中文摘要：

采用小波变换进行肌电信号预处理与多尺度分解,并采用小波系数最大值与平均能量值作为肌电信号特征,采用支持向量机进行特征分类识别的运动解码,并用此方法进行了腕部动作识别的实验.与时域特征、频域特征、AR参数特征提取方法以及神经网络识分类别方法进行对比,结果表明:基于支持向量机的小波特征提取方法可以较好地区分不同腕部动作,具有最高的分类精度,极大改善前臂假肢的操纵性能.

英文摘要：

Wavelet transform was used to preprocess electromyography (EMG)signal and extract classification features,and the maximum value of wavelet coefficient and the average energy value were used as the characteristics of EMG signal.Then using the support vector machine (SVM)for features classification of the motion decoding method.The wrist action recognition experiments were carried on with this method.Finally,it was compared with the time domain features,frequency do-main features,autoregressive model (AR)parameter feature extraction method and neural network i-dentification method.Results show that this method can well distinguish different wrist action,get the highest classification accuracy and greatly improve the handling performance of the forearm pros-thesis.