🛡️ CrowdShield - AI-Driven Crowd Management System

Unified AI-Powered Crowd & Disaster Management for Nashik Kumbh Mela 2027

🎯 Overview

CrowdShield is a comprehensive AI-driven crowd management system designed specifically for mega gatherings like the Nashik Kumbh Mela 2027. The system combines the power of Flask backend with FastAPI features and Satellite Analysis to provide real-time crowd monitoring, intelligent alerts, and emergency response coordination.

🆕 New: Satellite Crowd Analysis Module

The satellite_analysis/ module provides advanced AI-powered analysis of satellite imagery for crowd density estimation and prediction. This module is specifically designed for large-scale events like Kumbh Mela and can be used independently or integrated with the main CrowdShield system.

Key Features:

🛰️ Multi-Method Analysis: Edge density, texture analysis, blob detection, and hybrid approaches
🤖 Machine Learning: Random Forest-based crowd count prediction
📊 Data Management: JSON-based structured data storage
🎯 Event Prediction: Predict crowd counts for recurring events
📈 Visualization: Comprehensive analysis and prediction visualizations

🚀 Quick Start

Prerequisites

Python 3.8+
Git
Modern web browser

Installation

Clone the repository

git clone https://github.com/NMOLE08/CrowdShield.git
cd CrowdShield

Install dependencies

pip3 install -r requirements.txt
pip3 install -r satellite_analysis/requirements.txt

Start the unified backend
```
python3 main.py
```
Access the system
- 🌐 Frontend Dashboard: http://localhost:5002/frontend
- 📋 API Documentation: http://localhost:5002/api/docs
- 🏥 Health Check: http://localhost:5002/health

Demo Mode

python3 scripts/run_demo.py

🤖 AI Models & Technologies

Computer Vision & AI

YOLOv8 (Ultralytics): Real-time person detection and crowd counting
OpenCV: Video processing and frame analysis
PyTorch: Deep learning framework for model inference
NumPy: Numerical computing for data processing
Pillow: Image processing and manipulation

Machine Learning & Analytics

Scikit-learn: Machine learning algorithms for predictive analytics
Pandas: Data manipulation and analysis
SciPy: Scientific computing and statistical analysis

Real-time Communication

Flask-SocketIO: WebSocket communication for real-time updates
Python-SocketIO: Real-time bidirectional communication
Paho MQTT: IoT device communication protocol

🌐 APIs & External Services

Mapping & Geospatial

OpenStreetMap (OSM) API: Free location services, geocoding, route optimization
OSRM: Open Source Routing Machine for directions
Nominatim: Free geocoding service
Mapbox: Interactive maps and heatmap visualization
Folium: Python mapping library for data visualization
Geopy: Geocoding and distance calculations

Communication & Notifications

Twilio: SMS and WhatsApp alert delivery
WebSocket: Real-time data streaming to frontend

Database & Storage

SQLite: Lightweight database for data persistence
SQLAlchemy: Database ORM and management
Flask-SQLAlchemy: Flask database integration

🏗️ System Architecture

Backend Components

┌─────────────────────────────────────────────────────────────┐
│                    CrowdShield Unified Backend              │
├─────────────────────────────────────────────────────────────┤
│  Flask Core (AI & Detection)    │  FastAPI Features         │
│  ├── YOLOv8 Detection          │  ├── WebSocket Server      │
│  ├── CCTV Management           │  ├── Twilio Integration    │
│  ├── Nashik Simulation         │  ├── Real-time Updates     │
│  ├── Route Optimization        │  └── Push Notifications    │
│  ├── Satellite Analysis        │                            │
│  └── Database Management       │                            │
└─────────────────────────────────────────────────────────────┘

Core Modules

crowd_detection.py: YOLOv8-based people detection
cctv_manager.py: Multi-stream video processing
enhanced_alert_system.py: Master branch's threshold engine
mapbox_heatmap_generator.py: Interactive map visualization
nashik_kumbh_simulator.py: Venue-specific simulation
route_optimizer.py: Crowd-aware navigation
database.py: Data persistence and management
satellite_crowd_analyzer.py: Satellite image analysis
enhanced_crowd_predictor.py: ML-based crowd prediction

