Narrow Results

Nowadays, it is essential to capture and evaluate student action in the physical education classroom to assess their behavior. Every student’s performance is unique in physical activity. Every time, the staff or trainer cannot watch and evaluate the students individually. At the university level, the use of classroom capture systems is becoming more widespread. However, due to technology’s recent growth and application, the research on classroom capture systems’ efficacy in university classrooms has been minimal. This paper is proposed for the student action capture and evaluation system. Image preprocessing is the process of preparing pictures for use in model training and inference. This covers resizing, orienting, and color adjustments, among other things. As a result, a change that can be an augmentation in certain cases can be better served as a pretreatment step in others. The DL-IF uses cloud technology for data storage and evaluation. DL-IF uses the imaging technology to monitor students’ actions and responses in the classroom. The image data are evaluated based on the trained set of data provided in DL-IF’s Artificial Neural Network (ANN). The evaluation of unique individuality in every student’s performance is reported to the respective trainer. The simulation analysis of the proposed method DL-IF proves that it can monitor, capture and evaluate every student’s action in all physical activity classrooms. Hence, it proved that this framework could work with high accuracy and minimized mean square error rate.