Smart Visual Sensing for Overcrowding in COVID-19 Infected Cities Using Modified Deep Transfer Learning

Currently, COVID-19 is circulating in crowded places as an infectious disease. COVID-19 can be prevented from spreading rapidly in crowded areas by implementing multiple strategies. The use of unmanned aerial vehicles (UAVs) as sensing devices can be useful in detecting overcrowding events. Accordingly, in this article, we introduce a real-time system for identifying overcrowding due to events such as congestion and abnormal behavior. For the first time, a monitoring approach is proposed to detect overcrowding through the UAV and social monitoring system (SMS). We have significantly improved identification by selecting the best features from the water cycle algorithm (WCA) and making decisions based on deep transfer learning. According to the analysis of the UAV videos, the average accuracy is estimated at 96.55%. Experimental results demonstrate that the proposed approach is capable of detecting overcrowding based on UAV videos' frames and SMS's communication even in challenging conditions.

I. Introduction

In Public places and societies, overcrowding may signify an unusual community, riot, or public event [1], [2]. Crowded areas can contribute to the spread of infectious diseases due to social concerns. Today, different cameras, such as CCTV cameras, are used in society for video surveillance in crowded environments [3]. Monitoring the crowd from overcrowding is also challenging because abnormal events occur infrequently, and the people's congestion may sometimes exhibit different behaviors. In society, the detection of overcrowding plays a crucial role in preventing problems such as the emergence of abnormal population behaviors [4]. It is also necessary to provide a continuous and real-time automated monitoring system based on social computing in smart cities to prevent errors caused by fatigue and inefficiency. The detection of overcrowding in smart environments is one of the novel social systems that can improve population management. Due to the rapid growth of urbanization in modern cities, intelligent approaches are needed to handle essential challenges such as civil infrastructure, healthcare, and transportation [5]–[7]. In light of the above, recent breakthroughs in computer technology, networking, and sensing offer us the opportunity to further integrate information systems with our physical society. To develop more efficient and optimal density management methods, we can describe overcrowding using automated learning methods [8]–[10]. Intelligent systems can be improved in smart cities and diseases can be prevented by using artificial intelligence, on the other hand. Drones, otherwise known as unmanned aerial vehicles (UAVs), operate without a pilot. In contrast to CCTV cameras (which are often fixed and not portable), drones can record environmental conditions at high altitudes and near the ground. As a result, UAVs can be employed to address challenges such as overcrowding and unusual behaviors among crowds. Network UAV applications can be highly effective in detecting overcrowding as well as monitoring population behavior. Moreover, control centers process information by taking time, being accurate, and being aware.