AICTE Virtual Internship Project

Waste Classification Using CNN

📌 Project Overview

Waste Classification Using CNN is a deep learning-based project that classifies waste into two categories: Organic and Recyclable using a Convolutional Neural Network (CNN). The model is trained using TensorFlow/Keras with ImageDataGenerator for preprocessing.

The project includes a Streamlit app for user interaction and image classification.

📌 Dataset

Download Dataset from this link: Kaggle - Waste Classification Data

🏗 Project Architecture

1. Model Development (Python & TensorFlow/Keras)

   • Preprocessing waste images using ImageDataGenerator
   • Training a CNN model to classify waste
   • Overcoming overfitting issues

2. Frontend (Streamlit Web App)

   • Streamlit UI to allow users to upload images for classification

3. Deployment

   • Streamlit app for quick access

🚀 Features

✔️ Classifies waste as Organic or Recyclable
✔️ CNN model trained with TensorFlow/Keras
✔️ Streamlit App for easy access
✔️ Deployed on Hugging Face Spaces

🛠 Technologies Used

• Python (TensorFlow, Keras, OpenCV, NumPy, Pandas)
• Streamlit (Web App UI)

📂 Project Structure

Waste-Classification-Using-CNN/
│── Week1/  # Initial implementation and dataset exploration
│── Week2/  # Model development and training
│── Week3/  # Model evaluation and optimization
│── Final_week/  # Finalized project files and deployment setup
    │── app.py  # Streamlit-based UI
    │── requirements.txt  # Requirements
    │── CNN_model/  # Trained CNN model files
│── .gitignore  # Git ignore file
│── README.md  # Project Documentation

🛠 Setup & Installation

🔹 Streamlit App

cd Final_week
pip install -r requirements.txt
streamlit run app.py

📌 Deployment Links

• Streamlit App : https://waste-classification-mani.streamlit.app/

---

Contributions & feedback are welcome! 😊