Towards Proactive Surveillance through CCTV Cameras under Edge-Computing and Deep Learning

Table 1

Yolo results table.

	Epochs	Precision	Recall	F1 score	Mean avg precision (@50%)	Avg. intersection over union (@50%)
Yolo-v4	1000	75	81	80	85.3	56.82
(Loss 0.6)	2000	85	82	86	88.13	67.78
	3000	90	84	87	90.22	71.18
Yolo-v4 (csp)	1000	74	75	74	73.85	54.26
Loss (6.24)	2000	79	86	82	86.64	62.75
	3000	76	88	81	87.69	61.22
Yolo-v4 (tiny)	1000	56	54	55	54.56	37.59
(Loss 13.6)	2000	82	75	79	82.14	62.02
	3000	79	81	80	85.01	60.1