基于计算机视觉的手势识别康复系统研究与应用

针对目前医学上手部康复治疗的康复器械功能单一,训练过程动作反复且枯燥无趣,恢复过程比较缓慢等问题,提出了一种基于计算机视觉的手势识别康复系统;首先介绍了图像采集、分割、平滑处理、分类、识别等计算机视觉关键技术,其次,重点阐述了手势康复系统的实现细节;通过摄像头采集不同年龄,性别人群手势样本数据,建立康复手势数据库,利用计算机视觉,卷积神经网络,PyQt图形界面等技术来构建康复系统,给出具有趣味性并且高效便利的康复训练方案,治疗方法弥补传统康复治疗的缺陷;实验结果表明,系统运行可靠、准确率高达96％,不但可以提高患者对康复训练的兴趣,积极性,而且价格相对于其他康复器械更加低廉,应用前景更广阔.

Research and Application of Gesture Recognition and Rehabilitation System Based on Computer Vision

Aiming at the problems of single function of rehabilitation equipment for hand rehabilitation in medicine, repeated and boring movements during training, and slow recovery process, a gesture recognition rehabilitation system based on computer vision is proposed. First, it introduces the key technologies of computer vision such as image acquisition, segmentation, smoothing, classification, and recognition. Second, it focuses on the implementation details of the gesture rehabilitation system. Collect gesture sample data of people of different ages and genders through the camera, establish a rehabilitation gesture database, use computer vision, convolutional neural network, PyQt graphical interface and other technologies to build a rehabilitation system, and provide interesting, efficient and convenient rehabilitation training programs and treatments Methods to make up for the shortcomings of traditional rehabilitation therapy. The experimental results show that the system is reliable and get 96% accuracy, not only can increase the patient''s interest and enthusiasm for rehabilitation training, but also the price is lower than other rehabilitation equipment, and the application prospect is broader.