| 基于表面肌电信号的腰背动作识别新方法 |
| |
| 引用本文： | 刘二宁,邹任玲,姜亚斌,胡秀枋,卢旭华,王海滨,范虓杰,张安林.基于表面肌电信号的腰背动作识别新方法[J].教育技术导刊,2009,19(11):71-74. |
| |
| 作者姓名： | 刘二宁 邹任玲 姜亚斌 胡秀枋 卢旭华 王海滨 范虓杰 张安林 |
| |
| 作者单位： | 1. 上海理工大学 医疗器械与食品学院，上海 200093；2. 上海长征医院，上海 200003 |
| |
| 基金项目： | 国家自然科学基金项目（61473193）；上海理工大学医工交叉项目（1019308505） |
| |
| 摘 要： | 为提高基于表面肌电信号的人体腰背动作识别率，提出一种基于小波包能量与改进NARX神经网络的分类识别新方法。利用小波包变换对动作部位进行表面肌电信号特征提取，并采用改进NARX神经网络进行分类识别。选取8名实验者分别在扭腰、弯腰、侧弯腰3种动作下进行表面肌电信号数据采集，选择db4小波包函数对信号进行6层分解，得到第6层64个频带的小波包分解系数，代表各个动作信息的特征向量，作为改进NARX神经网络的输入进行分类识别。对照实验组中，改进NARX神经网络的识别率较高，总体识别率达到96.7%。实验结果表明，利用该识别方法对腰部动作进行分类识别，分类准确，且识别率更高。
|
| 关 键 词： | 表面肌电信号 动作识别 小波包变换 改进NARX神经网络 |
| 收稿时间： | 2020-04-04 |
A New Method for Recognition of Lumbar and Back Movements Based on Surface EMG Signals |
| |
| LIU Er-ning,ZOU Ren-ling,JIANG Ya-bin,HU Xiu-fang,LU Xu-hua,WANG Hai-bin,FAN Xiao-jie,ZHANG An-lin.A New Method for Recognition of Lumbar and Back Movements Based on Surface EMG Signals[J].Introduction of Educational Technology,2009,19(11):71-74. |
| |
| Authors: | LIU Er-ning ZOU Ren-ling JIANG Ya-bin HU Xiu-fang LU Xu-hua WANG Hai-bin FAN Xiao-jie ZHANG An-lin |
| |
| Institution: | 1. School of Medical Instrument and Food，University of Shanghai for Science and Technology，Shanghai 200093，China；2. Shanghai Changzheng Hospital， Shanghai 200003， China |
| |
| Abstract: | In order to improve the recognition rate of human low back motion pattern based on surface EMG， this paper proposes a new classification and recognition method based on wavelet packet energy and improved NARX neural network.The wavelet transform is used to extract the surface EMG signal features of the action part， and the improved NARX neural network is used for classification and identification. Eight experimental subjects were selected to perform surface EMG signal data acquisition under the lumbar motion of twisting， bending and side bending. The db4 wavelet packet function was used to decompose the signal in 6 layers to obtain the decomposition coefficient of the 64 wavelet band in the 6th layer. The coefficients represent the feature vector of each action information are used as an input to the improved NARX neural network for classification and recognition. In the control group， the improved NARX neural network has a higher recognition rate， and the overall recognition rate reaches 96.7%. The experimental results show that the waist movement is classified and recognized by this recognition method with accurate classification and higher recognition rate. |
| |
| Keywords: | surface EMG signal motion recognition wavelet packet transform improved NARX neural network |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
