用于无人机遥感图像的高精度实时语义分割网络
|
|
魏新雨,饶蕾,范光宇,陈年生,程松林,杨定裕
|
High-precision real-time semantic segmentation network for UAV remote sensing images
|
|
Xinyu WEI,Lei RAO,Guangyu FAN,Niansheng CHEN,Songlin CHENG,Dingyu YANG
|
|
| 表 7 Potsdam测试集上分割算法的性能比较 |
| Tab.7 Performance comparison of segmentation algorithms on Potsdam test set % |
|
| 算法 | IoU | mIoU | | 不透水表面 | 建筑物 | 低植被 | 树木 | 车辆 | | FCN 8S | 86.1 | 91.5 | 76.3 | 77.4 | 90.7 | 84.4 | | BiSeNet | 87.2 | 91.7 | 76.9 | 78.9 | 91.0 | 85.1 | | PSPNet | 87.9 | 92.1 | 77.1 | 79.7 | 91.5 | 85.7 | | DeepLab V3+ | 87.7 | 92.2 | 77.4 | 79.4 | 91.2 | 85.5 | | BANet | 88.9 | 92.7 | 78.2 | 79.8 | 91.4 | 86.2 | | CoaT | 88.5 | 92.9 | 78.5 | 80.4 | 91.9 | 86.1 | | SegFormer | 87.9 | 92.6 | 78.8 | 79.9 | 91.6 | 86.0 | | DC-Swin | 88.7 | 93.3 | 78.0 | 80.1 | 92.0 | 86.5 | | UNetFormer | 89.0 | 93.5 | 79.2 | 80.5 | 91.9 | 86.8 | | MobileViT V3 | 88.9 | 93.7 | 79.0 | 80.2 | 92.1 | 86.8 | | RSSFormer | 89.1 | 93.7 | 79.8 | 80.4 | 91.7 | 86.9 | | SSFNet | 89.3 | 93.9 | 79.1 | 80.9 | 92.2 | 87.1 |
|
|
|