Intelligent Analysis And Recognition Of Teachers' Body Movements In Middle School Classroom Teaching Videos

In recent years,artificial intelligence technology,represented by face recognition,speech recognition and image processing,has been widely used in the field of education,bringing new changes to the development of intelligent education.For students,the classroom is the most important place to study.In the classroom teaching scene,in addition to the teacher's own teaching level and other factors,the teacher's body movements will also have an impact on the teaching effect.Therefore,it is helpful for teachers to timely understand the body movements that they use in classroom teaching,which correspond to the teaching content and improve the teaching strategies,so as to improve teaching skills and promote the improvement of classroom teaching quality.However,the current classification system of teacher's body movements is not perfect.The traditional method of teacher's body movements recognition and analysis relies on artificial observation,which is inefficient and consumes human resources.How to intelligently identify and analyze teacher's teaching body movements in class has gradually become a hot topic.To solve the above problems,based on the latest educational theories and the new generation of artificial intelligence technology,based on the real classroom learning environment,this paper proposes to apply the deep learning method to the recognition of teachers' teaching body movements,in order to better serve the teaching.The main work of this paper is as follows:Firstly,based on the existing research,the author constructs the body movement analysis and coding system of middle school classroom teachers.This article in view of the current physical teaching behavior code table,etc.Analysis of the existing research,overcome traditional teacher body movements fuzzy classification of the difficulties,to build out of the classroom environment and conforms to the analysis of teachers' body movements coding system,clear body movements to teachers to teach,blackboard writing,instructions,tour guide,learners demonstrate five categories on stage.Secondly,a database of teachers' body movements is constructed.There are only a few databases of teachers' body movements,but a high quality database plays an important role in training effective recognition models and accurately understanding teachers' teaching behaviors.Therefore,this paper extracts images from the open middle school classroom teaching videos in the way of one frame per second,cleans and classifies the images,and constructs the database required by the current research,which contains about 20,000 original images.Thirdly,a deep learning-based method is proposed to identify teachers' body movements.At first,the self-built database was divided into training set,verification set and test set.After training Vgg16,Vgg19 and other network models through the training set,it was found that the accuracy of Vgg16 and Vgg19 network models was not much different.But the number of parameters of Vgg16 was significantly less than that of Vgg19.Therefore,this paper adopted Vggl6 to identify teachers' body movements.Finally,the paper compares the recognition effect of different methods on teachers' body movements with the artificial method,and verifies the effectiveness of the method in this paper.