Method, device, terminal and medium for analyzing students' psychological state based on audio and video

Abstract

The present invention relates to the field of artificial intelligence technology, and provides an audio-video-based method, device, terminal, and medium for analyzing students' mental states, including: performing transfer learning and training students' emotional expression recognition models based on open source emotional expression data sets and private emotional expression data sets ; Distortion correction and grid mapping are performed on the student panoramic video to obtain the student target video of each student; the student target video is recognized by the student emotional expression recognition model to obtain the activation degree and pleasure; the student target video is calculated to obtain the activity degree; according to the student's audio Calculate negative emotions; analyze the mental state of students based on activation, pleasure, activity and negative emotions. The invention can be applied to wisdom education, can comprehensively analyze the mental state of students in multiple modes, and has high accuracy. In addition, the present invention also relates to the field of block chain technology, and the mental state can be stored in the block chain.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an audio-video-based method, device, terminal and medium for analyzing students' mental states. Background technique [0002] With the continuous improvement of the informatization of the society, schools are paying more and more attention to the mental health of students. In different growth environments, students may have psychological allergies due to mental or physical changes, and thus induce mental illness. If it is not detected and treated in time, it will lead to more serious consequences. [0003] In the prior art, although the expression changes of the students are analyzed by collecting the video of each student, and the psychological state of the students is identified based on the expression changes, this technology has the following disadvantages: collecting the video of each student is time-consuming and laborious; Different students have ...

AI Technical Summary

Problems solved by technology

If not detected and treated in time, it will lead to more serious consequences

[0003] In the prior art, although the expression changes of the students are analyzed by collecting the video of each student, and the psychological state of the students is identified based on the expression changes, this technology has the following disadvantages: collecting the video of each student is time-consuming and laborious; Different students have different personalities. Even if some students already have mental illness, it is difficult to express it from the expression, resulting in poor accuracy of the analyzed psychological state, which cannot truly represent the psychological state of the student, or cannot perform expression analysis at all.

Images