Automatic Weapon Detection in CCTV systems Using Deep Learning

Jadhav, Rohit Subhash (2022) Automatic Weapon Detection in CCTV systems Using Deep Learning. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (9MB) | Preview
Preview	PDF (Configuration manual) Download (21MB) | Preview

Abstract

Security surveillance is becoming highly crucial as the frequency of school shootings, armed robberies, and terrorist operations are increasing constantly. By catching these activities early on, there will be a lot less collateral harm, which will lower the number of offenses. This work introduces real-time weapon identification approach using surveillance systems based on computer-vision. For the purpose of finding the weapons and confusion objects in this study, different YOLO approaches are used. The study concentrates on Scaled-YOLOv4 and YOLOv4 to efficiently differentiate between confusion objects and real weaponry. With mean average accuracy (mAP@0.50) of 86.19%, Precision of 79%, and F1-score of 77%, YOLOv4 performs moderately better than Scaled YOLOv4. YOLOv4 also shows superior confidence score in pictures as well as video footage.

Item Type:	Thesis (Masters)
Subjects:	H Social Sciences > HV Social pathology. Social and public welfare > Criminology > Crimes and Offences Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > TR Photography Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Tamara Malone
Date Deposited:	26 Jan 2023 17:13
Last Modified:	03 Mar 2023 11:10
URI:	https://norma.ncirl.ie/id/eprint/6141

Actions (login required)

View Item