Model-based co-clustering for hyperspectral images

Julien Jacques (1, 2) , Cyril Ruckebusch (3)

1 ERIC - Entrepôts, Représentation et Ingénierie des Connaissances
2 MODAL - MOdel for Data Analysis and Learning
3 LASIRE - Laboratoire Avancé de Spectroscopie pour les Intéractions la Réactivité et l'Environnement - UMR 8516

A model-based co-clustering algorithm for hyperspectral images is presented. This algorithm, which relies on the probabilistic latent block model for continuous data, aims to cluster both the pixels and the spectral features of the images. This approach has been applied to a benchmark Raman imaging dataset and revealed relevant information for spatial-spectral exploratory investigation of the data.

Mots clés: co-clustering; latent block model; hyperspectral images

Domaines: Statistiques [math.ST]; Chimie analytique

Julien Jacques, Cyril Ruckebusch. Model-based co-clustering for hyperspectral images. Journal of Spectral Imaging, 2016, 5, pp.a3-1 - a3-6. ⟨10.1255/jsi.2016.a3⟩. ⟨hal-01367941⟩