基于个性化特征分析的表面肌电手势识别  

Surface Electromyography Gesture Recognition Based on Personalized Feature Analysis

作　　者：刘栋博 刘韵 房玉 易民升 王维博 LIU Dongbo;LIU Yun;FANG Yu;YI Minsheng;WANG Weibo(School of Electrical and Electronic Information,Xihua University,Chengdu,Sichuan 610039,China;Shaoyang Polytechnic,Shaoyang,Hunan 422004,China)

机构地区：[1]西华大学,电气与电子信息学院,四川成都610039 [2]邵阳职业技术学院,湖南邵阳422004

出　　处：《计量学报》2024年第12期1900-1908,共9页Acta Metrologica Sinica

基　　金：国家自然科学基金(61901393);教育部春晖计划科研项目(Z2018118)。

摘　　要：使用表面肌电信号对手势动作进行识别是实现人工智能假肢控制的重要方法。表面肌电信号是从皮肤表层采集的肌电信号,在一定程度上可以反映神经肌肉的活动。由于其随机非平稳的特性,利用表面肌电信号控制人工智能假肢始终无法达到理想的效果,特别是体现在性别等个体差异的影响上。通过差异化性别特征分析,针对7类手势动作提出了一种基于表面肌电信号的手势识别方法。实验结果表明,该方法在5种分类器中的识别准确率都有明显提升。其中线性判别分析的提升最为显著,达到了15.84%;支持向量机和极端梯度提升在手势识别中效果最好,准确率达到了98.41%。To realize artificial limb control,it is an important method to use surface electromyography(sEMG)signal to recognize gestures.sEMG signals are electromyographic signals collected from the surface layer of the skin,which can reflect the neuromuscular activities to a certain extent.Due to its random and non-smooth characteristics,the use of sEMG signals to control artificial intelligent prosthetic limbs has never been able to achieve the desired results,especially reflected in the influence of individual differences such as gender.A gesture recognition method based on surface EMG signals is proposed for seven types of gesture movements through differentiated gender characterization.Experimental results show that the recognition accuracy of the method is significantly improved in all five classifiers.Among them,linear discriminant analysis has the most significant improvement,reaching 15.84%;support vector machine and extreme gradient boosting are the most effective in gesture recognition,with an accuracy of 98.41%.

关 键 词：手势识别 表面肌电 个性化差异 能量补偿 

分 类 号：TB99[一般工业技术—计量学] TB973[机械工程—测试计量技术及仪器]