Research On Students' Interest In Classroom Based On Video

As education enters the era of intelligence,school information construction has entered the application stage from the basic construction stage.In the classroom teaching,mastering the students' listening status is the realistic basis for the efficient conduct of teaching activities,and it also has important reference significance for adjusting teaching strategies.The traditional way to grasp the state of students' listening is to record the lectures by experts.This way of evaluating the state of students' listening requires a lot of personnel and time and has certain subjective interference,which obviously cannot meet the needs of modern personalized education.In view of the lack of teachers' observation methods for students,exploring how to use the existing infrastructure to intelligently evaluate students' listening status has become an urgent problem in smart education.This article uses the technology of image recognition to analyze student classroom videos,mainly researches on student behavior recognition,student classroom interest measurement,modeling,and application of student interest model in classroom videos.Through the establishment of student interest model based on student classroom behavior,Effective evaluation of students' classroom status.The completed research content of the thesis is as follows:(1)Aiming at the current analysis and evaluation of students' classroom status,a "fuzzy level analysis" method based on students' classroom behavior is proposed,which quantifies students ' classroom status through comprehensive evaluation of students 'classroom behaviors,and quantifies the The value is called "interest study".(2)After studying how to quantify the state of students ' classroom listening,combined with the behavior information of the students,the state information of a certain period of time is expanded into a set of vectors containing behavior information and interest values,and a vector is constructed based on this vector.Interest matrix,that is,the student's classroom interest model.(3)On the basis of the student's classroom interest model,constructs the early warning judgment of student's classroom behavior based on decision tree,and the prediction of student's interest degree based on the expanded behavior vector.Using the surveillance video of the classroom as experimental data,after analysis examined the students' interest in reasonable and practical model established.