基于改进小波包与样本熵的表面肌电信号特征提取

针对表面肌电信号（s EMG）进行小波包分解后子空间维数较大导致部分子空间的特征信息被减弱和相邻子空间的频率接近会导致不同程度的信息冗余问题,提出了一种基于改进小波包与样本熵相结合的特征提取方法。在对原始s EMG进行小波变换的同时,对其某一高频子空间也进行小波变换,提取改进小波包分解后四个低频子空间的样本熵作为特征。在智能轮椅平台上进行实验,实验结果显示采用基于该算法的轮椅系统不仅正确识别率较高,而且稳定性更好。...

Full description

Saved in:
Bibliographic Details
Published in计算机应用研究 Vol. 32; no. 3; pp. 701 - 704
Main Author 张毅 冯晓林 罗元
Format Journal Article
LanguageChinese
Published 重庆邮电大学信息无障碍工程研发中心,重庆,400065%重庆邮电大学光纤通信技术重点实验室,重庆,400065 2015
Subjects
小波
小波包
样本熵
特征提取
肌电信号
特征提取
样本熵
小波包
feature extraction
sample entropy
小波
wavelet
EMG(electromyogram)
wavelet packet
肌电信号
Online AccessGet full text
ISSN1001-3695
DOI10.3969/j.issn.1001-3695.2015.03.014

Cover

More Information
Summary:针对表面肌电信号（s EMG）进行小波包分解后子空间维数较大导致部分子空间的特征信息被减弱和相邻子空间的频率接近会导致不同程度的信息冗余问题,提出了一种基于改进小波包与样本熵相结合的特征提取方法。在对原始s EMG进行小波变换的同时,对其某一高频子空间也进行小波变换,提取改进小波包分解后四个低频子空间的样本熵作为特征。在智能轮椅平台上进行实验,实验结果显示采用基于该算法的轮椅系统不仅正确识别率较高,而且稳定性更好。
Bibliography:51-1196/TP
After wavelet packet decomposed surface electromyography （sEMG） , feature information of some subspace will weaken because subspace dimension is too large, and adjacent subspace will lead to information redundancy because their fre- quencies is proximity. To solve these problems, this paper proposed an improved wavelet packet decomposition algorithm, which made a high-frequency subspace wavelet transform based on the main wavelet transform. And then it extracted sample entropy feature from four subspaces of improved wavelet packet decomposition. Conducting experiments in the intelligent wheelchair platform, and experiment results show that wheelchair system based on improved wavelet packet decomposition al- gorithm has a higher correct recognition rate and a better stability.
EMG （eleetromyogram） ; wavelet; wavelet packet; sample entropy; feature extraction
ZHANG Yi, FENG Xiao-lin, LUO Yuan（ a. Information Accessibility Engineering R＆D Center, b. Key Laboratory of Optical Fiber Communication Technolog
ISSN:1001-3695
DOI:10.3969/j.issn.1001-3695.2015.03.014