A modern web application for real-time waste classification using YOLO object detection. Identify recyclable, non-recyclable, and hazardous waste items through image upload or live camera feed.
✨ AI Detection - Real-time YOLO-based waste classification
📱 Real-time Processing - Instant results with confidence scores
📸 Image Upload - Upload photos for quick classification
📹 Live Camera - Use your webcam for live detection
💡 Eco Tips - Get sustainability recommendations for each waste type
🌍 Sustainable Impact - Learn about waste reduction and recycling
Visit: https://earthify-waste-classifier.onrender.com
cd Waste-Classification-Web
pip install -r requirements.txt
python app.pyVisit http://localhost:5000 in your browser.
docker build -t earthify .
docker run -p 5000:5000 earthifyWaste-Classification-Web/
├── app.py # Flask application
├── static/
│ ├── app.js # Client-side functionality
│ ├── style.css # UI styling (black & white theme)
│ └── uploads/ # User uploaded images
├── templates/
│ └── index.html # Main web interface
├── requirements.txt # Python dependencies
├── runtime.txt # Python version (3.12.7)
├── Procfile # Render deployment config
└── Dockerfile # Docker configuration
../weights/
├── best.onnx # YOLO model (ONNX format)
├── best.pt # YOLO model (PyTorch)
└── last.pt # Training checkpoint
The model classifies waste into 12 categories:
- Battery
- Biological
- Brown Glass
- Cardboard
- Clothes
- Green Glass
- Metal
- Paper
- Plastic
- Shoes
- Trash
- White Glass
- Backend: Flask 3.0.3
- ML Framework: ONNX Runtime (YOLOv8)
- Image Processing: Pillow, NumPy
- Frontend: HTML5, CSS3, Vanilla JavaScript
- Deployment: Render, Docker
- Production Server: Gunicorn
Karthick V
- GitHub: @Gorghs
- LinkedIn: karthickv4
- Email: karthick.venkatachalem@gmail.com
- Portfolio: karthick-rnen.onrender.com
MIT License - Feel free to use this project for educational and commercial purposes.