On combining multi-normalization and ancillary measures for the optimal score level fusion of fingerprint and voice biometrics

EURASIP Journal on Advances in Signal Processing

Table 1 Estimation of ancillary measures and the optimal integration weights

No.	SNR	Integration weight estimation with ancillary measures
	(dB)	ρ	λ	d ^′	γ	GSx_opt	GAx_opt	PSOx_opt	GSβ	GAβ	PSOβ
1	20	0.6358	0.7421	0.7059	0.6946	1.0000	0.9872	0.9822	0.6946	0.6857	0.6822
2	15	0.6479	0.7086	0.6786	0.6784	1.0000	0.9613	0.9043	0.6784	0.6521	0.6134
3	10	0.6128	0.7035	0.6386	0.6516	0.8200	0.8848	0.8841	0.5343	0.5766	0.5761
4	5	0.6033	0.6875	0.5899	0.6269	0.6500	0.7596	0.8649	0.4075	0.4762	0.5422
5	0	0.4886	0.6720	0.5159	0.5588	0.5000	0.6670	0.8052	0.2794	0.3727	0.4500
6	-5	0.3236	0.6518	0.4136	0.4630	0.5000	0.6632	0.7712	0.2315	0.3071	0.3571
7	-10	0.1855	0.6365	0.2210	0.3477	0.5000	0.4487	0.7163	0.1738	0.1560	0.2490