The world continues to witness several waves of COVID-19 spread due to the emergence of new variants of the SARS-CoV-2 virus. Stopping the spread requires synergistic efforts that include the use of technologies such as unmanned aerial vehicles and machine learning. This paper presents a novel system for detecting disease symptoms from a distance using unmanned aerial vehicles equipped with thermal and visual image sensors. A hardware/software system that uses thermography to accurately calculate the skin temperature of targeted individuals using thermal cameras is developed. In addition, machine vision algorithms are developed to recognize human actions such as coughing and sneezing, which are paramount symptoms of respiratory infections. The proposed system is implemented and tested in outdoor environments. The results of experiments showed that the system can determine the skin temperature of multiple targeted individuals simultaneously with an error of less than 1 °C. The field experiments showed that the developed system is capable of simultaneously measuring the temperature of more than 10 individuals in less than 5 seconds. Just to give a perspective, it takes at least 3 seconds to measure one individual's temperature if this was done using traditional methods. Furthermore, the results showed that the system has accurately detected actions such as coughing and sneezing with almost 96% accuracy at a real-time performance of 28 frames/second.
|Journal||Remote Sensing Applications: Society and Environment|
|Publication status||Published - Aug 1 2022|
ASJC Scopus subject areas
- Geography, Planning and Development
- Computers in Earth Sciences