Morphological segmentation of hyperspectral images

Guillaume Noyel (1) , Jesus Angulo (1) , Dominique Jeulin (1)

1 CMM - Centre de Morphologie Mathématique

The present paper develops a general methodology for the morphological segmentation of hyperspectral images, i.e., with an important number of channels. This approach, based on watershed, is composed of a spectral classification to obtain the markers and a vectorial gradient which gives the spatial information. Several alternative gradients are adapted to the different hyperspectral functions. Data reduction is performed either by Factor Analysis or by model fitting. Image segmentation is done on different spaces: factor space, parameters space, etc. On all these spaces the spatial/spectral segmentation approach is applied, leading to relevant results on the image.

Mots clés: factor analysis; watershed segmentation; Mathematical Morphology; hyperspectral imagery; hyperspectral image analysis; multivariate image analysis; Classification

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement des images [eess.IV]; Traitement du signal et de l'image [eess.SP]

Guillaume Noyel, Jesus Angulo, Dominique Jeulin. Morphological segmentation of hyperspectral images. Image Analysis & Stereology, 2007, 26 (3), pp.101-109. ⟨10.5566/ias.v26.p101-109⟩. ⟨hal-01220414v2⟩