基于自回归模型和学习矢量量化神经网络的手指动作识别

手指的力量和动作是反映手指协同运动、评价手部运动机能的重要参数。本文提出了一种以自回归（Auto-regressive，AR）模型和学习矢量量化（Learning Vector Quantization，LVQ）网络相结合的表面肌电信号处理方法。13名受试者参与了目标力量为4N、6N、8N等三个力量等级的指力跟踪实验，对指力信号和前臂指浅屈肌（flex digitorum superficials，FDS）、指伸肌（extensor digitorum，ED）的表面肌电信号进行了同步记录；通过对采集到的肌电信号进行预处理，提取AR系数作为其特征值；然后设计了一个LVQ神经网络，对同等力量水平下食指、中指的动作进行模式分类，分类正确率在80％以上。实验表明，表面肌电信号（surface Eleetromyography，sEMG）与手指动作具有相关性，使用AR结合LVQ的sEMG有较高的识别率。

Motion Identification of Finger Based on Auto-regressive Model and Learning-vector-quantization Neural Network

Finger force was an important parameter to reflect finger synergetic motion and evaluate hand movement function. In this paper, a method to process surface electromyography signal was presented. It was based on auto-regressive (AR) model and learning - vector - quantization (LVQ) neural network. First, 13 subjects volunteered to take part in the test which required them to complete finger force tracing experiment at three force level, the surface Electromyography (sEMG) signal of flex digitorum superficials (FDS) and extensor digitorum (ED) were recorded , second, AR model coefficient of sEMG signal was calculated, then, designed a LVQ neural network to classify the motion of index finger and middle finger at the same force level, the classify correct rate was above 80%. The results of experiment showed that, a correlation between finger motion and sEMG signal had been represented, and this method had a high recognition rate.