CMADS v1.0: Crowd Monitoring and Anomaly Detection System Using YOLOv8, ByteTrack, and Vision Transformer
Article content
Abstract
Managing crowds and noticing unusual behaviors are important for ensuring public safety at large events, particularly in areas with multiple gates where movement can be crowded and unpredictable. Traditional surveillance systems often fall short in accurately identifying individuals, gauging crowd density, or detecting abnormal activity. To tackle these issues, we propose an AI-based automated crowd monitoring and anomaly detection system, CMADS v1.0, that integrates YOLOv8 for people detection, ByteTrack for robust tracking, and Vision Transformers (ViT) for gender classification. This system can estimate the number of people present, including gender distribution, and detect unusual behaviors such as loitering, sudden surges, or slow movement—all of which are useful for monitoring gate zones at large events. The system achieves 13-15 frames per second and 89% mean Average Precision (mAP) in people detection, with 98.7% accuracy in gender classification. It also successfully identifies loitering behavior with 87% accuracy, abnormal speeds with 90% accuracy, and crowd surges with 92% accuracy. These preliminary results are promising, demonstrating that the system can perform the tasks effectively in real time. The system aims to support security teams by providing detailed information on crowd movements and possible hazards, finally enhancing crowd control efforts.
Keywords
Anomaly detection, ByteTrack, crowd behavior analysis, gender classification, multi-person tracking, vision transformer (ViT), you only look once version 8 (YOLOv8)
Downloads
References
N. Kwak and B. Lee, “Deep learning applied abnormal human behavior detection in video surveillance systems – A survey,” Int. J. Contents, vol. 20, no. 4, pp. 84–95, 2024.
S. Abba, A. M. Bizi, J. A. Lee, S. Bakouri, and M. L. Crespo, “Real-time object detection, tracking, and monitoring framework for security surveillance systems,” Heliyon, vol. 10, no. 15, 2024.
Ultralytics, “YOLOv8,” GitHub repository, 2023. [Online]. Available: https://github.com/ultralytics/ultralytics/blob/main/docs/en/models/yolov8.md
I. Zhang, “ByteTrack: Multi-object tracking by associating every detection box,” GitHub repository, 2022. [Online]. Available: https://github.com/ifzhang/ByteTrack
Google, “Vision transformer (ViT) model,” Hugging Face, 2023. [Online]. Available: https://huggingface.co/google/vit-base-patch16-224-in21k
D. Sudharson et al., “Proactive Headcount and Suspicious Activity Detection using YOLOv8,” Procedia Comput. Sci., vol. 230, pp. 61–69, 2023.
A. Vijayakumar and S. Vairavasundaram, “YOLO-based object detection models: A review and its applications,” Multimedia Tools Appl., pp. 1–40, 2024.
X. Cheng et al., “Ship imaging trajectory extraction via an aggregated YOLO model,” Eng. Appl. Artif. Intell., vol. 130, p. 107742, 2024.
D. Ferreira and M. Basiri, “Dynamic Target Tracking and Following with UAVs Using Multi-Target Information: Leveraging YOLOv8 and MOT Algorithms,” Drones, vol. 8, p. 488, 2024.
J. Chen, “Crowd Counting and People Density Detection: An Overview,” in Proc. 3rd Int. Conf. Eng. Manage. Inf. Sci. (EMIS 2024), Springer Nature, 2024, p. 434.
P. Wu et al., “Weakly supervised video anomaly detection and localization with spatio-temporal prompts,” in Proc. 32nd ACM Int. Conf. Multimedia, Oct. 2024, pp. 9301–9310.
R. Nasir, Z. Jalil, M. Nasir, T. Alsubait, M. Ashraf, and S. Saleem, “An Enhanced Framework for Real-Time Dense Crowd Abnormal Behavior Detection Using YOLOv8,” Artificial Intelligence Review, Vol. 58, No. 202, 2025.
Mokayed, Hamam, et al. "Real-time human detection and counting system using deep learning computer vision techniques." Artificial Intelligence and Applications. Vol. 1. No. 4. 2023.
R. Padilla et al., “A comparative analysis of object detection metrics with a companion open-source toolkit,” Electronics, vol. 10, no. 3, p. 279, 2021. doi: 10.3390/electronics10030279.
S. K. Jayasingh et al., “Integrated Crowd Counting System Utilizing IoT Sensors, OpenCV and YOLO Models for Accurate People Density Estimation in Real-Time Environments,” in Proc. 1st Int. Conf. Cognit., Green and Ubiquitous Comput. (IC-CGU), IEEE, Mar. 2024, pp. 1–6.
Y. Liu, L. Kennedy, H. Amiri, and A. Züfle, “Neural collaborative filtering to detect anomalies in human semantic trajectories,” in Proc. 1st ACM SIGSPATIAL Int. Workshop Geospatial Anomaly Detection (GeoAnomalies’24), Atlanta, GA, USA, 2024.
P. Nguyen et al., “Multi-camera multi-object tracking on the move via single-stage global association approach,” Pattern Recognition, vol. 152, p. 110457, 2024.
G. A. Noghre, A. D. Pazho, and H. Tabkhi, “An exploratory study on human-centric video anomaly detection through variational autoencoders and trajectory prediction,” in Proc. IEEE/CVF Winter Conf. Appl. Comput. Vision, 2024, pp. 995–1004.
E. Solano-Carrillo et al., “UTrack: Multi-Object Tracking with Uncertain Detections,” arXiv preprint, arXiv:2408.17098, 2024.
J. Cao et al., “Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking,” arXiv preprint, 2023.
J. Fei and M. Han, “Real-time object detection methods: An evaluation of CMCA-YOLO for surveillance,” Electronics, vol. 12, no. 4, pp. 222–238, 2023.
H. Xie, Z. Xiao, W. Liu, and Z. Ye, “PVNet: A Used Vehicle Pedestrian Detection, Tracking, and Counting Method,” Sustainability, vol. 15, pg. 14326, 2023.
Article details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
You must read and accept the copyright terms and conditions(click here for copyright terms).