Key result
ECG biometric algorithms degrade with different-session data, requiring multiple training sessions for reliable authentication.
ECG biometric authentication algorithms experience performance degradation when training and testing data are from different sessions, requiring multiple training sessions to mitigate.
Answers are grounded in this paper and cite their sources.
Session-dependent ECG biometric degradation cautions against single-enrollment use; leaves open optimal multi-session protocols for clinical deployment.
The electrocardiogram (ECG) is one of the latest biometric modality. This paper takes into account most of the technique which are applied to use the electrocardiogram as a biometric identification. Here we differentiate the approach based on features and classification scheme finally a comparative study of authentication performance of few ECG biometric is presented by using available database. The comparative analysis includes the cases where training and testing data is received from different sessions of same day or from different days. In this authentication process it is observed that most of the algorithm perform as per requirement when training data is received from the same session but a degradation occur when data is received from the different session. Due to degradation in performance, multiple training sessions are used.
No takes yet. Share an insight, caveat, or question.
Kumar et al. (2017) reported a review. ECG biometric identification was evaluated on Authentication performance. ECG biometric identification algorithms perform well with same-session training data, but performance degrades with different-session data, necessitating multiple training sessions.
Synapse has enriched 10 related papers. Consider them for comparative context: