Recognition of Student Emotions Through Facial Expressions in Classrooms Using Deep Learning Techniques

Abstract

Classroom assessments offer crucial insights into teaching and learning by helping teachers analyze their methods and identify areas for improvement. This paper reviews recent research on detecting student emotions in educational settings and presents a deep learning-based approach for emotion recognition. Using the FER2013 facial emotion database, we trained a convolutional neural network (CNN) model, leveraging transfer learning with the VGG16 architecture pre-trained on the Cohn-Kanade (CK+) database. The system uses a camera to detect and classify emotions like sadness, happiness, anger, and more, providing real-time feedback on classroom mood. Our approach, which can be applied in various settings such as video conferencing, improves the accuracy and speed of emotion recognition, benefiting the overall learning experience