Automated Video Analysis for Crowd Management

Abstract

Video surveillance is increasing in many areas, including facilities management, traffic monitoring, crowd control and urban security. Despite the growing demand for CCTV and related systems in public spaces, challenges remain regarding the ease of surveillance installation. This research demonstrates a cost-effective way to use computer technology to identify, track and count the flow of people through cameras. This study investigates two software applications and compares their performance. The system's performance has been proven in controlled and uncontrolled global environments. Additionally, the system integrates with existing monitoring systems, improving its scalability and adaptability. It demonstrates high performance in processing and analysing video data and demonstrates the speed and accuracy of group identification and tracking. The system's user-friendly interface and potential for adoption across a wide range of industries and sectors demonstrates its impact on increasing safety and improving well-being in public spaces. Future research directions include development and application of the proposed methodology in other clinical settings.