@article{ART003176659},
author={Hoi-Min Park and Jae-Woong Kim and Joon-Yong Kim},
title={A real-time abnormal behavior detection model using Yolov5 is proposed},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2025},
volume={30},
number={2},
pages={91-98}
TY - JOUR
AU - Hoi-Min Park
AU - Jae-Woong Kim
AU - Joon-Yong Kim
TI - A real-time abnormal behavior detection model using Yolov5 is proposed
JO - Journal of The Korea Society of Computer and Information
PY - 2025
VL - 30
IS - 2
PB - The Korean Society Of Computer And Information
SP - 91
EP - 98
SN - 1598-849X
AB - In this paper, a system that provides fast and accurate object detection performance and detects abnormal behavior occurring in an indoor environment in real time using the YOLOv5 model, which is a model suitable for detecting abnormal behavior in real time, was proposed and implemented. This system receives real-time images from CCTV cameras or video files, and detects abnormal behavior by inputting each frame into the YOLOv5 model. The detected abnormal behavior is visualized with a bounding box and label and recorded in a log file together with the occurrence time. As a result of the experiment, the proposed system was able to detect abnormal behavior with a high accuracy of 93% in various indoor environments, and the performance was improved by reducing the IOU loss by about 0.12. Abnormal behavior detection is important in security and safety management. For example, in residential areas or hospitals, it is possible to detect and respond to intrusion or accidents early, thereby reducing human damage and enhancing safety. Therefore, the technology that detects abnormal behavior quickly and accurately in an indoor environment plays an important role in building a safe environment. This paper proposed the expansion of datasets and the improvement of model performance in the future.
KW - YOLOv5;Dataset;Deep learning;Object Detection
DO -
UR -
ER -
Hoi-Min Park, Jae-Woong Kim and Joon-Yong Kim. (2025). A real-time abnormal behavior detection model using Yolov5 is proposed. Journal of The Korea Society of Computer and Information, 30(2), 91-98.
Hoi-Min Park, Jae-Woong Kim and Joon-Yong Kim. 2025, "A real-time abnormal behavior detection model using Yolov5 is proposed", Journal of The Korea Society of Computer and Information, vol.30, no.2 pp.91-98.
Hoi-Min Park, Jae-Woong Kim, Joon-Yong Kim "A real-time abnormal behavior detection model using Yolov5 is proposed" Journal of The Korea Society of Computer and Information 30.2 pp.91-98 (2025) : 91.
Hoi-Min Park, Jae-Woong Kim, Joon-Yong Kim. A real-time abnormal behavior detection model using Yolov5 is proposed. 2025; 30(2), 91-98.
Hoi-Min Park, Jae-Woong Kim and Joon-Yong Kim. "A real-time abnormal behavior detection model using Yolov5 is proposed" Journal of The Korea Society of Computer and Information 30, no.2 (2025) : 91-98.
Hoi-Min Park; Jae-Woong Kim; Joon-Yong Kim. A real-time abnormal behavior detection model using Yolov5 is proposed. Journal of The Korea Society of Computer and Information, 30(2), 91-98.
Hoi-Min Park; Jae-Woong Kim; Joon-Yong Kim. A real-time abnormal behavior detection model using Yolov5 is proposed. Journal of The Korea Society of Computer and Information. 2025; 30(2) 91-98.
Hoi-Min Park, Jae-Woong Kim, Joon-Yong Kim. A real-time abnormal behavior detection model using Yolov5 is proposed. 2025; 30(2), 91-98.
Hoi-Min Park, Jae-Woong Kim and Joon-Yong Kim. "A real-time abnormal behavior detection model using Yolov5 is proposed" Journal of The Korea Society of Computer and Information 30, no.2 (2025) : 91-98.
