs the segmentation scale
Fig. 2. The ED3Modified values vs the segmentation scale
Image segmentation results at different scales
Fig. 3. Image segmentation results at different scales
Evaluation of the optimal feature spaces dimensions
Fig. 4. Evaluation of the optimal feature spaces dimensions
Extraction results of random forest tea plantations
Fig. 5. Extraction results of random forest tea plantations
Comparison of the object-oriented and pixel-based classifications
Fig. 6. Comparison of the object-oriented and pixel-based classifications
类别数据层对象特征指标指标数
光谱
绿

近红外
各个波段光谱反射率
均值/标准差
14
归一化植被指数
归一化水体指数
燃烧面积指数
调整土壤亮度的植被指数
亮度指数
植被指数均值
对象内部最大差值其他指数均值
纹理对比度
相异度
角二矩阵
相关性

同质性
灰度共生矩阵6
空间分割对象紧致度
面积
长/宽
3
Table 1. The features of object-oriented classification
类型贝叶斯分类决策树分类随机森林分类
生产者精度/%使用者精度/%生产者精度/%使用者精度/%生产者精度/%使用者精度/%
茶园67.2375.3368.8483.8370.5487.13
森林96.8293.7296.7191.4497.8393.72
农田44.3152.7454.6266.2169.6874.44
不透水层80.0065.8291.4283.6298.4184.93
水体82.1388.5380.8692.3488.5484.64
统计Kappa=0.70
总体精度/%=87.73
Kappa=0.72
总体精度/%=88.52
Kappa=0.78
总体精度/%=91.23
Table 2. Comparison of the accuracies of object-oriented supervision classification
类型基于像元的多分类类型面向对象的二分类
生产者精度/%使用者精度/%生产者精度/%使用者精度/%
茶园55.4363.71茶园76.5392.74
森林74.5480.23其他96.8194.63
农田49.4053.74
不透水层63.6371.12
水体78.8283.94
统计Kappa=0.58
总体精度/%=71.42
Kappa=0.85
总体精度/%=94.74
Table 3. Comparison of the accuracies of pixel-wise RF and object-oriented RF for tea plantations extraction