Benchmarking Segmentation Results Using a Markov Model and a Bayes Information Criterion

Fionn Murtagh, Xiaoyu Qiao, Danny Crookes, Paul Walsh, P. A.Muhammed Basheer, Adrian Long

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Citations (Scopus)


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.

Original languageEnglish
Title of host publicationOpto-Ireland 2002
Subtitle of host publicationOptical Metrology, Imaging, and Machine Vision
EditorsAndrew Shearer, Fionn D. Murtagh, James Mahon, Paul F. Whelan
Number of pages7
ISBN (Print)0819446580, 9780819446589
Publication statusPublished - 19 Mar 2003
Externally publishedYes
EventOpto-Ireland 2002: Optical metrology, Imaging, and Machine Vision - Galway, Ireland
Duration: 5 Sep 20026 Sep 2002

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
ISSN (Print)0277-786X


ConferenceOpto-Ireland 2002: Optical metrology, Imaging, and Machine Vision
Internet address


Dive into the research topics of 'Benchmarking Segmentation Results Using a Markov Model and a Bayes Information Criterion'. Together they form a unique fingerprint.

Cite this