Deep Learning-Based Real-Time Weapon Detection with YOLOv8 for Intelligent Surveillance

Deep Learning-Based Real-Time Weapon Detection with YOLOv8 for Intelligent Surveillance

1Dr.Shalima Sulthana

Associate Professor, Dept of Computer Science and Engineering Vignan's Institute of Management and Technology for Women, Hyd.

Email: m.s.sulthana2012@gmail.com

3Sunkari Ashwini

UG Student, Department of Computer Science and Engineering Vignan's Institute of Management and Technology for Women, Hyd.

Email: sunkariashwini95@gmail.com

2Musku Keerthana Reddy

UG Student, Department of Computer Science and Engineering Vignan's Institute of Management and Technology for Women, Hyd.

Email: keerthanareddymusku93@gmail.com

4Gottipati Reshma

UG Student, Department of Computer Science and Engineering Vignan's Institute of Management and Technology for Women, Hyd.

Email: gottipatireshma167@gmail.com

Abstract— The escalation of security threats has highlighted the implementation of intelligent surveillance systems with the ability to identify threats in advance. This study suggests a real-time deep learning-based weapon detection system with the YOLOv8, an object detection deep model. The system inspects real-time video streams for identifying and locating weapons such as guns and knives at high accuracy and low latency. By leveraging the improved-optimized architecture of YOLOv8, the model achieves high accuracy of detection with computational performance maintained high enough for real-time applications. Deep training on weapon dataset annotations enables strong performance in a broad surveillance setting. The developed methodology enhances situational awareness and provides secure autonomous response, and thus it is a great candidate for use in safety- critical environments like public transportation centers, schools, and critical infrastructure facilities.

Index Terms — Real-Time Surveillance, Weapon Detection, Deep Learning, YOLOv8, Object Detection, Smart Surveillance, Public Safety, Threat Detection.