电子学报 ›› 2017, Vol. 45 ›› Issue (11): 2735-2741.DOI: 10.3969/j.issn.0372-2112.2017.11.022

所属专题: 机器学习—特征选择

• 学术论文 • 上一篇    下一篇

融合表面肌电和加速度信号的下肢运动模式识别研究

席旭刚, 汤敏彦, 张自豪, 张启忠, 罗志增   

  1. 杭州电子科技大学智能控制与机器人研究所, 浙江杭州 310018
  • 收稿日期:2016-07-15 修回日期:2017-02-27 出版日期:2017-11-25 发布日期:2017-11-25
  • 通讯作者: 席旭刚
  • 作者简介:汤敏彦,女,1992年生于浙江湖州.现为硕士研究生,主要研究方向为信号处理、模式识别.E-mail:11045404@hdu.edu.cn
  • 基金资助:
    浙江省自然科学基金(No.LY17F030021);国家自然科学基金(No.61671197)

Lower Limb Motion Recognition Based on the Fusion of sEMG and Acceleration Signal

XI Xu-gang, TANG Min-yan, ZHANG Zi-hao, ZHANG Qi-zhong, LUO Zhi-zeng   

  1. Intelligent Control & Robotics Institute of Hangzhou Dianzi University, Hangzhou, Zhejiang 310018, China
  • Received:2016-07-15 Revised:2017-02-27 Online:2017-11-25 Published:2017-11-25

摘要: 为了提高下肢运动模式识别率,本文设计了一种融合表面肌电和加速度信号的下肢运动模式识别方法.首先,用局部均值分解将表面肌电信号分解为多个乘积函数(Product Functions,PFs),再计算PF成分的多尺度排序熵.然后,通过拉普拉斯权重(Laplacian score,LS)特征选择算法选定每路肌电信号的一个尺度排序熵为特征,并把该特征和加速度信号的排序熵组成特征向量.最后,根据类内欧氏距离和类间样本分布,设计了改进的二叉树支持向量机,把特征向量输入该支持向量机进行下肢运动模式分类.实验结果表明所提方法对七个日常动作的平均识别率达到98.62%,相较于其他方法有较高的识别率.

关键词: 下肢运动模式识别, 表面肌电信号, 加速度信号, 多尺度排序熵, 改进二叉树支持向量机

Abstract: In order to improve the recognition rate of lower limb motion pattern,(a novel lower limb motion recognition method was designed by fusion of surface electromyography (sEMG) signal and acceleration signal.Firstly,the sEMG signal was decomposed into a set of product functions(PFs)by Local mean decomposition(LMD),and the multiscale permutation entropy(MPE) of PFs was calculated.Then,one scale permutation entropy was selected as the feature of sEMG by the Laplacian score.The feature vector is composed by this sEMG feature and the permutation entropy of acceleration signal.Finally,based on the combination of inter-class Euclidean distance and intra-class sample distribution,an improved support vector machine based binary tree(ISVM-BT) was designed.The feature vector was inputted into this SVM to recognize the lower limb motion.The experimental results indicate that the proposed method achieved 98.62% at the average recognition rate for seven daily activities,and has higher accuracy than other methods.

Key words: lower limb motion pattern recognition, surface electromyography (sEMG), acceleration signal, multiscale permutation entropy(MPE), improved support vector machine based binary tree(ISVM-BT)

中图分类号: