Table 3 Performance comparison of object detection models.
From: YOLO11-WLBS: an efficient model for pavement defect detection
Models | P | R | F1 | mAP@0.50 | mAP@0.50–0.95 | Para/M |
|---|---|---|---|---|---|---|
SSD | 0.572 | 0.388 | 0.462 | 0.531 | 0.332 | 32.77 |
Faster R-CNN | 0.636 | 0.451 | 0.528 | 0.598 | 0.391 | 26.21 |
Mask R-CNN | 0.659 | 0.486 | 0.560 | 0.611 | 0.403 | 25.32 |
YOLOv5 | 0.763 | 0.524 | 0.621 | 0.731 | 0.549 | 22.18 |
YOLOv8 | 0.823 | 0.624 | 0.714 | 0.775 | 0.619 | 23.27 |
YOLOv9 | 0.832 | 0.625 | 0.714 | 0.765 | 0.613 | 16.78 |
YOLOv10 | 0.831 | 0.632 | 0.718 | 0.769 | 0.596 | 16.58 |
YOLO11 | 0.853 | 0.698 | 0.768 | 0.797 | 0.624 | 20.06 |