基于表面肌电信号的手腕动作模式识别

doi:10.3969/j.issn.0258-8021.2013.03.01

摘要
图/表
参考文献
相关文章 (8)

全文: PDF (868 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要基于表面肌电信号的肢体动作模式识别是假手仿生控制的基础。为提高动作模式识别率，从肌电信号的产生机理出发，选取分别表征肌电信号形态特征——细节复杂度和整体自似性的近似熵指标和分维数指标，作为模式识别的特征向量；同时提出一种具有增量学习能力的K最近邻(KNN)模型增量学习算法作为模式识别的分类器。在对10位受试者手腕的4个精细动作(腕伸、腕屈、腕内旋、腕外旋)的识别实验中,取得了92.5 %以上的正确识别率。同时对增量学习能力对分类器动作模式识别率的影响做对比实验，当假肢使用者生理变化时，以 KNN模型增量学习算法作为分类器比采用不具增量学习能力的KNN模型算法的识别率高4.5%。实验表明，该肌电信号动作模式的识别方法方案合理，具有应用价值。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张启忠^* 席旭刚马玉良罗志增佘青山

关键词 ：肌电信号, 肌电假肢, KNN模型增量学习算法, 近似熵, 分维数

Abstract：Action pattern recognition of limbs using sEMG is the basis for bionic control of a prosthetic hand. In consideration of the generation mechanism of sEMG, the approximate entropy and the fractal dimension, which feature the sEMG's morphological characteristics including complexity and overall selfsimilarity,　was chosen as the feature vector of pattern recognition to improve action mode recognition rate.　In the meantime, a K nearest neighbor (KNN) model incremental learning method with incremental learning ability, was presented as a classifier of pattern recognition. In pattern recognition experiment of classifying four fine movements of the wrist (namely wrist extension, wrist flexion, wrist pronation, wrist supination) with 10 participants, the correct mode recognition rate is above 92.5%.　In a contrast experiment that was designed to evaluate the effects of the increment learning ability to the action mode recognition rate, the correct recognition rate is 4.5 percent higher than KNN mode arithmetic without incremental learning ability when the prosthetic users changed physiologically. The above experimental result shows action mode recognition method based on the EMG is reasonable and practical.

Key words： electromyography myoelectric prosthesis KNN model incremental learning algorithm approximate entropy fractal dimension

基金资助:国家自然科学基金(61172134)；浙江省自然科学基金（Y1111189， LY12F03007）; 浙江省科技计划项目(2012C33075)

引用本文:

张启忠^* 席旭刚马玉良罗志增佘青山. 基于表面肌电信号的手腕动作模式识别[J]. 中国生物医学工程学报, 2013, 32(3): 257-265.
ZHANG Qi Zhong^* XI Xu Gang MA Yu Liang LUO Zhi Zeng SHE Qing Shan. Study on Wrists MultiMovement Pattern Recognition Based on sEMG. journal1, 2013, 32(3): 257-265.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021.2013.03.01 或 http://cjbme.csbme.org/CN/Y2013/V32/I3/257

［1］Li Guanlin, Li Yaonan, Yu Long, et al. Conditioning and sampling issues of EMG signals in motion recognition of multifunctional myoelectric prostheses［J］. Ann Biomed Eng, 2011, 39(6): 1779-1787.
［2］Smith LH, Hargrove LJ, Lock BA, et al. Determining the optimal window length for pattern recognitionbased myoelectric control: Balancing the competing effects of classification error and controller delay［J］. IEEE Trans Neural Syst Rehabil Eng, 2011, 19(2): 186-192.
［3］Hahne JM, Graimann B, Muller KR. Spatial Filtering for Robust Myoelectric Control［J］. IEEE Trans Biomed Eng, 2012, 59(5): 1436-1443.
［4］Khushaba RN, Kodagoda S, Takruri M, et al. Toward improved control of prosthetic fingers using surface electromyogram (EMG) signals［J］. Expert Syst　Appl, 2012, 39(12):10731-10738.
［5］Gini G, Arvetti M, Somlai I, et al. Acquisition and analysis of EMG signals to recognize multiple hand movements for prosthetic applications［J］. Appl Bion Biomechan, 2012, 9(2): 145-155.
［6］Li Guanlin, Schultz AE, Kuiken TA. Quantifying pattern recognitionbased myoelectric control of multifunctional transradial prostheses［J］. IEEE Trans Neural Syst Rehabil Eng, 2010, 18(2): 185-192.
［7］Losier Y, Englehart K, Hudgins B. Evaluation of shoulder complex motionbased input strategies for endpoint prostheticlimb control using dualtask paradigm［J］. J Rehabil Res Dev, 2011, 48: 669-678.
［8］Bunderson NE, Kuiken TA. Quantification of feature space changes with experience during electromyogram pattern recognition control［J］. IEEE Trans Neural Syst Rehabil Eng, 2012, 20(3): 239-246.
［9］Duchene J, Hogrel J Y. A model of EMG generation［J］. IEEE Trans Biomed Eng, 2000, 47(2): 192-201.
［10］Basmajian JV, Deluca CJ. Electromyographyphysiology, engineering and noninvasive applications［M］. New York::IEEE Press, 2004.
［11］Pincus SM. Approximate entropy (ApEn) as a complexity measure［J］. Chaos, 1995, 5(1): 110-117.
［12］Cover TM, Hart PE. Nearest neighbor pattern classification［J］. IEEE Trans Inf Theory, 1967, 13(1): 21-27.
［13］Guo Gongde, Wang Hui, Bell D,et al. Using kNN model for automatic text categorization［J］. Soft Comput, 2006, 10(5): 423-430.
［14］陈黎飞，郭躬德. 最近邻分类的多代表点学习算法［J］.模式识别与人工智能，2011，24（6）：883-888.
［15］刘加海, 王健, 罗晓芳. 局部肌肉疲劳的表面肌电信号复杂度和熵变化［J］. 生物物理学报, 2004, 20(3): 198-202.
［16］蔡立羽, 王志中, 李凌. 肌肉疲劳过程中的表面肌电信号特征研究［J］. 中国康复医学杂志, 2000, 15(2): 94-95.
［17］Arjunan SP, Kumar DK. Fractal properties of surface electromyogram for classification of low level hand movements from singlechannel forearm muscle activity［J］. J Mech Med Biol, 2011,11(3):581-590.
［18］Arjunan SP, Kumar DK. Decoding subtle forearm flexions using fractal features of surface electromyogram from single single and multiple sensors［J］. J Neur Rehabil, 2010, 7(1): 1-10.
［19］Fukuda O, Tsuji T, Kaneko M, et al. A humanassisting manipulator teleoperated by EMG signals and arm motions［J］. IEEE Trans Robot Autom, 2003, 19(2): 210-222.