【摘要】 为降低康复训练设备的成本，促进手部康复训练技术融入家居生活，本文提出了一种基于视觉的手部康复训练实验系统设计方法。该系统由PC机与普通摄像头作为硬件支撑，根据人体肤色在YCrCb颜色空间中的分布特点，基于颜色筛选提取手部轮廓，采用反向传播神经网络实现手部轮廓的识别。基于游戏机制，开发与手部康复训练相匹配的游戏系统，开展康复训练实验测试。结果表明，本系统手势识别准确率较高，实时性好，适用于居家手势康复训练，为学生在计算机视觉技术的知识理解与技术应用方面提供教学案例，能够满足手部康复训练的实验研究。更多还原

【Abstract】 In order to reduce the cost of rehabilitation training equipment and promote the integration of hand rehabilitation training technology into home life, a design method of hand rehabilitation training experimental system based on computer vision is proposed. This system is supported by PC and ordinary camera as hardware. According to the distribution characteristics of human skin tone in the YCrCb color space, the hand contour is extracted based on color screening and then is recognized by the back propagation neural network. Based on the game mechanism, the game system matching the hand rehabilitation training is developed and the experimental test of rehabilitation training is carried out. The results show that the system has high recognition accuracy, good real-time performance, can be considered as a teaching case for students in the knowledge understanding and technology application of computer vision technology. The system is suitable for home gesture rehabilitation training and can meet the experimental research of hand rehabilitation training.更多还原