Number of the records: 1
A statistical approach to local evalution of a single texture image
- 1.
SYSNO ASEP 0025872 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title A statistical approach to local evalution of a single texture image Title Statistická metoda lokálního vyhodnocování textury Author(s) Grim, Jiří (UTIA-B) RID, ORCID
Somol, Petr (UTIA-B) RID
Haindl, Michal (UTIA-B) RID, ORCID
Pudil, Pavel (UTIA-B) RIDSource Title Proceedings of the Sixtheenth Annual Symposium of the Pattern Recognition Association of South Africa. - Cape Town : University of Cape Town, 2005 / Nicolls F. Pages s. 171-176 Number of pages 6 s. Action Annual Symposium of the Pattern Recognition Association of South Africa /16./ Event date 23.11.2005-25.11.2005 VEvent location Langebaan Country ZA - South Africa Event type WRD Language eng - English Country ZA - South Africa Keywords fintite mixtures ; EM algorithm ; texture modelling ; texture evaluation Subject RIV BB - Applied Statistics, Operational Research R&D Projects 1M0572 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) GA402/03/1310 GA ČR - Czech Science Foundation (CSF) CEZ AV0Z10750506 - UTIA-B (2005-2011) Annotation Assuming local and shift-invariant texture properties we describe the statistical dependencies between pixels by a joint probalibity density of pixels within a suitably chosen observation window. We estimate the unknow density in the form of a Gaussian mixture from data obtained by shifting the window. At each position of the window the normal mixture density is computed and the corresponding log-likelihood value is presented as a grey level at the central pixel of the window to detect unusual places. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2006
Number of the records: 1