Table 7 Results summary of proposed model for normal-case and hard-case test dataset category-wise.

Train Dataset	Test Dataset	Categories	Images	Instances	Precision	Recall	mAP50	mAP50-90
train ( normal Dataset \(\cup\) hard Dataset)	test normal Dataset	All	816	816	0.88	0.89	0.92	0.77
		Blood	83	83	0.83	0.81	0.87	0.60
		Gun	149	149	0.95	0.92	0.96	0.84
		Insulting Gesture	69	69	0.97	0.97	0.99	0.83
		Knife	354	354	0.91	0.82	0.89	0.76
		Alcohol	79	79	0.71	0.91	0.90	0.81
		Cigarette	82	82	0.91	0.89	0.94	0.79
	test hard Dataset	All	722	2145	0.84	0.74	0.82	0.62
		Blood	149	455	0.77	0.57	0.69	0.47
		Gun	84	135	0.87	0.67	0.81	0.60
		Insulting Gesture	40	62	0.92	0.89	0.93	0.75
		Knife	123	224	0.82	0.66	0.77	0.62
		Alcohol	146	586	0.87	0.76	0.84	0.67
		Cigarette	230	683	0.86	0.79	0.85	0.65

Quick links

Search