Gabor transform applied to segmentation and skeletonization of digital images

Issue title: IV Congress of Theoretical and Computational Physical Chemistry Venezuela, 2012

Guest editors: Fernando Ruettex, Anibal Sierraltax and Morella Sánchezxy

Article type: Research Article

Authors: Perez, Ronald^{a; d; *} | Lasso, William^{b; d} | Jimenez, Carlos^c | Mattos, Lorenzo^d | Torres, Cesar O.^d

Affiliations: [a] Guajira University, Riohacha, Colombia | [b] Popular University of Cesar, Cesar, Colombia | [c] Applied Mathematics Group (GIMA), Guajira University, Riohacha, Colombia | [d] Optics and Informatics Group (LOI), Popular University of Cesar, Cesar, Colombia | [x] Centro de Química, Instituto Venezolano de Investigaciones Científicas, IVIC, Caracas, Venezuela | [y] Departamento de Química, Instituto Universitario de Tecnología FRP, Caracas, Venezuela

Correspondence: [*] Corresponding author: Ronald Perez, Postgraduate Physics Student, Guajira University, Riohacha, Colombia. E-mail: ropeji2@gmail.com.

Abstract: This paper implements a digital algorithm that allows the segmentation and skeletonization of digital images, this method is particularly important in the processing of retinographies where fundus analysis enables early detection of diseases such as diabetic retinopathy. The algorithm implemented using the Fourier transforms of two-dimensional Gabor and with the aid of conventional filters. The performance of the algorithm implemented under the Matlab mathematical platform has been compared with conventional methods resulting in obtaining filtered images with better contrast and higher resolution.

Keywords: Gabor transform, segmentation, skeletonizer

DOI: 10.3233/JCM-140498

Journal: Journal of Computational Methods in Sciences and Engineering, vol. 14, no. 1-3, pp. 219-225, 2014