PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Unsupervised Texture Segmentation Using a nonlinear Energy Optimisation
Sasan Mahmoodi and Bayan Sharif
Journal of Electronic Imaging Volume 15, Number 3, pp. 1-8, 2005.

Abstract

nonlinear functional is considered in this paper for segmentation of images containing structural textures. A structural texture pattern in an image is characterized by a certain amplitude spectrum, and segmentation of different patterns is obtained by detecting different regions with different amplitude spectra. A gradient descent-based algorithm is proposed in this paper by deriving equations minimising the functional. This algorithm implementing the solutions minimising the functional is based on the level set method. An effective method employed in this algorithm is shown to be robust in a noisy environment. Experimental results demonstrate that the proposed method outperforms segmentation obtained by using the simulated annealing algorithm based on Gaussian Markov Random Fields

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
Theory & Algorithms
ID Code:4246
Deposited By:Sasan Mahmoodi
Deposited On:20 December 2008