Research And Implementation Of Electronic Yoga Teaching System Based On Kinect

With the rapid development of social economy and science and technology,the pursuit of quality of life also has been promoted for people,leisure sports gradually become a trend,yoga as a traditional and fashionable fitness style favored by more and more people.However,the teaching time of the traditional way of training yoga is fixed,there is a lot of inconvenience for people,but according to the teaching video to learn can not guarantee the accuracy of action,may not be able to achieve the training effect.Therefore,in this situation,the intelligent electronic yoga teaching has an important significance for improving the level of training.In this system,Kinect is selected as the sensing carrier of human body movement information,and a set of electronic yoga teaching system is designed.The system integrates the function of motion information acquisition and action evaluation.This article through Kinect collection of a large number of standard yoga movements,the establishment of standard yoga action database,as yoga training comparison template.The trainer only need to make yoga before Kinect action,the system will automatically collect the trainer action data and identify the action name,the system through the dynamic time regular algorithm and the Hausdorff distance measurement algorithm will train the athletic data stream and the standard database.And proposes a similarity evaluation method based on the fusion of the two algorithms.By calculating the similarity between the two,the quality of the trainer is evaluated.At the same time,the system proposes an algorithm for the evaluation of the action of the human joint angle measurement,which can accurately point out the joint point where the action is not in place.Through the test,the system can carry out real-time posture recognition,and can achieve the purpose of evaluating the quality of training,to meet the basic training needs,has a certain application value.