Land Use Change Detection Project

Land use change detection and monitoring system in Córdoba, Argentina, using satellite imagery.

System Requirements

Software Versions

• Python 3.11.8
• Node.js 18
• Redis 7.0

Development Environment Setup

1. Clone the repository:

git clone https://github.com/OmdenaAI/CordobaArgentinaChapter_MonitoringLandUseTransformation.git

2. Navigate to project directory:

cd CordobaArgentinaChapter_MonitoringLandUseTransformation

3. Install Python 3.11.8:

For MacOS (using pyenv):

# Install pyenv if you haven't already
brew install pyenv

# Install Python 3.11.8
pyenv install 3.11.8

# Set Python 3.11.8 as local version for this project
pyenv local 3.11.8

# Reload shell configuration
source ~/.zshrc    # If using zsh
# OR
source ~/.bashrc   # If using bash

# Verify installation
python --version  # Should show Python 3.11.8

For Windows:

# Using winget (Windows Package Manager)
winget install Python.Python.3.11

# Or download the installer from official website
# https://www.python.org/downloads/release/python-3117/

# Refresh environment variables (open a new terminal)
# OR run this command in PowerShell:
$env:Path = [System.Environment]::GetEnvironmentVariable("Path","Machine") + ";" + [System.Environment]::GetEnvironmentVariable("Path","User")

# Verify installation
python --version  # Should show Python 3.11.8

4. Set up Python virtual environments:

Preprocessing Service:

python -m venv preprocessing_service/venv
source preprocessing_service/venv/bin/activate # Linux/Mac
.\preprocessing_service\venv\Scripts\activate # Windows

Install dependencies (if they exist):

pip install -r preprocessing_service/requirements.txt

Model Service:

python -m venv model-service/venv
source model_service/venv/bin/activate # Linux/Mac
.\model_service\venv\Scripts\activate # Windows

Install dependencies (if they exist):

pip install -r model_service/requirements.txt

API Gateway:

python -m venv api-gateway/venv
source api_gateway/venv/bin/activate # Linux/Mac
.\api_gateway\venv\Scripts\activate # Windows

Install dependencies (if they exist):

pip install -r api_gateway/requirements.txt

Queue Service:

python -m venv queue-service/venv
source queue_service/venv/bin/activate # Linux/Mac
.\queue_service\venv\Scripts\activate # Windows

Install dependencies (if they exist):

pip install -r queue_service/requirements.txt

Set up Frontend:

cd frontend
npm install

Configure Redis:

docker run -d --name redis-stack-server -p 6379:6379 redis/redis-stack-server:latest

Project Structure

change_detection_project/
├── preprocessing_service/ # Image preprocessing service
├── model_service/        # Model inference service
├── api_gateway/          # API Gateway (FastAPI)
├── queue_service/        # Queue service (Redis + Celery)
├── frontend/            # Frontend (React)
└── common/              # Shared code between services

Local Development

Running Services Individually

1. Preprocessing Service:

cd preprocessing_service
source venv/bin/activate
python src/main.py

2. Model Service:

cd model_service
source venv/bin/activate
python src/main.py

3. API Gateway:

cd api_gateway
source venv/bin/activate
uvicorn src.main:app --reload

4. Queue Service:

cd queue_service
source venv/bin/activate
celery -A src.workers worker --loglevel=info

5. Frontend:

cd frontend
npm run dev

Using Docker Compose

docker-compose up -d

Code Conventions

• Follow PEP 8 for Python
• ESLint with Airbnb config for JavaScript/TypeScript
• Conventional Commits for commit messages
• Google Style Python Docstrings for documentation

Git Workflow

1. Create branch from develop:

git checkout develop
git pull origin develop
git checkout -b feature/feature_name

2. Commit changes:

git add .
git commit -m "feat: description of change"
git push origin feature/feature_name

Create PR in GitHub from feature -> develop

Additional Documentation

• Development Guide
• API Documentation
• Deployment Guide

Contact

For questions or suggestions, contact:

• Tech Lead: Name
• Project Manager: Name

Available Satellites

Satellite Data Sources

The project can work with imagery from the following satellites:

Satellite	Resolution	Revisit Time	Bands	Best Use Case
Sentinel-2	10m, 20m, 60m	5 days	13 bands	Land use, agriculture, forest monitoring
Landsat 8-9	15m, 30m, 100m	16 days	11 bands	Historical analysis, long-term changes
MODIS	250m, 500m, 1km	Daily	36 bands	Large-scale monitoring, rapid changes
Planet	3-5m	Daily	4 bands	High-resolution monitoring (commercial)