基于多源时序SAR数据的涿州洪涝淹没动态监测

庄会富, 王鹏, 苏亚男, 张祥, 范洪冬

Dynamic monitoring of flood inundation in Zhuozhou, Hebei Province based on multi-temporal SAR data

ZHUANG Huifu, WANG Peng, SU Yanan, ZHANG Xiang, FAN Hongdong

表3 对比方法精度评估

Tab.3 Accuracy assessment of comparative methods

方法 Kappa F1 OA P R 验证集Ⅰ KI 0.300 6 0.453 6 0.751 2 0.378 7 0.565 4 OTSU 0.040 6 0.268 1 0.618 8 0.206 5 0.382 4 K-means 0.040 6 0.268 1 0.618 8 0.206 5 0.382 4 LR 0.438 4 0.510 1 0.862 5 0.729 6 0.392 1 MR 0.772 3 0.809 2 0.937 4 0.912 9 0.726 6 DDNet 0.679 0 0.725 8 0.916 6 0.908 4 0.604 4 PNLI 0.834 8 0.866 1 0.949 1 0.833 1 0.901 9 本文方法T1→T2 0.853 5 0.882 0 0.953 8 0.827 0 0.944 8 本文方法T2→T1 0.859 2 0.886 5 0.955 8 0.835 3 0.944 3 验证集Ⅱ KI 0.719 2 0.758 0 0.931 4 0.654 8 0.899 7 OTSU 0.719 2 0.758 0 0.931 4 0.654 8 0.899 7 K-means 0.719 2 0.758 0 0.931 4 0.654 8 0.899 7 LR 0.727 0 0.763 2 0.936 5 0.687 3 0.857 9 MR 0.788 3 0.813 1 0.956 3 0.829 3 0.797 6 DDNet 0.717 8 0.747 5 0.946 0 0.846 1 0.669 6 PNLI 0.795 6 0.820 1 0.956 9 0.816 1 0.824 1 本文方法T1→T2 0.843 2 0.862 2 0.966 5 0.847 5 0.877 4 本文方法T2→T1 0.844 5 0.863 8 0.966 2 0.833 1 0.896 8