Computer Vision Based Face Mask and Safety Gear Detection

Abstract

This paper presents a Computer Vision Based Face Mask and Safety Gear Detection System that utilizes Artificial Intelligence (AI) and Deep Learning techniques to automatically detect whether individuals are wearing face masks through real-time video captured from a webcam. The primary objective of the proposed system is to improve safety compliance by reducing the dependence on manual monitoring in workplaces, hospitals, educational institutions, industries, and public places. Traditional face mask monitoring methods rely on human observation, which is time-consuming, laborintensive, prone to errors, and inefficient in crowded environments. To overcome these limitations, the proposed system performs automated real-time detection with high accuracy and minimal human intervention. The system is developed using Python, Flask, OpenCV, TensorFlow, Keras, HTML, CSS, and JavaScript. OpenCV is used for image acquisition and face detection, while a pretrained MobileNetV2 deep learning algorithm is employed to classify each detected face as Mask or No Mask. Flask provides a web-based interface that enables users to access the application and view live detection results. The processed video stream displays prediction labels and confidence scores for each detected face, allowing instant monitoring and decision-making. The proposed system offers fast processing speed, reliable performance, and cost-effective deployment.