欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对表面肌电信号非线性、噪声强等特点，设计一种快速有效的表面肌电信号手部多运动模式识别方法，用于肌电假手的实时控制。提出了一种基于经验模态分解样本熵和聚类分析的表面肌电信号多运动模式识别方法。该方法对动作持续阶段的信号首先进行经验模态分解，将其分解为多个平稳的固有模态函数（ IMF），再依据频率有效度选取若干个包含肌电信号有效信息的IMF分量求和后，计算其样本熵。以尺侧腕伸肌和尺侧腕屈肌两路肌电信号对应的EMD样本熵作为特征向量，设计了主轴核聚类算法的聚类分类器进行模式识别。成功识别了展拳、握拳、腕上翻和腕下翻4种动作，平均识别率达到93%。该方法取得了较高的识别率，抗干扰能力强，计算量少，可用于肌电假手的控制。
英文摘要：
According to the chaotic and nonlinear characteristics of surface electromyography （sEMG）, a fast and efficient hands movement sEMG pattern recognition method for real-time control of myoelectric prosthetic hand is designed. A multi-modeling pattern recognition method of sEMG features based on the empirical mode decomposition （EMD） sample entropy and clustering analysis is proposed. First, it decomposes the sEMG signal into a set of intrinsic mode functions （IMF）, then combines some of the IMF which contains the useful information according to frequency effectiveness, and calculates the sample entropy of the combination. The sample entropy of two sEMG of the extensor carpi ulnaris and flexor carpi ulnaris constitute the feature vector, the clustering classifier which based on principal axis clustering arithmetic is applied to classify the four hand movements. The result shows that four movements （hand extension, hand grasps, wrist spreads and wrist bends） ale successfully identified. The average recognition rate is 93%. The method achieved high recognition rate, anti-interference ability and less computation, that is suitable for the control of the myoelectric prosthetic hand.
同期刊论文项目
同项目期刊论文
期刊信息
位置：