Evaluating the Effects of Feature set and Hyperparameter Optimization on sEMG-Based Gesture Recognition

• Gabriel Molina de Lima UTFPR
• André L. D. Rossi UTFPR
• José Jair Alves Mendes Junior UTFPR
• Daniel Prado Campos UTFPR
• Rafael Gomes Mantovani UTFPR

DOI: https://doi.org/10.5753/eniac.2025.11792

Abstract

Gesture recognition using myoelectric signals (sEMG) is a powerful tool for Human-Machine Interfaces (HMIs). This study investigates the classification of sEMG signals with varying sets of gestures and predictive features using Bayesian Optimization for hyperparameter tuning of Machine Learning (ML) algorithms employed for this task. Experiments were conducted on the Ninapro DB2 and DB3 datasets, covering data from both intact and amputated individuals and exploring different ML algorithms. Even though Support Vector Machines (SVMs) showed the most notable improvement through hyperparameter tuning, the best experimental results were achieved by the Random Forest (RF) classifier on both datasets, with an average F-Score of 0.853 in DB2 and 0.720 in DB3, using only five gestures and nine features. Overall, a larger feature set enhanced signal representation, particularly in reduced gesture sets, while data from amputated individuals posed greater classification challenges.

References

Atzori, M., Gijsberts, A., Castellini, C., Caputo, B., Hager, A.-G. M., Elsig, S., Giatsidis, G., Bassetto, F., and Müller, H. (2014). Electromyography data for non-invasive naturally-controlled robotic hand prostheses. Scientific Data, 1.

Atzori, M., Gijsberts, A., Heynen, S., Hager, A.-G. M., Deriaz, O., van der Smagt, P., Castellini, C., Caputo, B., and Müller, H. (2012). Building the ninapro database: A resource for the biorobotics community. In 2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), pages 1258–1265.

Aviles, M., Rodríguez-Reséndiz, J., and Ibrahimi, D. (2023). Optimizing emg classification through metaheuristic algorithms. Technologies, 11(4).

Bi, L., Feleke, A. G., and Guan, C. (2019). A review on emg-based motor intention prediction of continuous human upper limb motion for human-robot collaboration. Biomedical Signal Processing and Control, 51:113–127.

Challa, K., AlHmoud, I. W., Kamrul Islam, A. K. M., and Gokaraju, B. (2023). Emgbased hand gesture recognition using individual sensors on different muscle groups. In 2023 IEEE Applied Imagery Pattern Recognition Workshop (AIPR), pages 1–4.

de Lima, G., Campos, D., and Mantovani, R. (2024). A review on the recent use of machine learning for gesture recognition using myoelectric signals. In Anais do XXI Encontro Nacional de Inteligência Artificial e Computacional, pages 180–191, Porto Alegre, RS, Brasil. SBC.

Demšar, J. (2006). Statistical comparisons of classifiers over multiple data sets. The Journal of Machine Learning Research, 7:1–30.

Fajardo, J. M., Gomez, O., and Prieto, F. (2021). Emg hand gesture classification using handcrafted and deep features. Biomedical Signal Processing and Control, 63:102210.

Geng, W., Du, Y., Jin, W., Wei, W., Hu, Y., and Li, J. (2016). Gesture recognition by instantaneous surface emg images. Scientific Reports, 6(1):36571.

Ghaffar Nia, N., Kaplanoglu, E., and Nasab, A. (2023). Emg-based hand gestures classification using machine learning algorithms. In SoutheastCon 2023, pages 787–792.

Grattarola, A., Mora, M. C., Cerdá-Boluda, J., and Ortiz, J. V. G. (2025). Grasp pattern recognition using surface electromyography signals and bayesian-optimized support vector machines for low-cost hand prostheses. Applied Sciences, 15(3).

Huang, D. and Chen, B. (2019). Surface emg decoding for hand gestures based on spectrogram and cnn-lstm. In 2019 2nd China Symposium on Cognitive Computing and Hybrid Intelligence (CCHI), pages 123–126.

Hudgins, B., Parker, P., and Scott, R. (1993). A new strategy for multifunction myoelectric control. IEEE Transactions on Biomedical Engineering, 40(1):82–94.

Jia, G., Lam, H.-K., Liao, J., and Wang, R. (2020). Classification of electromyographic hand gesture signals using machine learning techniques. Neurocomputing, 401:236–248.

Khan, M. U., Khan, H., Muneeb, M., Abbasi, Z., Abbasi, U. B., and Baloch, N. K. (2021). Supervised machine learning based fast hand gesture recognition and classification using electromyography (emg) signals. In 2021 International Conference on Applied and Engineering Mathematics (ICAEM), pages 81–86.

Krishnan, S. and Athavale, Y. (2018). Trends in biomedical signal feature extraction. Biomedical Signal Processing and Control, 43:41–63.

Marsland, S. (2015). Machine learning: an algorithmic perspective. CRC press.

Merletti, R. and Cerone, G. L. (2020). Tutorial. surface EMG detection, conditioning and pre-processing: Best practices. J Electromyogr Kinesiol, 54:102440.

Nia, N. G., Kaplanoglu, E., and Nasab, A. (2023). Emg-based hand gestures classification using machine learning algorithms. In 2023 IEEE SoutheastCon, pages 787–792.

Ozdemir, M. A., Kisa, D. H., Guren, O., and Akan, A. (2022). Hand gesture classification using time–frequency images and transfer learning based on cnn. Biomedical Signal Processing and Control, 77:103787.

Phinyomark, A., N. Khushaba, R., and Scheme, E. (2018). Feature extraction and selection for myoelectric control based on wearable emg sensors. Sensors, 18(5).

Qiao, X., Chen, Z., and Liang, S. (2024). semg-based fine-grained gesture recognition via improved lightgbm model.

Sandoval-Espino, J. A., Zamudio-Lara, A., Marbán-Salgado, J. A., Escobedo-Alatorre, J. J., Palillero-Sandoval, O., and Velásquez-Aguilar, J. G. (2022). Selection of the best set of features for semg-based hand gesture recognition applying a cnn architecture. Sensors, 22(13).

Snoek, J., Larochelle, H., and Adams, R. P. (2012). Practical bayesian optimization of machine learning algorithms. In Pereira, F., Burges, C., Bottou, L., and Weinberger, K., editors, Advances in Neural Information Processing Systems, volume 25. Curran Associates, Inc.

Welch, P. (1967). The use of fast fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Transactions on Audio and Electroacoustics, 15(2):70–73.