International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | AI-Based Crowd Management System |
|---|---|
| ABSTRACT | Managing large crowds in public environments has become a major obstacle in modern society due to increasing population density and the frequent organization of large-scale events such as festivals, concerts, sports matches, religious gatherings, transportation hubs, and public rallies. Ensuring safety, preventing overcrowding, and avoiding dangerous situations like stampedes or panic incidents require continuous monitoring and quick decision-making. Conventional manual surveillance techniques are often limited by human error, delayed response, and difficulty in handling multiple crowded areas simultaneously. To overcome these restrictions, the proposed AI-Based Crowd Management The system makes use of innovative technology like machine learning and artificial intelligence, as well as computer vision. The system's purpose is to analyse real-time video feeds collected from surveillance cameras placed in crowded locations. It processes visual data to estimate crowd density, track movement patterns, and identify unusual or dubious behaviour. Techniques such as object identification and motion analysis are used to understand how individuals and groups behave within a monitored area. The system continuously evaluates crowd conditions and detects early warning signs such as sudden crowd buildup, unusual movement flow, or panic-like behaviour. When any risky situation is identified, the system generates instant alerts and sends notifications to authorities or security personnel. This enables faster emergency response and helps in taking preventive actions before the situation becomes critical. In addition, predictive analytics can be used to forecast crowd behaviour based on historical and real-time data, supporting better planning and resource allocation during events. |
| AUTHOR | SAHANA G, MAHESHWARI M DESAI PG Student, Dept. of MCA, City Engineering College, Bengaluru, Karnataka, India Assistant Professor, Dept. of MCA, City Engineering College, Bengaluru, Karnataka, India |
| VOLUME | 184 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1405023 |
| pdf/23_AI-Based Crowd Management System.pdf | |
| KEYWORDS | |
| References | [1] J. Redmon and A. Farhadi, “YOLOv8: Real-Time Object Detection and Tracking Improvements,” Ultralytics Technical Report, 2023. [2] S. Li, Y. Wang, and H. Chen, “Deep learning-based crowd counting and density estimation: A review,” IEEE Access, vol. 10, pp. 115000–115020, 2022. [3] M. Zhao and K. Singh, “Vision Transformer approaches for crowd analysis and surveillance systems,” arXiv preprint arXiv: 2208.XXXX, 2022. [4] R. Gupta and P. Sharma, “Real-time crowd behavior detection using deep neural networks,” International Journal of Image and Graphics, vol. 24, no. 3, 2023. [5] L. Zhang, X. Liu, and J. Wang, “An AI-based intelligent surveillance system for public crowd safety,” IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 2, pp. 980–992, 2024. [6] A. Patel and S. Mehta, “Crowd anomaly detection using computer vision and machine learning techniques,” Procedia Computer Science, vol. 218, pp. 1450–1457, 2023. [7] H. Kim et al., “Smart surveillance framework for crowd management in smart cities,” IEEE Smart Cities Conference, 2025. |
