论文部分内容阅读
心音是人体心脏搏动产生的具有个体特征的生理参数。实验利用小波变换实现对心音信号的去噪处理,选择Mel频率倒谱系数(MFCC)作为特征参量,并经过主成分分析(PCA)降维处理后用于个体分类识别中。旨在对心音作为生物识别特征参数可行性和识别方法做一个初步研究。实验结果显示,在选定的实验条件下,系统可以达到90%以上的识别率。可以为心音识别技术的进一步研究提供参考。
Heart sound is a physiological parameter that has individual characteristics generated by the beats of the human heart. The wavelet transform was used to denoise the heart sound signal. Mel frequency cepstrum coefficient (MFCC) was selected as the feature parameter, which was reduced by the Principal Component Analysis (PCA) and used in the classification of individuals. The aim is to make a preliminary study on the feasibility and identification of heart sound as a feature parameter of biometrics. Experimental results show that under the selected experimental conditions, the system can achieve more than 90% recognition rate. It can provide reference for the further research of heart sound recognition technology.