【作者】 罗元； 杨杨； 张毅； 黎胜晖； 陈凯；

【Author】 LUO Yuan,YANG Yang,ZHANG Yi,LI Shenghui,CHEN Kai (Key Laboratory of Optical Fiber Communication Technology,Research Center of Intelligent System and Robot, Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

【机构】 重庆邮电大学 智能系统及机器人研究所,光纤通信技术重点实验室；

【摘要】 目前智能轮椅的手势控制系统多基于通用PC机,针对其携带不方便、功耗大、成本高等缺点,设计了一种嵌入式手势控制系统。该系统采用S3C2440为微处理器的ARM9核心板作为硬件平台,Linux操作系统作为软件平台,运用AdaBoost算法和Kalman滤波器实现了手势识别,并利用Qt/Embedded图形开发环境完成了嵌入式人机交互界面的设计。实验证明,该嵌入式手势控制系统能很好地应用在智能轮椅上,平均识别率为95.3%。更多还原

【Abstract】 The hand gesture recognition control system of intelligent wheelchair is mostly based on PC.In order to overcome its short comings as inconvenience to carry,large power consumption and higher cost,a design of embedded hand gesture recognition control system was introduced.The system uses the core board ARM9 whose microprocessor is S3C2440 as hardware platform.Linux operating system is used as software platform.Hand gesture recognition is realized by using AdaBoost algorithm and Kalman filter.Qt/Embedded graphical development environment is used for the interface design of embedded human-wheelchair interaction.The experiment indicates that the embedded hand gesture recognition control system can be well applied in the intelligent wheelchair,and the average recognition rate arrives 95.3%.更多还原