An automatic detection system that can scan people entering a public or private space and determine whether they are wearing a protective face covering or not and whether that face covering is being worn correctly is discussed in the International Journal of Ad Hoc and Ubiquitous Computing. Such a detection system could hook into an alert system on the smartphones of visitors to a space or allow stewards to offer guidance to those entering the space who need to be advised on the wearing of a face covering.
Despite the fact COVID-19 remains an ongoing health threat to people, many public and private spaces have, at the time of writing, forsaken the need for visitors and employees to wear a face covering in order to reduce the spread of infection. However, given that the virus undergoes constant evolutionary change, there may well be an urgent need once again to “mask up”. Notwithstanding the emergence of another airborne pathogen in the future.
Vishnu Kumar Kaliappan and Dugki Min of Konkuk University in Seoul, South Korea, and Rajasekaran Thangaraj, P. Pandiyan, and K. Mohanasundaram of the KPR Institute of Engineering and Technology, S. Anandamurugan of Kongu Engineering College in Tamil Nadu, India, point out that wearing a face mask in public places, particularly enclosed spaces, is one of the most effective strategies for protecting individuals from this disease, according to the World Health Organization (WHO). The WHO is yet to lower the status of COVID-19 from pandemic. At the time of writing, there were around 1.4 million new cases in the previous seven days and almost 10000 deaths reported in that time.
The team’s algorithmic approach can determine whether a person is wearing a mask and whether or not that mask is in place correctly (not below the nose or chin, for instance). It is based on an object detection model known as YOLO version 5. YOLO is an abbreviation in this context for “you only look once”. It has an accuracy of 99.4 percent. Such precision would make much easier the job of stewards or ushers in a space there to ensure correct mask wearing. The system copes with masks of different sizes, shapes, and colours.
Kaliappan, V.K., Thangaraj, R., Pandiyan, P., Mohanasundaram, K., Anandamurugan, S. and Min, D. (2023) ‘Real-time face mask position recognition system using YOLO models for preventing COVID-19 disease spread in public places’, Int. J. Ad Hoc and Ubiquitous Computing, Vol. 42, No. 2, pp.73–82.