PASCAL NDVI Block

© 2025 AustralMetrics SpA. All rights reserved.

A high-performance module for calculating NDVI, NDRE, and SAVI indices from Sentinel-2/Landsat imagery, focusing on simplicity and ISO 42001 compliance.

Project Status

• Version: 1.0.0
• Compatibility: Windows, Linux, macOS
• Python: 3.7+

Features

• Simplified Input: Process satellite imagery without complex configuration files
• Automatic Detection: Smart identification of spectral bands based on image type
• Multiple Indices: Calculate NDVI, NDRE, and SAVI in a single pass
• Intuitive CLI: Simple commands for all operations
• Automated Pipeline: Complete processing with a single command

Normative Dependencies

This project follows ISO 42001 compliance requirements and integrates with other PASCAL ecosystem blocks. For detailed information about normative dependencies and compliance:

• Normative Dependencies Documentation
• ISO 42001 Compliance
• Security Policy

Quality Control & CI/CD

This project implements robust Continuous Integration (CI) and Continuous Delivery (CD) practices:

• Automated Testing: Every push and pull request triggers:
  • Unit tests with pytest
  • Style verification with flake8
  • Type checking with mypy
• Quality Control:
  • 100-character line limit
  • Consistent formatting
  • Mandatory documentation

Traceability & Audit

ISO 42001-compliant logging system:

• Location: results/logs/
  • Main logs: pascal_ndvi_YYYYMMDD_HHMMSS.log
  • Automatic backups: backup/pascal_ndvi_*_backup.log
  • Integrity verification: SHA-256 hashes in *.sha256
• Logged Information:
  • Precise timestamps
  • User and process data
  • Input parameters
  • Results and errors

Installation

# Clone repository
git clone https://github.com/australmetrics/pascal-ndvi-block.git
cd pascal-ndvi-block

# Create virtual environment (recommended)
python -m venv venv
source venv/bin/activate  # Linux/Mac
# or
venv\Scripts\activate  # Windows

# Install dependencies
pip install -r requirements.txt

Basic Usage

Simple Index Calculation

# Calculate all indices for an image
python -m src.main indices --image=data/my_image.tif

# Specify output directory
python -m src.main indices --image=data/my_image.tif --output=results

Process with Clipping (optional)

# Clip image using shapefile
python -m src.main clip --image=data/my_image.tif --shapefile=data/area.shp

# Complete processing: clip + indices
python -m src.main auto --image=data/my_image.tif --shapefile=data/area.shp

Automated Processing

# Simplest option: image only, no clipping
python -m src.main auto --image=data/my_image.tif

Project Structure

pascal-ndvi-block/
├── data/                  # Input imagery folder
├── results/              # Generated results (default)
├── src/
│   ├── __init__.py
│   ├── main.py            # Entry point and CLI
│   ├── preprocessor.py    # Preprocessing functions (clipping)
│   └── indices.py         # Vegetation indices calculation
├── tests/
│   └── test_main.py       # Unit tests
├── README.md
└── requirements.txt

Configuration

1. Copy the environment variables example file:

Copy-Item .env.example .env

2. Edit the .env file with your settings:
   • USERNAME: Your username for audit logs
   • LOG_LEVEL: Logging level (DEBUG, INFO, WARNING, ERROR, CRITICAL)

The .env file will not be uploaded to Git for security.

ISO 42001 Compliance

This module is designed following ISO 42001 principles for artificial intelligence management systems:

• Simplicity: Minimal interfaces and clear workflows
• Documentation: Fully documented code and APIs
• Maintainability: Modular structure and automated testing
• Reproducibility: Consistent results with the same input data
• Traceability: Detailed processing logs using loguru

Requirements

• Python 3.7+
• rasterio
• geopandas
• numpy
• typer
• loguru

Support and Contact

• Internal Repository: Create GitHub issue to report problems
• Documentation: See documentation in the docs/ folder
• Contact: Contact AustralMetrics SpA development team

License and Usage

This software is property of AustralMetrics SpA. Its use, modification, and distribution is restricted to authorized company personnel.