Table 6 Performance comparison of the baseline and our proposed model on the hold-out set of 75 authentic defect images.
From: Balancing complexity and accuracy for defect detection on filters with an improved RT-DETR
Model | Precision (%) | Recall (%) | mAP@0.5 (%) | mAP@0.5:0.95(%) |
|---|---|---|---|---|
RT-DETR (baseline) | 93 | 92.2 | 92.65 | 68.27 |
RT-DETR-LGA (ours) | 97.1 | 97.9 | 98.7 | 78.11 |
Performance gain | +4.1 | +5.9 | +6.05 | +9.84 |
