Title: Fuzzy homogeneity measures for path-based colour image segmentation

Authors: J. Chamorro-Martinez, D. Sanchez, B. Prados-Suarez, E. Galan-Perales

Addresses: Department of Computer Science and Artificial Intelligence, University of Granada, C/Periodista Daniel Saucedo Aranda s/n, 18071 Granada, Spain. ' Department of Computer Science and Artificial Intelligence, University of Granada, C/Periodista Daniel Saucedo Aranda s/n, 18071 Granada, Spain. ' Department of Computer Science and Artificial Intelligence, University of Granada, C/Periodista Daniel Saucedo Aranda s/n, 18071 Granada, Spain. ' Department of Computer Science and Artificial Intelligence, University of Granada, C/Periodista Daniel Saucedo Aranda s/n, 18071 Granada, Spain

Abstract: In this paper we propose a set of measures to model the concept of homogeneity in path-based image segmentation. We introduce the idea of path homogeneity as the aggregation of resemblances between consecutive pixels in the path. This resemblance is obtained from a measure of resemblance between neighbour pixels. In order to aggregate these resemblance values we propose the use of certain families of t-norms that verify a set of intuitive properties. We have studied the performance and behaviour of these functions through a set of experiments. Finally, we have applied these proposals to obtain fuzzy segmentations from real images.

Keywords: image segmentation; fuzzy segmentation; path-based segmentation; fuzzy connectivity; fuzzy colour homogeneity; colour images; image processing; uncertainty.

DOI: 10.1504/IJISTA.2006.009910

International Journal of Intelligent Systems Technologies and Applications, 2006 Vol.1 No.3/4, pp.301 - 318

Published online: 01 Jun 2006 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article