基于多源遥感时序特征和卷积神经网络的露天矿区土地利用分类
刘昊, 杜守航, 邢江河, 李军, 高天琳, 尹程弘

Land use classification of open-pit mining areas based on multi-source remote sensing time series features and convolutional neural networks
LIU Hao, DU Shouhang, XING Jianghe, LI Jun, GAO Tianlin, YIN Chenghong
表3 ResNet50的有效性
Tab.3 Effectiveness of ResNet50
指标 OA/% Kappa系数
ResNet50 95.36 0.942 1
RF 94.08 0.926 4
ANN 86.64 0.833 6
KNN 92.92 0.912 2
SVM 91.52 0.894 8
贝叶斯 77.35 0.721 9