Machine and Deep learning in Biometric Authentication: A Review

Abstract

Biometric authentication is getting privileged as it allows authentication of a legitimate user without entering a personal identification number or password. Only a glance of any physical characteristics at a camera proves the user’s identity making it secure from shoulder suffering attacks and theft. This review will explain the various biometric authentication techniques available for user authentication. Biometric authentication is generally based on two categories 1. Based on physical traits 2. Based on behavioral traits. A comprehensive comparison of all methods is discussed with their merits and demerits. Based on the comparison, we will provide insight into how machine learning algorithms help develop biometric authentication systems. We will also explore the various applications based on biometric authentication.