Ai-powered Intelligent Surveillance System Designed for Real-time Detection and Tracking of Individuals
Year: 2026
Type: article
Abstract: Traditional security systems largely depend on continuous manual monitoring, where personnel watch multiple CCTV feeds for extended periods. This approach often leads to fatigue, delayed responses, and overlooked suspicious activities. To address these limitations, this study presents a smart surveillance system powered by real-time AI-based person detection and centroid tracking. The system utilizes YOLOv8 for accurate detection and a lightweight tracking method to assign unique IDs, ensuring consistent identification across frames. A Flask-based web interface enables live monitoring while recording movement data in a local SQLite database for auditing and analysis. Operating entirely on local hardware, the system ensures enhanced privacy, reduced latency, and easy access via a local network. Experimental findings demonstrate reliable performance in dynamic environments, making it suitable for campuses, offices, parking areas, and other public spaces. Keywords—AI Surveillance, YOLOv8, Object Tracking, Flask Dashboard, Real-Time Monitoring, Centroid Tracking, Computer Vision, Security Automation.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7220008865