Založeno v roce 2005 s podporou MŠMT ČR (projekt 1M0572)

Publikace

A Subspace Approach to Texture Modelling by Using Gaussian Mixtures

Typ:
Konferenční příspěvek
Autoři publikace:
Název sborniku:
Proceedings of the 18th International Conference on Pattern Recognition. ICPR 2006
Nakladatel:
IEEE Press
Místo vydání:
Hong Kong
Rok:
2006
Strany:
235-238
ISBN:
0-7695-2521-0
ISSN:
1051-4651
Klíčová slova:
texture modelling, Gaussian mixtures, subspace approach
Anotace:
Assuming local and shift-invariant texture properties we describe the statistical dependencies between pixels by a joint probability density of gray-levels within a suitably chosen observation window. We estimate the unknown multivariate density in the form of a Gaussian mixture of product components from data obtained by shifting the observation window by considering a subspace approach based on a structural mixture model.
 
Copyright 2005 DAR XHTML CSS