基于表面肌电图手势动作意图识别的系统综述

doi:10.3969/j.issn.1006-9771.2022.09.005

《中国康复理论与实践》 ›› 2022, Vol. 28 ›› Issue (9): 1032-1038.doi: 10.3969/j.issn.1006-9771.2022.09.005

基于表面肌电图手势动作意图识别的系统综述

朱旭,刘静,董泽萍,仇大伟()

山东中医药大学智能与信息工程学院,山东济南市 250355

收稿日期:2022-03-18 修回日期:2022-06-08 出版日期:2022-09-25 发布日期:2022-10-08
通讯作者: 仇大伟 E-mail:dwqiu@foxmail.com
作者简介:朱旭(1998-),男,汉族,河南信阳市人,硕士研究生,主要研究方向：表面肌电与手部康复。
基金资助:
国家自然科学基金面上项目(82174528);国家自然科学基金面上项目(82074579);山东省中医药科技发展计划项目(2019-0056)

Gesture action intent recognition based on surface electromyography: a systematic review

ZHU Xu,LIU Jing,DONG Zeping,QIU Dawei()

College of Intelligent and Information Engineering, Shandong University of Traditional Chinese Medicine, Ji'nan, Shandong 250355, China

Received:2022-03-18 Revised:2022-06-08 Published:2022-09-25 Online:2022-10-08
Contact: QIU Dawei E-mail:dwqiu@foxmail.com
Supported by:
National Natural Science Foundation of China (General)(82174528);National Natural Science Foundation of China (General)(82074579);Shandong Science and Technology Development Program of Traditional Chinese Medicine(2019-0056)

摘要/Abstract

摘要：

目的系统综述基于表面肌电图的手势动作意图识别的研究进展。
方法通过检索中国知网、万方数据库、PubMed、Web of Science,搜集基于表面肌电图的手势动作意图识别的实验研究,检索时限为建库至2021年12月。根据实验内容和质量筛选文献,并对其分类方法和其他影响因素进行总结性分析。
结果共返回文献735篇,最终纳入25篇,发表时间主要集中于2012年至2021年,研究对象为正常受试者或截肢者,分类模型包含传统机器学习模型和深度学习模型,其他影响因素包含采集方式、噪声干扰和滑动窗口大小。
结论目前基于表面肌电信号的传统机器学习模型已得到成熟应用,深度学习模型的手势识别技术具有很大潜力。受试者的个体差异、手势分类的实时性需求和肌电设备的稳定性需求仍有待解决。

关键词: 人机交互, 表面肌电图, 手势识别, 机器学习, 系统综述

Abstract:

Objective To systematicly review the researches of gesture action intent recognition based on surface electromyography (sEMG).
Methods Experimental researches on gesture action intention recognition based on sEMG were retrieved from CNKI, Wanfang Data, PubMed and Web of Science. The literatures were screened, and the classification methods and other related factors were summarized.
Results A total of 735 researches were returned, and 25 researches were finally included. The publication time was mainly from 2012 to 2021. The subjects were healthy people or amputees. The classification model included traditional machine learning models and deep learning models. Other related factors included acquisition, noise interference and sliding window size.
Conclusion Traditional machine learning models based on sEMG signals have been maturely applied, and gesture recognition with deep learning models are of great potential. The individual differences of subjects, the real-time requirements of gesture classification and the stability requirements of sEMG devices still need to be addressed.

Key words: human-machine interaction, surface electromyography, gesture recognition, machine learning, systematic review

中图分类号:

R496

朱旭,刘静,董泽萍,仇大伟. 基于表面肌电图手势动作意图识别的系统综述[J]. 《中国康复理论与实践》, 2022, 28(9): 1032-1038.

ZHU Xu,LIU Jing,DONG Zeping,QIU Dawei. Gesture action intent recognition based on surface electromyography: a systematic review[J]. 《Chinese Journal of Rehabilitation Theory and Practice》, 2022, 28(9): 1032-1038.

