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 5 UAVid测试集上分割算法的性能比较 |
| Tab.5 Performance comparison of segmentation algorithms on UAVid test set % |
|
| 算法 | IoU | mIoU | | 杂物 | 建筑物 | 道路 | 树木 | 植被 | 移动车辆 | 静止车辆 | 行人 | | FCN 8S | 63.6 | 84.6 | 76.1 | 77.6 | 60.2 | 62.3 | 47.1 | 14.5 | 60.8 | | BiSeNet | 64.7 | 85.7 | 77.3 | 78.3 | 61.1 | 63.4 | 48.6 | 17.5 | 62.1 | | PSPNet | 65.4 | 85.7 | 79.5 | 79.2 | 61.5 | 72.6 | 49.4 | 19.4 | 64.1 | | DeepLab V3+ | 65.3 | 86.1 | 80.0 | 78.1 | 60.3 | 71.4 | 49.1 | 21.8 | 64.0 | | BANet | 66.7 | 85.4 | 80.7 | 78.9 | 62.1 | 69.3 | 52.8 | 21.0 | 64.6 | | CoaT | 69.0 | 88.5 | 80.0 | 79.3 | 62.0 | 70.0 | 59.1 | 18.9 | 65.8 | | SegFormer | 66.6 | 86.3 | 80.1 | 79.6 | 62.3 | 72.5 | 52.5 | 28.5 | 66.0 | | DC-Swin | 67.5 | 86.5 | 80.4 | 80.1 | 61.9 | 72.2 | 54.3 | 27.3 | 66.3 | | UNetFormer | 68.4 | 87.4 | 81.5 | 80.2 | 63.5 | 73.6 | 56.4 | 31.0 | 67.8 | | MobileViT V3 | 67.7 | 87.3 | 81.4 | 80.1 | 63.6 | 73.8 | 54.8 | 29.7 | 67.3 | | RSSFormer | 67.2 | 86.8 | 81.1 | 79.9 | 63.3 | 72.5 | 58.4 | 30.8 | 67.8 | | SSFNet | 68.7 | 88.4 | 81.5 | 80.5 | 63.9 | 77.1 | 56.7 | 31.5 | 68.5 |
|
|
|