简约支持向量机分类算法在下肢动作识别中的应用研究

中国机械工程 ›› 2011, Vol. 22 ›› Issue (4): 433-438.

简约支持向量机分类算法在下肢动作识别中的应用研究

吴剑锋1;吴群2;孙守迁1

1.浙江大学,杭州,310027

2.浙江理工大学,杭州,310018

出版日期:2011-02-25 发布日期:2011-03-15
基金资助:
国家自然科学基金资助项目(60703041)；广东省教育部产学研结合项目(2009B090600139)；浙江理工大学科研启动基金资助项目(1009831-Y)
National Natural Science Foundation of China（No. 60703041）

Research on Classification Algorithm of Reduced Support Vector Machine for Low Limb Movement Recognition

Wu Jianfeng1;Wu Qun2;Sun Shouqian1

1.Zhejiang University,Hangzhou,310027

2.Zhejiang Sci-Tech University,Hangzhou,310018

Online:2011-02-25 Published:2011-03-15
Supported by:

National Natural Science Foundation of China（No. 60703041）

摘要/Abstract

摘要：

为提高多模式人体下肢动作识别的准确性,提出了一种基于简约支持向量机算法的下肢动作识别方法。通过动作分解将人体日常下肢动作行为分解为不同的动作片段以组成识别目标集;以下肢肌肉表面肌电信号为信息源,综合短时统计时域特征值和Mallat小波时频域特征值建立识别特征向量空间;采用核聚类简化的方法降低计算复杂度,提高算法的鲁棒性。起立、平地常速行走以及上下楼梯等四个日常下肢动作识别实验的结果证明了该方法的有效性。



关键词:

下肢动作识别, 表面肌电信号, 支持向量机, 多元分类

Abstract:

In order to improve precision of the multi-motion pattern recognition of lower limb movements, a recognition method based on reduced

SVM was proposed. The daily movements were decomposed into different segments to form the recognition goals, and the surface electromyography signals of low limb muscles were used as information sources. The eigenvectors were extracted using Mallat wavelet transform and statistical characteristics in time-domain. Then the simplified method of kernel clustering was used to reduce the computational complexity and improve the robustness of algorithm. Four patterns were identified herein,i.e. standing, normal speed walking, up-stairs and down-stairs. The experimental results show the effectiveness of this method.

Key words: lower-limb movement recognition, surface electromyo graphy(EMG), support vector machine(SVM), multi- classification

中图分类号:

TB391.4

吴剑锋, 吴群, 孙守迁.

简约支持向量机分类算法在下肢动作识别中的应用研究

[J]. 中国机械工程, 2011, 22(4): 433-438.

TUN Jian-Feng, TUN Qun, SUN Shou-Qian.

Research on Classification Algorithm of Reduced Support Vector Machine for Low Limb Movement Recognition

[J]. China Mechanical Engineering, 2011, 22(4): 433-438.

参考文献

[1]陈鹰,杨灿军.人机智能系统理论与方法[M].杭州:浙江大学出版社,2006.
[2]Leuthardt E C, Schalk G, Wolpaw J R, et al. A Brain--computer Interface Using Electrocortico- graphic Signals in Humans[J]. Journal of Neural Engineering, 2004,1 (2) : 63-71.
[3]Li Yong, Gao Xiaorong, Liu H, et al. Classification of Single--trial Electroencephalogram During Finger Movement[J]. IEEE Transactions on Bio--medical Engineering, 2004,51(6) : 1019-1025.
[4]Zhou H, Hu H, Harris N D, et al. Applications of Wearable Inertial Sensors in Estimation of Upper I.imb Movements[J]. Journal of Biomedical Signal Processing and Control, 2006,1 ( 1 ) : 22-32.
[5]Sazonov E S, Bumpus T, Zeigler S, et al. Classification of Plantar Pressure and Heel Acceleration Patterns Using Neural Networks[C]//Proceedings of 2005 IEEE International Joint Conference on Neural Networks. Montreal, Canada, 2005 : 3007-3010.
[6]Kuzelicki J, Zefran M, Burger H, et al. Synthesis of Standing--up Trajectories Using Dynamic Optimization[J]. Gait & Posture,2005,21(1) :1-11.
[7]Castellini C, Smagt P. Surface EMG in Advanced Hand Prosthetics[J]. Biological Cybernetics, 2009, 100:35-47.
[8]Yan Zhiguo,Wang Zhizhong, Xie Hongbo. The Application of Mutual Information--based Feature Selection and Fuzzy LS--SVM Based Classifier in Motion Classification[J]. Computer Methods and Programs in Biomedicine, 2008,90:275-284.
[9]Lucas M F,Gaufriau A S,Pascual S,et al. Multi-- channel Surface EMG Classification Using Support Vector Machines and Signal--based Wavelet Optimization[J]. Biomedical Signal Processing and Control,2008,3(2) :169-174.
[10]席旭刚[1],李仲宁[1],罗志增[1].基于相关性分析和支持向量机的手部肌电信号动作识别[J].电子与信息学报,2008,30(10):2315-2319.
[11]佘青山[1],孟明[1],罗志增[1],马玉良[1].基于多核学习的下肢肌电信号动作识别[J].浙江大学学报：工学版,2010(7):1292-1297.
[12]颜志国.基于粗糙集和支持向量机的表面肌电特征约简和分类研究 [D].上海交通大学,2008:
[13]曾志强.支持向量分类机的训练与简化算法研究 [D].浙江大学计算机科学与技术学院,2007:
[14]吴剑锋.基于肌电信号的人体下肢运动信息获取技术研究 [D].浙江大学计算机科学与技术学院,2008:
[15]何乐生.基于肌电信号的人机接口技术的研究[D].南京:东南大学,2007.