Integrating YOLO Algorithms for Dual-Functionality Surveillance

Abstract

Since the outbreak in December 2019, the COVID-19 pandemic has significantly impacted Malaysia, affecting both the nation and its citizens on a substantial scale. Consequently, the government has put forth several recommended standard operating procedures (SOPs) aimed at mitigating the pandemic's impact. These measures include mask-wearing, frequent handwashing, and adhering to social distancing guidelines. Maintaining social distance is crucial in reducing the transmission of this lethal virus from an individual carrying the infection to a person in good health. The method encompasses the utilization of the You Only Look Once (YOLO) object detection algorithm for the widespread application of identifying and overseeing adherence to social distancing practices within monitoring systems. In addition to its effectiveness in object detection, YOLO has gained extensive global utilization in systems for detecting human presence. Thus, this study provides a review of YOLO-based monitoring systems that have been developed previously. From that, this paper introduces a framework integrating thermal imaging for combined social distancing and body temperature monitoring. From the results, only 14% of the total numbers of visitors at the exhibition complied with the social distancing. The other 12.5% were in danger when they broke the rules while 73% of individuals were observed to be in a situation of moderate risk. However, most of the recorded body temperatures were normal.