Abstract
This paper is based on a methodology for segmentation of the main retinal layers in Optical Coherence Tomography (OCT) images. The input image is transformed into a geometric graph and the layers to be detected will be given by its minimum-cost closed set. The main problem in this method is the selection of the appropriate cost functions associated to the graph, because of the variety of anomalies that images from patients might have. © 2013 Springer-Verlag Berlin Heidelberg.
Author supplied keywords
Cite
CITATION STYLE
González, A., Penedo, M. G., Vázquez, S. G., Novo, J., & Charlón, P. (2013). Cost function selection for a graph-based segmentation in OCT retinal images. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8112 LNCS, pp. 125–132). Springer Verlag. https://doi.org/10.1007/978-3-642-53862-9_17
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
