摘要
为提高手腕动作的识别率,提出了一种将主成分分析(PCA)和极限学习机(ELM)相结合的手腕动作肌电信号识别方法。该方法提取手腕4种动作(内翻、外翻、握拳、展拳)的肌电信号,运用小波变化提取小波特征构造特征矢量,利用PCA算法对特征矢量进行降维,摒弃冗余信息,实现肌电信号特征参数的降维,最后运用ELM对降维后的数据进行识别分类。实验结果表明:将PCA和ELM相结合的方法有着更高的手腕动作识别率,验证了该方法的可行性。
In order to improve the recognition rate of wrist movements,a new method of wrist movement recognition based on principal component analysis(PCA)and extreme learning machine(ELM)is proposed.This method extracts the EMG signals of four wrist movements(wrist down,wrist up,hand grasps,hand extension).Wavelet transform is used to extract wavelet features and construct feature vectors.To realize the dimension reduction of EMG characteristic parameters,the PCA algorithm is used to reduce the dimension of feature vectors and discard the redundant information.Finally,ELM is used to recognize and classify the low-dimensional data.The experimental results show that the combination of PCA and ELM has a higher recognition rate of wrist movements,which verifies the feasibility of the method.
作者
景甜甜
洪洁
JING Tiantian;HONG Jie(School of Mechanical and Electrical Engineering,Anhui Jianzhu University,Hefei 230601,China;Anhui Jianghuai Automobile Group limited company,Hefei 230601,China)
出处
《重庆理工大学学报（自然科学）》
CAS
北大核心
2019年第12期96-100,共5页
Journal of Chongqing University of Technology：Natural Science
基金
安徽省高校省级自然科学基金研究项目“欠驱动自适应柔性机器人手的研究与设计分析”(KJ2018JD24)
关键词
表面肌电信号
模式识别
主成分分析
极限学习机
surface electromyography
pattern recognition
principal component analysis
extreme learning machine
