基于Leap Motion的三维动态手势识别研究

Research on 3D dynamic gesture recognition based on Leap Motion

手势识别作为人机交互的重要手段,其识别率的提高是当今重要的研究方向。通过分析当今三维手势识别技术的发展状况,采用Leap Motion设备针对三维动态手势进行数据采集和处理,使用SVM与PNN模式识别算法,分别进行了两类手势区分以及多类手势区分的仿真实验。结果表明,经过优化的PNN对动态手势均较SVM能达到更高的识别率,且经过PCA处理后运算效率较高。

Gesture recognition is an essential means of human-computer interaction.The improvement of recognition rate is an important research direction today.By analyzing the development of today’s 3 D gesture recognition technology,Leap Motion equipment is used for data acquisition and processing of 3 D dynamic gestures.Using SVM and PNN pattern recognition algorithms,two kinds of gesture differentiation and multi-class gesture differentiation simulation experiments are carried out.The results show that the optimized PNN can achieve higher recognition rate than dynamic SVM,and the computational efficiency is higher after PCA processing.