Affiliations: Department of Police training, Liaoning Police College, Liaoning, Dalian, China
Correspondence:
[*]
Corresponding author. Lixin Yan, Department of Police training, Liaoning Police College, Liaoning, Dalian, China. E-mail: yanlixin517@sina.com.
Abstract: In traditional teaching, people can quickly guess the emotion type of the other person based on facial expressions but understanding the human facial expression by computer is a very complicated problem. There may be some differences in the expression of each person. Therefore, how to make the computer ‘read and understand’ the emotional state of the person according to the facial expression of the person is the research focus of this paper. In this paper, the expression recognition based on dynamic sequence is developed, and the mapping relationship between basic expression and emotion is studied to construct the emotional model, and the emotional state recognition of the learner is realized by the research on facial expression. In the intelligent teaching environment, the teacher adjusts the teaching strategy according to the emotional state recognition test results, improves the teaching efficiency, and realizes the wisdom teaching. Moreover, combined with the actual situation, an expression recognition algorithm based on the ultra-wide regression network model for unsupervised learning classroom education is constructed. Through experimental analysis, we can know that this research algorithm has certain advantages in facial expression recognition.
