EMG hand gesture classification using handcrafted and deep features

Abstract Currently, electromyographic (EMG) signal gesture recognition is performed with devices of many channels. Each channel gives a signal that must be filtered and processed, which sometimes can be a slow process that requires high-cost hardware to process all the data quickly enough. This paper presents a combined feature approach method for EMG classification using handcrafted features obtained from time-spectral discrete analysis and deep features extracted from a convolutional neural network (CNN), which classifies signals recorded from a single channel device. The method proposed only requires 100 signals from each gesture for training, thus the time needed to train the system is reduced. The proposed approach combines handcrafted features from a time-spectral analysis, like mean absolute value (MAV), slope sign changes (SSC), peak frequencies, wavelet transform (WT) coefficients, etc, and deep features to create the feature vector. The feature vector is then classified using a multi-layer perceptron classifier (MLPC). Experimental results showed an average classification accuracy of 81.54%, 88.54%, and 94.19% for 8, 6, and 5 gesture-classes, respectively. The results could serve as a basis for a real implementation of EMG signal gesture recognition with a device of only one channel.

[1]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[2]  Hong Liu,et al.  Robust EMG Pattern Recognition with Electrode Donning/Doffing and Multiple Confounding Factors , 2017, ICIRA.

[3]  Tanu Sharma,et al.  A novel feature extraction for robust EMG pattern recognition , 2016, Journal of medical engineering & technology.

[4]  Guido Bugmann,et al.  A note on the probability distribution function of the surface electromyogram signal☆ , 2013, Brain Research Bulletin.

[5]  Sachin Taran,et al.  An efficient approach for physical actions classification using surface EMG signals , 2020, Health Inf. Sci. Syst..

[6]  M. Knaflitz,et al.  Analysis of myoelectric signals recorded during dynamic contractions , 1996 .

[7]  Zhihua Wang,et al.  Hand Gestures Recognition from Multi-channel Forearm EMG Signals , 2016, ICCSIP.

[8]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[9]  An Application of the Two-Factor Mixed Model Design in Educational Research , 2019 .

[10]  M. Alexander,et al.  Principles of Neural Science , 1981 .

[11]  A.R. Ismail,et al.  Continuous wavelet transform application to EMG signals during human gait , 1998, Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284).

[12]  Yusuke Yamanoi,et al.  EMG-based posture classification using a convolutional neural network for a myoelectric hand , 2020, Biomed. Signal Process. Control..

[13]  Guanglin Li,et al.  Selection of sampling rate for EMG pattern recognition based prosthesis control , 2010, 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology.

[14]  Abdulhamit Subasi,et al.  Comparison of Bagging and Boosting Ensemble Machine Learning Methods for Automated EMG Signal Classification , 2019, BioMed research international.

[15]  U. Sadiya,et al.  Comparison on Trapezoidal and Simpson’s Rule for Unequal Data Space , 2019, International Journal of Mathematical Sciences and Computing.

[16]  Novruz Allahverdi,et al.  Deep Belief Networks Based Brain Activity Classification Using EEG from Slow Cortical Potentials in Stroke , 2016 .

[17]  S Christopher,et al.  A Comprehensive Study on EMG Feature Extraction and Classifiers , 2018 .

[18]  J. L. Torres,et al.  Classification of Myoelectric Signals Using Multilayer Perceptron Neural Network with Back Propagation Algorithm in a Wireless Surface Myoelectric Prosthesis , 2022 .

[19]  Erik Scheme,et al.  Electromyogram pattern recognition for control of powered upper-limb prostheses: state of the art and challenges for clinical use. , 2011, Journal of rehabilitation research and development.

[20]  C.S. Pattichis,et al.  Time-scale analysis of motor unit action potentials , 1999, IEEE Transactions on Biomedical Engineering.

[21]  Cristhian Mauricio Potes.  Assessment of human muscle fatigue from surface EMG signals recorded during isometric voluntary contractions by using a cosine modulated filter bank , 2008 .

