Your browser doesn't support javascript.
Alert System for Face Mask Detection using CNN
5th International Conference on IoT in Social, Mobile, Analytics and Cloud (I-SMAC) ; : 764-769, 2021.
Article in English | Web of Science | ID: covidwho-1779068
ABSTRACT
The pandemic of the Coronavirus has a significant impact on people's health and lives. There is a question asked to ourselves whether it is possible to return to normalcy? To do this, a risk-free environment must be developed. COVID-19 virus spreads through the respiratory mainly from infected persons who come in close contact with other beings. Wearing a Face Mask will act as a physical barrier for these respiratory droplets. The main aim of this research paper is to create a face mask detection system using TensorFlow, Keras, and OpenCV that identifies whether or not a person is wearing a mask by monitoring a live video feed and notifying them with a beep sound. An alert system has been integrated with face mask detection system in three phases. In the first phase, the dataset is trained with CNN (using Keras/TensorFlow) with an accuracy of 98.38% The second phase is focused on identifying whether a person is wearing a mask on a live video stream using Haar Cascade Classifier. In the third phase, the alert system is developed. If a person is detected not wearing a mask, he/she gets alerted via beep sound. This face mask detection system can be used at the entrance of universities, airports, hotels, public buildings, Railway Stations, or any major places huge crowds are expected.
Keywords

Full text: Available Collection: Databases of international organizations Database: Web of Science Language: English Journal: 5th International Conference on IoT in Social, Mobile, Analytics and Cloud (I-SMAC) Year: 2021 Document Type: Article

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: Databases of international organizations Database: Web of Science Language: English Journal: 5th International Conference on IoT in Social, Mobile, Analytics and Cloud (I-SMAC) Year: 2021 Document Type: Article