北京生物医学工程

基于对抗适应网络的跨个体肌电手势识别方法

Cross-subject hand gesture recognition with surface electromyography based on adversarial adaptation network

作者：朱九英米红林付佳杰

单位：上海电子信息职业技术学院（上海 201411）<br />通信作者：朱九英。E-mail: 82881135@qq.com

分类号：R318.04

出版年·卷·期（页码）：2023·42·2（124-129）

摘要：

目的表面肌电信号可以直接反映用户的动作意图，近年来已经成为手势识别等人机交互任务的主要控制信号。然而，个体差异性使得用户模型不能通用，限制了其应用与发展。为了解决这个问题，论文提出了一种新的跨个体对抗适应网络(cross-subject adversarial adaptation network, CAAN)。方法该网络包括特征编码器、手势分类器和个体分类器3个子模块，使用了新的对抗性适应训练方法训练网络，达到分离出个体私有特征的目标。CAAN网络在采集的数据集上进行训练和测试，数据集包括11名受试者的6种手势。结果实验结果表明，方法的手势识别准确率达到88.08%，通过比较，该方法的性能优于现有的方法。结论本文提出的CAAN网络可有效进行跨个体手势识别，为人机交互提供可靠的技术。

Objective Intra-subject hand gestures recognition based on surface electromyography has been extensively researched in current years, however, the gesture recognition on cross-subject tasks has more broad application prospects. Methods For addressing this problem, we propose a novel cross-subject adversarial adaptation network (CAAN), which fulfills the intra-subject gesture recognition task. An adversarial adaptation training method is developed to train the network to encourage the emergence of the features that are discriminative and subject-independent. The CAAN is evaluated on the collected dataset (including six gestures from eleven subjects). Results The experimental results show that the proposed method outperforms state-of-arts, which achieving offline accuracies at 88.08% respectively. Conclusions The proposed CAAN can effectively carry out cross-subject gesture recognition and provide reliable technology for human-computer interaction.

参考文献：

[1] Zhang Z, Yang K, Qian J, et al. Real-time surface EMG pattern recognition for hand gestures based on an artificial neural network[J]. Sensors, 2019, 19(14): 3170.?
[2] Zhang Z, He C, Yang K. A novel surface electromyographic signal-based hand gesture prediction using a recurrent neural network[J]. Sensors, 2020, 20(14): 3994.
[3] Geng W, Du Y, Jin W, et al. Gesture recognition by instantaneous surface EMG images[J]. Scientific Reports, 2016, 6(1): 36571.
[4] Hu Y, Wong Y, Wei W, et al. A novel attention-based hybrid CNN-RNN architecture for sEMG-based gesture recognition[J]. PLoS One, 2018, 13(10): e0206049.
[5] C?té-Allard U, Fall CL, Drouin A, et al. Deep learning for electromyographic hand gesture signal classification using transfer learning[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27(4): 760-771.
[6] Chen X, Li Y, Hu R, et al. Hand gesture recognition based on surface electromyography using convolutional neural network with transfer learning method[J]. IEEE Journal of Biomedical and Health Informatics, 2020, 25(4): 1292-1304.
[7] Yasen M, Jusoh S. A systematic review on hand gesture recognition techniques, challenges and applications[J]. PeerJ Computer Science, 2019, 5: e218.
[8] Xie B, Meng J, Li B, et al. Biosignal-based transferable attention Bi-ConvGRU deep network for hand-gesture recognition towards online upper-limb prosthesis control[J]. Computer Methods and Programs in Biomedicine, 2022, 224: 106999.
[9] Ganin Y, Ustinova E, Ajakan H, et al. Domain-adversarial training of neural networks[J]. The Journal of Machine Learning Research, 2016, 17(1): 2096-2030.
[10] Karpathy A, Toderici G, Shetty S, et al. Large-scale video classification with convolutional neural networks[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2020). Seattle, WA, USA: IEEE Press, 2014: 1725-1732.
[11] Benalcázar ME, Motoche C, Zea JA, et al. Real-time hand gesture recognition using the myo armband and muscle activity detection[C]//2017 IEEE Second Ecuador Technical Chapters Meeting (ETCM). Salinas, Ecuador: IEEE Press, 2017: 1-6.
[12] Atzori M, Cognolato M, Müller H. Deep learning with convolutional neural networks applied to electromyography data: a resource for the classification of movements for prosthetic hands[J]. Frontiers in Neurorobotics, 2016, 10: 9.
[13] Colli-Alfaro JG, Ibrahim A, Trejos AL. Design of user-independent hand gesture recognition using multilayer perceptron networks and sensor fusion techniques[C]// 2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR). Toronto, ON, Canada: IEEE Press, 2019: 1103-1108.
[14] Du Y, Jin W, Wei W, et al. Surface EMG-based inter-session gesture recognition enhanced by deep domain adaptation[J]. Sensors, 2017, 17(3): 458.
[15] Kim M, Chung WK, Kim K. Subject-independent sEMG pattern recognition by using a muscle source activation model[J]. IEEE Robotics and Automation Letters, 2020, 5(4): 5175-5180.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录