Student Classroom Behavior Management Based on Computer Vision Using Quantum Evolutionary Algorithm with DenseNet 121 Model
Fang Wu, Tao Xu · 2024
Student recognition and the evaluation of classroom education have both grown to depend heavily on the student classroom behavior management in recent years. The method used to assess and analyze student behavior in the classroom establishes whether the student's giving attention or not. However, because of the complexity of classroom conduct, it was difficult to identify intelligent students. Therefore, in this research Quantum Evolutionary Algorithm with DenseNet 121 (QEA-DenseNet 121) is suggested by studying computer vision of student behavior in the classroom. Usually, the recommended system design is used to evaluate the system's testing and training procedures. The final input images are then subjected to human location estimate using a camera to capture subsequent frames. The error correcting system is integrated with the body position estimate and person recognition algorithms. Lastly, a model known as QEA-DenseNet-121 is recommended as a practical resource for precisely evaluating student behavior in the classroom. Results showed that the suggested approach outperformed the existing models such as Skeleton Pose Estimation (SPE), YOLO-v4 and Intelligent Real-Time Vision (IRTV) with relative gains in Average Accuracy of 99.64%, Precision of 99.53%, Recall of99.71 %, and F1-measure of 99.49%.