To search, Click below search items.


All Published Papers Search Service


Fusion of Face and Voice: An improvement


Y. M. Fouda


Vol. 12  No. 4  pp. 37-43


Multimodal biometric systems provide a better recognition performance compared to systems based on a single biometric modality. The fusion technique is necessary and effective for combing information in multimodal biometric system. In this paper, we propose a new fusion technique based on simple sum rule and product rule. The double sigmoid normalization technique is used after adjusting the parameter t to give us better results. The proposed fusion scheme is compared with some fusion schemes such as sum rule, product rule, maximum rule and minimum rule. Experimental are carried out on two different build multimodal biometric databases. Experimental results indicate that proposed fusion scheme achieves higher performance as compared with other fusion techniques.


Multimodal biometric, score fusion, normalization, face, voice