Title: Performance evaluation of several anisotropic diffusion filters for fundus imaging

Authors: Mariem Ben Abdallah; Jihene Malek; Ahmad Taher Azar; Hafedh Belmabrouk; Julio Esclarín Monreal

Addresses: Electronics and Micro-electronics Laboratory, Monastir University, Tunisia ' Electronics and Micro-electronics Laboratory, Monastir University, Tunisia ' Faculty of Computers and Information, Benha University, Egypt ' Electronics and Micro-electronics Laboratory, Monastir University, Tunisia ' Imaging Technology Center (CTIM), Las Palmas-Gran Canaria University, Spain

Abstract: In image processing by partial differential equations, the first and simplest models to have and use are based on linear diffusion. The common difficulty of linear filters is the excessive smoothing that makes track edges difficult. Therefore, we can affirm that any improvement of these linear models must be carried out inside the diffusion operator, thus sacrificing their linearity. The work achieved in this context will make the subject of the following paper. We will see how these difficulties can be overcome by the use of the nonlinear models. This document treats the automatic preprocessing of a retinal vascular network in fundus images, using various anisotropic diffusion filters, in order to improve the interpretation of the images for the doctor's diagnosis. To evaluate the chosen methods, we have performed image enhancement parameters, mean preservation and variance reduction, and edge preservation.

Keywords: anisotropic diffusion filters; filtering; retinal images; performance evaluation; fundus imaging; image processing; partial differential equations; PDEs; nonlinear models; retinal vascular networks; medical diagnosis; image enhancement parameters; mean preservation; variance reduction; edge preservation.

DOI: 10.1504/IJIEI.2015.069100

International Journal of Intelligent Engineering Informatics, 2015 Vol.3 No.1, pp.66 - 90

Received: 17 Nov 2014
Accepted: 19 Dec 2014
Published online: 27 Apr 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article