Title :
ECG analysis: a new approach in human identification
Author :
Biel, Lena ; Pettersson, Ola ; Philipson, Lennart ; Wide, Peter
Author_Institution :
Dept. of Sci. & Technol., Orebro Univ., Sweden
fDate :
6/1/2001 12:00:00 AM
Abstract :
A new approach in human identification is investigated. For this purpose, a standard 12-lead electrocardiogram (ECG) recorded during rest is used. Selected features extracted from the ECG are used to identify a person in a predetermined group. Multivariate analysis is used for the identification task. Experiments show that it is possible to identify a person by features extracted from one lead only. Hence, only three electrodes have to be attached on the person to be identified. This makes the method applicable without too much effort
Keywords :
biometrics (access control); electrocardiography; feature extraction; medical signal processing; principal component analysis; sensor fusion; signal classification; ECG analysis; PCA; SIMCA classifier; data fusion; feature extraction; human identification; multivariate analysis; Electric potential; Electric variables measurement; Electrocardiography; Electrodes; Face detection; Feature extraction; Fingerprint recognition; Foot; Heart; Humans;
Journal_Title :
Instrumentation and Measurement, IEEE Transactions on
