基于磁场刺激的肌电信号模式识别的研究

对于人体表面肌电（SEMG）信号提出一种新的研究方法，即在磁场刺激下，采用小波变换的方法，对从掌长肌、肱桡肌、尺侧腕屈肌和肱二头肌上采集的4路表面肌电信号进行分析，并提取其6级小波分解系数绝对值累加和的平均值作为信号的特征．构建特征矢量．输入神经网络分类器进行模式识别，经过训练能够成功地识别出握举、展拳、腕内旋、腕外旋、屈腕、伸腕、前臂内旋、前臂外旋8种运动模式．实验结果表明，该方法识别率高，所需数据量少．运算速度快，实时性好，为肌电等生物电信号的研究提供了一种新方法．

Study of Surface EMG Pattern Recognition Based on Magnetic Stimulation

A new pattern recognition technique is proposed for SEMG.Four channel SEMG signals from four muscles(palmaris longus,brachioradialis,flexor carpi ulnaris and biceps brachii) under magnetic stimulation are analyzed with wavelet transformation.The average of absolute value summation of 6 layers wavelet decomposition coefficients are distilled,and are regarded as signal characteristics to compose eigenvector.A neural network classifier is adopted to identify different motions.The eight motions,hand grasp,hand extension,wrist pronation,wrist supination,wrist flexion,wrist extension,forearm pronation and forearm supination,can be successfully identified after training.Experiments show that the method has good performance in real time processing,high rates of calculation and identification.