An Efficient Score level Multimodal Biometric System using ECG and Fingerprint

Abstract

Biometric system is a security system that uses human’s unique traits to identify and authenticate the user. Biometrics refers to biological traits of a human that are often categorized as physiological traits like fingerprint, iris, face and behavioral characteristics like signature style, voice and typing rhythm. The Biological signals like Electrocardiography (ECG), Electromyography(EMG), and Electroencephalography (EEG) have not been explored to biometric applications as their scope was limited to medical applications only. Recent survey suggests that these biological signals can be explored as a part of the biometric application. The main objective of this paper is to explore the possibility of using the ECG as a part of multimodal biometric. ECG has lower accuracy but fusing it with a traditional biometric like fingerprint yields a higher accuracy rate and it is really difficult to spoof the system. The proposed multimodal biometrics system has an accuracy of 98% with the false acceptance rate of 2% and almost 0% of false rejection rate.