图/表 2

图1

表1

纳入文献的一般情况"

纳入文献	受试者	手势数	分类模型	结论
Wu等^[12](2020)	4例男性、2例女性	6	BPNN	与sEMG特征数量相比,测量位置数量变化对识别结果影响更大
Naik等^[13](2016)	5例截肢受试者	12	ICA-SSO	减少肌电传感器的数量,提高肌电控制系统的鲁棒性
Du等^[14](2017)	23例健康受试者	22	DBN	先进的分类器使用高密度sEMG数据集比稀疏多通道sEMG得到更高准确率
Chen等^[15](2020)	18例健康受试者	8	3D-CNN	3D-CNN比2D-CNN能更好识别高密度sEMG数据集,但处理时间也增加
Duan等^[16](2018)	4例男性、4例女性	9	LDA	使用3个传感器识别9个手势,平均准确率91.7%
Zhang等^[17](2019)	8例男性、4例女性	5	ANN	手势识别率98.7%
Saeed等^[18](2021)	10例健康受试者、6例截肢受试者	11	LDA、ANN	与原始信号相比,特征组合作为输入能得到更高识别率
Tang等^[19](2012)	6例男性	11	LDA、KNN、NB	通过定义新的特征和设计新的级联结构提高手势识别率
Xue等^[20](2021)	8例男性、2例女性	13	CCA-OT	减少不同受试者肢体之间概率函数漂移的必要性
Sun等^[21](2019)	8例男性、2例女性	52	GFM	GFM分解特征因子对提高肌电手势分类准确率有积极作用
Cao等^[22](2021)	36例健康受试者	4	AMPSO-SVM	遗传算法用于解决高维和冗余问题
Junior等^[23](2020)	8例男性、5例女性	6	ELM、KNN、SVM-RBF等	降维和特征量缩减可以有效提高手势识别率
Jie等^[24](2021)	7例男性、1例女性	5	PSO	少量特征仍然可以达到高密度sEMG信号的手势识别率
Sharif等^[28](2020)	6例男性、2例女性	16	SVM、CNN	动态手势的手势识别精度不如静态手势
Zhai等^[29](2017)	40例健康受试者、11例截肢受试者	49	CNN	重新校准后的CNN可以补偿sEMG基准漂移
Cheng等^[30](2021)	20例男性、7例女性	52	CNN	sEMG多特征图像能够更好实现信息互补,提高手势分类准确率
Asif等^[31](2020)	18例健康受试者	10	CNN	单个网络会明显偏好个别手势
许留凯等^[32](2021)	40例健康受试者	49	CNN	CNN结合能量核相图有着较好的手势识别精度
李沿宏等^[33](2021)	20例男性、7例女性	52	多流卷积网络	多流卷积网络学习sEMG信号的时序信息
Nasri等^[34](2019)	35例健康受试者	6	GRU	GRU实时评估系统能区分大约80%训练手势
Zhang等^[35](2020)	8例男性、5例女性	21	GRU	文献提出的模型有着更好的实时性能和准确性
Yu等^[36](2019)	8例男性	150	DBN	特征级融合作为网络的输入有着最好的手势识别率

