Skip to main navigation Skip to search Skip to main content

Semi-automatic seed points selection in fuzzy connectedness approach to image segmentation

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

12 Citations (Scopus)

Abstract

A new method improving the fuzzy connectedness approach to medical image segmentation is described. The segmentation based on fuzzy connectedness relies on a fuzzy connectivity scene creation by assigning a strength of connectedness to each possible path between some predefined seed point located inside an object and any other image element and performing the thresholding. The new idea is to automatically choose more seed points inside, as well as outside the segmented structure, in order to improve the method effectiveness and to reduce the computational time. The selection is based on two points marked manually. The method has been tested on a set of 3D Computed Tomography (CT) lung images with delineated nodules. Examples and results of this method applied to the segmentation of lung nodules are shown.

Original languageEnglish
Title of host publicationComputer Recognition Systems 2
EditorsMarek Kurzynski, Michal Wozniak, Andrzej Zolnierek, Edward Puchala
Pages679-686
Number of pages8
DOIs
Publication statusPublished - 2007

Publication series

NameAdvances in Soft Computing
Volume45
ISSN (Print)1615-3871
ISSN (Electronic)1860-0794

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Computational Mechanics
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Semi-automatic seed points selection in fuzzy connectedness approach to image segmentation'. Together they form a unique fingerprint.

Cite this