Binary partition tree-based local spectral unmixing

Lucas Drumetz (1) , Miguel Angel Veganzones (1) , Ruben Marrero (2) , Guillaume Tochon (1) , Mauro Dalla Mura (1) , Antonio Plaza (3) , Jocelyn Chanussot (4, 1)

1 GIPSA-SIGMAPHY - GIPSA - Signal Images Physique
2 Planeto
3 Hyperspectral Computing Laboratory (Caceres, Spain)
4 University of Iceland [Reykjavik]

The linear mixing model (LMM) is a widely used methodology for the spectral unmixing (SU) of hyperspectral data. In this model, hyperspectral data is formed as a linear combination of spectral signatures corresponding to macroscopically pure materials (endmembers), weighted by their fractional abundances. Some of the drawbacks of the LMM are the presence of multiple mixtures and the spectral variability of the endmembers due to illumination and atmospheric effects. These issues appear as variations of the spectral conditions of the image along its spatial domain. However, these effects are not so severe locally and could be at least mitigated by working in smaller regions of the image. The proposed local SU works over a partition of the image, performing the spectral unmixing locally in each region of the partition. In this work, we first introduce the general local SU methodology, then we propose an implementation of the local SU based on a binary partition tree representation of the hyperspectral image and finally we give an experimental validation of the approach using real data.

Mots clés: Hypespectral imaging; local spectral unmixing; spectral bundles; binary partition trees

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

Lucas Drumetz, Miguel Angel Veganzones, Ruben Marrero, Guillaume Tochon, Mauro Dalla Mura, et al.. Binary partition tree-based local spectral unmixing. WHISPERS 2014 - 6th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, Jun 2014, Lausanne, Switzerland. pp.n/c. ⟨hal-01010427⟩