<![CDATA[This research discusses the development of a face mask detection system using a Convolutional Neural Network (CNN) for automatic door access in hospitals. Considering the high risk of infectious disease transmission in hospital environments, the implementation of strict health protocols, including mandatory mask usage, is essential. Manual supervision of mask compliance has limitations; therefore, an automated system is required to improve monitoring effectiveness. The dataset used in this study was collected using an ESP32 Cam, consisting of 1,186 images of masked and unmasked faces. The CNN model achieved an average training accuracy of 96.60%, with Precision and Recall values of 0.98. The automatic door system was evaluated through real-time testing involving six subjects, each undergoing 15 trials with masks and 15 trials without masks, resulting in a total of 180 trials. The system achieved a detection accuracy of 90.00% for masked faces and 74.44% for unmasked faces, with an overall system accuracy of 82.22%. These results indicate that the proposed system is capable of reliably supporting automatic door access control based on face mask compliance in hospital environments.]]>
Copyrights © 2026
