东北大学学报（自然科学版）

东北大学学报(自然科学版) ›› 2006, Vol. 27 ›› Issue (3): 280-283.DOI: -

• 论著 • 上一篇    下一篇

支持向量机在表面肌电信号模式分类中的应用

崔建国;王旭;李忠海;张大千;   

  1. 东北大学教育部暨辽宁省流程工业综合自动化重点实验室;东北大学教育部暨辽宁省流程工业综合自动化重点实验室;沈阳航空工业学院自动控制系;沈阳航空工业学院自动控制系 辽宁沈阳110004;辽宁沈阳110004;辽宁沈阳110034;辽宁沈阳110034
  • 收稿日期:2013-06-23 修回日期:2013-06-23 出版日期:2006-03-15 发布日期:2013-06-23
  • 通讯作者: Cui, J.-G.
  • 作者简介:-
  • 基金资助:
    国家自然科学基金资助项目(50477015)

Application of support vector machine in pattern classification of surface EMG

Cui, Jian-Guo (1); Wang, Xu (1); Li, Zhong-Hai (2); Zhang, Da-Qian (2)   

  1. (1) Key Laboratory of Process Industry Automation of MOE and Liaoning Province, Northeastern University, Shenyang 110004, China; (2) Department of Automation Control, Shenyang Institute of Aeronautical Engineering, Shenyang 110034, China
  • Received:2013-06-23 Revised:2013-06-23 Online:2006-03-15 Published:2013-06-23
  • Contact: Cui, J.-G.
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

985

摘要： 采用小波变换的方法对实验采集的原始四通道表面肌电信号(sEMG)进行了分析,并提取小波分解系数的奇异值构建特征矢量,利用“一对一”分类策略和二叉树设计的多类支持向量机(SVM)分类器,很好地实现了对前臂8种运动表面肌电信号的模式分类,8种运动模式的平均识别率为98.75%.研究表明SVM分类准确率明显优于传统的BP神经网络、Elman神经网络和RBF神经网络分类器,且识别精度高,鲁棒性好,对肌电信号及其他非平稳生理电信号的模式识别,提供了一种具有良好应用前景的新方法.

关键词: 表面肌电信号, 小波变换, 支持向量机, 模式分类

Abstract: A new pattern recognition method of surface electromyography (sEMG) is proposed, based on wavelet transform and support vector machine (SVM). The original four-channel sEMG signals from four corresponding muscles are analyzed with wavelet transform, then the singular values of wavelet decomposition coefficients are extracted to be the signal characteristics to construct eigenvector. A new multi-class SVM classifier is designed according to the 'one by one' classification strategy and binary tree. Experiment results show that the eight forearm movement patterns can be well recognized after training by the SVM with average recognition rate up to 98.75%, and that the SVM can sort out sEMG eight movement patterns more accurately than conventional BP neural, Elman neural and RBF neural networks, with high robustness also provided. The method proposed can be directly applied to the pattern recognition of other nonstationary bioelectric signals.

中图分类号: