基于Rapid Eye数据的北京生态涵养区土地利用分类及变化研究
|
|
郑琪,邸苏闯,潘兴瑶,刘洪禄,朱永华,张岑,周星
|
Study of Land Use Classification and Changes in the Ecological Conservation Region of Beijing based on Rapid Eye Images
|
|
Qi Zheng,Suchuang Di,Xingyao Pan,Honglu Liu,Yonghua Zhu,Cen ZHang,Xing Zhou
|
|
表4 各分类法精度对比一览表
|
Table 4 Comparison of accuracy of each classification
|
|
| 精度 | 分类方法 | 乔木 | 灌木 | 耕地 | 水体 | 建筑区 | 矿石堆 | 砂石坑 |
|---|
| 制图精度/% | 决策树分类法 | 66.67 | 60.00 | 85.71 | 100.00 | 80.00 | 71.43 | 100.00 | | 最邻近分类 | 76.19 | 69.23 | 38.46 | 100.00 | 69.23 | 50.00 | 100.00 | | 耦合分类法 | 88.00 | 88.46 | 88.89 | 95.24 | 81.82 | 100.00 | 100.00 | | 用户精度/% | 决策树分类法 | 64.00 | 72.00 | 60.00 | 90.00 | 80.00 | 100.00 | 80.00 | | 最邻近分类 | 64.00 | 72.00 | 50.00 | 80.00 | 90.00 | 80.00 | 60.00 | | 耦合分类法 | 88.00 | 92.00 | 80.00 | 100.00 | 90.00 | 80.00 | 80.00 | | 总体精度/% | 决策树分类法 | 75.00 | | 最邻近分类 | 71.00 | | 耦合分类法 | 90.00 | | Kappa系数 | 决策树分类法 | 0.69 | | 最邻近分类 | 0.71 | | 耦合分类法 | 0.90 |
|
|
|