摘要
心电信号在医疗方面的应用有着广泛的研究,然而心电信号反映了人体心脏的生理活动,不同人的心电信号各不相同,因此可以用作身份识别。该文采用Gabor原子库对心电信号进行基于MP的稀疏分解,分解后所得的原子参数和投影值中包含该信号的重要信息,将其作为特征参数,并采用支持向量机对其进行分类。通过对20个人的心电信号进行实验,识别系统的识别率可达95.3%。
Analysis of the electrocardiogram (ECG) signal has been in the spotlight of study in the clinical field for the past two decades, ECG Signals reflect cardiac electrical activity, and vary from person to person, so it could Use for biometrics identifieatlon. This:paper Uses MP sparse decomposition with Gabor dictionary to get the atoms parameters and projections which contain ECG signals important information as ECG signals characteristics, and use support vector machine (SVM) to classify the signal. Experimental results show that the performance of tile system over 20 subjects is 95. 3%.
出处
《杭州电子科技大学学报(自然科学版)》
2011年第4期98-101,共4页
Journal of Hangzhou Dianzi University:Natural Sciences
关键词
心电信号
身份识别
支持向量机
稀疏分解
匹配追踪
electrocardiogram
biometric identification
support vector machine
sparse decomposition
mzttt:-hing pursuit