表1

参考文献 41

[1]	SAMADANI A A, KULIC D. Hand gesture recognition based on surface electromyography[C]. 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, 2014: 4196-4199.
[2]	OUDAH M, AL-NAJI A, CHAHL J. Hand gesture recognition based on computer vision: a review of techniques[J]. J Imaging, 2020, 6(8): 73. doi: 10.3390/jimaging6080073
[3]	ATZORI M, GIJSBERTS A, CASTELLINI C, et al. Electromyography data for non-invasive naturally-controlled robotic hand prostheses[J]. Sci Data, 2014, 1(1): 1-13. doi: 10.3233/DS-170013
[4]	DISSELHORST-KLUG C, SCHMITZ-RODE T, RAU G. Surface electromyography and muscle force: limits in sEMG-force relationship and new approaches for applications[J]. Clin Biomech (Bristol, Avon), 2009, 24(3): 225-235. doi: 10.1016/j.clinbiomech.2008.08.003
[5]	BITTIBSSI T M, GENEDY M A, MAGED S A. sEMG pattern recognition based on recurrent neural network[J]. Biomed Signal Process Control, 2021, 70: 103048. doi: 10.1016/j.bspc.2021.103048
[6]	曹梦琳, 陈宇豪, 王珏, 等. 基于表面肌电图的人体运动意图识别研究进展[J]. 中国康复理论与实践, 2021, 27(5): 595-603.
	CAO M L, CHEN Y H, WANG J, et al. Advance in human motion intention recognition based on surface electromyography[J]. Chin J Rehabil Theory Pract, 2021, 27(5): 595-603.
[7]	胡进, 侯增广, 陈翼雄, 等. 下肢康复机器人及其交互控制方法[J]. 自动化学报, 2014, 40(11): 2377-2390.
	HU J, HOU Z G, CHEN Y X, et al. Lower limb rehabilitation robots and interactive control methods[J]. Acta Autom Sin, 2014, 40(11): 2377-2390.
[8]	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.
	DING Q C, XIONG A B, ZHAO X G, et al. A review on researches and applications of sEMG-based motion intent recognition methods[J]. Acta Autom Sin, 2016, 42(1): 13-25.
[9]	AFSHARIPOUR B, SOEDIRDJO S, MERLETTI R. Two-dimensional surface EMG: the effects of electrode size, interelectrode distance and image truncation[J]. Biomed Signal Process Control, 2019, 49: 298-307. doi: 10.1016/j.bspc.2018.12.001
[10]	JUNIOR J J A M, FREITAS M, SIQUEIRA H, et al. Comparative analysis among feature selection of sEMG signal for hand gesture classification by armband[J]. IEEE Lat Am Trans, 2020, 18(6): 1135-1143. doi: 10.1109/TLA.2020.9099752
[11]	LI W, SHI P, YU H. Gesture recognition using surface electromyography and deep learning for prostheses hand: state-of-the-art, challenges, and future[J]. Front Neurosci, 2021, 15: 621885. doi: 10.3389/fnins.2021.621885
[12]	WU C, YAN Y, CAO Q, et al. sEMG measurement position and feature optimization strategy for gesture recognition based on ANOVA and neural networks[J]. IEEE Access, 2020, 8: 56290-56299. doi: 10.1109/ACCESS.2020.2982405
[13]	NAIK G R, AL-TIMEMY A H, NGUYEN H T. Transradial amputee gesture classification using an optimal number of sEMG sensors: an approach using ICA clustering[J]. IEEE Trans Neural Syst Rehabil Eng, 2016, 24(8): 837-846. doi: 10.1109/TNSRE.2015.2478138
[14]	DU Y, JIN W, WEI W, et al. Surface EMG-based inter-session gesture recognition enhanced by deep domain adaptation[J]. Sensors (Basel), 2017, 17(3): 458. doi: 10.3390/s17030458
[15]	CHEN J, BI S, ZHANG G, et al. High-density surface EMG-based gesture recognition using a 3D convolutional neural network[J]. Sensors (Basel), 2020, 20(4): 1201. doi: 10.3390/s20041201
[16]	DUAN F, REN X, YANG Y. A gesture recognition system based on time domain features and linear discriminant analysis[J]. IEEE Trans Cogn Dev Syst, 2018, 13(1): 200-208. doi: 10.1109/TCDS.2018.2884942
[17]	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 (Basel), 2019, 19(14): 3170. doi: 10.3390/s19143170
[18]	SAEED B, ZIA-UR-REHMAN M, GILANI S O, et al. Leveraging ANN and LDA classifiers for characterizing different hand movements using EMG signals[J]. Arab J Sci Eng, 2021, 46(2): 1761-1769. doi: 10.1007/s13369-020-05044-x
[19]	TANG X, LIU Y, LV C, et al. Hand motion classification using a multi-channel surface electromyography sensor[J]. Sensors (Basel), 2012, 12(2): 1130-1147. doi: 10.3390/s120201130
[20]	XUE B, WU L, WANG K, et al. Multiuser gesture recognition using sEMG signals via canonical correlation analysis and optimal transport[J]. Comput Biol Med, 2021, 130: 104188. doi: 10.1016/j.compbiomed.2020.104188
[21]	SUN W, LIU H, TANG R, et al. sEMG-based hand-gesture classification using a generative flow model[J]. Sensors (Basel), 2019, 19(8): 1952. doi: 10.3390/s19081952
