Multimodal biometric traits protection using dynamic graphs and modified cipher techniques

Abstract

Human identification is one of the prime knowledge to provide access to all kind of applications automatic user recognition could be implemented using everlasting biometric traits biometrics consists of discriminative bodily and behavioral attributes which facilitate to authenticate whether a particular individual is authorized or not fingerprint palm print face and iris are some of the instances of inborn biometric modalities that provide maximum user convenience the earlier user authentication systems employed passwords tokens smart cards and personal identification numbers for verification purposes and were largely affected by spoof attacks brute force attacks dictionary attacks and key stolen attacks biometrics overcome several drawbacks of the traditional alidation systems further digital biometric evidence could not be shared with others could not be lost or forgot by the user and there is no need to remember anything however if the biometric details are stolen by an intruder that leads to the permanent loss of the identity since biometrics are irrevocable hence template security is the major concern in the biometric research community to ensure the privacy and safety of an individual without affecting matching accuracy newline