International Journal of Scientific Research and Engineering Development

Traditional surveillance is limited and relies heavily on manual monitoring. These limitations have been exposed by the growing demand for intelligent security systems. In this paper, we propose an intelligent computer vision framework for real-time surveillance and intruder detection called NeuroSentinel. The system utilises live video feeds, face detection, and facial recognition technologiesto identify authorised and unauthorised individuals. If an intruder is detected, the framework automatically captures evidence and generates real-time alerts to improve security response. NeuroSentinel is an ef icient, scalable, and cost-ef ective surveillance solution built using Python, OpenCV, Flask, and MySQL. The proposed system improves security monitoring accuracy, decreases human ef ort, and enables automated detection of threats across diverse real-world environments including educational institutions, of ices, and residential areas.

Karunamurthy A, T. Thamizhselvan,"NeuroSentinel: An Intelligent Computer Vision Framework for Real-Time Surveillance and Intruder Detection" International Journal of Scientific Research and Engineering Development, V9(3): Page(2118-2121) May-June 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.