Multi-normalization
S. M. Anzar, Puthumangalathu Savithri Sathidevi · 2012
The effectiveness of a multibiometric system can be improved by modifying the scores obtained from the degraded modalities in an appropriate manner. Score normalization is a promising method to improve the system performance. In this paper, we propose a new method entitled as multi-normalization based fusion, to improve the robustness and efficiency of a multibiometric system, under various noise conditions. As the match score values from the individual matchers follow nonhomogeneous statistical distributions. we propose different score normalization techniques for the complementary modalities employed. The performance of the proposed technique is analysed in the context of fingerprint and voice biometrics using sum rule of fusion. Experimental results show that this fusion strategy performs considerably better, than the baseline systems. The biometric solution that we provided could be easily integrated with any multibiometric system with score level fusion.