@article{ART003329485},
author={Hoi-Min Park and Seong-Hyun Park},
title={A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2026},
volume={31},
number={4},
pages={129-135}
TY - JOUR
AU - Hoi-Min Park
AU - Seong-Hyun Park
TI - A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments
JO - Journal of The Korea Society of Computer and Information
PY - 2026
VL - 31
IS - 4
PB - The Korean Society Of Computer And Information
SP - 129
EP - 135
SN - 1598-849X
AB - In the modern society, CCTV plays an essential role in crime prevention and safety management, but the existing abnormal behavior detection method has limitations in its application in various environments as it often relies on a method of defining and classifying specific behavior types in advance. In particular, since the form of abnormal behavior is not fixed in the actual environment and can appear in various forms depending on lighting, camera angle, and user behavior pattern, a more generalized detection method is required. Against this background, this study proposed a method of detecting abnormal symptoms based on changes in normal behavior patterns. The proposed method detects human objects using YOLOv5s and extracts behavioral characteristics by accumulating movement information based on center coordinates in time window units.
Compared with the pattern formed in the normal behavior section, the section in which a certain level of change or higher occurred was judged as an abnormal symptom. As a result of the experiment, the characteristic value remained stable in the normal section, while the variability tended to increase in the behavior change section. In addition, the F1-score improved from 0.835 to 0.908 compared to the frame unit method, confirming about 7%p performance improvement, and false detection due to temporary noise also decreased.
KW - Anomaly Detection;Normal Behavior Pattern;Object Detection;YOLO;Time Window Analysis
DO -
UR -
ER -
Hoi-Min Park and Seong-Hyun Park. (2026). A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments. Journal of The Korea Society of Computer and Information, 31(4), 129-135.
Hoi-Min Park and Seong-Hyun Park. 2026, "A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments", Journal of The Korea Society of Computer and Information, vol.31, no.4 pp.129-135.
Hoi-Min Park, Seong-Hyun Park "A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments" Journal of The Korea Society of Computer and Information 31.4 pp.129-135 (2026) : 129.
Hoi-Min Park, Seong-Hyun Park. A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments. 2026; 31(4), 129-135.
Hoi-Min Park and Seong-Hyun Park. "A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments" Journal of The Korea Society of Computer and Information 31, no.4 (2026) : 129-135.
Hoi-Min Park; Seong-Hyun Park. A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments. Journal of The Korea Society of Computer and Information, 31(4), 129-135.
Hoi-Min Park; Seong-Hyun Park. A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments. Journal of The Korea Society of Computer and Information. 2026; 31(4) 129-135.
Hoi-Min Park, Seong-Hyun Park. A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments. 2026; 31(4), 129-135.
Hoi-Min Park and Seong-Hyun Park. "A Study on Abnormal Sign Detection Based on Normal Behavior Pattern Changes in Indoor CCTV Environments" Journal of The Korea Society of Computer and Information 31, no.4 (2026) : 129-135.
