🌱 Clean2Earn — Rewarding Real-World Environmental Impact

Clean2Earn is an AI-powered sustainability platform that rewards users with digital coins for performing real-world environmental activities like cleaning surroundings.

Users submit before & after images, and our AI verification system ensures authenticity before rewarding them.

---

✨ Features

• 📸 Upload Before/After Cleanup Images
• 🤖 AI-based Cleaning Verification
• 🪙 Earn Eco-Coins for verified actions
• 📊 User Dashboard (coins, impact, activities)
• 🔐 JWT Authentication System
• 📍 Location tagging using geolocation
• 🚫 Fraud Detection using AI
• 📈 Impact Score Tracking

---

🧠 Problem Statement

People want to contribute to the environment but lack:

• Motivation
• Reward systems
• Trust-based validation

Fake claims also reduce the credibility of such systems.

---

💡 Solution

Eco-Coins solves this by:

• Providing incentives (coins) for eco-actions
• Using AI to verify authenticity
• Building a trust-based ecosystem

---

⚙️ Tech Stack

Frontend

• React (Vite)
• Tailwind CSS

Backend

• Node.js
• Express.js
• MongoDB
• JWT Authentication

AI Integration

• Groq API (LLaMA / Vision Model)
• Image-based verification system

---

📁 Project Structure

Eco-Coins/
│
├── frontend/              # React frontend
├── backend/
│   ├── src/
│   │   ├── controllers/
│   │   ├── routes/
│   │   ├── models/
│   │   ├── middleware/
│   │   ├── services/      # AI logic (Groq)
│   │   └── config/
│   │
│   └── server.js
│
└── README.md

---

🔐 Authentication Flow

1. User registers / logs in
2. JWT token is generated
3. Token is used to access protected routes
4. Middleware verifies user identity

---

🤖 AI Verification Workflow

Step 1: User Submission

User uploads:

• Before Image
• After Image
• Location metadata

---

Step 2: Backend Processing

Request goes to:

POST /api/verify-cleaning

---

Step 3: AI Verification (Core Logic)

• Images are sent to Groq AI service

• AI analyzes:

  • Difference between images
  • Whether actual cleaning happened
  • Signs of manipulation

---

Step 4: AI Response Format

AI returns:

{
  "verdict": "CLEANED | FRAUD_DETECTED",
  "confidence": "HIGH | MEDIUM | LOW",
  "details": "Explanation"
}

---

Step 5: Decision Engine

• If verdict = CLEANED:

  • ✅ User earns 25 coins
  • 📈 Impact score increases
  • 🧹 Cleanup count increments

• Else:

  • ❌ No reward
  • ⚠️ Marked as fraud

---

Step 6: Database Update

• Verification stored in Verification collection

• User stats updated:

  • ecoCoins
  • cleanups
  • impactScore
  • activity history

---

Step 7: Daily Limit Protection

• Max 2 cleanups per day
• Prevents abuse/spam

---

📍 Location Handling

• Uses OpenStreetMap Reverse Geocoding
• Converts coordinates → readable city name
• Stored with each activity

---

🧪 Future Enhancements

• 🎁 Reward redemption (NGO tie-ups)
• 📱 Mobile app
• 🌍 Community challenges
• 🔗 Blockchain-based proof

---

🌟 Why This Project Stands Out

• Real-world + AI integration
• Fraud-resistant reward system
• Scalable sustainability model
• Social impact + tech innovation

---

👨‍💻 Author

Bhumika Tyagi

---

⭐ Support

If you like this project:

• Star ⭐ the repo
• Share 🚀
• Contribute 🤝

---

Sneak-Peak