CNN-Based Personal Identification System Using Resting State Electroencephalography

Table 3

Comparison of the performance of the proposed personal identification systems with 14-, 32-, and 64-channel EEG signals (positions of the electrodes are shown in Figure 5).

Session	Channels	Rank-1 (%)	FRR (%)	FAR (%)	EER (%)
EO	14	99.04 ± 0.95	0.25 ± 0.21	0.25 ± 0.21	0.25 ± 0.21
	32	99.29 ± 0.81	0.19 ± 0.16	0.19 ± 0.16	0.19 ± 0.16
	64	99.29 ± 0.85	0.21 ± 0.22	0.21 ± 0.22	0.21 ± 0.22
EC	14	99.11 ± 0.85	0.18 ± 0.15	0.19 ± 0.16	0.19 ± 0.15
	32	99.31 ± 0.90	0.15 ± 0.23	0.17 ± 0.24	0.16 ± 0.23
	64	99.44 ± 0.75	0.16 ± 0.19	0.16 ± 0.19	0.16 ± 0.19
EO&EC	14	99.32 ± 0.60	0.18 ± 0.15	0.18 ± 0.15	0.18 ± 0.15
	32	99.64 ± 0.35	0.09 ± 0.06	0.09 ± 0.06	0.09 ± 0.06
	64	99.78 ± 0.23	0.06 ± 0.08	0.07 ± 0.08	0.07 ± 0.08