AI-Based Face Mask Detection for Enhanced Security in Facial Biometrics

This paper introduces a novel AI-based method for the identification of masks. This approach has the potential to improve the security and adaptability of face biometric systems. It is particularly effective in situations where coverings obscure significant facial features. The pervasive use of masks significantly impedes conventional face recognition algorithms, particularly those that were developed after COVID-19, as they are reliant on visible facial structures. The proposal is a face mask identification method that utilizes a convolutional neural network (CNN) to improve its accuracy and efficiency. This enables the biometric system to dynamically modify its identifying algorithms or activate alternative authentication methods, such as voice or iris recognition. The model's remarkable accuracy and the capacity to process data in real-time are a result of its training on a diverse and extensive dataset that includes both concealed and unmasked faces. Experimental data indicate that the rates of incorrect acceptance and rejection have decreased as a result of a substantial improvement in identification accuracy under obscured conditions. The incorporated disguise detection technology enhances the security, reliability, and adaptability of facial biometrics, a significant concern for contemporary security systems.