[22]	CAO L, ZHANG W, KAN X, et al. A novel adaptive mutation PSO optimized SVM algorithm for sEMG-based gesture recognition[J]. Sci Programming, 2021, 2021: 9988823.
[23]	JUNIOR J J A M, FREITAS M L B, SIQUEIRA H V, et al. Feature selection and dimensionality reduction: an extensive comparison in hand gesture classification by sEMG in eight channels armband approach[J]. Biomed Signal Process Control, 2020, 59: 101920. doi: 10.1016/j.bspc.2020.101920
[24]	JIE J, LIU K, ZHENG H, et al. High dimensional feature data reduction of multichannel sEMG for gesture recognition based on double phases PSO[J]. Complex Intell Syst, 2021, 7(4): 1877-1893. doi: 10.1007/s40747-020-00232-6
[25]	MENDES JUNIOR J J A, FREITAS M L B, CAMPOS D P, et al. Analysis of influence of segmentation, features, and classification in sEMG processing: a case study of recognition of brazilian sign language alphabet[J]. Sensors (Basel), 2020, 20(16): 4359. doi: 10.3390/s20164359
[26]	TRIWIYANTO T, PAWANA I P A, PURNOMO M H. An improved performance of deep learning based on convolution neural network to classify the hand motion by evaluating hyper parameter[J]. IEEE Trans Neural Syst Rehabil Eng, 2020, 28(7): 1678-1688. doi: 10.1109/TNSRE.2020.2999505
[27]	CHEN L, FU J, WU Y, et al. Hand gesture recognition using compact CNN via surface electromyography signals[J]. Sensors (Basel), 2020, 20(3): 672. doi: 10.3390/s20030672
[28]	SHARIF H, SEO S B, KESAVADAS T K. Hand gesture recognition using surface electromyography[C]. 2020 42nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, 2020: 682-685.
[29]	ZHAI X, JELFS B, CHAN R H M, et al. Self-recalibrating surface EMG pattern recognition for neuroprosthesis control based on convolutional neural network[J]. Front Neurosci, 2017, 11: 379. doi: 10.3389/fnins.2017.00379
[30]	CHENG Y, LI G, YU M, et al. Gesture recognition based on surface electromyography‐feature image[J]. Concurr Comp-Pract E, 2021, 33(6): e6051.
[31]	ASIF A R, WARIS A, GILANI S O, et al. Performance evaluation of convolutional neural network for hand gesture recognition using EMG[J]. Sensors (Basel), 2020, 20(6): 1642. doi: 10.3390/s20061642
[32]	许留凯, 张克勤, 徐兆红, 等. 基于表面肌电信号能量核相图的卷积神经网络人体手势识别算法[J]. 生物医学工程学杂志, 2021, 38(4): 621-629.
	XU L K, ZHANG K Q, XU Z H, et al. Convolutional neural network human gesture recognition algorithm based on phase portrait of surface electromyography energy kernel[J]. J Biomed Eng, 2021, 38(4): 621-629.
[33]	李沿宏, 江茜, 邹可, 等. 融合注意力机制的多流卷积肌电手势识别网络[J]. 计算机应用研究, 2021, 38(11): 3258-3263.
	LI Y H, JIANG X, ZOU K, et al. Multi-stream convolutional myoelectric gesture recognition networks fusing attentional mechanisms[J]. Appl Res Comput, 2021, 38(11): 3258-3263.
[34]	NASRI N, ORTS-ESCOLANO S, GOMEZ-DONOSO F, et al. Inferring static hand poses from a low-cost non-intrusive sEMG sensor[J]. Sensors (Basel), 2019, 19(2): 371. doi: 10.3390/s19020371
[35]	ZHANG Z, HE C, YANG K. A novel surface electromyographic signal-based hand gesture prediction using a recurrent neural network[J]. Sensors (Basel), 2020, 20(14): 3994. doi: 10.3390/s20143994
[36]	YU Y, CHEN X, CAO S, et al. Exploration of Chinese sign language recognition using wearable sensors based on deep belief net[J]. IEEE J Biomed Health Inform, 2019, 24(5): 1310-1320. doi: 10.1109/JBHI.2019.2941535
[37]	WEI W, WONG Y, DU Y, et al. A multi-stream convolutional neural network for sEMG-based gesture recognition in muscle-computer interface[J]. Pattern Recogn Lett, 2019, 119: 131-138. doi: 10.1016/j.patrec.2017.12.005
[38]	AMMA C, KRINGS T, BÖER J, et al. Advancing muscle-computer interfaces with high-density electromyography[C]. Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (CHI), 2015: 929-938.
[39]	MATRAN-FERNANDEZ A, RODRÍGUEZ MARTÍNEZ I J, POLI R, et al. SEEDS, simultaneous recordings of high-density EMG and finger joint angles during multiple hand movements[J]. Sci Data, 2019, 6(1): 186. doi: 10.1038/s41597-019-0200-9
[40]	ENGLEHART K, HUDGINS B. A robust, real-time control scheme for multifunction myoelectric control[J]. IEEE Trans Biomed Eng, 2003, 50(7): 848-854. doi: 10.1109/TBME.2003.813539
[41]	HUDGINS B, PARKER P, SCOTT RN. A new strategy for multifunction myoelectric control[J]. IEEE Trans Biomed Eng, 1993, 40(1): 82-94. doi: 10.1109/10.204774

