Feasibility Research on the Application of Facial Expression Recognition in College Students' Mental Health Interview

Abstract

Using computer expression recognition technology to learn the different reflections of mental sub-health status, and then explore the possibility of using facial expression recognition method in new interviews to identify mental sub-health status through this method. Methods the sub-health self-assessment scale and symptom checklist 90 (SCL-90) were used for questionnaire survey. Twenty-one subjects were selected and divided into experimental group 1, experimental group 2 and control group through pairing. An experimental study of facial expression feedback was carried out. Results compared with the control group, the somatization, interpersonal sensitivity, anxiety and psychosis of the experimental group 1 were significantly lower than those of the control group (t=2.25, -2.45, -2.42, -2.39; p<0.05); Compared with the control group, the interpersonal sensitivity and anxiety of the experimental group 2 were significantly lower than those of the control group (t=-2 06, -2.16, -2.23; p<0.05); There was no significant difference between experimental group 1 and experimental group 2. Through the computer analysis test of the sampled data, the conclusion shows that the use of computer facial expression recognition can identify the possibility of mental sub-health status of college students.