Table 2 Comparison results for baseline model selection.
From: Metal surface defect detection using SLF-YOLO enhanced YOLOv8 model
Dataset | Metric | YOLOv8n | YOLOv8s |
|---|---|---|---|
NEU-DET | P(%) | 76.2 | 75.1 (↓ 1.1) |
R(%) | 71.2 | 71.8 (↑ 0.6) | |
mAP50(%) | 76.6 | 75.9 (↓ 0.7) | |
mAP50:95(%) | 43.0 | 41.1 (↓ 1.9) | |
AL10-DET | P(%) | 82.7 | 83.4 (↑ 0.7) |
R(%) | 76.3 | 80.1 (↑ 3.8) | |
mAP50(%) | 80.6 | 83.0 (↑ 2.4) | |
mAP50:95(%) | 63.3 | 65.3 (↑ 2.0) | |
Param(M) | 3.01 | 11.12 | |
Flops(G) | 8.1 | 28.5 |
