基于卷积神经网络和高分辨率影像的湿地群落遥感分类——以洪河湿地为例
|
孟祥锐, 张树清, 臧淑英
|
Remote Sensing Classification of Wetland Communities Based on Convolutional Neural Networks and High Resolution Images: A Case Study of the Honghe Wetland
|
Xiangrui Meng, Shuqing Zhang, Shuying Zang
|
|
表4 CNN、TSP-SVM和SP-SVM方法在研究区B的分类精度对比(%) |
Table 4 Classification accuracy evaluation of CNN, TSP-SVN and SP-SVM in study site B(%) |
|
| 分类 | 水体 | 林地 | 旱生杂草 | a | b | c | d | CNN | 用户精度 | 99.25 | 98.73 | 93.17 | 98.42 | 60.72 | 84.52 | 99.07 | 制图精度 | 95.33 | 90.43 | 88.44 | 70.98 | 98.78 | 89.17 | 85.05 | 分类精度 | 97.29 | 94.58 | 90.81 | 84.7 | 79.75 | 86.85 | 92.06 | TSP-SVM | 用户精度 | 97.59 | 99.73 | 80.62 | 96.68 | 44.23 | 73.14 | 99.87 | 制图精度 | 77.32 | 92.96 | 53.09 | 92.26 | 81.04 | 91.24 | 97.07 | 分类精度 | 87.45 | 96.34 | 66.85 | 94.47 | 62.64 | 82.17 | 98.47 | SP-SVM | 用户精度 | 97.72 | 98.84 | 78.2 | 95.68 | 43.75 | 71.94 | 99.83 | 制图精度 | 78.76 | 92 | 52.9 | 90.91 | 77.65 | 90.18 | 96.57 | 分类精度 | 88.24 | 95.42 | 65.55 | 93.29 | 60.7 | 81.06 | 98.2 |
|
|
|