基于表面肌电图手势动作意图识别的系统综述

Gesture action intent recognition based on surface electromyography: a systematic review

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 2

参考文献 41

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	王航宇, 葛可可, 范永红, 都丽露, 邹敏, 封磊. 基于ICD-11和ICF主动式音乐疗法改善认知障碍老年人认知功能的系统综述[J]. 《中国康复理论与实践》, 2024, 30(1): 36-43.
[2]	闻嘉宁, 金秋艳, 张琦, 李杰, 司琦. 认知参与型身体活动对发展儿童青少年执行功能的效果：基于ICF的系统综述[J]. 《中国康复理论与实践》, 2024, 30(1): 44-53.
[3]	葛可可, 范永红, 王航宇, 都丽露, 李长江, 邹敏. 失眠老年人正念干预健康效益的系统综述[J]. 《中国康复理论与实践》, 2024, 30(1): 54-60.
[4]	张婧雅, 邹敏, 孙宏伟, 孙昌隆, 朱峻同. 听障儿童青少年焦虑或抑郁情绪心理干预效果的系统综述[J]. 《中国康复理论与实践》, 2023, 29(9): 1004-1011.
[5]	王俊宇, 杨永, 袁逊, 谢婷, 庄洁. 高强度间歇训练对健康儿童青少年执行功能效果的系统综述[J]. 《中国康复理论与实践》, 2023, 29(9): 1012-1020.
[6]	魏晓微, 杨剑, 魏春艳. 特殊教育学校孤独症谱系障碍儿童参与适应性瑜伽活动的心理与行为效益的系统综述[J]. 《中国康复理论与实践》, 2023, 29(9): 1021-1028.
[7]	杨亚茹, 杨剑. 基于WHO-HPS架构学校身体活动相关健康服务及其健康效益：系统综述的系统综述[J]. 《中国康复理论与实践》, 2023, 29(9): 1040-1047.
[8]	史佳伟, 李凌宇, 杨浩杰, 王琴潞, 邹海欧. 预康复对全膝关节置换术后患者的有效性：系统综述的系统综述[J]. 《中国康复理论与实践》, 2023, 29(9): 1057-1064.
[9]	蔡华年, 费思先, 张忆晨, 孙青, 郭帅, 宋韬. 基于导纳控制的双边康复机器人运动辅助分析[J]. 《中国康复理论与实践》, 2023, 29(9): 1104-1109.
[10]	蒋长好, 黄辰, 高晓妍, 戴元富, 赵国明. 神经反馈训练对老年人认知功能效果的系统综述[J]. 《中国康复理论与实践》, 2023, 29(8): 903-909.
[11]	魏晓微, 杨剑, 魏春艳, 贺启令. 学校环境下适应性体育课程促进智力与发展性残疾儿童心理运动发展的系统综述[J]. 《中国康复理论与实践》, 2023, 29(8): 910-918.
[12]	崔尧, 丛芳, 黄富表, 曾明, 颜如秀. 不同镜像神经元训练策略下脑与肌肉的活动特征：基于近红外光谱与表面肌电图技术[J]. 《中国康复理论与实践》, 2023, 29(7): 782-790.
[13]	王少璞, 陈钢. 基于世界卫生组织健康促进学校架构的心理行为健康服务及其健康效益：系统综述的系统综述[J]. 《中国康复理论与实践》, 2023, 29(7): 800-807.
[14]	蒋长好, 高晓妍. 短时身体活动对儿童认知功能影响的系统综述[J]. 《中国康复理论与实践》, 2023, 29(6): 667-672.
[15]	李丹, 王剑雄, 黄茂茂, 胥方元, 曾秋, 李佶钖, 李洋, 夏翠宏, 郑雅丹, 胥章彧, 方雯凤, 万腾刚. 健康中老年女性上下楼梯时下肢肌肉的表面肌电图表现[J]. 《中国康复理论与实践》, 2023, 29(6): 731-737.