A teacher classroom performance intelligent evaluation method based on deep learning

Abstract

The invention discloses a teacher classroom performance intelligent evaluation method based on deep learning, and the method comprises the steps: carrying out the facial expression recognition, voicerecognition and behavior recognition of a teaching video of a teacher to be evaluated; Integrating vector information obtained by facial expression recognition, speech recognition and behavior recognition of the to-be-evaluated teacher to obtain an emotion state vector and a teaching state vector of the to-be-evaluated teacher in the class; Training a deep learning network for evaluation based onthe data set marked and scored by the expert; And inputting the emotion state vector and the teaching state vector of the teacher to be evaluated in the classroom into the trained deep learning network to obtain an evaluation result of the classroom performance of the teacher. According to the method, the classroom performance of a teacher is simply and objectively evaluated by combining facial expressions and limb actions, voice information is introduced in the recognition process, the result accuracy is improved through combination of images and voices, the pressure of a supervision group isreduced, and the evaluation result is more real and objective.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and more specifically, relates to a method for intelligently evaluating teachers' classroom performance based on deep learning. Background technique [0002] Colleges and universities have a large scale, and the daily teaching activities are very rich. At present, for the teaching work of the majority of teachers, the school has a variety of evaluation systems such as the supervision team evaluation and student evaluation. However, there is a lack of direct and effective evaluation methods for teachers' performance in the classroom. Due to limited energy and time, the supervisory team is unable to conduct long-term, full-coverage evaluations of all teachers' courses. Student evaluation contains subjective factors, which may be unfair to some teachers who are strict in teaching and have difficult exams. Moreover, the student evaluation feedback cycle is long and not real-time. ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects of the prior art, the purpose of the present invention is to solve the technical problems of incomplete and low accuracy evaluation of teachers' classroom performance in the prior art

Images