Enhanced Weapon Detection and Classification Using YOLO-Based Object Detection Frameworks and Emergency Alert System
SRIVATHSAN P
, SHAWN KUMAR B , SARVESH S S , Sachin Aditiya B
using the YOLOV5 object detection framework, this work proposes a real-time, AI-
powered system intended for intelligent surveillance and public safety by identification
and classification of firearms. Rising frequency of armed threats in public areas calls
for quick, highly accurate threat detection systems automatically. From both still photos
and live video streams, the suggested system detects many kinds of weapons including
rifles, handguns, knives, and other dangerous objects using YOLOV5. Nine separate
weapon categories were created from bespoke data augmentation methods including
rotation, scale, and brightness modification. Implementing Python, OpenCV, and the
Ultralytics interface for YOLOV5, the system achieves real-time performance at 30 FPS
with a mean Average Precision (mAP@0.5) of 0.84. F1-score analysis, precision-recall
curves, and confusion matrix evaluation show great short latency detection power. Built
to be lightweight, modular, and scalable, this solution may be installed on mobile
security systems, smart CCTV cameras, and edge devices. The technology offers major
advances in proactive security measures for important areas such airports, schools,
transportation hubs, and public events by merging modern deep learning with effective
surveillance systems.
"Enhanced Weapon Detection and Classification Using YOLO-Based Object Detection Frameworks and Emergency Alert System", TIJER - TIJER - INTERNATIONAL RESEARCH JOURNAL (www.TIJER.org), ISSN:2349-9249, Vol.12, Issue 4, page no.b132-b140, April-2025, Available :https://tijer.org/TIJER/papers/TIJER2504159.pdf
Volume 12
Issue 4,
April-2025
Pages : b132-b140
Paper Reg. ID: TIJER_157663
Published Paper Id: TIJER2504159
Downloads: 000298
Research Area: Science and Technology
Country: Chennai, TamilNadu, India
ISSN: 2349-9249 | IMPACT FACTOR: 8.57 Calculated By Google Scholar | ESTD YEAR: 2014
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 8.57 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: TIJER (IJ Publication) Janvi Wave
