Fused Iris Biometrics for Person Identification

Abstract: The conventional authentication methods such as proxy based and biometrics based are not user centric and endangers security and privacy, system provides automatic identification of an individual based on a unique feature or characteristic possessed by the individual. Iris authentication is a biometric modality which can be used to identify a person. In this paper, proposes a method in which a reference subject (RS) is securely fused with user's biometrics, developing a Biocapsule (BC) from the fused biometrics for authentication. Selection of the reference subject can be a physical RS or a logical RS. The pre-processing techniques are done before the fusion process. Features of user biometrics and RS biometrics are extracted. Individual features make the proposed BC mechanism a user-centric authentication approach. Fusing the transformed user biometrics and RS biometrics, a BC is developed from fused biometrics. The generated BC is matched with BC in the database which is already stored during the enrolment stage. If the matching is true, user identity is correct else wrong identification. Fusion aims to increase the security of the biometrics because through the fusion, user biometrics is hidden by RS biometrics, thus provides secure biometric that is privacy preserving. BC carries no hints that the user is weighted more than the RS and equally treats user and RS. This biometric system possesses various properties such as security, privacy preservation, cross matching resistance, etc. Keywords: Authentication, privacy-preserving, Biocapsule (BC), secure fusion, Reference subject (RS).