北京生物医学工程

基于无线表面肌电信号采集的上肢动作识别

Wireless surface EMG acquisitionfor upper limb motion recognition

作者：吴志文李晓欧

单位：上海理工大学医疗器械与食品学院(上海200093)

分类号：R318.04

出版年·卷·期（页码）：2016·35·6（593-598）

摘要：

目的为识别上肢动作并应用于人机交互领域以及为相关患者提供上肢康复训练，设计一个无线表面肌电信号采集及识别系统。方法系统主要由硬件部分与软件部分组成。硬件设计方面，由增强型80C51作为各个模块的控制中心。贴片电极采集的肌电信号，经仪表放大器AD8422放大处理，并进行A/D转换，最后通过无线方式将信号发送给接收盒并传送至PC。软件设计方面，在VC平台下，通过均方根、自回归系数提取特征值，利用支持向量机算法进行动作模式识别。结果设备的采集部分体积为37mm×27mm×15mm，可方便地实现穿戴式，上位机部分则可以满足对信号的各种分析以及作为人机交互界面。结论该系统可实现对患者的康复训练，也可扩展到游戏娱乐。

Objective To recognize upper limb motion and apply it in the human-computer interaction field as well as provide upper limb rehabilitation training for the related patients, we designed a wireless surface electromyography (EMG) signal-acquisition and recognition system. Methods The system is consisted of hardware and software. In the aspect of hardware design, enhanced 80C51 MCU is the control center of each module. The front-end signal is collected by the surface electrode patch, amplified by the instrument amplifier AD8422, with A/D conversion, finally, sent to the receiving box and the upper monitoring through the wireless mode. In the aspect of software design, in the VC platform, the characteristic value is extracted through root mean square, AR coefficients and average power. Action pattern recognition is carried out with support vector machine. Results The volume of the device is 37mm×27mm×15mm, which can be easily worn on the surface skin of the human body. Conclusions The part of the upper monitoring can realize rehabilitation training for the patients and be extended to game entertainment.

参考文献：

［1］Yee Mon Aung, Adel Al-Jumaily. sEMG based ANN for shoulder angle prediction［J］. Procedia Engineering, 2012, 41：1009-1015.

［2］Al-Mulla MR, Sepulveda F, Colley M. Evolved pseudo-wavelet function to optimally decompose sEMG for automated classification of localized muscle fatigue［J］. Medical Engineering & Physics, 2011,33：411-417.

［3］王凯，于鸿洋，张萍. 基于AdaBoost算法和光流匹配的实时手势识别［J］. 微电子学与计算机,2012,29(4) :138-141.

Wang Kai, Yu Hongyang, Zhang Ping. Real-time gesture recognition based on AdaBoost algorithm and optical flow matching［J］. Microelectronics & Computer, 2012, 29(4):138-141.

［4］赵大威，姜力，黄海，等．多自由度仿人型假手设计［J］.哈尔滨工业大学学报，2008，40(7)：1067-1070．

Zhao Dawei，Jiang Li，Huang Hai，et al. Development of a multi·DOF anthropomorphic prosthetic hand［J］. Journal of Harbin Institute of Technology, 2008,40(7)：1067-1070．

［5］胡巍，赵章琰，路知远，等.无线多通道表面肌电信号采集系统设计木［J］. 电子测量与仪器学报，2009,23（11）：30-35.

Hu Wei, Zhao Zhangyan, Lu Zhiyuan, et al. Design of wireless multi—channel surface EMG acquisition system［J］. Journal of Electronic Measurement and Instrument, 2009,23（11）：30-35.

［6］王从政，陈香，董中飞，等.一种基于DSP的实时手势交互系统［J］.传感技术学报，2011,24（5）：688-693.

Wang Congzheng，Chen Xiang，Dong Zhongfei，et al. A real—time DSP-based gesture interaction system［J］. Chinese Journal of Sensors and Actuators, 2011,24（5）：688-693.

［7］王人成，郑双喜，蔡付文,等.基于表面肌电信号的手指运动动作模式识别系统［J］.康复医学工程, 2008,23（5）：410-412.

Wang Rencheng, Zheng Shuangxi, Cai Fuwen, et al. Recognition system of finger movement pattern based on sEMG［J］. Chinese Journal of Rehabilitation Medicine, 2008,23(5):410-412.

［8］艾青松，卢英，刘泉.高斯径向基函数重构特征对表面肌电信号识别［J］.计算机工程与应用，2013，49（12）：182-186.

Ai Qingsong，Lu Ying，Liu Quan. Recognition of sEMG based on reconstructed feature by Gaussian radial basis function［J］. Computer Engineering and Applications，2013，49(12)：182-186.

［9］李琳，王建辉，顾树生. 一种改进的基于信号能量阈值的表面肌电信号自动分割方法［J］.计算机科学，2013，40（6A）：188-191.

Li Lin, Wang Jianhui, Gu Shusheng. Improved automatic segmentation method of sEMG based on signals’ energy value［J］.Computer Science，2013，40（6A）：188-191.

［10］李瑞辉，范志坚，赵翠莲，等.利用sEMG能量高斯分布特性提取动作信号的方法［J］ .中国医疗器械杂志， 2014，38（3）：177-180.

Li Ruihui, Fan Zhijian, Zhao Cuilian, et al. Motion signal extraction method based on sEMG energy gauss distribution characteristics［J］. Chinese Journal of Medical Instrumentation,2014，38（3）：177-180.

［11］李晶皎.模式识别［M］. 4版. 北京：电子工业出版社，2010:293.

Li Jingjiao. Pattern Recognition［M］. 4th ed. Beijing: Publishing House of Electronics Industry, 2010:293.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录