Table 9 Comparisons between our proposed model and the other YOLO models.
From: Lightweight enhanced YOLOv8n underwater object detection network for low light environments
Model | Parameter/M | GFlops | mAP@0.5 | mAP@0.5:0.95 | FPS |
|---|---|---|---|---|---|
Faster-RCNN | 13.72 | 18.12 | 67.4 | 44.3 | 75.6 |
YOLOv3 | 12.44 | 28.91 | 69.6 | 46.8 | 68.9 |
YOLOv5s | 26.58 | 16.80 | 73.8 | 54.7 | 305.8 |
YOLOv7 | 35.69 | 103.22 | 74.9 | 56.5 | 215.7 |
YOLOv8n | 6.65 | 8.71 | 79.6 | 58.2 | 318.7 |
YOLOv9-c | 25.3 | 108.3 | 84.4 | 60.7 | 197.9 |
YOLOv10n | 7.86 | 17.6 | 85.9 | 56.6 | 286.8 |
Our | 5.62 | 8.96 | 86.1 | 60.8 | 297.8 |
