We examine the efficiency of four machine learning algorithms for the fusion of several biometrics modalities to create a\r\nmultimodal biometrics security system. The algorithms examined are Gaussian Mixture Models (GMMs), Artificial Neural\r\nNetworks (ANNs), Fuzzy Expert Systems (FESs), and Support VectorMachines (SVMs). The fusion of biometrics leads to security\r\nsystems that exhibit higher recognition rates and lower false alarms compared to unimodal biometric security systems. Supervised\r\nlearning was carried out using a number of patterns from a well-known benchmark biometrics database, and the validation/testing\r\ntook place with patterns from the same database which were not included in the training dataset. The comparison of the algorithms\r\nreveals that the biometrics fusion system is superior to the original unimodal systems and also other fusion schemes found in the\r\nliterature.
Loading....