融合混合注意力机制与多尺度特征增强的高分影像建筑物提取

曲海成, 梁旭

Building extraction from high-resolution images using a hybrid attention mechanism combined with multi-scale feature enhancement

QU Haicheng, LIANG Xu

表6 马萨诸塞州建筑数据集定量对比实验结果

Tab.6 Quantitative comparison experiment results of Massachusetts Buildings Dataset

方法 IoU/% F1/% Precision/% Recall/% Param/MB VGG16-U-net 66.92 80.65 79.15 82.21 18.26 DeepLab V3+ 69.16 81.38 83.85 79.05 30.61 BANet 72.20 83.86 83.07 84.66 38.72 ST-UNet 71.26 — 86.49 — 26.47 BMNet 75.79 85.58 87.02 84.19 21.34