EcoSort-AI

Garbage Material Classification

♻️ EcoSortAI – Smart Waste Classification & Recycling Guide

EcoSortAI is an AI-powered waste classification system that uses Deep Learning (CNN) to automatically identify different types of waste from images and provide detailed recycling & disposal guidance.

The project includes:

• ✅ A trained TensorFlow/Keras model
• ✅ An interactive Tkinter desktop GUI
• ✅ A Streamlit Web App for deployment
• ✅ Automatic PDF report generation with recycling details

---

🚀 Features

• 🖼 Upload waste image → Get instant classification
• 📊 Confidence distribution chart for model predictions

♻️ EcoSort-AI — Smart Waste Classification & Recycling Guide

EcoSort-AI is an AI-powered waste classification project that uses deep learning (CNN) to identify types of waste from images and provide recycling and disposal guidance.

The project in this repository includes:

• A TensorFlow / Keras model for classification
• A Tkinter desktop GUI (GUI.py)
• A Streamlit web app (app.py)
• PDF report generation and utility scripts

Demo

Place screenshots in an assets/ folder and reference them here (optional):

Features

• Upload an image and get an automatic waste-class prediction
• Prediction confidence scores and simple visualization
• Recycling/disposal information per class
• Exportable PDF reports for each prediction

Dataset

The dataset used to train models was split into train/validation/test and contains classes such as: Battery, Biological, Brown Glass, Cardboard, Clothes, Green Glass, Metal, Paper, Plastic, Shoes, Trash, White Glass.

Tech Stack

• TensorFlow / Keras
• Tkinter (desktop GUI)
• Streamlit (web UI)
• ReportLab (PDF generation)
• Pandas, NumPy, Matplotlib

Installation

Clone the repo and install dependencies:

git clone https://github.com/rajshinde9909/EcoSort-AI.git
cd EcoSort-AI-main
pip install -r requirements.txt

Run the desktop GUI:

python GUI.py

Run the Streamlit app:

streamlit run app.py

Results

Typical final model accuracy and training curves are included in the project artifacts (if present). Your mileage depends on dataset, training settings, and preprocessing.

License

MIT License © 2025 Rushikesh Kadam

Acknowledgements

• TensorFlow team
• Streamlit community
• Open-source datasets and contributors

If you want, I can also create an assets/ folder and a curated requirements.txt tuned for Streamlit Cloud deployment. 🏙 Smart cities for automated waste sorting