A Deep Learning Framework for Crowd Internet of Things (Crowd-IoT)

Close-circuit television (CCTV) cameras are typically used to monitor crowd gatherings in a variety of locations, including hospitals, parks, stadiums, airports, and sites of cultural and religious significance. CCTV camera downsides include a small coverage area, installation issues, portability, high power consumption, and frequent operator monitoring. Because of this, many academics are now focusing on computer vision and deep learning, which have solved these problems by reducing the need for human input. The Internet of Things (IoT) is increasing the use of crowd control, which gives cities new ways to use data online for monitoring, managing crowds, and controlling the mechanisms of devices that allow them to handle large amounts of data in context. This paper extensively reviews crowd classifications and presents the most recent developments in crowd monitoring with embedded hardware and also discusses the prospective IoT implementation of crowd analysis and the types of cloud platforms used for crowd-IoT.