Satellite Image Classification (CS826 - Group Project)

This project implements satellite image classification using three deep learning models:

1. Deep Neural Network (DNN)
2. Convolutional Neural Network (CNN)
3. Transfer Learning Model (EfficientNet_B0)

Dataset: Public satellite image dataset with four slasses;

• Cloudy
• Desert
• Green Area
• Water

The project includes:

• Data preprocessing
• Model training
• Evaluation (accuracy, confusion matrix)
• Comparison of model performance

The idea is to built a system which classifies the full satellite images into 4 environment categories and then mearure success by how accurately it does.

A colab notebook will be provided for final submission.