An LDA-based Relative Hysteresis Classifier with Application to Segmentation of Retinal Vessels

A. P. Condurache, F. Muller, A. Mertins

Abstract

In a pattern classification setup, image segmentation is achieved by assigning each pixel to one of two classes: object or background. The special case of vessel segmentation is characterized by a strong disproportion between the number of representatives of each class (i.e. class skew) and also by a strong overlap between classes. These difficulties can be solved using problem-specific knowledge. The proposed hysteresis classification makes use of such knowledge in an efficient way. We describe a novel, supervised, hysteresis-based classification method that we apply to the segmentation of retina photographies. This procedure is fast and achieves results that comparable or even superior to other hysteresis methods and, for the problem of retina vessel segmentation, to known dedicated methods on similar data sets.
Original languageEnglish
Title of host publication2010 20th International Conference on Pattern Recognition
Number of pages4
Place of PublicationIstanbul, Turkey
PublisherIEEE
Publication date01.08.2010
Pages4202-4205
Article number5597743
ISBN (Print)978-1-4244-7541-4
ISBN (Electronic)978-1-4244-7542-1
DOIs
Publication statusPublished - 01.08.2010
Event2010 20th International Conference on Pattern Recognition - Istanbul, Turkey
Duration: 23.08.201026.08.2010

Fingerprint

Dive into the research topics of 'An LDA-based Relative Hysteresis Classifier with Application to Segmentation of Retinal Vessels'. Together they form a unique fingerprint.

Cite this