Multiresolution approach to biomedical image segmentation with statistical models of appearance

Spela Ivekovic, Aleš Leonardis

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)peer-review

1 Citation (Scopus)

Abstract

Structural variability present in biomedical images is known to aggravate the segmentation process. Statistical models of appearance proved successful in exploiting the structural variability information in the learning set to segment a previously unseen medical image more reliably. In this paper we show that biomedical image segmentation with statistical models of appearance can be improved in terms of accuracy and efficiency by a multiresolution approach. We outline two different multiresolution approaches. The first demonstrates a straightforward extension of the original statistical model and uses a pyramid of statistical models to segment the input image on various resolution levels. The second applies the idea of direct coefficient propagation through the Gaussian image pyramid and uses only one statistical model to perform the multiresolution segmentation in a much simpler manner. Experimental results illustrate the scale of improvement achieved by using the multiresolution approaches described. Possible further improvements are discussed at the end. © Springer-Verlag Berlin Heidelberg 2003.

Original languageEnglish
Title of host publicationScale Space Methods in Computer Vision
Subtitle of host publication4th International Conference, Scale Space 2003 Isle of Skye, UK, June 10–12, 2003 Proceedings
Pages667-682
Number of pages16
Volume2695
ISBN (Electronic)978-3-540-44935-5
DOIs
Publication statusPublished - 2003
Event4th International Conference Scale Space Methods in Computer Vision - Isle of Skye, Isle of Skye, United Kingdom
Duration: 10 Jun 200312 Jun 2003

Publication series

NameLecture Notes in Computer Science
Volume2695
ISSN (Print)0302-9743

Conference

Conference4th International Conference Scale Space Methods in Computer Vision
CountryUnited Kingdom
CityIsle of Skye
Period10/06/0312/06/03

Fingerprint Dive into the research topics of 'Multiresolution approach to biomedical image segmentation with statistical models of appearance'. Together they form a unique fingerprint.

Cite this