Unsupervised texture segmentation in a deterministic annealing framework

T. Hofmann, J. Puzicha, J.M. Buhmann
1998 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We present a novel optimization framework for unsupervised texture segmentation that relies on statistical tests as a measure of homogeneity. Texture segmentation is formulated as a data clustering problem based on sparse proximity data. Dissimilarities of pairs of textured regions are computed from a multi scale Gabor lter image representation. We discuss and compare a class of clustering objective functions which is systematically derived from invariance principles. As a general optimization
more » ... ramework we propose deterministic annealing based on a mean eld approximation. The canonical way to derive clustering algorithms within this framework as well as an e cient implementation of mean eld annealing and the closely related Gibbs sampler are presented. We apply both annealing variants to Brodatz like micro texture mixtures and real word images. T. Hofmann, J. Puzicha, J.M. Buhmann: Unsupervised Texture Segmentation 1
doi:10.1109/34.709593 fatcat:ohmp67gezvfx5nfewhsrfhpctm