An ECG classifier designed using modified decision based neural networks
Computers and Biomedical Research
Large-Scale Evaluation of Multimodal Biometric Authentication Using State-of-the-Art Systems
IEEE Transactions on Pattern Analysis and Machine Intelligence
Pattern Recognition
Correlation-based classification of heartbeats for individual identification
Soft Computing - A Fusion of Foundations, Methodologies and Applications - Special issue on Digital Information Forensics
A taxonomy of biometric system vulnerabilities and defences
International Journal of Biometrics
Evaluating the use of ECG signal in low frequencies as a biometry
Expert Systems with Applications: An International Journal
Multimodal biometric system combining ECG and sound signals
Pattern Recognition Letters
Hi-index | 0.10 |
This paper explores the effectiveness of a novel multibiometric system that is resulted from the fusion of the electrocardiogram (ECG) with an unobtrusive biometric face and another biometric fingerprint which is known to be a least obtrusive for efficient individual authentication. The unimodal systems of the face and the fingerprint biometrics are neither secure nor they can achieve the optimum performance. Using the ECG signal as one of the biometrics offer advantage to a multibiometric system that ECG is inherited to an individual which is confidential, secured and difficult to be forged. It has an inherent feature of vitality signs that ensures a strong protection against spoof attacks to the system. Transformation based score fusion technique is used to measure the performance of the fused system. In particular, the weighted sum of score rule is used where weights are computed using equal error rate (EER) and match score distributions of the unimodal systems. The performance of the proposed multibiometric system is measured using EER and receiver operating characteristic (ROC) curve. The results show the optimum performance of the multibiometric system fusing the ECG signal with the face and fingerprint biometrics which is achieved to an EER of 0.22%, as compared to the unimodal systems that have the EER of 10.80%, 4.52% and 2.12%, respectively for the ECG signal, face and fingerprint biometrics.