协同Sentinel-1/2多特征优选的甘蔗提取方法

卢献健, 张焕铃, 晏红波, 黎振宝, 郭子扬

A method for sugarcane information extraction based on multi-feature optimal selection of Sentinel-1/2 image data

LU Xianjian, ZHANG Huanling, YAN Hongbo, LI Zhenbao, GUO Ziyang

表3 甘蔗种植区地物分类精度统计

Tab.3 Statistics of ground object classification accuracy in sugarcane planting area

地物类型 方案1 方案2 方案3 方案4 方案5 方案6 UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% UA/% PA/% 香蕉 86.36 80.28 96.96 82.05 93.24 76.66 92.98 79.10 94.52 88.46 94.44 98.55 水稻 98.38 88.40 98.55 94.44 97.22 94.59 98.24 93.33 98.11 91.22 100.00 96.15 甘蔗 90.15 93.42 90.54 95.80 91.45 97.96 91.88 96.94 94.75 97.15 97.59 98.61 裸地 86.11 93.93 95.77 94.44 97.05 91.66 96.47 88.17 93.75 93.75 92.50 97.36 林地 83.91 82.47 80.45 86.34 85.65 88.64 86.84 89.59 88.28 88.97 92.48 89.95 水体 97.88 99.28 98.63 97.29 98.61 97.26 98.70 98.06 100.00 97.93 100.00 97.45 建设用地 88.41 86.82 93.82 85.87 92.90 90.00 90.37 88.94 92.46 93.40 91.71 93.25 OA/% 89.53 91.33 92.21 92.27 93.61 95.49 Kappa 0.87 0.89 0.90 0.90 0.92 0.94