Design of Intelligent Assisted Teaching System for Civics and Political Science Classes in Colleges and Universities Based on Image Processing Algorithms and Methods of Analyzing Students’ Interactive Behavior

In traditional teaching, it is time-consuming and labor-intensive to ensure the teaching quality only through teachers' observation of students' behavioral status.Therefore, deep learning algorithms and image processing algorithms are used to construct an intelligent assisted teaching system, so as to realize classroom target detection, interactive behavior recognition and classification.By analyzing students' interactive behaviors and monitoring the classroom status in real time, the quality of education and teaching is further improved.Taking 14 examples of Civics and Political Science classes in a university as research samples for empirical analysis, it can be seen that the accuracy rate of the algorithm proposed in this paper for detecting standing behavior reaches 84%, and the accuracy rate of the overall behavior detection system is 81.4%, which is a good effect.Teacher-student interaction behavior is most often characterized by "instruction-passive response".Student-teacher interaction behavior "Passive Response-Lecture" appeared most frequently.The student-student interaction behavior is most often "debriefing-debriefing".