MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641046810 A) filed by Jennifer D; Dr. P. Deepa; Sanjana V S; Varshini P; and Varshini K, Chennai, Tamil Nadu, on April 12, for 'vigilanteye: next-gen smart monitoring for weaponized threat prevention.'
Inventor(s) include Dr. P. Deepa; Sanjana V S; Varshini P; and Varshini K.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "Firearm-related crimes pose significant challenges to law enforcement, requiring advanced surveillance solutions beyond traditional CCTV monitoring. This paper presents a deep learning-based firearm detection system utilizing Faster Region- Based Convolutional Neural Networks (Faster RCNN) and EfficientDet architectures for enhanced accuracy. To optimize detection performance, an ensemble scheme integrating Non-Maximum Suppression, Non-Maximum Weighted, and Weighted Box Fusion techniques is implemented. Among these, the Weighted Box Fusion-based approach demonstrates superior results, achieving mean average precisions (mAP) of 77.02%, 16.40%, and 29.73% for mAP0.5, mAP0.75, and mAP[0.50:0.95], respectively, outperforming other models in firearm and human face detection. Extensive testing with unknown images and video clips validates the system's reliability, supporting law enforcement by enabling quick incident detection and intelligence gathering for timely preventive action. Additionally, the proposed approach extends its applicability to detecting gun-related content in social media videos. The ensemble-based strategy consistently enhances primary model performance, making it a valuable tool for crime prevention and public safety."
Disclaimer: Curated by HT Syndication.
