Abstract

This paper is concerned with the effective evaluation of multimodal biometric database and discusses data\ncollection method with testing protocols, for Electrocardiogram (ECG), face and fingerprint biometrics of 50 individuals.\nDatabase of biometric images, and the numerical modules or templates derived from them, are often encrypted with the\nintension of inhibiting their compromise in bulk. But compromise of individual measures cannot always be prevented by\nprotecting database and transmission channels because biometric measures, although privately owned, are sometimes\npublicly observable (e.g. a photo of a person’s face can be taken with a camera or downloaded from a web page.). In general,\nbiometric measures are not secret, even if it might be quiet complicated to acquire usable copies (e.g. ECG map) without\ncooperation of the person when used for security and privacy, biometric characteristics are more like public keys than private\nkeys. Unlike public keys, however, biometric measures cannot be revoked if stolen or mimicked. Hence Multi-modal\nbiometrics has gained considerable attention because they enhance the accuracy of biometric person authentication as well as\nsecurity against attack. Combinations of such multi-modal biometrics are diverse: fingerprint and iris; iris and face ; face and\nocular ; face and gait ; gait and the height . In this research paper, we proposed an alternative method for database privacy to\nauthenticate a person for live-ness testing using multimodal biometrics such as ECG-Finger Print-Face in a reliable way as\ncompared to unimodal biometric traits.