Ameliorating the Accuracy and Mitigating the Error rate of Multimodal biometrics using Minimum Cost Matcher and Autoencoder

In present era, enhancing the fast and accurate recognition of an individual is becoming the prominent role. Multimodal biometrics addresses different vulnerabilities of single modal biometrics and achieves the better recognition accuracy. Multimodal biometrics gains the attention of fusion of two or more traits for accurate recognition of an individual. Intruders face the difficult situation in hacking the multimodal biometric template as it gives the hard time for intruders to gain the information simultaneously. This fusion helps to enhance recognition rate of an individual and also achieves the safe strategy. Eye biometric is adaptable for providing security in large scale applications. This paper focuses on enhancing fast and accurate recognition of an individual by mitigating the recognition error rate. This paper addresses the feature level fusion of retina and iris of an individual. These biometric features are extracted through Gray level co-occurrence matrix and the deep learners of unsupervised technique Autoencoder. Minimum Cost Matching method enhances the accuracy rate of individual with a minimal error rate. MCM achieves the high matching scores when it is compared with the various traditional similarity and dissimilarity measures.