基于小波特征分析的手指动作识别研究

目的：本文利用表面肌电（sEMG）信号来研究多种手指组合动作的识别问题。方法：在对采集的四个通道sEMG信号进行降噪预处理的基础上，采用移动加窗处理方法来提取关于手指运动状态的信号活动段，再分析各个信号活动段的小波系数统计特征，进而利用多类支持向量机（SVM分类算法来实现手指组合动作的识别。结果：动作识别率最高达到100％。结论：所采用方法能够有效地识别多种手势动作，并为后续基于肌电信号的实时人机接口系统的研究奠定了理论基础。

Finger Gesture Recognition Based on Wavelet Features Analysis

Objective： The recognition problem of finger gestures using the multi-channel sEMG signals was explored in this paper. Methods： Based on the pre-processing of the collected four-channel sEMG signals, the moving-window method was utilized to extract the activities of fingers actions from the sEMG signals. Then, the statistical features of wavelet parameters were analyzed, and the SVM was used for the recognition of all the finger gestures. Results： The highest recognition rate can be reached up to 100%. Conclusion： The experimental results showed that the method we used could recognize multiple gestures effectively,and this preparatory work could be applied for the study of human-machine interface in our future work.