Research On Key Technology Of Hand Function Rehabilitation Training Evaluation System Based On Leap Motion

With the rapid development of virtual reality technology,people have higher and higher requirements for human-computer interaction.They always hope to achieve it in a more direct,natural and realistic way,especially in the scene of human-computer interaction through gestures.Therefore,it is of great significance to explore and improve gesture recognition in human-computer interaction.Nowadays,the number of patients with hand dysfunction caused by stroke is increasing.Based on the existing hand function rehabilitation equipment,mechanical equipment will not only increase the burden of patients’ hands,but also lack the necessary immersion experience,and it can not track and evaluate the rehabilitation process.Therefore,this paper proposes an immersive rehabilitation training system in which patients can carry out rehabilitation training independently.Patients do not need to wear any weight-bearing equipment,they only need to perform the corresponding gestures to carry out rehabilitation training.The details of the research are as follows:(1)The system mainly uses Leap Motion binocular vision sensor is used to collect human hand information.Through the data fusion technology of improved principal component analysis(PCA),the real-time data collected by the sensor is fused,and more complete information of human hand posture is obtained with less principal components.The stability and accuracy of posture data are improved,and the overall operation efficiency of the system is also improved.(2)Then,the improved support vector machine(SVM)and k-nearest neighbor algorithm(KNN)are used to recognize and classify gesture data to make Leap Motion can obtain the feature information of hand more accurately and quickly,which reduces the heavy workload.The optimized SVK improves the accuracy and effectiveness of gesture recognition,so that patients can more effectively carry out the rehabilitation training of autonomous hand function.(3)Finally,the evaluation system of hand function rehabilitation training based on leap motion is built on unity 3D platform.Through the idea of analytic hierarchy process(AHP),the gesture data of rehabilitation training for patients are effectively evaluated,and then the results of rehabilitation training are recorded in the local database in real time,which is used as the necessary reference factor for attending doctors,so as to facilitate the tracking treatment of patients.The experimental results show that the hand function rehabilitation system can accurately detect the complete information of the hand and trigger the corresponding rehabilitation actions and the effect of actions in the PC,which effectively realizes the implementation of the hand function rehabilitation training system and has a good application prospect.