📡 API Endpoints

Core Detection

POST /api/detect - Real-time crowd detection
GET /api/heatmap - Generate crowd density heatmap
GET /api/alerts - Get active alerts
POST /api/routes/optimize - Optimize routes based on crowd

Nashik Kumbh Mela

POST /api/nashik/simulate - Generate simulation scenarios
GET /api/nashik/status - Get venue status
GET /api/nashik/map-data - Get interactive map data
GET /api/nashik/realtime-map - Real-time map updates

Satellite Analysis

POST /api/satellite/analyze - Analyze satellite imagery
GET /api/satellite/predict - Get crowd predictions
GET /api/satellite/status - Satellite analysis status
GET /api/satellite/data - Historical analysis data

Alert System

GET /api/alerts - Get current active alerts
GET /api/alerts/status - Get alert system status
GET /api/alerts/history - Get alert history
POST /api/alerts/test - Test alert system with Twilio

🎪 Nashik Kumbh Mela Features

Venue Integration

5 Strategic Gates: Ramkund, Godavari Ghat, Main Bridge, Temporary Bridge, Parking Area
Real Coordinates: Accurate Nashik location mapping
Capacity Management: Per-gate crowd limits and monitoring
Emergency Routes: Optimized evacuation paths

Simulation Scenarios

Normal Crowd Flow: Regular festival attendance
Peak Hour Rush: High-density periods
Emergency Situations: Stampede, fire, medical emergencies
Weather Conditions: Rain, heat impact on crowd behavior

🚨 Alert System

Threshold Engine

Density-Based Alerts: GREEN, YELLOW, RED, CRITICAL levels
Smart Cooldown: 5-minute cooldown between repeated alerts
Hourly Limits: Maximum 10 alerts per hour per location
Real-time Processing: Automatic alert generation from CCTV data

Alert Features

Multi-level Alerts: Based on crowd density and capacity
Location-Specific: Gate-specific alerting for Nashik Kumbh Mela
WebSocket Notifications: Real-time alert delivery
Twilio Integration: SMS/WhatsApp alert delivery
Alert History: Complete tracking with timestamps

🗺️ Map Integration

Mapbox Features

Interactive Heatmaps: Color-coded crowd density visualization
Real-time Updates: Live crowd data overlay
Gate Markers: All 5 Nashik Kumbh Mela gates with popups
Landmark Integration: Godavari River, Ramkund, Main Ghat
Alert Visualization: Real-time alert markers on map

OpenStreetMap Integration

Route Optimization: Crowd-aware navigation using OSRM
Geocoding: Free location services using Nominatim
Distance Matrix: Real-time distance calculations
Place Details: Venue information and details

📹 CCTV & Video Processing

Multi-Stream Processing

5 CCTV Feeds: Real-time video streams from all gates
YOLOv8 Detection: Person detection with bounding boxes
Frame Generation: Simulated video frames with overlays
Performance Metrics: Real-time processing statistics

Video Features

Detection Overlays: Bounding boxes, labels, timestamps
Crowd Counting: Real-time person count per feed
Status Indicators: Visual status for each camera
Frame Caching: Optimized frame delivery

🎨 Frontend Dashboard

Modern UI Features

Glass-morphism Design: Modern glass-like interface
Responsive Layout: Works on all devices
Real-time Updates: Live data without page refresh
Interactive Maps: Mapbox integration with heatmaps
CCTV Thumbnails: Live camera feed previews

Dashboard Components

Live Statistics: Real-time crowd count and alerts
Interactive Map: Nashik Kumbh Mela venue visualization
Alert Panel: Active alerts with priority levels
CCTV Feeds: Clickable camera thumbnails
System Status: Health monitoring and status indicators

