The Integration of Artificial Intelligence and Computer Vision in Large-Scale Video Surveillance of Railway Stations

Video surveillance equipment has become increasingly popular in the past two decades for security and human behavior insights. However, monitoring multiple cameras increases costs and makes it difficult for operators to maintain focus. To automate this process, surveillance cameras can be integrated with artificial intelligence (AI) and computer vision (CV) technologies. The project aims to enhance safety and provide a secure environment through intelligent surveillance across a large network of cameras. The system presented in this paper offers real-time monitoring based on YOLOv8 and difference detection algorithms, capable of handling live recordings from over 1000 security cameras simultaneously. Key features include intelligent event detection, allowing for the definition of regions of interest (ROI) within camera views for targeted monitoring and assigning varying priority levels to alarms. The architecture supports scalability, allowing the addition of more cameras and the parallel execution of multiple AI-based scripts for video stream analysis. In this paper, an intelligent video surveillance system is presented as an example of railway station surveillance, but the technology has a wide range of applications in different domains.