Automated Thermal Screening for COVID-19 Using Machine Learning

In the last two years, millions of lives have been lost due to COVID-19. Despite vaccination programmes for a year, hospitalization rates and deaths are still high new variants Stringent guidelines COVID-19 screening measures such as temperature check mask at all public places helping reduce spread Visual inspections ensure these can be taxing erroneous. Automated inspection ensures an effective accurate screening. Traditional approaches involve identification faces masks from visual camera images followed by extraction values thermal imaging cameras. Use primary modality limits applications only good-lighting conditions. The use alone makes system invariant illumination. However, lack open source datasets is issue develop systems. this paper, we discuss our work on using machine learning over video streams face detection subsequent in passive non-invasive way that enables automated method places. We NTIC dataset was used training models collected 8 different locations. Our results show presence This performance stays same even under low-lighting conditions, whereas with trained classifiers more than 50% degradation.