Back to Search
Start Over
A COMPARATIVE STUDY OF CONTEXTUAL SEGMENTATION METHODS FOR DIGITAL ANGIOGRAM ANALYSIS.
- Source :
-
Cybernetics & Systems . Jan/Feb2004, Vol. 35 Issue 1, p63-83. 21p. - Publication Year :
- 2004
-
Abstract
- This paper presents a comparative study of several well-known and thoroughly tested techniques for the segmentation of textured images, including two algorithms belonging to the adaptive Bayesian family of restoration and segmentation methods, and a novel approach based on the recently introduced concept of the frequency histogram of connected elements (FHCE). The paper first introduces the parameters that define a connected element and then details the sensitivity analysis of these parameters, showing that the grayscale intensity histogram of a digital image is a particular case of the FHCE. The application domain chosen for comparison purposes is the problem of medical images segmentation and, more specifically, as a particularly illustrative case the segmentation of digital angiograms is analyzed in detail. To get a comparative evaluation of FHCE performance, two well-established adaptive or contextual Bayesian segmentation algorithms have been applied to the segmentation of digital angiograms as well. The paper ends with a brief discussion of the comparative performances. [ABSTRACT FROM AUTHOR]
- Subjects :
- *DIAGNOSTIC imaging
*ANGIOGRAPHY
*ALGORITHMS
*BAYESIAN analysis
Subjects
Details
- Language :
- English
- ISSN :
- 01969722
- Volume :
- 35
- Issue :
- 1
- Database :
- Academic Search Index
- Journal :
- Cybernetics & Systems
- Publication Type :
- Academic Journal
- Accession number :
- 11763123
- Full Text :
- https://doi.org/10.1080/01969720490246849