Table 1.

Accuracy, EER, and AUC of discriminative classification models according to feature selections.

		Accuracy (%)				EER (%)				AUC
	Feature	Finger	Ratio	u-Limb	Ratio + u-Limb	Finger	Ratio	u-Limb	Ratio + u-Limb	Finger	Ratio	u-Limb	Ratio + u-Limb
Classifier
RF		70.83	88.21	89.98	96.43	7.5460	2.1570	3.0275	0.6050	0.9793	0.9974	0.9968	0.9998
LDA		72.53	88.75	87.52	96.00	6.6380	3.0731	3.0117	1.3022	0.9826	0.9930	0.9947	0.9967
kNN		60.66	87.46	74.86	96.31	20.2827	6.4651	12.9619	1.9015	0.7972	0.9353	0.8704	0.9810
SVM		63.55	85.74	75.29	92.99	12.7228	3.7646	6.2097	2.7668	0.9500	0.9938	0.9872	0.9970
CNN		77.62	94.40	95.70	97.60	5.0524	1.1063	0.9834	0.4917	0.9891	0.9990	0.9993	0.9999