Citation

Rajah, Daveena Sri Michael and Ramly, Athirah Mohd and Khamayseh, Yahya and Chen, Charis Kwan Shwu and Amphawan, Angela and Neo, Tse Kian (2025) CCTV armed robbery detection with YOLOv8. In: 4th International Conference on Computer, Information Technology and Intelligent Computing, CITIC 2024, 23 July 2024 - 25 July 2024, Virtual, Online.

Text c-6.pdf - Published Version Restricted to Repository staff only Download (204kB)

Abstract

Crimes such as armed robbery are prevalent in large cities. Inevitably, there are armed robbery cases in which victims do not achieve justice or a rightful verdict against culprits due to CCTV evidence destruction or modification, deeming the CCTV evidence invalid in court. To address this, this paper proposes the usage of YOLOv8, a refined iteration of the You Only Look Once (YOLO) architecture designed for the specific task of real-time armed robbery detection in surveillance videos. Leveraging the efficiency of the YOLO algorithm, our model demonstrates heightened precision of 97.8%, recall of 94.3 and with a 99 FPS suitable for real time detection through training a custom armed robbery dataset. Notably, YOLOv8 incorporates an innovative evidence collection mechanism by storing timestamped screenshots of detected scenes in a secure database. This integration not only enhances post-incident investigations but also acts as a deterrent, contributing to the overall effectiveness of public safety and crime prevention efforts. Extensive benchmarking on diverse datasets underscores the model's proficiency in accurately identifying armed robbery instances while maintaining real-time processing capabilities.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Armed robbery detection, YOLOv8, real-time video surveillance, object detection
Subjects:	H Social Sciences > HV Social pathology. Social and public welfare. Criminology > HV7231-9960 Criminal justice administration > HV7551-8280.7 Police. Detectves. Constabulary > HV7935-8025 Administration and organization
Divisions:	Faculty of Creative Multimedia (FCM)
Depositing User:	Nor Afiqah Mohd Adnan
Date Deposited:	09 Dec 2025 05:07
Last Modified:	13 Dec 2025 01:54
URII:	http://shdl.mmu.edu.my/id/eprint/14989

Downloads

Edit (login required)