IoT temperature scanning and facemask detection entry system
Date
2022Author
Mutatiina, Atuhaire Calvin
Namale, Gloria
Kiggundu, Jonathan
Ssemugga, Ronald Ssemyalo
Metadata
Show full item recordAbstract
Due to the emergence and rapid spread of COVID-19, different governments worldwide made it a requirement for each person to wear a face mask to reduce and control the spread of the coronavirus that causes COVID-19 disease. To enforce this, security personnel have been hired at entrances of different public places like malls, offices, schools, and other public areas to enforce face mask wearing by the public and temperature detection before being granted access to these places. However, this implementation method became time-consuming and tiresome, with some personnel even ignoring their duties.
To address this, this report proposes a machine learning model that can detect in real time whether a person is wearing a face mask and has their body temperature recorded. The model has been trained and tested on various datasets with 1000 images where 60% of the images were tested on faces with masks and 40% on images with masks therefore 91%, 87%, and 89% precision were achieved, recall and F1 score for images with masks on and an 86%, 90% and 88% precision, recall and F1 score for images without a facemask, hence proved a reliable model for implementation with minimal error.
This model makes use of concepts of machine learning to carry out its functions and also shows how prototyping hardware is used to implement the model and be used and tested in real-time by the common user who does not need to know the workings of the model.