REAL-TIME FACEMASK RECOGNITION WITH ALARM SYSTEM UTILIZING DEEP LEARNING

ABSTRACT

The Coronavirus disease (COVID-19) pandemic has created a global health crisis, and the use of face masks has become a critical preventive measure to control the spread of the virus. Enforcing the use of face masks in public places can be challenging, especially in crowded areas. The project aims to develop a real-time face mask recognition system utilizing computer vision techniques to detect individuals who are not wearing masks in public places. The system is designed to operate in real time, utilizing a camera and a pretrained convolutional neural network model for detecting faces and classifying them as masked or unmasked. The objective of this project is to achieve a high classification accuracy for the system by evaluating it using two publicly available datasets. The project is divided into two main modules: face detection and mask classification. The face detection module uses a pre-trained deep learning model to detect faces in the input video stream, and the mask classification module uses another pre-trained deep learning model to classify each detected face as masked or unmasked. The proposed system is implemented using the Python programming language and several computer vision libraries, including OpenCV and TensorFlow.The system is evaluated using two publicly available datasets, the Labeled Faces in the Wild (LFW) dataset and the Face Detection Data Set and Benchmark (FDDB) dataset.