CleanAndGreen

Environmental AI research leveraging Computer Vision for automated waste classification and sorting.

🌍 The Mission

The goal of CleanAndGreen is to optimize recycling processes by using deep learning to accurately identify and categorize waste materials (plastic, glass, paper, etc.) from images. This helps reduce contamination in recycling streams.

🔬 Research & Implementation

• Data Collection: Curated a dataset of thousands of waste images.
• Model Training: Utilized Convolutional Neural Networks (CNNs) and transfer learning to build a robust classifier.
• Results: Achieved high accuracy in distinguishing between recyclable and non-recyclable items.

📂 Repository Structure

• waste-classify/: Contains the core classification models and training scripts.
• nb/: Jupyter notebooks with EDA (Exploratory Data Analysis) and model experiments.

🛠️ Tech Stack

• Python: Primary language for data science and model development.
• PyTorch / TensorFlow: Deep learning frameworks used for model training.
• OpenCV: For image preprocessing and augmentation.
• Jupyter: For interactive experimentation and visualization.

🚀 Getting Started

1. Clone the repository

   git clone https://github.com/yashmahe2020/CleanAndGreen.git
   cd CleanAndGreen

2. Explore the Notebooks Open the Jupyter notebooks in nb/ or waste-classify/ to see the model training process and results.

---

Note: This is a research-focused repository demonstrating applied AI in sustainability.