Intelligent systems for arms base identification: A survey on YOLOv3 and deep learning approaches for real-time weapon detection

Soppari, Kavitha and Kalam, Muhammad Abul and Garugu, Sri Sudha and Mannem, Dinesh and Pagidimari, Aravind and Aluvala, Poojitha (2025) Intelligent systems for arms base identification: A survey on YOLOv3 and deep learning approaches for real-time weapon detection. World Journal of Advanced Research and Reviews, 25 (1). pp. 2058-2066. ISSN 2581-9615

Weapon detection using computer vision is a crucial component of modern security systems, ensuring the safety of public spaces by identifying dangerous objects like firearms and knives. With advancements in artificial intelligence, particularly in deep learning, weapon detection systems can now operate in real-time, providing faster and more accurate results than ever before. In this work, we propose the development of a cloud-based weapon detection system that leverages deep learning techniques, such as YOLO (You Only Look Once), to detect weapons in images and video streams. The system is designed to process both static and dynamic visual data, providing real-time alerts and detailed monitoring for security personnel. The system will be equipped with an object detection pipeline that incorporates pre-trained models to identify weapons and monitor their presence across various environments. This allow for easy tracking and auditing of security incidents. By implementing this weapon detection system, organizations can significantly improve their security measures, providing faster identification of threats and reducing the risk of violence in sensitive areas.