Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Parameters selection of morphological scale-space decomposition for hyperspectral images using tensor
 

Summary: Parameters selection of morphological scale-space
decomposition for hyperspectral images using tensor
modeling
Santiago Velasco-Forero and Jes´us Angulo
Mines ParisTech, Center de Morphologie Math´ematique, Fontainebleau, France
ABSTRACT
Dimensionality reduction (DR) using tensor structures in morphological scale-space decomposition (MSSD) for
HSI has been investigated in order to incorporate spatial information in DR. We present results of a comprehensive
investigation of two issues underlying DR in MSSD. Firstly, information contained in MSSD is reduced using
HOSVD but its nonconvex formulation implicates that in some cases a large number of local solutions can be
found. For all experiments, HOSVD always reach an unique global solution in the parameter region suitable to
practical applications. Secondly, scale parameters in MSSD are presented in relation to connected components
size and the influence of scale parameters in DR and subsequent classification is studied.
Keywords: U nsupervised feature extraction, mathematical morphology, tensor analysis, dimensional reduction,
classification, principal component, hyperspectral imagery.
1. INTRODUCTION
Hyperspectral images (HSI) are acquired by a sensor that captures radiant flux data in typically hundreds of
contiguous bands, producing a signature for each pixel in the image. It is accepted that HSI includes many
highly correlated bands presenting considerable amounts of spectral redundancy. Consequently, reducing the
dimensionality of hyperspectral data without losing important information is one of the main subject of interest

  

Source: Angulo,Jesús - Centre de Morphologie Mathématique, Ecole des Mines de Paris

 

Collections: Computer Technologies and Information Sciences