Table 4.

Performance and parameters of base YOLOv7 and YOLOv8 models on the validation dataset and test dataset.

Model	Precision val | test	Recall val | test	mAP50 val | test	mAP50-95 val | test	Parameters
YOLOv7	0.964 | 0.978	0.941 | 0.889	0.955 | 0.926	0.716 | 0.718	37.21 M
YOLOn8n	0.898 | 0.910	0.809 | 0.828	0.844 | 0.821	0.562 | 0.577	3.01 M
YOLOv8s	0.932 | 0.914	0.817 | 0.832	0.886 | 0.841	0.599 | 0.661	11.14 M
YOLOv8m	0.919 | 0.912	0.865 | 0.868	0.916 | 0.916	0.702 | 0.721	25.86 M
YOLOv8l	0.929 | 0.904	0.870 | 0.875	0.924 | 0.915	0.689 | 0.711	43.63 M
YOLOv8x	0.938 | 0.918	0.872 | 0.847	0.923 | 0.910	0.694 | 0.723	68.16 M

Bold values signify the value of the best performing base model under each metric.