UrbanEye

Civic Infrastructure Intelligence Platform

Live Demo · API Docs · Report Bug

---

What is UrbanEye?

UrbanEye is a full-stack civic issue management system that connects citizens, field officers, and government administrators. Citizens report infrastructure problems (potholes, garbage, broken streetlights) via a mobile app or web interface. The system uses Google Gemini to analyze uploaded images, auto-categorize issues, and predict potential infrastructure failures based on weather data and historical patterns.

Core workflow:

1. Citizen uploads photo of issue → Gemini analyzes and categorizes it
2. Report is geo-tagged and routed to the appropriate department
3. Field officers receive assignments and update status
4. Admins monitor via live heatmap and analytics dashboard

---

Project Structure

Urbaneye/
├── mobileappUrbanEye/     # Flutter mobile app (Android/iOS)
├── webapp-urbaneye/       # React web dashboard
└── UE_backend-main/       # Flask REST API

---

Features

Image Analysis (Gemini)

• Upload any image of a civic issue
• Gemini 2.5 Flash identifies: potholes, garbage, sewage, drainage, streetlight damage, traffic signals, illegal dumping, waterlogging
• Auto-assigns severity (low/medium/high) and department

Predictive Intelligence

• Correlates historical report data with Open-Meteo weather API
• Scrapes local news via DuckDuckGo for context (protests, construction, events)
• Outputs predictions with reasoning: "Heavy rain forecast + 3 drainage complaints in sector → Deploy suction truck"

Role-Based Access

Role	Capabilities
Civilian	Report issues, track status, view leaderboard
Field Officer	Receive tasks, update status, verify resolutions
Dept Head	Manage department-specific reports (Roads/Water/Waste)
Gov Admin	Full analytics, HRMS, AI predictions, heatmap
Super Admin	User management, system configuration
Gig Worker	On-demand task pickup
Social Worker	NGO collaboration on social issues

Voice Commands (Web Dashboard)

The admin dashboard supports Web Speech API:

• "Show heatmap" → Opens map view
• "AI predictions" → Triggers prediction engine
• "How many critical reports?" → Speaks count aloud
• "Status report" → AI voice summary of current state

HRMS Module

Built-in human resource management for government staff:

• Employee directory
• Attendance tracking
• Payroll management
• Recruitment pipeline

---

Tech Stack

Layer	Technology
Mobile	Flutter 3.7, Firebase Auth, Cloud Firestore
Web	React 18, Vite 5, Tailwind CSS, Framer Motion
Maps	Leaflet.js, React-Leaflet
Charts	Recharts
Backend	Flask 3.0, Flask-RESTx, SQLAlchemy
Database	PostgreSQL (production), SQLite (dev)
AI	Google Gemini 2.5 Flash, LangChain
Auth	JWT (web/API), Firebase Auth (mobile)

---

Installation

Prerequisites

• Python 3.10+
• Node.js 18+
• Flutter SDK 3.7+
• PostgreSQL 14+ (optional, SQLite works for dev)
• Google Gemini API key (get one here)

Backend

cd UE_backend-main

# Create virtual environment
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Additional packages for predictions
pip install langchain langchain-community langchain-google-genai duckduckgo-search

# Configure environment
cp sample.env .env

Edit .env:

GEMINI_API_KEY=your_key_here
DATABASE_URL=postgresql://user:pass@localhost:5432/urbaneye
JWT_SECRET_KEY=your_secret_key

# Initialize database
python -c "from app import app, db; app.app_context().push(); db.create_all()"

# Seed sample data (optional)
python seed_db.py

# Run server
python app.py

API runs at http://localhost:5000. Swagger docs at http://localhost:5000/docs/.

Web Dashboard

cd webapp-urbaneye

npm install
npm run dev

Runs at http://localhost:5173.

Mobile App

cd mobileappUrbanEye

flutter pub get

# Configure Firebase
# Add google-services.json (Android) and GoogleService-Info.plist (iOS)

flutter run

---

API Documentation

Base URL: /api/v1

Authentication

Endpoint	Method	Description
/auth/register	POST	Create new user
/auth/login	POST	Get JWT token

Reports

Endpoint	Method	Auth	Description
/reports	GET	Optional	List all reports
/reports	POST	Required	Create report
/reports/<id>	PUT	Required	Update status
/reports/leaderboard	GET	-	Top reporters

Detection

Endpoint	Method	Description
/detection/analyze	POST	Upload image for Gemini analysis

Government Admin

Endpoint	Method	Auth	Description
/gov/predictions	GET	gov_admin	AI infrastructure predictions
/gov/analytics	GET	gov_admin	Dashboard statistics
/gov/team	GET	gov_admin	Team members

HRMS

Endpoint	Method	Description
/hr/candidates	GET	Recruitment pipeline
/hr/attendance	GET	Attendance records
/hr/payroll	GET	Payroll data

Full Swagger documentation available at /docs/ when running the backend.

---

Environment Variables

Variable	Required	Description
GEMINI_API_KEY	Yes	Google Gemini API key
GEMINI_API_KEYS	No	Comma-separated keys for rotation
GEMINI_MODEL	No	Model name (default: gemini-2.5-flash)
DATABASE_URL	Yes	PostgreSQL connection string
JWT_SECRET_KEY	Yes	Secret for JWT signing

---

Database Schema

Users → id, email, password_hash, name, role, department

Reports → id, category, department, description, severity, status, user_id, assigned_to, latitude, longitude, image_url, created_at

ReportLogs → id, report_id, status, message, updated_by, timestamp

Workers → id, user_id, type (mcd/gig/ngo), skills, vehicle_type, current_lat, current_lng, is_available, rating

---

Scripts

Backend utility scripts in UE_backend-main/:

Script	Purpose
seed_db.py	Populate database with sample data
reset_db.py	Drop and recreate all tables
mock_data.py	Generate mock reports
check_db.py	Verify database connection
verify_users.py	List all users

---

Deployment

Backend (Render/Railway)

1. Set environment variables in dashboard
2. Build command: pip install -r requirements.txt
3. Start command: gunicorn app:app

Frontend (Vercel/Netlify)

1. Build command: npm run build
2. Publish directory: dist
3. Set VITE_API_URL environment variable

---

License

MIT

---

Contributors

• Ayan Ahmed Khan