Design And Implementation Of Teaching Evaluation System Based On Video Analysis

In modern quality assessment,teaching evaluation is the most important evaluation index.Today,with the improvement of teaching system information,the use of information technology to improve teaching evaluation is more and more concerned by schools and education departments.Modern education has formed a comprehensive evaluation system after several decades of development.Computer vision technology has provided a good technical theoretical basis for video analysis after rapid development in recent years.Due to the popularization of information-based teaching concepts,each university has its own large and complete teaching platform.Under the background of the development of information technology,with the improvement of software development system and technological innovation,a series of educational platform subsystems such as teaching evaluation began to develop in the direction of informationization and automation.The teaching evaluation system is divided into two parts: teacher evaluation and student evaluation.The system extracts the parts of the two parts that can be quantized by the computer for design implementation to improve the efficiency and accuracy of the evaluation.The system adopts C/S framework for design and implementation.The client includes functions such as teacher teaching audio and video information collection and uploading,student attendance information and student behavior information collection.The server side analyzes the audio and video of the teacher's teaching to obtain the teacher's teaching style and scoring model,and obtains the student's seat table for the student's attendance and behavior information.The teacher-side information analysis is divided into two parts:(1)video analysis of the teacher's character detection and face detection in the video;(2)audio segmentation(VAD)and MFCC(mel frequency cepstrum coefficient for the teacher's teaching audio)Audio feature extraction and feature classification to assist video analysis to obtain accurate indicators of teacher behavior.The student behavior consists of three parts:(1)attendance and attendance by video analysis of student attendance;(2)combined with attendance information and student's global feature detection method to track student path by student path;(3)analyze student video,use The face detection algorithm analyzes student participation in class.The indicators analyzed by the integrated student and teacher behavior were trained using the XGBoost model and the evaluation results were calculated.Teaching evaluation is a systematic project that requires a lot of manpower,material resources and financial resources.With the aid of the teaching evaluation system,the evaluation efficiency and accuracy can be greatly improved,and the consumption of human,material and financial resources can be reduced,and the evaluation results can be managed in a unified manner.At the same time,the large amount of data collected by the system will also provide a basis for in-depth analysis and decision-making of teaching evaluation.