Firearm detection using Yolov7

Shaik, Rizwana (2023) Firearm detection using Yolov7. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Firearms have been a constant issue and the biggest contributor to disrupting public safety worldwide. This is an important issue that cannot be overlooked. An autonomous visual gun detection model can help provide surveillance and monitoring to all public places. In earlier works, gun detection has always faced problems achieving the appropriate accuracy or speed in real time. A dependable gun detection model will allow a quicker response and propose safety measures. We look into different papers and their work for a robust gun detection model using Yolo algorithms. I utilized the Yolo algorithms with multiscale concatenation and prediction heads for our paper. We train and validate the Yolo variants on a curated gun image dataset acquired from various sources. The Yolo model for gun detection achieved 87% precision and 70% recall, making it a reliable and well-performing model for different images of firearms and their orientations. This detection model approaches the state-of-the-art in the targeted deep neural architectures for security applications. In a real-time scenario, the latest model for gun detection using Yolo enables automated surveillance and alert systems to detect firearm threats faster. The performance of this model is sufficient for the video and embedded application in CCTVs (Closed-circuit Television). The main challenges faced are the scenarios where illumination is not proper and partial visibility of the firearm makes it difficult for the model to detect the object. This has caused a few true negative and false positive scenarios.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Tomer, Vikas UNSPECIFIED
Uncontrolled Keywords:	Yolov8; Convolutional neural network; Real-time detection; Security systems; Computer Vision
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 Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence > Computer vision Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence > Computer vision
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	21 May 2025 11:41
Last Modified:	21 May 2025 11:41
URI:	https://norma.ncirl.ie/id/eprint/7605

Actions (login required)

View Item