Built and trained a Convolutional Neural Network (CNN) with augmented datasets to improve accuracy and generalization. Integrated OpenCV's Haar Cascade for face detection and implemented live mask/no-mask classification with real-time bounding boxes. Evaluated the model using confusion matrix and classification metrics, then deployed it with an interactive interface for live camera monitoring. The solution is suitable for healthcare facilities, transportation hubs, workplaces, and other public environments requiring automated mask compliance monitoring.
