Department of Pattern Recognition

(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.

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