gms | German Medical Science

102. Jahrestagung der DOG

Deutsche Ophthalmologische Gesellschaft e. V.

23. bis 26.09.2004, Berlin

Automatic analysis of retina images regarding Papille and Vessel

Meeting Abstract

Search Medline for

  • corresponding author D. Paulus - Computervisualistik, Universität Koblenz-Landau, Koblenz
  • S. Chastel - Computervisualistik, Universität Koblenz-Landau, Koblenz

Evidenzbasierte Medizin - Anspruch und Wirklichkeit. 102. Jahrestagung der Deutschen Ophthalmologischen Gesellschaft. Berlin, 23.-26.09.2004. Düsseldorf, Köln: German Medical Science; 2004. Doc04dogFR.12.05

The electronic version of this article is the complete one and can be found online at:

Published: September 22, 2004

© 2004 Paulus et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Detection of the papilla region and vessel detection on images of the retina are problems that can be solved by pattern recognition techniques. Fundus images can be used as well as topographic image as provided, e.g. by the HRT device. We present methods for papilla detection based on surface approximations of the papilla region; alternatively, a combination of line detection, image morphology, and Hough transform can be used in reflectivity images.

Vessels can be found with matched filters (wavelets, derivatives of the Gaussian, etc.). The parameter of these models can be chosen fixed which leads to an accurate detection of vessels of a certain width. An automatic adaptation of the parameters allows for the detection of vessels with varying diameters.

An alternative to these parametric approaches can be found when image morphology is applied. In our contribution we describe and compare these approaches.