[22]  Giancarlo Ferrigno,et al.  Artificial neural network EMG classifier for functional hand grasp movements prediction , 2016, The Journal of international medical research.

[23]  Levi J. Hargrove,et al.  A Comparison of Surface and Intramuscular Myoelectric Signal Classification , 2007, IEEE Transactions on Biomedical Engineering.

[24]  Aboul Ella Hassanien,et al.  A Hybrid Classification Model for EMG Signals Using Grey Wolf Optimizer and SVMs , 2015, AISI.

[25]  D Graupe,et al.  Multifunctional prosthesis and orthosis control via microcomputer identification of temporal pattern differences in single-site myoelectric signals. , 1982, Journal of biomedical engineering.

[26]  Arnab Bhattacharjee,et al.  Deep Learning Approach to Cardiovascular Disease Classification Employing Modified ECG Signal from Empirical Mode Decomposition , 2019, Biomed. Signal Process. Control..

[27]  Clément Gosselin,et al.  Deep Learning for Electromyographic Hand Gesture Signal Classification Using Transfer Learning , 2018, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[28]  Yakup Kutlu,et al.  Deep learning with 3D-second order difference plot on respiratory sounds , 2018, Biomed. Signal Process. Control..

[29]  Huosheng Hu,et al.  Support Vector Machine-Based Classification Scheme for Myoelectric Control Applied to Upper Limb , 2008, IEEE Transactions on Biomedical Engineering.

[30]  Boualem Boashash,et al.  A methodology for detection and classification of some underwater acoustic signals using time-frequency analysis techniques , 1990, IEEE Trans. Acoust. Speech Signal Process..

[31]  Tim Schaeps,et al.  Analyzing an sEMG signal using wavelets , 2009 .

[32]  Xinjun Sheng,et al.  Common spatial-spectral analysis of EMG signals for multiday and multiuser myoelectric interface , 2019, Biomed. Signal Process. Control..

[33]  Kevin B. Englehart,et al.  A robust, real-time control scheme for multifunction myoelectric control , 2003, IEEE Transactions on Biomedical Engineering.

[34]  Novruz Allahverdi,et al.  Deep Learning on Computerized Analysis of Chronic Obstructive Pulmonary Disease , 2020, IEEE Journal of Biomedical and Health Informatics.

[35]  S. Karlsson,et al.  Mean frequency and signal amplitude of the surface EMG of the quadriceps muscles increase with increasing torque--a study using the continuous wavelet transform. , 2001, Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology.

[36]  Shalin Savalia,et al.  Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks , 2018, Bioengineering.

[37]  Eugenio Orosco,et al.  Two-dimensional myoelectric control of a robotic arm for upper limb amputees , 2007 .

[38]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[39]  R.N. Scott,et al.  A new strategy for multifunction myoelectric control , 1993, IEEE Transactions on Biomedical Engineering.

[40]  Ganesh R. Naik,et al.  Single-Channel EMG Classification With Ensemble-Empirical-Mode-Decomposition-Based ICA for Diagnosing Neuromuscular Disorders , 2016, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[41]  Roberto Merletti,et al.  Electromyography. Physiology, engineering and non invasive applications , 2005 .

[42]  Abhishek Das,et al.  Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization , 2016, 2017 IEEE International Conference on Computer Vision (ICCV).

[43]  Pradip Sircar,et al.  Automated emotion recognition based on higher order statistics and deep learning algorithm , 2020, Biomed. Signal Process. Control..

[44]  Dario Farina,et al.  Myoelectric Control of Artificial Limbs¿Is There a Need to Change Focus? [In the Spotlight] , 2012, IEEE Signal Process. Mag..

[45]  K Kanosue,et al.  The number of active motor units and their firing rates in voluntary contraction of human brachialis muscle. , 1979, The Japanese journal of physiology.

[46]  Beth Jelfs,et al.  Self-Recalibrating Surface EMG Pattern Recognition for Neuroprosthesis Control Based on Convolutional Neural Network , 2017, Front. Neurosci..