Simulation of classroom student behavior recognition based on PSO-kNN algorithm and emotional image processing

Abstract

Students’ classroom behavior recognition and emotion recognition effects directly determine the degree of teachers’ control of the classroom teaching process. At present, teachers and students belong to two groups in traditional teaching, and teachers cannot effectively mobilize students’ learning emotions. In order to improve the teaching effect, this paper combines the PSO algorithm and the KNN algorithm to obtain the PSO-KNN joint algorithm, and combines with the emotional image processing algorithm to construct an artificial intelligence-based classroom student behavior recognition model. Moreover, based on the image processing technology, this paper uses key frame detection for feature recognition, and this paper improves the recognition process based on the inter-frame similarity measurement algorithm and initial cluster center selection in the key frame extraction method of clustering. In addition, this paper analyzes the effect of the model constructed on the behavior recognition and emotion recognition of students. The research results show that the joint algorithm constructed in this paper has a high accuracy rate for students’ emotion recognition and behavior recognition, and can meet the actual teaching needs.