EMG Hand-Gesture Classification Using Time, Frequency and Time–Frequency Domain Features with Machine-Learning Techniques
DOI:
https://doi.org/10.61467/2007.1558.2026.v17i3.1251Keywords:
EMG, Hand-gesture classification, Feature extraction SVMAbstract
This paper investigates single-channel surface electromyography for seven hand gestures using a compact feature-based machine-learning pipeline. EMG from twelve participants was segmented into contraction-level windows and described by time-, frequency- and time–frequency-domain features, complemented with simple feature-engineering transformations. Support vector machines with linear, RBF and polynomial kernels and a multilayer perceptron were evaluated. First, models were trained on the full feature set, obtaining a no-PCA baseline where linear and polynomial SVM achieved accuracies slightly above ninety-two percent and the neural network reached about eighty-nine percent. Then, Principal Component Analysis was applied, retaining two, five, nine and twelve components. Very low dimensionality degraded all models, whereas retaining nine to twelve components slightly improved RBF SVM and clearly benefited the neural network, which reached almost ninety-four percent accuracy.
Spanish-language metadata / Metadatos en español
Título en español:
Clasificación de gestos de la mano mediante EMG utilizando características de los dominios del tiempo, la frecuencia y el tiempo-frecuencia con técnicas de aprendizaje automático
Resumen:
Este artículo analiza la electromiografía de superficie monocanal para siete gestos de la mano mediante un proceso compacto de aprendizaje automático basado en características. Se segmentó el EMG de doce participantes en ventanas a nivel de contracción y se describió mediante características en los dominios del tiempo, la frecuencia y el tiempo-frecuencia, complementadas con transformaciones sencillas de ingeniería de características. Se evaluaron máquinas de vectores de soporte con núcleos lineales, RBF y polinómicos, así como un perceptrón multicapa. En primer lugar, los modelos se entrenaron con el conjunto completo de características, obteniéndose una referencia sin PCA en la que las SVM lineales y polinómicas alcanzaron precisiones ligeramente superiores al noventa y dos por ciento, mientras que la red neuronal alcanzó alrededor del ochenta y nueve por ciento. A continuación, se aplicó un análisis de componentes principales, conservando dos, cinco, nueve y doce componentes. Una dimensionalidad muy baja deterioró todos los modelos, mientras que conservar entre nueve y doce componentes mejoró ligeramente el rendimiento de la máquina de vectores de soporte RBF y benefició claramente a la red neuronal, que alcanzó una precisión de casi el noventa y cuatro por ciento.
Palabras Claves:
Clasificación de gestos con las manos, extracción de características, SVM, ANN, PCA.
Smart citations:
https://scite.ai/reports/10.61467/2007.1558.2026.v17i3.1251
Dimensions.
Open Alex.
References
Alhatemi, R. A. J., & Savaş, S. (2024). A weighted ensemble approach with multiple pre-trained deep learning models for classification of stroke. Medinformatics, 1(1), 10–19. https://doi.org/10.47852/bonviewMEDIN32021963
Anam, K., Ismail, H., Hanggara, F. S., Avian, C., & Worsito, S. B. (2021). Cross validation configuration on k-NN for finger movements using EMG signals. In 2021 7th International Conference on Instrumentation, Control, and Automation (ICA) (pp. 17–21). IEEE. https://doi.org/10.1109/ICA52848.2021.9625699
Asghari Oskoei, M., & Hu, H. (2007). Myoelectric control systems—A survey. Biomedical Signal Processing and Control, 2(4), 275–294. https://doi.org/10.1016/j.bspc.2007.07.009
Chattopadhyay, S. (2024). Towards predicting recurrence risk of differentiated thyroid cancer with a hybrid machine learning model. Medinformatics. Advance online publication. https://doi.org/10.47852/bonviewMEDIN42024441
Chen, C., Yu, Y., Sheng, X., Meng, J., & Zhu, X. (2023). Real-time hand gesture recognition by decoding motor unit discharges across multiple motor tasks from surface electromyography. IEEE Transactions on Biomedical Engineering, 70(7), 2058–2068. https://doi.org/10.1109/TBME.2023.3234642
