The well-being and academic performance of university students are significantly influenced by their emotional experiences within learning environments. This research presents a novel application of Facial Emotion Recognition (FER) using web cameras and Convolutional Neural Networks (CNN) to enhance classroom and laboratory settings. The system is designed to monitor students’ emotional states in real time, enabling educators to adapt learning environments for improved engagement and well-being. Initially, an image dataset was collected through web cameras placed in classrooms, capturing students’ facial expressions during lectures and activities. The dataset was preprocessed to account for variations in lighting, facial orientations, and occlusions. A CNN-based model was developed and trained to classify emotions such as happiness, sadness, anger, and neutrality. Hyperparameter tuning and data augmentation techniques were employed to optimize model performance. The model was evaluated using metrics such as accuracy, precision, recall, and F1-score to ensure its effectiveness across diverse conditions. Additionally, the system was optimized to handle real-time emotion detection in varying classroom environments. This real-time FER system allows continuous monitoring of students’ emotional states, providing actionable insights for educators to create more supportive and adaptive learning environments. The results demonstrate the potential of FER technologies to transform educational environments by fostering a deeper understanding of student emotions and promoting well-being. This research offers a foundation for the integration of AI-driven tools in modern education to enhance learning experiences.