摘要
分析上臂动作与上臂肌肉的关系,通过表面肌电信号正确识别上臂的动作,是实现上肢功能修复的关键.设计了上肢曲臂、伸臂、水平外摆、水平内收、手臂垂直外旋和手臂垂直内旋6个动作,分别同时记录三角肌、肱二头肌和肱三头肌的表面肌电信号,采用时域和频域的方法提取特征值,通过人工神经网络进行识别,识别率达到90%以上.结果表明,通过上肢肱二头肌、肱三头肌和三角肌的表面肌电信号识别上臂的运动是可行的,为应用生物电信号控制机械假肢和实现脊髓损伤功能障碍修复奠定理论基础.
The loss of upper limb function caused by spinal cord injury can be restored partly through functional elec- trical stimulation. By means of the analysis of the relationship between the movements and muscles of upper arm, cor- rect recognition of upper arm movements by surface electromyography is the key to realize function restoration, six movements are set, as bending arm, straightening the arm, swaying outward horizontally, swaying inward horizontal- ly, rotating outward vertically and rotating inward vertically, to record the surface electromyography of deltoid, bi- ceps and triceps by three channels simultaneously, by using time domain and frequency domain method for picking up the characteristic value, recognizing through artificial neural network. The recognition rate is above 90%. It confirms that it＇s feasible to recognize upper limb movements by surface electromyography of deltoid, biceps and triceps, which lays a theoretical foundation that biological electricai signals control mechanical prosthesis and can realize the restoration of the function caused by spinal cord injury.
出处
《南通大学学报（自然科学版）》
CAS
2013年第1期14-17,共4页
Journal of Nantong University(Natural Science Edition)
基金
江苏省高校自然科学研究项目(11KJB510022)
东南大学生物电子学国家重点实验室开放研究基金项目(2011E05)
南通市科技计划项目(K2009037)
关键词
表面肌电信号
上臂动作
识别
人工神经网络
surface electromyography
upper ann
recognition
artificial neural network