🔧 Configuration

Environment Variables

# Application
SECRET_KEY=your_secret_key
FLASK_DEBUG=True
PORT=5002

# OSM API (OpenStreetMap - Free, no API key required)
OSM_ENABLED=True

# Mapbox
MAPBOX_ACCESS_TOKEN=your_mapbox_access_token

# Twilio (Push Notifications)
TWILIO_ACCOUNT_SID=your_twilio_account_sid
TWILIO_AUTH_TOKEN=your_twilio_auth_token
TWILIO_PHONE_NUMBER=your_twilio_phone_number

# MQTT (optional)
MQTT_ENABLED=False
MQTT_BROKER=localhost
MQTT_PORT=1883

🧪 Testing

Test Scripts

test_enhanced_alerts.py: Enhanced alert system validation
test_cctv.py: CCTV functionality testing
test_api_endpoints.py: API endpoint verification
run_demo.py: Comprehensive system demo
test_setup.py: Satellite analysis environment testing

Test Coverage

Alert System: Threshold engine and cooldown logic
CCTV Integration: Multi-stream processing
Map Integration: Mapbox and Google Maps APIs
API Endpoints: All REST endpoints and WebSocket
Real-time Updates: Live data streaming
Satellite Analysis: ML model validation and testing

📁 Project Structure

CrowdShield/
├── src/
│   ├── api/           # Flask API layer
│   ├── core/          # Core AI & detection modules
│   └── config/        # Configuration management
├── frontend/          # Modern dashboard UI
├── backend/           # FastAPI features (integrated)
├── satellite_analysis/ # Satellite image analysis module
├── data/              # Generated data & models
├── scripts/           # Demo & utility scripts
├── tests/             # Test suites
└── docs/              # Documentation

🎯 Use Cases

For Authorities

Real-time Monitoring: Live crowd density tracking
Emergency Response: Instant alert system
Route Management: Optimized traffic flow
Resource Allocation: Data-driven decision making
Satellite Analysis: Large-scale crowd monitoring

For Event Organizers

Capacity Planning: Predictive crowd modeling
Safety Management: Proactive risk mitigation
Communication: Automated alert distribution
Analytics: Historical data analysis
Predictive Insights: ML-based crowd forecasting

🤝 Contributing

Fork the repository
Create a feature branch (git checkout -b feature/amazing-feature)
Commit your changes (git commit -m 'Add amazing feature')
Push to the branch (git push origin feature/amazing-feature)
Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

Ultralytics: YOLOv8 implementation
Mapbox: Interactive mapping
Twilio: Communication services
OpenCV: Computer vision library
Scikit-learn: Machine learning framework
Nashik Kumbh Mela Authorities: Venue information and requirements

📞 Support

For support and questions:

📧 Email: support@crowdshield.ai
🐛 Issues: GitHub Issues
📖 Documentation: Wiki

Built with ❤️ for Nashik Kumbh Mela 2027

Name		Name	Last commit message	Last commit date
Latest commit History 25 Commits
backend		backend
cctv_feeds		cctv_feeds
data		data
docs		docs
examples		examples
frontend		frontend
nashik_simulation		nashik_simulation
perplexcity reports		perplexcity reports
satellite_analysis		satellite_analysis
scripts		scripts
src		src
tests		tests
trained_models		trained_models
virat_dataset		virat_dataset
.DS_Store		.DS_Store
.gitignore		.gitignore
BACKEND_INTEGRATION_SUMMARY.md		BACKEND_INTEGRATION_SUMMARY.md
CROWDSHIELD_COMPLETION_SUMMARY.md		CROWDSHIELD_COMPLETION_SUMMARY.md
README.md		README.md
RESTRUCTURE_SUMMARY.md		RESTRUCTURE_SUMMARY.md
h -u origin master		h -u origin master
load_sample_videos.py		load_sample_videos.py
main.py		main.py
requirements.txt		requirements.txt
run.py		run.py
start.bat		start.bat
start.sh		start.sh
start_server.py		start_server.py
test_alert_functionality.html		test_alert_functionality.html
yolov8n.pt		yolov8n.pt

NMOLE08/CrowdShield

Folders and files

Latest commit

History

Repository files navigation