基于双树复小波分解的Boosting集成学习土地覆被分类研究
李润祥,高小红,汤敏

Study on Boosting Ensemble Learning Land Cover Classification based on Dual-Tree Complex Wavelet Transform
Runxiang Li,Xiaohong Gao,Min Tang
表3 Sentinel-2A影像分类精度
Table 3 Classification accuracy of Sentinel-2A image
土地覆被类型RFGBDTDTCWT-GBDTXGBoostDTCWT-XGBoostLightGBMDTCWT-LightGBM
用户精度/%制图精度/%用户精度/%制图精度/%用户精度/%制图精度/%用户精度/%制图精度/%用户精度/%制图精度/%用户精度/%制图精度/%用户精度/%制图精度/%
耕地82.6381.2185.0683.2686.6383.6287.7586.3288.8486.9889.6489.0190.9686.68
有林地91.5490.1284.5080.2390.5192.3292.6392.7693.6992.4593.6893.8693.8492.69
灌木林地86.2388.2387.3688.2687.5489.3188.2389.3688.3689.3589.4589.2389.8590.21
疏林地、其他林地90.1293.6390.1588.4592.3591.9493.3291.8892.8893.4594.6395.5496.5695.34
高覆盖草地84.2690.2384.6385.6684.7684.5585.4683.5286.6584.7886.8183.6587.8286.45
中覆盖草地88.2380.5290.3682.7793.4285.6391.3886.2392.6691.5693.3691.3995.3592.36
低覆盖草地83.1284.3385.4281.5686.0685.4386.8284.6787.2686.4587.8284.8288.3685.42
城乡工矿居住90.2985.3691.6592.3491.7891.2391.9391.6292.1693.4593.3594.2694.4295.32
建设用地
河流94.7595.2695.3196.3696.5696.2396.7696.6796.5396.4895.6996.2696.5995.36
水库坑塘96.1193.1697.4693.2397.3295.1397.1597.2398.3695.0698.3399.1499.2198.36
未利用土地80.1280.0783.5381.2684.2182.1484.6883.4787.9382.6388.4585.3689.1887.66
总体精度/%87.8589.1991.2390.5892.4691.7293.25
Kappa0.860.870.900.880.890.900.91