Classification of Hand Gestures Based on Single-channel sEMG Decomposition

doi:10.3901/JME.2016.07.006

Abstract

Abstract: Surface electromyography (sEMG) has been applied extensively in gestures recognition. In order to improve the recognition accuracy, multi-channel sEMG is conventionally sampled, which also increases the complexity of applications. To solve the problem, a novel gesture recognition method based on sEMG decomposition is proposed. Sampling sEMG signals from the muscle of human upper limb by a single-channel electrode; then decomposing the sEMG into six motor unit action potential trains (MUAPTs) and the decomposition process includes 2-order differential filtering, threshold calculation, spike detection and hierarchical clustering. Afterwards, the features, including integral of absolute value, maximum value, median of non-zero value and semi-window energy, are extracted to form a feature matrix, whose dimension is then reduced by the principal component analysis. Finally, support vector machine is employed to recognize five different hand gestures, and 80.4% of accuracy can be obtained, while only about 70% of recognition accuracy can be achieved by traditional methods without sEMG decomposition.

Key words: hierarchical clustering, motor unit action potential trains, principal component analysis, sEMG, support vector machine

CLC Number:

TG156

XIONG Anbin, DING Qichuan, ZHAO Xingang, HAN Jianda, LIU Guangjun. Classification of Hand Gestures Based on Single-channel sEMG Decomposition[J]. Journal of Mechanical Engineering, 2016, 52(7): 6-13.

