Research Article
Attention-Based Convolutional Neural Network for Pavement Crack Detection
Table 6
Yantai Dataset–CE loss.
| | Model | Precision/(%) | Mean IoU/(%) | Recall/(%) | F1/(%) |
| | ENet | 90.88 | 81.62 | 92.22 | 91.55 | | ExFuse | 91.76 | 82.27 | 90.64 | 91.20 | | FCN | 91.56 | 82.52 | 90.80 | 91.18 | | LinkNet | 91.34 | 82.48 | 92.11 | 91.72 | | SegNet | 91.73 | 81.98 | 89.73 | 90.72 | | UNet | 93.54 | 78.63 | 89.24 | 91.34 |
| | CrackResAttentionNet | 94.64 | 83.28 | 92.05 | 93.33 |
|
|