Archive · 12 Aug 2025
Students Develop AI System to Predict Crowd Risks
A group of college students created an AI system to keep people safe in crowded places.
The system uses cameras and AI to count how many people are in an area.
If it gets too crowded, it sends alerts to security and the public, like a quick warning system.
The system runs without needing the internet and works on small computers, so it's fast and cheap.
People can also scan a code to see how crowded a place is.
This technology can help prevent accidents in places like malls, stadiums, and festivals, making crowded spaces safer for everyone.
Students developed an AI system to predict crowd risks in public spaces.
The system uses the YOLOv5s AI model and Raspberry Pi for real-time crowd detection.
It sends automated alerts via WhatsApp, Telegram, or SMS when overcrowding is detected.
A companion app provides real-time monitoring, heatmaps, and live video feeds.
The system is designed for use in smart cities and rural areas to improve safety.
- Who
- Developed by five students from Mohan Babu University.
- What
- An AI-based system for predicting and managing crowd risks.
- When
- The system is designed to operate in real-time.
- Where
- Malls, stadiums, transit hubs, and other crowded areas.
- Why
- To provide early warnings and improve safety in crowded public spaces.
Public Safety Focus
Adaptability and Inclusivity
Deployment
Public Safety Focus
Prioritizes real-time alerts and accessible information for quick responses.
Adaptability and Inclusivity
Emphasizes the importance of adapting the system to varied environments.
- Core Technology
- YOLOv5s AI model on Raspberry Pi 4
- Alerts
- WhatsApp, Telegram, SMS
- Public Access
- QR codes for crowd status
- Application Areas
- Malls, stadiums, railway stations
- Risk Classification
- Low, Medium, High