Unsupervised segmentation of textured images by pairwise data clustering

Yin, H. and Allinson, N. M. (1996) Unsupervised segmentation of textured images by pairwise data clustering. In: 3rd IEEE International Conference on Image Processing , 16-19 september 1996, Lausanne.

Full text not available from this repository.

Official URL: http://dx.doi.org/10.1109/ICIP.1996.560389

Abstract

We propose a novel approach to unsupervised texture segmentation, which is formulated as a combinatorial optimization problem known as pairwise data clustering with a sparse neighborhood structure. Pairwise dissimilarities between texture blocks are measured in terms of distribution differences of multi-resolution features. The feature vectors are based an a Gabor wavelet image representation. To efficiently solve the data clustering problem a deterministic annealing algorithm based on a meanfield approximation is derived. An application to Brodatz-like microtexture mixtures is shown. We statistically adress the questions of adequacy of the proposed cost function and the quality of the deterministic annealing algorithm compared with its stochastic variants.

Item Type:Conference or Workshop Item (Paper)
Additional Information:We propose a novel approach to unsupervised texture segmentation, which is formulated as a combinatorial optimization problem known as pairwise data clustering with a sparse neighborhood structure. Pairwise dissimilarities between texture blocks are measured in terms of distribution differences of multi-resolution features. The feature vectors are based an a Gabor wavelet image representation. To efficiently solve the data clustering problem a deterministic annealing algorithm based on a meanfield approximation is derived. An application to Brodatz-like microtexture mixtures is shown. We statistically adress the questions of adequacy of the proposed cost function and the quality of the deterministic annealing algorithm compared with its stochastic variants.
Keywords:neural networks, segmentation
Subjects:G Mathematical and Computer Sciences > G730 Neural Computing
Divisions:College of Science > School of Computer Science
ID Code:5083
Deposited By: Bev Jones
Deposited On:20 Apr 2012 14:14
Last Modified:20 Apr 2012 14:14

Repository Staff Only: item control page