基于特征优选和时空融合算法的黄河三角洲湿地类别制图方法研究
|
冯倩, 张佳华, 邓帆, 吴贞江, 赵恩灵, 郑培鑫, 韩杨
|
A mapping methodology for wetland categories of the Yellow River Delta based on optimal feature selection and spatio-temporal fusion algorithm
|
FENG Qian, ZHANG Jiahua, DENG Fan, WU Zhenjiang, ZHAO Enling, ZHENG Peixin, HAN Yang
|
|
表9 分类精度统计
|
Tab.9 The statistics of classification accuracy
|
|
类别 | 方案1 | 方案2 | 方案3 | 方案4 | 方案5 | 方案6 | 方案7 | PA/% | UA/% | PA/% | UA/% | PA/% | UA/% | PA/% | UA/% | PA/% | UA/% | PA/% | UA/% | PA/% | UA/% | 浅海水域 | 95.94 | 97.35 | 99.15 | 96.41 | 98.31 | 94.13 | 99.54 | 98.35 | 99.36 | 97.13 | 97.50 | 96.86 | 99.12 | 94.88 | 泥质海滩 | 81.07 | 84.69 | 88.02 | 86.08 | 85.79 | 86.31 | 95.48 | 88.61 | 93.44 | 87.20 | 94.69 | 85.55 | 98.60 | 96.98 | 水库坑塘 | 80.85 | 70.03 | 75.16 | 81.01 | 79.07 | 80.73 | 83.16 | 88.91 | 75.16 | 85.68 | 77.06 | 90.21 | 83.67 | 90.33 | 河流 | 86.37 | 89.07 | 86.92 | 96.18 | 87.46 | 93.93 | 93.80 | 96.70 | 91.26 | 89.52 | 88.52 | 96.41 | 89.36 | 99.86 | 水田 | 91.89 | 89.52 | 93.30 | 89.51 | 94.28 | 85.79 | 94.84 | 90.83 | 90.03 | 94.42 | 90.87 | 92.17 | 96.86 | 90.02 | 农田 | 93.79 | 75.57 | 89.40 | 97.91 | 82.72 | 99.72 | 90.14 | 95.31 | 91.76 | 95.78 | 93.91 | 81.30 | 84.90 | 96.12 | 建筑用地 | 62.53 | 79.68 | 70.69 | 74.55 | 70.45 | 69.63 | 81.10 | 81.30 | 85.70 | 66.65 | 76.24 | 69.00 | 86.37 | 86.30 | 盐田 | 76.51 | 84.65 | 89.06 | 82.07 | 87.43 | 88.62 | 90.95 | 90.76 | 82.22 | 85.51 | 93.53 | 85.92 | 96.00 | 97.77 | 草本沼泽 | 54.71 | 73.11 | 84.28 | 57.51 | 74.05 | 59.39 | 74.83 | 77.93 | 66.87 | 70.95 | 83.54 | 72.93 | 87.74 | 86.33 | 灌木沼泽 | 64.30 | 81.72 | 70.51 | 82.77 | 63.36 | 86.17 | 70.76 | 89.77 | 74.43 | 84.23 | 53.84 | 83.50 | 71.49 | 70.85 | 总精度/% | 85.11 | 89.05 | 87.39 | 91.90 | 90.01 | 88.89 | 92.28 | Kappa系数 | 0.83 | 0.86 | 0.85 | 0.91 | 0.88 | 0.87 | 0.91 |
|
|
|