Leveraging convolutional neural networks for real-time student attendance tracking

This study presents an advanced Convolutional Neural Network (CNN) for facial detection and identification, outperforming traditional models. The method uses a channel wise separable CNN to extract key characteristics from images, paired with Softmax classifiers and Support Vector Machine (SVM) for precise categorization. The system was applied in a smart classroom to monitor student attendance, achieving a 98.11% success rate on the labeled faces in the Wild database. The cloud-based edge computing architecture demonstrated potential to enhance security and effectiveness in educational contexts, with notable improvements in accuracy and efficiency compared to existing methods.