Satellite Image Analysis — BTech Final Year Project

A Streamlit web app that detects land-cover changes between two satellite images using SVM and CNN models.

Project Structure

Project/
├── app.py               ← Main Streamlit application
├── svm_model.pkl        ← Pre-trained SVM model (scikit-learn)
├── cnn_model.h5         ← Pre-trained CNN model (Keras/TensorFlow)
├── requirements.txt     ← Python dependencies
├── .gitignore
└── README.md

Setup & Run

1. Install dependencies

python -m pip install -r requirements.txt

2. Run the app

streamlit run app.py

3. Open browser

http://localhost:8501

How to Use

Page	What it does
1	Choose SVM or CNN model
2	Upload Before & After satellite images + dates
3	View aligned image comparison
4	View change detection heatmap
5	View land classification & calamity detection
6	View feature correlation & model accuracy

Models

• SVM — Uses NDVI, NDWI, Brightness features. Accuracy: ~82%
• CNN — Binary classifier (128x128 input, sigmoid output). Accuracy: ~91%

Push to GitHub

git init
git add .
git commit -m "Initial commit"
git branch -M main
git remote add origin https://github.com/YOUR_USERNAME/YOUR_REPO.git
git push -u origin main