Table 3 Comparison results with other methods on the potsdam dataset
From: GLE-net: global-local information enhancement for semantic segmentation of remote sensing images
Method | IoU\(\%\) | Evaluation Index\(\%\) | ||||||
|---|---|---|---|---|---|---|---|---|
Imp.surf. | Buliding | Lowveg. | Tree | Car | MIoU | Mean.F1 | OA | |
FCN | 83.48 | 89.45 | 75.56 | 74.42 | 82.55 | 86.51 | 79.06 | 81.11 |
UNet | 83.56 | 86.30 | 75.92 | 75.01 | 81.75 | 87.03 | 77.35 | 83.06 |
Deeplab V3+ | 85.73 | 89.95 | 76.05 | 74.58 | 84.92 | 87.68 | 78.97 | 83.85 |
PSPNet | 86.22 | 87.02 | 75.94 | 76.13 | 86.36 | 87.09 | 81.99 | 83.26 |
Trans-UNet | 85.98 | 85.15 | 76.05 | 75.14 | 87.25 | 88.28 | 82.56 | 82.97 |
Swin-UNet | 84.77 | 79.98 | 76.02 | 76.33 | 87.47 | 87.24 | 83.04 | 83.52 |
GLE-Net | 87.43 | 89.89 | 77.53 | 77.83 | 87.46 | 87.33 | 85.55 | 86.04 |
