Table 2.

Tuned hyperparameters used in each YOLO model.

Parameter	YOLOv7-tiny	YOLOv8s	YOLOv9-S	YOLOv10-S
Input Size	640 × 640 px	640 × 640 px	640 × 640 px	640 × 640 px
Batch Size	16	16	16	16
Initial Learning Rate	0.01	0.01	0.01	0.01
Final Learning Rate	0.0002	0.0002	0.0002	0.002
Optimizer	AdamW	AdamW	AdamW	AdamW
Cos_lr Function	False	True	True	True
Cls_mosaic Parameter	-	50	50	50