A smart, AI-powered waste classification system that sees trash... and thinks clean! Built to sort six types of waste using powerful transfer learning models.
- 🧭 Why This Project?
- 🔍 What Can It Detect?
- 🧠 Inside the Model
- 📂 Project Layout
- 🚀 Live Demo
- 💡 Features At A Glance
- 🛠 Built With
- 💬 Community & Feedback
- 🤝 How to Contribute
- 📄 License
- 🏆 Recognition
Every piece of garbage matters.
Improper sorting leads to overflowing landfills and wasted recyclables.
This project brings machine learning to the front lines of sustainability, helping automate and simplify waste classification.
🧠 Built during the Shell-Edunet Skills4Future Internship (June–July 2025).
The AI model classifies any uploaded garbage image into:
| Category | Example Items |
|---|---|
| 🟫 Cardboard | Boxes, cartons |
| 🟡 Plastic | Bottles, containers |
| 📰 Paper | Newspapers, wrappers |
| 🔩 Metal | Cans, utensils |
| 🟢 Glass | Jars, shattered pieces |
| 🗑️ Trash | Everything else non-recyclable |
| Feature | Description |
|---|---|
| 📦 Architecture | EfficientNetV2B2 (state-of-the-art) |
| 🔄 Transfer Learning | Pretrained on ImageNet, fine-tuned for trash |
| 📱 Interface | Gradio / Streamlit for live predictions |
| 📈 Accuracy | 98% on validation |
| 🆚 Baseline | Compared against MobileNetV2 |
Garbage_Classification/
├── Week1/ # Research & setup
├── Week2/ # Model experimentation
├── Week3/ # Evaluation & UI
├── Dataset/ # Preprocessed images
├── app.py # Gradio or Streamlit app
├── model_efficientnet.h5 # Trained weights
└── README.md🎯 Try It Out Yourself
Run locally:
# Install required libraries
pip install -r requirements.txt
# Launch the app
streamlit run app.py✅ Real-time garbage prediction
🌍 Contributes to smart waste segregation
⚙️ Based on clean modular code
📊 Great for learning CNN + Transfer Learning
🚮 Encourages environmental awareness
| Tool | Role |
|---|---|
| 🐍 Python | Core scripting language |
| 🔬 TensorFlow | Model training + inference |
| 🧰 Keras | Transfer learning pipelines |
| 💬 Gradio/Streamlit | Web deployment & UI |
Got feedback? Found a bug? Want to contribute?
| 📌 Platform | Use Case |
|---|---|
| GitHub Issues | Bug reports, feature requests |
| Discussions | Ideas, questions, suggestions |
We love meaningful contributions!
# 1. Fork it
# 2. Create a new branch
git checkout -b feature/amazing-feature
# 3. Make your changes
git commit -m "✨ Add amazing feature"
# 4. Push & submit PR
git push origin feature/amazing-feature📘 Check out CONTRIBUTING.md for more.
📜 Open-source under the MIT License — free to use, improve, and distribute.
This project was proudly developed as part of the:
🛢️ Shell-Edunet Skills4Future Internship
Supporting real-world AI innovation for sustainability.
Made with 🧠 and ♻️ by Aditix Anand & Contributors