Biometric Jammer: A Security Enhancement Scheme using SVM Classifier

Abstract

A new privilege of biometrics help to reduce the stress of user, Which comes along with the traditional access methods of passwords and token. Using the biometrics limitations and weaknesses can be knocked out. However, biometrics has raise privacy risks and new security since they cannot be easily revoked. Due to the spoofing attack on biometrics. Thus, to protect biometric traits against spoofing attack a multimodal biometric jammer scheme for the security enhancement have been developed and suggested in this paper. Firstly, we analyze why the multimodal biometric system have attracted attention for high security-demanding schemes. Secondly, security of biometric system is increasing and prevented it from spoofing attack developing a machine learning system model. We show that these machine learning algorithms perform pre-processing of biometric traits images. Further we analyze user identification with the increase precision and reliability using biometric features. Where feature extraction of each one trait of biometric is done and then all features are concatenation to get a single feature. With the aid of machine learning classifier using extracted features the algorithm predict the result of the system.