基于视觉和肌电信息融合的手势识别方法-《郑州大学学报(工学版)》

文章信息/Info

Title:: Fusion of Vision and EMG Signal Information for Gesture Recognition ba＜x＞sed on Support Vector Machine

Author(s):: Peng Jinzhu; Dong Mengchao; Yang Yang;; School of Electrical Engineering, Zhengzhou University;

Keywords:: Gesture recognition; the direction gradient straight recipe map; muscle signal; time domain characteristics; feature fusion

摘要:: 针对人机交互技术对手势识别提出的种类多样性和识别精确度的问题，本文提出一种基于视觉方向梯度直方图（HOG）特征和肌电信号（EMG）时域特征的融合及支持向量机（SVM）分类器的手势识别方法。利用视觉传感器和智能臂环分别采集手势图像信息和肌电信号，预处理后提取对应的HOG特征和时域特征；采用串行融合的方式将两种特征进行特征级融合；以SVM为多类分类器完成手势识别模型的训练和检验，并进行对比分析。实验结果表明，采用融合特征的36类手势识别模型的总体正确率达到了96%，能显著减少特征数据量，并有效提高多种类手势识别正确率。

Abstract:: Aiming at the problem of variety and accuracy of gesture recognition in human-computer interaction, this paper proposes a gesture recognition method ba＜x＞sed on fusion features of visual histogram of orientation gradient (HOG) and time-domain features of electromyography (EMG), where support vector machine (SVM) is uses as a classifier. Visual sensors and smart armbands are used to collect gesture image information and EMG signals respectively, and the corresponding HOG features and time domain features are then extracted after preprocessing . The two features are fused at feature level by employing serial fusion. And SVM is used as a multi-class classifier to complete the training and verification of gesture recognition model. Experimental results show that the overall accuracy of 36 types of gesture recognition models reaches 96% by using the proposed fusion features, which can effectively reduce the amount of feature data and significantly improve the accuracy of multiple types of gesture recognition.