Common biometric systems like fingerprint and face recognition are more convenient in daily applications, but biological and behavioral characteristics of the biometrics features can be fabricated and digitally stolen. Thus, biometrics features with liveness detection such as the electrocardiogram (ECG) have been introduced as its features are hidden and difficult to forge. This study presents a review of ECG biometrics based on deep learning and generalization issues in deep learning. Based on the review, deep learning methods such as recurrent neural networks (RNN) and long short-term memory networks (LSTM) with attention mechanisms can be employed to improve the performance and generalization ability of ECG biometrics systems.
A Review of ECG Biometrics: Generalization in Deep Learning with Attention Mechanisms
Lect. Notes Electrical Eng.
International Conference on Robotics, Vision, Signal Processing and Power Applications ; 2021 April 05, 2021 - April 06, 2021
Proceedings of the 12th International Conference on Robotics, Vision, Signal Processing and Power Applications ; Chapter : 57 ; 453-458
2024-03-31
6 pages
Article/Chapter (Book)
Electronic Resource
English
Security - Biometrics - New biometrics products
Online Contents | 2003
Biometrics-access control - Looking biometrics in the eye
Online Contents | 2009
British Library Online Contents | 2006
|