🗑️ Garbage Classification Model — CS549

A Convolutional Neural Network trained to classify garbage images into 7 categories:

• Cardboard
• Glass
• Metal
• Paper
• Plastic
• Trash
• Biodegradable

🔍 Use Case

Helps automate waste sorting for better recycling. Great for demos, PoC apps, or smart bin integration.

🧠 Model Overview

• Architecture: CNN (Conv2D, MaxPooling, Dense)
• Input: 224x224 RGB
• Output: 7-class probability vector
• Accuracy: ~92% (validation)
• Explainability: Supports Grad-CAM

📦 How to Use

from tensorflow.keras.models import load_model
from PIL import Image
import numpy as np

model = load_model("GarbageMLModel_CS549.h5")

img = Image.open("example.jpg").resize((224, 224))
img_array = np.expand_dims(np.array(img) / 255.0, axis=0)

pred = model.predict(img_array)
classes = ['cardboard', 'glass', 'metal', 'paper', 'plastic', 'trash', 'biodegradable']
print("Prediction:", classes[np.argmax(pred)])

📁 Files

• GarbageMLModel_CS549.h5 – Trained model
• label_map.json – Label mapping

👤 Author

Vincent Huynh
📧 vintendohuynh@gmail.com
🔗 LinkedIn | 🔗 GitHub