Research Article
LE-YOLOv5: A Lightweight and Efficient Road Damage Detection Algorithm Based on Improved YOLOv5
Table 3
Neck network ablation experiment.
| Models | Params (M) | GFLOPS | Precision (%) | Recall (%) | mAP0.5 (%) | F1-score (%) |
| Baseline | 4.37 | 9.9 | 57.2 | 52.1 | 52.0 | 54.4 | Baseline-GSConv | 4.03 | 9.3 | 58.6 | 51.7 | 52.3 | 54.9 | Baseline-GSConv-VCACSP | 3.38 | 6.9 | 59.6 | 52.5 | 53.6 | 55.8 |
|
|