Research On Limb Teaching Information Management System Based On Artificial Intelligence

In the beginning of 2020,a COVID-19 erupted in Wuhan has brought big pain and disaster to the whole country.In such a war without the smoke of gunpowder,the continued epidemic had a great impact on education and teaching,and online teaching has become an vital platform for "teaching" and "learning".The online education based on media,teachers can't monitor students and their learning test is limited.Therefore,more accurate and rapid teaching management in the discipline has become an urgent need,which is of great significance to the scientific control of teaching.At the same time,China's education department issued the "educational informationization 2.0 action plan",various teaching platforms based on the Internet have been invented and applied on a large scale,effectively promoting the development of educational informationization.At present,the education and teaching management system is mainly based on resource sharing and teacher-student interaction.It is mainly aimed at classroom teaching subjects and realized by means of multimedia.It effectively breaks the current situation of teacher-student disconnection and slow information interaction under the traditional education mode.However,to give priority to with body movement of subjects such as sports and dance,its teaching training,assignments,interaction between teachers and students,and students are unable to use words and images and other forms of traditional management,assessment and supervision,meanwhile video data is subject to different shooting angle and filming equipment can cause excessive pressure to transmission network when upload video data,bring great difficulties to the evaluation and teaching control.At the same time,in the body movement discipline,the wrong movement can cause great harm to the human body.At present,the accuracy evaluation of motion mainly has the following characteristics:in the aspect of motion acquisition,video and depth sensor are the main features,while the portable sensor information acquisition is less used for motion accuracy evaluation.In terms of evaluation algorithm,dynamic time adjustment algorithm,Euclidean distance and cosine similarity are the main methods,while classical machine learning algorithm and neural network algorithm are not used more.As for the teaching control system,its functions are mainly teacher-student interaction and data sharing,and it is few involved the analysis of character recognition and action evaluation.In order to solve the above problems,this paper designs and implements a teaching control system based on artificial intelligence,which uses artificial intelligence technology to digitize the body movement information and analyze the movement accurately.The system can be implemented for limb movements of dynamic monitoring in the class discipline and accuracy evaluation,character recognition and the improvement of movement technology,interface display action evaluation,action results such as the curve,can satisfy the requirements of the sports training and teaching,to provide information and precision of body movements disciplines teaching quantifiable basis and control means.The specific research contents are as follows:(1)The design of teaching control system.Taking the management of body movement as the research focus,a teaching control system framework based on artificial intelligence is proposed and designed,which consists of four parts:action collector,wireless access point,data processing,server and terminal.(2)In terms of related algorithms,the first is the research on the method of character recognition.Considering the intra-class differences between individual actions,the feature extraction algorithm is proposed based on LR to ensure the accuracy of character recognition.As far as the two kinds of hand movement character recognition experiments are concerned,the recognition accuracy is more than 99%,which can distinguish the characters well.Secondly,a feature extraction algorithm based on SVM is proposed to perceive the motion difference.In the three kinds of hand motion evaluation experiments,the proposed feature extraction algorithm has a good effect,and the evaluation accuracy has reached more than 90%,which proves the effectiveness of the algorithm.Finally,as for the analysis and research on the improvement of movement technology in the body movement discipline,the deep neural network(GRU)is used to analyze the movement.Taking the two-handed shooting as an example,the author tries to find the subtle differences between missed and missed shots in shooting training,so as to continuously improve the accuracy of shooting in sports training and optimize the training results.(3)In terms of system development,a hardware platform composed of sensors,wireless access points and PC terminals was developed.The software platform consists of data transmission,data storage,data processing,action evaluation and other modules.Based on the experimental platform of the two,this paper completes the motion evaluation,including the character recognition,the motion evaluation and the motion technique improvement analysis.(4)Developed the teaching control system designed in this paper,and implemented the teaching control system based on C#and pycharm.The functions realized include student login,teacher management,character recognition,the motion evaluation and so on.