Table 4 Comparison of the performance of different models on the deepcrack.
From: An efficient semantic segmentation method for road crack based on EGA-UNet
Method | Dice(%) | IoUb(%) | IoUf(%) | mIoU(%) | Params(M) | FLOPs(G) | FPS |
|---|---|---|---|---|---|---|---|
FCN | 67.7 | 97.4 | 57.7 | 77.6 | 134.27 | 197.38 | 41 |
U-Net | 79.2 | 98.5 | 71.4 | 84.9 | 4.37 | 32.29 | 91 |
SegNet | 79.8 | 98.5 | 69.9 | 84.5 | 29.48 | 135.79 | 46 |
PSPNet | 77.5 | 98.3 | 68.6 | 83.4 | 46.59 | 141.54 | 43 |
DeepLabV3 | 78.9 | 98.4 | 70.1 | 84.2 | 73.22 | 130.82 | 49 |
EGA-UNet | 82.6 | 98.7 | 74.0 | 86.4 | 4.73 | 31.12 | 95 |
