基于小波分解和数据融合方法的ECG身份识别

Human Identification Based on ECG with Wavelet Decomposition and Data Fusion Method

目的研究一种新的基于ECG的身份识别方法。方法选取35位健康人ECG波形的波形特征、小波特征和融合特征作为特征向量,通过相关系数阈值法进行ECG身份识别。结果对35位被试者的另外40段数据进行身份识别验证,基于波形特征、小波特征、融合特征的身份识别正确率分别为82.5%、87.5%、95%。采用小波特征身份识别的正确率优于波形特征的正确率,而采用融合特征识别的正确率优于其他两种特征的正确率。结论实验表明,基于ECG的身份识别技术是可行的,可以和指纹等多种生物特征联合使用开发多导生物识别系统,并且本文所提方法算法简单,实时性好,准确度高。

Objective A new method is researched based on ECG for human identification.Methods Three features of wave feature,wavelet feature,fusion feature were selected as the feature vectors for 35 normal subject ECGs,and correlation coefficient discriminate analysis was used for human identification. Results The other 40 sessions from 35 subjects were used for verifying the accuracy of each feature.They were 82.5%、87.5% and 95% respectively for wave feature,wavelet feature and fusion feature. ECG wavelet feature outperformed ECG wave feature. While ECG fusion feature performed better than either feature. Conclusion The verification experiment demonstrates that the technology of human identification based on ECG is feasible. ECG can be combined with other biometric features such as fingerprint for developing the multi-biometric identification system. The method proposed in this paper is simple,real-time and exact for human identification.