Share Email Print

Proceedings Paper

Benchmarking segmentation results using a Markov model and a Bayes information criterion
Author(s): Fionn D. Murtagh; Xiaoyu Qiao; Danny Crookes; Paul Walsh; P. A. Muhammed Basheer; Adrian Long
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Features are derived from wavelet transforms of images containing a mixture of textures. In each case, the texture mixture is segmented, based on a 10-dimensional feature vector associated with every pixel. We show that the quality of the resulting segmentations can be characterized using the Potts or Ising spatial homogeneity parameter. This measure is defined from the segmentation labels. In order to have a better measure which takes into account both the segmentation labels and the input data, we determine the likelihood of the observed data given the model, which in turn is directly related to the Bayes information criterion, BIC. Finally we discuss how BIC is used as an approximation in model assessment using a Bayes factor.

Paper Details

Date Published: 19 March 2003
PDF: 7 pages
Proc. SPIE 4877, Opto-Ireland 2002: Optical Metrology, Imaging, and Machine Vision, (19 March 2003); doi: 10.1117/12.467441
Show Author Affiliations
Fionn D. Murtagh, Queen's Univ. Belfast (United Kingdom)
Xiaoyu Qiao, Queen's Univ. Belfast (United Kingdom)
Danny Crookes, Queen's Univ. Belfast (United Kingdom)
Paul Walsh, Queen's Univ. Belfast (United Kingdom)
P. A. Muhammed Basheer, Queen's Univ. Belfast (United Kingdom)
Adrian Long, Queen's Univ. Belfast (United Kingdom)

Published in SPIE Proceedings Vol. 4877:
Opto-Ireland 2002: Optical Metrology, Imaging, and Machine Vision
Andrew Shearer; Fionn D. Murtagh; James Mahon; Paul F. Whelan, Editor(s)

© SPIE. Terms of Use
Back to Top