Prospects for mitigating spectral variability in tropical species classification using self-supervised learning

Prieur Colin, Ait Ali Braham Nassim, Tresson Paul, Vincent Grégoire, Chanussot Jocelyn. 2024. Prospects for mitigating spectral variability in tropical species classification using self-supervised learning. In : 2024 14th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS). Washington : IEEE, 5 p. Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS 2024). 14, Helsinki, Finlande, 9 Décembre 2024/11 Décembre 2024.

Résumé : Airborne hyperspectral imaging is a promising method for identifying tropical species, but spectral variability between acquisitions hinders consistent results. This paper proposes using Self-Supervised Learning (SSL) to encode spectral features that are robust to abiotic variability and relevant for species identification. By employing the state-of-the-art Barlow-Twins approach on repeated spectral acquisitions, we demonstrate the ability to develop stable features. For the classification of 40 tropical species, experiments show that these features can outperform typical reflectance products in terms of robustness to spectral variability by 10 points of accuracy across dates.

Mots-clés libres : Remote Sensing, Forest management, Self- supervised learning

Auteurs et affiliations

• Prieur Colin, CNRS (FRA)
• Ait Ali Braham Nassim, Technical University of Munich (DEU)
• Tresson Paul, CIRAD-BIOS-UMR AMAP (FRA)
• Vincent Grégoire, IRD (FRA)
• Chanussot Jocelyn, Université Grenoble Alpes (FRA)

Source : Cirad-Agritrop (https://agritrop.cirad.fr/614235/)

Métriques

[ Page générée et mise en cache le 2025-09-22 ]