计算机工程2011，Vol.37Issue(1)：175-177,3.DOI:10.3969/j.issn.1000-3428.2011.01.061
基于心电图的身份识别方法
Human Identification Approach Based on ECG
摘要
Abstract
In view of the problems in current human identification based on Electrocardiography(ECG), such as small sample and multi-feature points detection, a novel approach using wavelet and Dynamic Time Warping(DTW) for human identification is investigated.The noise is reduced in preprccessing and the R peaks are detected by using wavelet transform.Then the reference templates of QRS wave and heart beat in limb leads are extracted and saved.Candidates are selected by the result of correction analysis between the test and templates of QRS waves.The only candidate is selected by the minimum distance between the test and templates of heart beat in limb leads using DTW.Experimental results show that the identification accuracy is 97.3％ in 112 individuals and individual identification time is 4.4 s, and the proposed method solves the problems of human identification based on ECG under single feature point detection and large sample.关键词
心电图/身份识别/小波变换/动态时间规整Key words
Electrocardiography(ECG)/ human identification/ wavelet transformation/ Dynamic Time Warping(DTW)分类
信息技术与安全科学引用本文复制引用
师黎,郭豹,李中健,赵云..基于心电图的身份识别方法[J].计算机工程,2011,37(1):175-177,3.基金项目
国家自然科学基金资助项目(60841004,60971110) （60841004,60971110）
