DUGi

Multivariate Association of Compositional Data Matrices with Applications in Comparing Hyperspectral Images

http://dugi.udg.edu/item/http:@@@@hdl.handle.net@@2072@@273626

It is well-known in image processing that, by varying the wavelength, any material reflects and absorbsin a different way the solar radiation. This is registered by hyperspectral sensors, which collectmultivariate discrete images in a series of contiguous wavelength bands, providing the spectral curves,which can distinguish between materials.In order to partition a multivariate image in regions belonging to different materials, we need tocompare these regions which are previously modelled by using compositional data matrices, where theentries in each row is a statistical discrete distribution of the radiance values (columns). These rowscorrespond to distinct but contiguos wavelengths. Thus the distribution in a row is very similar to thedistribution in close rows. To measure this proximity, we use Hellinger distance between rows, whichprovides a distance matrix.Given two hyperspectral regions of an image providing two compositional data matrices, we obtainthe corresponding distance matrices and, by using metric multidimensional scaling, we computetwo sets of principal coordinates, which are related by a multivariate association measure based oncanonical correlations.We ilustrate this approach comparing some multivariate regions of images captured by hyperspectralremote sensors

Universitat de Girona. Departament d’Informàtica i Matemàtica Aplicada