Predicting Human Stampedes in Public Events

Human stampedes in public events pose significant safety risks, and early prediction is crucial for preventing accidents and ensuring crowd safety. This project aims to predict human stampedes by leveraging computer vision techniques to analyze crowd density and human counting in real time. The system utilizes the COCO dataset, a well-established collection of images with labeled human figures, to train a YOLOv9 (You Only Look Once version 9) model for crowd monitoring. The primary objective is to predict stampede risks when crowd density exceeds a threshold of 20 individuals, signaling a potential hazard. The front-end interface is developed using HTML, CSS, and JavaScript for user interaction, while the back-end is implemented using Python and Flask for model deployment and prediction. By integrating advanced computer vision techniques, this project provides a scalable and effective solution for crowd management, offering real-time predictions to enhance public safety during large-scale events. The findings aim to improve crowd control measures, reduce the risk of stampedes, and contribute to better event management strategies.

Keywords:

Human Stampede Prediction, YOLOv9, Computer Vision, Crowd Density Analysis, Real-Time Monitoring, Public Safety, Flask, Python, Crowd Management, Event Safety, Machine Learning.