Fusion of Biometrics Based on D-S Theory

Human face and voice are often used biometric characteristics to establish personal identity. While an automatic personal identification system based sorely on faces or voices is often not able to meet the system performance requirements when the environment light changes or there exists environment noises. We have developed a personal identification system integrating human faces and voices. Experimental results show that the recognition accuracy of the fusion system is much higher than the single channel. D-S theory is one of the method often used in integrating multiple clues. It's essence is to undraw the basic probability of each class after fusion.