Research on gesture recognition method based on PSO optimized LSTM for surface muscle signals
Abstract
(sEMG) Surface EMG signals can quickly and accurately reflect the movement intention of human body, gesture recognition by sEMG has been widely studied in the field of human-robot interaction in rehabilitation robots, in order to improve the accuracy of gesture recognition. A gesture recognition method based on PSO optimised LSTM for surface muscle signals is investigated. Firstly, the S_Dbw method is used to evaluate the clustering validity of the time domain features of the surface EMG signals, and the feature (MAV)combination of absolute mean +(RMS)root mean square (RMS)+ (VAR) variance is selected as the feature space and input to the model of the depth algorithm. Secondly, in order to solve the problem of cumbersome parameter debugging and low accuracy of (LSTM)Long Short-Term Memory neural network (LSTM), Particle Swarm Optimization-Long Short-Term Memory PSO-LSTM algorithm is proposed and combined with convolutional neural network (convolutional neural network, CNN) and Long Short-Term Memory LSTM. The experimental results show that the PSO-LSTM model has an average recognition accuracy of 94.1%, which is 8.3% and 4.7% higher than the CNN and LSTM, and the average training time of the LSTM model is reduced by 35.52s compared to the LSTM. Therefore, it is advantageous to use PSO-LSTM model in gesture sEMG signal recognition and classification.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Zero Crossing Rate (ZCR) measures signal waveform zero-crossing frequency, allows rapid computation of muscle activation states, and is used for time-domain feature extraction for sEMG-based gesture recognition.
“| Zero Crossing Rate (ZCR) | Measure signal waveform zero-crossing | Rapid computation of muscle | Time-domain feature extraction for sEMG-based |”
From Recent advances in intelligent wearable systems: from multiscale biomechanical features towards human motion intent prediction (2026) · cites this paper as [244]
LSTM resolves long-term dependencies through gated mechanisms, dynamically parses continuous motion trajectories, and is used for exoskeleton joint angle prediction.
“| LSTM | Resolve long-term dependencies through gated | Dynamically parse continuous motion | Exoskeleton joint angle prediction 148,244,245 |”
From Recent advances in intelligent wearable systems: from multiscale biomechanical features towards human motion intent prediction (2026) · cites this paper as [244]
Cites
Hybrid speech recognition with Deep Bidirectional LSTM
Graves, A., Jaitly, N., & Mohamed, A. (2013) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru.2013.6707742 · Cited by 1,790 · see citing works
Understanding of Internal Clustering Validation Measures
Liu, Y., Li, Z., Xiong, H., Gao, X., & Wu, J. (2010) · Proceedings · doi:10.1109/icdm.2010.35 · Cited by 1,031 · see citing works
Clustering validity assessment: finding the optimal partitioning of a data set
Halkidi, M., & Vazirgiannis, M. (2002) · Proceedings · doi:10.1109/icdm.2001.989517 · Cited by 433 · see citing works
Characterizing EMG data using machine-learning tools
Yousefi, J., & Hamilton-Wright, A. (2014) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2014.04.018 · Cited by 159 · see citing works
Phukpattaranont, P., Thongpanja, S., Anam, K., Al-Jumaily, A. A., & Limsakul, C. (2018) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-018-1857-5 · Cited by 64 · see citing works
EMG-Based Motion Discrimination Using a Novel Recurrent Neural Network
Bu, N., Fukuda, O., & Tsuji, T. (2003) · Journal of Intelligent Information Systems · doi:10.1023/a:1024706431807 · Cited by 49 · see citing works
Hand Gesture Recognition Based on Deep Learning Method
Xing, K., Ding, Z., Jiang, S., Ma, X., Yang, K., Yang, C., Li, X., & Jiang, F. (2018) · IEEE International Conference on Data Science in Cyberspace · doi:10.1109/dsc.2018.00087 · Cited by 49 · see citing works
Qureshi, M. F., Mushtaq, Z., Zia‐ur‐Rehman, M., & Kamavuako, E. N. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3204121 · Cited by 43 · see citing works
Betthauser, J. L., Krall, J. T., Kaliki, R. R., Fifer, M. S., & Thakor, N. V. (2019) · International IEEE/EMBS Conference on Neural Engineering · doi:10.1109/ner.2019.8717169 · Cited by 29 · see citing works
The Study of Gesture Recognition Based on SVM with LBP and PCA
Wang, J., Xu, X., & Li, M. (2015) · Journal of Image and Graphics · doi:10.18178/joig.3.1.16-19 · Cited by 13 · see citing works
Liu, J., Furusawa, K., Tateyama, T., et al. (2019) · Journal of Image and Graphics · doi:10.18178/joig.7.2.45-49 · Cited by 13 · see citing works
Türkmen, C., Konca, M., & Yetim, B. (2022) · Acta Neurologica Belgica · doi:10.1007/s13760-022-01982-5 · Cited by 2 · see citing works
The Mapping Between Hand Motion States Induced by Arm Operation and Surface Electromyography
Hou, T., Qian, C., Lü, Y., et al. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-91122-9_27 · Cited by 1 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Chen, S., Peng, C., Yang, B., Lin, J., Zhou, L., Jiang, Z., Liu, Z., Liu, Y., & Tang, L. (2026) · npj Artificial Intelligence · doi:10.1038/s44387-026-00083-5 · Cited by 9 · see citing works
- Authors
- Dai, J., Lu, G., Qin, Z., Guo, X., Liu, X., & Yin, Q.
- Affiliations
- Tianjin University of Technology and Education (CN)
- Venue
- 2024 6th Asia Symposium on Image Processing (ASIP)
- Publisher
- IEEE
- Published
- 2024-06-13
- Pages
- 82-87
- OpenAlex
- W4404295185
- Cited by
- 1
- References
- 13
- Citations by year
- 2026: 1
- FWCI
- 0.94
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Asia Symposium on Image Processing (ASIP) (publishedVersion)
- Funders
- Tianjin Municipal Education Commission
- Topics
- Physical Sciences › Engineering › Biomedical Engineering › Muscle activation and electromyography studies
- Physical Sciences › Computer Science › Human-Computer Interaction › Hand Gesture Recognition Systems
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Emotion and Mood Recognition
- Keywords
- gesture recognition, surface electromyography, particle swarm optimization, LSTM, PSO-LSTM, feature selection, mean absolute value, root mean square, signal variance, time-domain features, rehabilitation robots, human-robot interaction, CNN comparison
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
