|Title:||Filtering and segmentation of retinal OCT images||Authors:||Alemán-Flores, Miguel
|UNESCO Clasification:||220990 Tratamiento digital. Imágenes||Issue Date:||2018||Journal:||Lecture Notes in Computer Science||Conference:||16th International Conference on Computer Aided Systems Theory, EUROCAST 2017||Abstract:||This work presents a method for the segmentation of optical coherence tomography images of the retina. Before segmenting the tomography, anisotropic diffusion is applied to reduce noise, but preserve the relevant edges. Afterward, the intensity profile of the images is analyzed to extract an initial approximation for the segmentation of three bands within the retina. Finally, a combination of attraction and regularization terms is used to refine the segmentation by fitting the limits of the bands to the highest gradients and smoothing their shapes to make them more regular. From the bands extracted in the different slices of the tomography, a three-dimensional reconstruction is performed for a better visualization of the results.||URI:||http://hdl.handle.net/10553/41431||ISBN:||9783319747262||ISSN:||0302-9743||DOI:||10.1007/978-3-319-74727-9_34||Source:||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)[ISSN 0302-9743], v. 10672 LNCS, p. 289-296|
|Appears in Collections:||Actas de congresos|
checked on Mar 7, 2021
Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.