基于LabVIEW的智能化中医脉象识别系统  

Intelligent recognition system of traditional Chinese medicine pulse based on LabVIEW software

作　　者：杨凌[1] 王汝旭[1] 张文博[1] 王光辉[1] 

机构地区：[1]兰州大学信息科学与工程学院,兰州730000

出　　处：《中国科技论文在线》2011年第1期49-53,58,共6页

基　　金：甘肃省教育厅科学基金资助项目(0906-02)

摘　　要：结合中医脉象理论和脉象形成机制,提取了脉象信号的升支斜率、降支斜率(绝对值)时域特征,利用小波包变换提取了脉象信号频域能量比特征,通过径向基函数(radial basis function简称RBF)神经网络进行训练和识别,并基于LabVIEW8.6软件,采用LabVIEW和Matlab混合编程的方法设计了集脉象信号实时采集、分析和分类识别功能于一体的智能化中医脉象识别系统。实验结果表明,该系统对细、平、洪3种脉象的识别正确率达到了94.29%,系统运行稳定。The intelligent recognition system of traditional Chinese medicine（TCM） pulse based on LabVIEW8.6 software is designed according to TCM theory and the mechanism of pulse formation,by using the method of LabVIEW software and Matlab software programming together.Real-time acquisition,analysis and recognition of pulse signals are the three main functions of the system.The features of the ascending branch of the slope and the descending slope（absolute value） of pulse signals are extracted in time-domain.Energy ratio features of pulse signals are extracted in frequency-domain by the wavelet packet.These values are trained and distinguished by RBF neural network.At present,the experimental results show that the operation of the system is stable,and the correct recognition percentage of three kinds of pulse（tiny pulse,normal pulse and flood pulse） is 94.29%.

关 键 词：人体脉象 特征提取 小波包 RBF神经网络 脉象识别 LABVIEW软件 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]