Multispectral texture segmentation

0411250 - UTIA-B 20030237 CZ eng K - Conference Paper (Czech conference)
Mikeš, Stanislav - Haindl, Michal
Multispectral texture segmentation.
Praha: MFF UK, 2003. ISBN 80-86732-18-5. In: WDS '03 Proceedings of Contributed Papers. - (Šafránková, J.), s. 221-225
[Week of Doctoral Students 2003. WDS'03. Praha (CZ), 10.06.2003-13.06.2003]
R&D Projects: GA AV ČR IAA2075302
Grant - others:Commission EC(XE) IST-2001-34744
Institutional research plan: CEZ:AV0Z1075907
Keywords : texture * unsupervised segmentation * Markov random fields
Subject RIV: BD - Theory of Information

An efficient and robust type of unsupervised multispectral texture segmentation method is presented. The algorithm starts with spectral factorization of an input multispectral texture image using the Karhunen-Loeve expansion. Monospectral factors of single texture patches are assumed to be modelled using a Gaussian Markov random field model. The texture segmentation is done by K-means algorithm in the Markov model parameter space evaluated for each pixel centered image window.
Permanent Link: http://hdl.handle.net/11104/0131335