Student Behavior Analysis in Intelligent Education Systems: Application and Evaluation of Object Detection Algorithms

This paper proposes a novel framework for real-time detection of student behaviors in classroom environments, with a focus on key actions such as hand raising, reading, and writing. The framework leverages an efficient convolutional neural network architecture, integrating multiple innovative modules to enhance behavior recognition capabilities. Extensive evaluations on two public student behavior datasets (SCB-Dataset3-S and SCB-Dataset5) demonstrate that the proposed framework outperforms existing state-of-the-art models in terms of accuracy and robustness, particularly for complex behaviors in dynamic classroom settings. The results indicate that this approach has significant practical implications for personalized teaching management, intelligent classroom systems, and educational performance assessment. The study suggests that the proposed framework provides a promising solution for the development of smart education systems.