Chen, Z., Min, H., Wang, D., Xia, Z., Sun, F., & Fang, B. (2023). A Review of Myoelectric Control for Prosthetic Hand Manipulation. Biomimetics, 8(3), Article 328. https://doi.org/10.3390/biomimetics8030328
Coskun, M., Yildirim, O., Demir, Y., & Acharya, U. R. (2022). Efficient deep neural network model for classification of grasp types using sEMG signals. Journal of Ambient Intelligence and Humanized Computing, 13(9), 4437–4450. https://doi.org/10.1007/s12652-021-03284-9
Joshi, M., & Singh, B. K. (2024). Deep learning techniques for brain lesion classification using various MRI (from 2010 to 2022): Review and challenges. Medinformatics. Advance online publication. https://doi.org/10.47852/bonviewMEDIN42021686
Khan, S. M., Khan, A. A., & Farooq, O. (2021). Pattern recognition of EMG signals for low level grip force classification. Biomedical Physics & Engineering Express, 7(6), Article 065012. https://doi.org/10.1088/2057-1976/ac2354
Lee, K. H., Min, J. Y., & Byun, S. (2022). Electromyogram-based classification of hand and finger gestures using artificial neural networks. Sensors, 22(1), Article 225. https://doi.org/10.3390/s22010225
Li, Y., Huang, Y., & Zhao, Q. (2022). Real-time collection and analysis of sports index time series based on multimodal sensor monitoring. Journal of Sensors, 2022, Article 7244856. https://doi.org/10.1155/2022/7244856
Lv, J., Yang, Y., Niu, L., Sun, X., Wang, L., Lin, W., Rong, H., & Zou, L. (2023). Prediction of hand grip strength based on surface electromyographic signals. Journal of King Saud University - Computer and Information Sciences, 35(5), Article 101548. https://doi.org/10.1016/j.jksuci.2023.04.001
Narayan, Y. (2021). Comparative analysis of SVM and Naive Bayes classifier for the SEMG signal classification. Materials Today: Proceedings, 37, 3241–3245. https://doi.org/10.1016/j.matpr.2020.09.093
Phinyomark, A., Phukpattaranont, P., & Limsakul, C. (2012). Feature reduction and selection for EMG signal classification. Expert Systems with Applications, 39(8), 7420–7431. https://doi.org/10.1016/j.eswa.2012.01.102
Rasnayake, R. M. P. K., Maduranga, M. W. P., & Sithara, J. P. D. M. (2022). Surface electromyography signal acquisition and classification using artificial neural networks (ANN). International Journal of Modern Education and Computer Science, 14(3), 64–75. https://doi.org/10.5815/ijmecs.2022.03.04
Sharma, A., Sharma, I., & Kumar, A. (2024). Signal acquisition and time-frequency perspective of EMG signal-based systems and applications. IETE Technical Review, 41(4), 466–485. https://doi.org/10.1080/02564602.2023.2265897
Subasi, A., & Qaisar, S. M. (2022). Surface EMG signal classification using TQWT, bagging and boosting for hand movement recognition. Journal of Ambient Intelligence and Humanized Computing, 13(7), 3539–3554. https://doi.org/10.1007/s12652-020-01980-6
Sun, B., Song, B., Lv, J., Chen, P., Sun, X., Ma, C., & Gao, Z. (2023). A multiscale feature extraction network based on channel-spatial attention for electromyographic signal classification. IEEE Transactions on Cognitive and Developmental Systems, 15(2), 591–601. https://doi.org/10.1109/TCDS.2022.3167042
Tepe, C., & Demir, M. C. (2022). Real-time classification of EMG Myo armband data using support vector machine. IRBM, 43(4), 300–308. https://doi.org/10.1016/j.irbm.2022.06.001
Unanyan, N. N., & Belov, A. A. (2021). Design of upper limb prosthesis using real-time motion detection method based on EMG signal processing. Biomedical Signal Processing and Control, 70, Article 103062. https://doi.org/10.1016/j.bspc.2021.103062
Zheng, Z., Wu, Z., Zhao, R., Ni, Y., Jing, X., & Gao, S. (2022). A review of EMG-, FMG-, and EIT-based biosensors and relevant human–machine interactivities and biomedical applications. Biosensors, 12(7), Article 516. https://doi.org/10.3390/bios12070516
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 International Journal of Combinatorial Optimization Problems and Informatics

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.