Recent Advancement in Cancelable Biometric for User Recognition: A Brief Survey

Abstract

Surreptitious duplication of biometric features poses a major threat to privacy and anonymity of an individual. Cancelable Biometrics-an emerging field of biometric template protection scheme can diminish these privacy threats. Generally cancelable Biometric approach transforms the original biometric data using non-invertible transformation function to generate pseudo-identities or cancelable templates. In contrast, application of invertible transformation function in conjunction with key can also generate cancelable templates. Mostly, cancelable templates are revocable and computationally complex to reverse. The necessity of preserving privacy and maintaining high recognition accuracy compulsorily needs storage and matching of transformed biometric data. In order to determine the research gap and future direction in the field of cancelable biometrics, in this survey, we have analysed and reviewed the recent work in field of cancelable biometric. For this, cancelable biometric approaches are classified into invertible and noninvertible methods. Also, the comparative analysis of cancelable approach is tabulated. The research gaps and future directions are also summarised.