Rapid Detection of Pilgrims Whereabouts During Hajj and Umrah by Wireless Communication Framework : An application AI and Deep Learning

Abstract

Human injuries and deaths occur often during public events like concerts, religious services, and political rallies because of a lack of proper crowd safety oversight. A small disaster can trigger panic in a huge crowd. Many intelligent video surveillance solutions can identify things, but despite the recent developments in artificial intelligence approaches and deep learning processes, it is very probable to track congested crowds and their mobility to avoid future detection disasters. Searching for points of interest makes use of movement analytics and classification to provide a superior platform for monitoring large crowds. The purpose of point-of-interest explorations is to aid in the management of the safety of moveable crowd events by assisting in the prediction and prevention of future disasters through the classification and analysis of real-time information gathered from crowds. Current surveillance cameras are insufficient for monitoring large crowds in outdoor locations due to their inability to scale. We believe that by using our proposed crowd analysis strategy, we may help enhance the current state of crowd safety management. Among the many aspects of crowd motion, we pay special attention to the difficulties of determining the identity, velocity, and direction of individuals inside the group. We then used these crowd-level semantics to monitor test POI searches in both a controlled lab environment and a real-world crowd. Findings from this study imply that POI searching can be utilized to help prevent harmful circumstances brought on by the movement of large crowds by recognizing the characteristics of mobile crowds in real time.