A Review of Visible and Concealed Weapon Detection Techniques Using Machine Learning and Deep Learning

The Gun violence facing the international society is an epidemic to which the world experts have been striving to find a solution to, since a long time. Gun violence endangers the fundamental human right to live. It affects all aspects of modern society. The availability and easy access to the weapo...

Full description

Saved in:

Bibliographic Details
Published in:	SN computer science Vol. 6; no. 6; p. 611
Main Authors:	Sandhu, Simratjit kaur, Bhatia, Rekha
Format:	Journal Article
Language:	English
Published:	Singapore Springer Nature Singapore 01.08.2025 Springer Nature B.V
Subjects:	Accuracy Algorithms Anomalies Artificial neural networks Automation Cameras Closed circuit television Computer Imaging Computer Science Computer Systems Organization and Communication Networks Computer vision Concealed weapons detection Crime Data Structures and Information Theory Datasets Deep learning Eigenvalues Gun violence Guns Information Systems and Communication Service Machine learning Mass murders Pattern Recognition and Graphics Public safety Review Article Sensors Software Engineering/Programming and Operating Systems Surveillance Synthetic data Violence Vision Weapons United States > US Optimized weapon detection Concealed weapon detection Visible weapon detection Faster R-CNN detection Weapon detection challenges Gun detection Weapon detection review YOLO detection Edge device detection Pistol detection CCTV surveillance SSD weapon detection Object detection Weapon datasets
ISSN:	2661-8907, 2662-995X, 2661-8907
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The Gun violence facing the international society is an epidemic to which the world experts have been striving to find a solution to, since a long time. Gun violence endangers the fundamental human right to live. It affects all aspects of modern society. The availability and easy access to the weapons these days appears to be the most contributing factor to atrocities committed with them. Numerous researches suggest that the mere awareness of access to a weapon increases the possibility that a crime is most likely to occur. The most dangerous crimes have been committed in countries with loose laws on the possession and ownership of guns. Governments and security agencies around the world have been working tirelessly to come up with efficient mechanisms to counter the ever-increasing threat of guns. Advancements in computer vision technology appear to be the silver lining that can guide the efforts towards making the world a safer place. After numerous trials and errors over the years, the researchers have zeroed in on the automatic weapon detection in public as well as private spaces to be the best solution to combat the increasing incidents of gun violence. CCTV cameras have long been used as monitoring devices, but with technological advancements they can be transformed into intelligent alerting devices that can sound alarms even in the absence of a human operator, in case of an anomaly detection. Autonomous detection through CCTV surveillance has replaced walk-through systems, since automatic systems are more proactive, less invasive, and less obvious to perpetrators of crime. As far as detection is concerned, Faster R-CNN and YOLO algorithms seem to be the preferred choices. From the research, it has been demonstrated that the techniques of pre-processing, augmentation, use of realistic and synthetic datasets improve the performance of detection. To help researchers in the field of automatic weapon detection, an extensive review has been done with the aim of covering the developments that have been done in the said field along with future directions and challenges associated with it. This review covers various machine learning, deep learning algorithms used for weapon detection, as well as the various datasets and the multiple challenges associated with the field.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	2661-8907 2662-995X 2661-8907
DOI:	10.1007/s42979-025-04173-0