Abstract: In order to control PowerPoint automatically without touching for users, a principal component analysis method is used on Kinect camera to realize the face recognition. By the results, the system can open the speaker’s files automatically. Then the Kinect camera is used to get the tracking data of palm. On the basis of the data, the two-dimensional image is divided into several function parts. The least square method is adopted to convert the two-dimensional tracking data at different depth and the intention of the speakers is gotten. The purpose of controlling the PowerPoint without touching is realized. The simulation results show that the method is effective.

Key words: face recognition, principal component analysis, PPT control, least square method

摘要： 为了实现演讲者无接触性操作情况下的PPT自动播放控制，系统利用Kinect摄像头通过主成分分析的方法对演讲人进行人脸识别，确定其姓名，自动为其打开与之相对应的PowerPoint文稿。利用Kinect对演讲者手心进行跟踪获得数据，并在固定深度时对手心在图像中的二维信息进行功能区域划分，在此基础上，利用最小二乘法实现在不同的深度信息下的二维信息进行转换，最终确定演讲者的动作意图，进而实现无接触性操作情况下的演讲人对PowerPoint的全自动人体控制。仿真结果表明基于该方法的控制有效性。

关键词: 人脸识别, 主成分分析, PPT控制, 最小二乘法