AetherGeo: A Spectral Analysis Interface for Geologic Mapping

• Type: Article in International Scientific Journal
• Date: 2025

Authors

Santos, G (Author) Other; Cardoso-Fernandes, J (Author) FCUP; Teodoro, AC (Author) Other

Journal

Title: Algorithms

Vol. 57

Final page: 378

ISSN: 1999-4893

Publisher: MDPI

Other information

Authenticus ID: P-018-YK7

DOI: 10.3390/a18070378

Abstract (EN): <jats:p>AetherGeo is a standalone piece of software (current version 1.0) that aims to enable the user to analyze raster data, with a special focus on processing multi- and hyperspectral images. Being developed in Python 3.12.4, this application is a free, open-source alternative for spectral analysis, something considered beneficial for researchers, allowing for a flexible approach to start working on the topic without acquiring proprietary software licenses. It provides the user with a set of tools for spectral data analysis through classical approaches, such as band ratios and RGB combinations, but also more elaborate techniques, such as endmember extraction and unsupervised image classification with partial spectral unmixing techniques. While it has been tested on visible and near-infrared (VNIR), short-wave infrared (SWIR), and VNIR-SWIR datasets, the functions implemented have the potential to be applied to other spectral ranges. On top of this, all results can be visualized within the software, and some tools allow for the inspection and comparison of spectra and spectral libraries. Providing software with these capabilities in a unified platform has the potential to positively impact research and education, as students and educators usually have limited access to proprietary software.</jats:p>

Language: English

Type (Professor's evaluation): Scientific