Table 4 Detection performance on environmental dataset.
From: An intelligent YOLO and CNN-BiGRU framework for road infrastructure based anomaly assessment
Model | mAP@0.5 (%) | mAP@0.5:0.95 (%) | Avg. AUROC | F1-score |
|---|---|---|---|---|
Mask R-CNN | 88.1 | 65.7 | 0.929 | 0.874 |
Detectron2 | 90.3 | 67.9 | 0.941 | 0.891 |
YOLOv7 | 89.5 | 67.2 | 0.943 | 0.889 |
YOLOv8 | 91.7 | 69.5 | 0.951 | 0.902 |
YOLOv9 | 93.3 | 71.8 | 0.961 | 0.915 |
YOLOv10 | 94.6 | 74.1 | 0.968 | 0.927 |
YOLOv11 | 95.4 | 76.3 | 0.971 | 0.938 |
Proposed model | 96.9 | 78.9 | 0.985 | 0.949 |
