TY  - JOUR
T1  - Texture Classification Using the Belief Net of a Segmentation Tree
AU - , M.A. Leo Vijilious AU - , J.P. Ananth AU - , V. Subbiah Bharathi 
JO  - Asian Journal of Information Technology
VL  - 6
IS  - 9
SP  - 929
EP  - 933
PY  - 2007
DA  - 2001/08/19
SN  - 1682-3915
DO  - ajit.2007.929.933
UR  - https://makhillpublications.co/view-article.php?doi=ajit.2007.929.933
KW  - Texture classification
KW  -segmentation
KW  -geometric
KW  -photometric properties
KW  -TSBN
AB  - This study presents a statistical approach to  texture classification from a single image obtained under unknownviewpoint and illumination. Unlike in prior work, in which texture primitives (textons) are defined in a filter-responsespace and texture classes modeled by frequency histograms of these textons, we seek to extract and model geometric and photometric properties of image regions defining the texture. To this end, texture images are first segmented bya multiscale segmentation algorithm and a universal set of texture primitives is specified over all texture classes in the domain of region geometric and photometric properties. Then, for each class, a Tree-Structured Belief Network (TSBN) is learned, where nodes represent the corresponding image regions and edges, their statistical dependecies. A given unknown texture is classified with respect to themaximum posterior distribution of the TSBN. Experimental results on the benchmark CUReT database demonstrate that our approach outperforms the state-of-the-artmethods.
ER  -