基于小波变换的多特征融合sEMG模式识别

sEMG Pattern Recognition Based on Multi Feature Fusion of Wavelet Transform

针对单一特征值表征能力差的情况,根据小波变换的多分辨分析思想,采用基于多种母小波的多特征融合的特征提取方法对表面肌电信号进行特征提取。本实验对十名测试人员进行肌电信号的采集,对日常生活中的四个基本下肢动作进行测试。首先,分别基于DB、Dmey和Bior三种不同的母小波,采用离散小波变换通过不同的分析方法对表面肌电信号进行多尺度分解。然后,通过分析发现,不同肌肉在不同特征提取方式下表征效果存在差异,为了结合不同特征方式的特点对基于不同小波基的特征值进行融合分析并比较。最后,将特征值分别输入到Elman神经网络和BP神经网络进行模式识别并比较分析。实验结果表明:通过对不同特征值进行识别比较,融合处理的特征值可以达到98.7%的识别率,并且,BP神经网络相较于Elman神经网络识别效果更好。

In view of the poor characterization of single feature value,multi feature fusion based on different waveletbasis was adopted to extract the surface EMG signal according to multi resolution analysis of wavelet transform. Theexperiment was conducted on ten testers and collected signals for four basic lower limb movements in daily life.First of all,discrete wavelet transform was used to decompose the surface EMG signals in multi-scale with DB,Dmey and Bior wavelet basis respectively. After that,it was founded that the characterization effects of differentmuscle vary by different extraction way. In order to combine the characteristics of different features,features werefused to analyze and compare. At last,the feature values were input to the Elman neural network and BP neural net-work for pattern recognition and comparison analysis. Experimental results showed that the recognition rate ob-tained by fusing the eigenvalues is higher than single feature with the accuracy up to 98.7%,and the BP neural net-work is better than the Elman neural network.