基于GMM-HMMs与Viterbi回溯的连续手势肌电信号预测与识别
作者:
作者单位:

1.黎明职业大学;2.中国科学院海西研究院泉州装备制造研究中心

基金项目:

福建省科技计划项目(2022L3094);泉州市科技计划项目(2021C021R);黎明职业大学2024年度规划项目(LZ202406)


Continuous gesture prediction and recognition using GMM-HMMs and Viterbi backtracking with sEMG
Author:
Affiliation:

1.Liming Vocational University;2.Quanzhou Institute of Equipment Manufacturing Haixi Institutes, Chinese Academy of Science

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    针对基于表面肌电信号(sEMG)的连续手势识别任务中,存在实时性较差和预测能力不足的问题,提出基于GMM-HMMs和Viterbi回溯的连续手势动作识别方法.采用滑动窗口对8通道肌电信号进行分窗,通过GMM-HMMs模型建立手势的空闲、上升、稳定和下降四个动作状态,提出改进的Viterbi滑动窗口边缘化策略,建立滑动窗口长期约束,实现连续手势动作状态预测.最终引入最大似然法动态阈值模型以区分手势类别.在由8位实验者完成的包含4种手势的12个连续双手势动作任务中,该方法的平均识别率为98.1%,预测时间为71ms,明显优于LSTM模型(94.2%,309ms)和GRU模型(93.8%,300ms),预测时间显著减少.

    Abstract:

    Addressing the issues of poor real-time performance and insufficient predictive capability in continuous gesture recognition tasks based on surface electromyography (sEMG), a method utilizing Gaussian Mixture Model-Hidden Markov Models (GMM-HMMs) and Viterbi traceback is proposed for continuous gesture recognition. This approach leverages Gaussian Mixture Models-Hidden Markov Models (GMM-HMMs) to classify hand gestures into idle, ascending, steady, and descending states. A refined Viterbi sliding window marginalization technique is implemented to ensure prolonged connections between adjacent windows, enabling anticipatory prediction of subsequent gesture states. Moreover, a dynamic threshold model based on maximum likelihood is incorporated to accurately categorize gestures. In a study where 8 participants performed 12 continuous dual-gesture tasks, the method proposed attained an average recognition rate of 98.1% with a prediction time of 71ms, surpassing the LSTM model (94.2%, 309ms) and the GRU model (93.8%, 300ms), resulting in a considerable decrease in prediction time.

    参考文献
    [1] 丁其川, 熊安斌, 赵新刚, 等. 基于表面肌电的运动意图识别方法研究及应用综述[J]. 自动化学报, 2016, 42(1): 13-25.
    [2] GAO Z, TANG R, HUANG Q, et al. A Multi-dof Prosthetic Hand Finger Joint Controller for Wearable semg Sensors by Nonlinear Autoregressive Exogenous Model[J]. Sensors, 2021, 21(8): 2576.
    [3] KERDJIDJ O, AMARA K, HARIZI F, et al. Implementing hand gesture recognition using EMG on the Zynq circuit[J]. IEEE Sensors Journal, 2023, 23(9): 10054-10061.
    [4] TRIGILI E, GRAZI L, CREA S, et al. Detection of movement onset using EMG signals for upper-limb exoskeletons in reaching tasks[J]. Journal of NeuroEngineering and Rehabilitation, 2019, 16(1): 45.
    [5] LI K, ZHANG J, WANG L, et al. A review of the key technologies for sEMG-based human-robot interaction systems[J]. Biomedical Signal Processing and Control, 2020, 62: 102074.
    [6] WANG S, HUANG L, JIANG D, et al. Improved multi-stream convolutional block attention module for sEMG-based gesture recognition[J]. Frontiers in Bioengineering and Biotechnology, 2022, 10.
    [7] WANG Y, WU Q, DEY N, et al. Deep back propagation–long short-term memory network based upper-limb sEMG signal classification for automated rehabilitation[J]. Biocybernetics and Biomedical Engineering, 2020, 40(3): 987-1001.
    [8] TORO-OSSABA A, JARAMILLO-TIGREROS J, TEJADA J C, et al. LSTM recurrent neural network for hand gesture recognition using EMG signals[J]. Applied Sciences, 2022, 12(19): 9700s
    [9] 丁其川, 赵新刚, 李自由, 等. 基于自更新混合分类模型的肌电运动识别方法[J]. 自动化学报, 2019, 45(8): 1464-1474.
    [10] YANG J, PAN J, LI J. sEMG-based continuous hand gesture recognition using GMM-HMM and threshold model[C]//2017 IEEE International Conference on Robotics and Biomimetics (ROBIO). Macau: IEEE, 2017: 1509-1514.
    [11] 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.
    [12] KIM E, SHIN J, KWON Y, et al. EMG-based dynamic hand gesture recognition using edge AI for human–robot interaction[J]. Electronics, 2023, 12(7): 1541.
    [13] WU L, CHEN X, CHEN X, et al. Rejecting novel motions in high-density myoelectric pattern recognition using hybrid neural networks[J]. Frontiers in Neurorobotics, 2022, 16: 862193.
    [14] XIONG D, ZHANG D, ZHAO X, et al. Deep learning for EMG-based human-machine interaction: a review[J]. IEEE/CAA Journal of Automatica Sinica, 2021, 8(3): 512-533.
    [15] YU B, ZHANG X, WU L, et al. A novel postprocessing method for robust myoelectric pattern-recognition control through movement pattern transition detection[J]. IEEE Transactions on Human-Machine Systems, 2020, 50(1): 32-41.
    [16] CHEN Y, ZHANG H, WANG C, et al. A hierarchical dynamic Bayesian learning network for EMG-based early prediction of voluntary movement intention[J]. Scientific Reports, 2023, 13(1): 4730.
    [17] ZHENG K, LIU S, YANG J, et al. sEMG-based continuous hand action prediction by using key state transition and model pruning[J]. Sensors, 2022, 22(24): 9949.
    [18] D'' ACCOLTI D, CLEMENTE F, MANNINI A, et al. Online classification of transient EMG patterns for the control of the wrist and hand in a transradial prosthesis[J]. IEEE Robotics and Automation Letters, 2023, 8(2): 1045-1052.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

杨进兴,刘帅,李俊.基于GMM-HMMs与Viterbi回溯的连续手势肌电信号预测与识别[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-09-05
  • 最后修改日期:2024-11-13
  • 录用日期:2024-11-13

地址:江苏省南京市宁六路219号    邮编:210044

联系电话:025-58731025    E-mail:nxdxb@nuist.edu.cn

南京信息工程大学学报 ® 2025 版权所有  技术支持:北京勤云科技发展有限公司