References

[1] 熊安斌,赵新刚,韩建达,等. 基于混沌理论的面瘫患者表面肌电信号分析[J]. 科学通报,2013,58(S2)：152-165.
XIONG Anbin,ZHAO Xingang,HAN Jianda,et al. Chaotic analysis of EMG on the patients of facial paralysis[J]. Chinese Science Bulletin,2013,58(S2)： 152-165.
[2] HARGROVE L J,SCHEME E J,ENGLEHART K B,et al. Multiple binary classifications via linear discriminant analysis for improved controllability of a powered prosthesis[J]. IEEE Trans. Neural Syst. Rehabil. Eng.,2010,18(1)：49-57.
[3] ZHANG D,ZHAO X,HAN J,et al. A comparative study on PCA and LDA based EMG pattern recognition for anthropomorphic robotic hand[C]// IEEE International Conference on Robotics and Automation,May 31-June 7,2014,HongKong,China. NJ：IEEE,2014：4850-4855.
[4] MATSUBARA T,MORINOTO J. Bilinear modelling of EMG signals to extract user-independent features for multi-user myoelectric interface[J]. IEEE Trans. Biomed. Eng.,2013,60(8)：2206-2213.
[5] 杨大鹏,赵京东,李楠,等. 基于预抓取模式识别的假手肌电控制方法[J]. 机械工程学报,2012,48(15)：1-8.
YANG Dapeng,ZHAO Jingdong,LI Nan,et al. Recognition of hand grasp pre-shaping patterns applied to prosthetic hand electromyography control[J]. Journal of Mechanical Engineering,2012,48(15)：1-8.
[6] TKACH D,HUANG H,KUIKEN T A. Study of stability of time-domain features for electromyographic pattern recognition[J]. Journal of Neuro Engineering and Rehabilitation,2010,7(21)：1-13.
[7] BOOSTANI R,MORADI M H. Evaluation of the forearm EMG signal features for the control of a prosthetic hand[J]. Physiological Measurement,2003,24：309-319.
[8] ENGLEHART K,HUDGIN B,PARKER P A. A wavelet-based continuous classification scheme for multifunction myoelectric control[J]. IEEE Trans. Biomed. Eng.,2001,48(3)：302-311.
[9] JU Z,OUYANG G,WILAMOWSKA-KORSAK M,et al. Surface EMG based hand manipulation identification via nonlinear feature extraction and classification[J]. IEEE Sensors Journal,2013,13(9)：3302-3311.
[10] SEGIL J,WEIR R. Design and validation of a morphing myoelectric hand posture controller based on principal component analysis of human grasping[J]. IEEE Trans. Neural Syst. Rehabil. Eng.,2014,22(2)：249-257.
[11] KHUSHABA R,AL-ANI A,AL-JUMAILY A. Orthogonal fuzzy neighborhood discriminant analysis for multifunction myoelectric hand control[J]. IEEE Trans. Biomed. Eng.,2010,57(6)：1410-1419.
[12] JIANG N,ENGLEHART K B,PARKER P A. Extracting simultaneous and proportional neural control information for multiple-DOF prostheses from the surface electromyographic signal[J]. IEEE Trans. Biomed. Eng.,2009,56(4)：1070-1080.
[13] AL-TIMEMY A,BUGMANN G,ESCUDERO J,et al. Classification of finger movements for the dexterous hand prosthesis control with surface electromyography[J]. IEEE J. Biomed. Health Inform,2013,17(3)：608-618.
[14] BU N,OKAMOTO M,TSUJI T. A hybrid motion classification approach for EMG-based human-robot interfaces using Bayesian and neural networks[J]. IEEE Trans. on Robotics,2009,25(3)：502-511.
[15] HUANG Y,ENGLEHART K B,HUDGINS B,et al. A Gaussian mixture model based on classification scheme for myoelectric control of powered upper limb prostheses[J]. IEEE Trans. Biomed. Eng.,2005,52(11)：1801-1811.
[16] SCHEME E J,ENGLEHART K B,HUDGINS B S. Selective classification for improved robustness of myoelectric control under nonideal conditions[J]. IEEE Trans. Biomed. Eng.,2011,58(6)：1698-1705.
[17] 丁其川,赵新刚,韩建达. 基于肌电信号容错分类的手部动作识别[J]. 机器人,2015, 37(1)：9-16.
DING Qichuan,ZHAO Xingang,HAN Jianda. Recognizing hand motions based on fault-tolerant classification with EMG signals[J]. Robot,2015,37(1)：9-16.
[18] DE LUCA C J,ADAM A,WOTIZ R,et al. Decomposition of surface EMG signals[J]. Journal of neurophysiology,2006,96(3)：1646-1657.
[19] KLEINE B U,Van DIJK J P,LAPATKI B G,et al. Using two-dimensional spatial information in decomposition of surface EMG signals[J]. Journal of Electromyography and Kinesiology. 2007,17(5)：535-548.
[20] GAZZONI M,FARINA D,MERLETTI R. A new method for the extraction and classification of single motor unit action potentials from surface EMG signals[J]. J. Neurosci Methods,2004,136(2)：165-177.
[21] NAWAB S H,CHANG S S,DE LUCA C J,High-yield decomposition of surface EMG signals[J]. Clinical Neurophysiology,2010,121(10)：1602-1615.
[22] MCGILL K C,CUMMINS K,DORFMAN L J. Automatic decomposition of the clinical electromyogram[J]. IEEE Trans. Biomed. Eng.,1985,32(7)：470-477.
[23] STASHUK D W. Decomposition and quantitative analysis of clinical electromyographic signals[J]. Medical Engineering & Physics,1999,21(6)：389-404.
[24] 李学军,李平,蒋玲莉. 类均值核主元分析法及在故障诊断中的应用[J]. 机械工程学报,2014,50(3)：123-129.
LI Xuejun,LI Ping,JIANG Lingli. Class mean kernel principal component analysis and its application in fault diagnosis[J]. Journal of Mechanical Engineering,2014,50(3)：123-129.
[25] BISHOP C M. Pattern recognition and machine learning[M]. Newyork：Springer-Verlag,2006.
[26] 潘礼正,宋爱国,徐国政,等. 基于SVM-GDFNN的上肢康复训练机器人处方诊断[J]. 机械工程学报,2013,49(13)：17-23.
PAN Lizhen,SONG Aiguo,XU Guozheng,et al. Prescription diagnosis for upper-limb rehabilitation training robot based on SVM-GDFNN[J]. Journal of Mechanical Engineering,2013,49(13)：17-23.
[27] CHIANG J,WANG J,MCKEOWN M J. A hidden Markov, multivariate autoregressive (HMM-mAR) network framework for analysis of surface EMG (sEMG) data[J]. IEEE Trans. Signal Proces.,2008,56(8)：4069-4081.