Publication details

Model-based texture segmentation

Conference Paper (international conference)

Haindl Michal, Mikeš Stanislav


serial: Image Analysis and Recognition. Proceedings, p. 306-313

action: International Conference ICIAR 2004, (Porto, PT, 29.09.2004-01.10.2004)

research: CEZ:AV0Z1075907

project(s): 507752, , IAA2075302, GA AV ČR, 1ET400750407, GA AV ČR

keywords: colour texture segmentation, Markov random fields

abstract (eng):

An efficient and robust type of unsupervised multispectral texture segmentation method is presented. Single decorrelated monospectral texture factors are assumed to be represented by a set of local Gaussian Markov random field (GMRF) models evaluated for each pixel centered image window and for each spectral band. The segmentation algorithm based on the underlying Gaussian mixture (GM) model operates in the decorrelated GMRF parametric space. The number of texture regions is adaptively changed.

abstract (cze):

Práce prezentuje efektivní a robustní metodu neřízené segmentace multispektrálních textur. Dekorelované monospektrální texturní složky jsou reprezentovány lokálními gauss-markovskými modely

Cosati: 09K

RIV: BD