Skip to content

README.md

Latest commit

 

History

History
64 lines (47 loc) · 1.85 KB

README.md

File metadata and controls

64 lines (47 loc) · 1.85 KB

STAMPede Detection System

🚀 Quick Start

# Install dependencies
pip install -r requirements.txt

# Start the system
python start_enhanced_system.py

📋 System Requirements

  • Python 3.8+
  • NVIDIA GPU with CUDA support (recommended)
  • 8GB+ RAM
  • Webcam or video files for testing

🎯 Features

  • Real-time Detection: YOLOv8 Large model with GPU acceleration
  • Professional Interface: Clean web dashboard with live metrics
  • Advanced Analytics: Multi-factor risk assessment and crowd flow analysis
  • Simple Visualization: Clear dots instead of cluttered bounding boxes
  • Smart Alerts: Conservative thresholds prevent false alarms

📁 Project Structure

person-detection/
├── web_server.py              # Main web application
├── stampede.py                # Core detection algorithm  
├── start_enhanced_system.py   # System startup script
├── train.py                   # Model training script
├── templates/
│   └── index.html            # Web interface
├── requirements.txt           # Dependencies
└── FINAL_PROJECT_DOCUMENTATION.md  # Complete documentation

🔧 Configuration

The system automatically detects and uses GPU acceleration if available. Key parameters:

  • Confidence: 0.15 (optimized for dense crowds)
  • Image Size: 1280px (high resolution)
  • Grid Resolution: 32x24 (fine analysis)
  • Risk Thresholds: Conservative to prevent false alarms

📖 Documentation

See FINAL_PROJECT_DOCUMENTATION.md for complete technical details, academic Q&A, and implementation specifics.

🎓 Academic Use

This project demonstrates:

  • Computer vision applications
  • Real-time processing systems
  • GPU acceleration techniques
  • Web application development
  • Risk assessment algorithms

Perfect for computer science, engineering, and AI/ML courses.