Connectivity-based segmentation of retinal vessels in eye fundus images

© Sociedad Española de Óptica. A new unsupervised method for segmentation of objects of diverse nature with the common feature of connectivity (e.g. branching trees or net-shaped objects) is proposed. A preferred application to the vasculature segmentation of retinal images has been illustrated usin...

Descripción completa

Detalles Bibliográficos
Autores: Rallo Capdevila, Miguel|||0000-0002-1842-5045, Millán Garcia-Varela, M. Sagrario|||0000-0001-6950-2373
Tipo de recurso: artículo
Fecha de publicación:2017
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/113942
Acceso en línea:https://hdl.handle.net/2117/113942
https://dx.doi.org/10.7149/OPA.50.4.49070
Access Level:acceso abierto
Palabra clave:Fundus oculi
Retina -- Imaging
Diagnostic imaging -- Digital techniques
Blood vessel segmentation
Computer-aided diagnosis
Digital image analysis
Eye fundus image
Retinal vasculature
Fons de l'ull
Retina -- Imatgeria
Imatgeria per al diagnòstic -- Tècniques digitals
Àrees temàtiques de la UPC::Ciències de la salut::Medicina::Oftalmologia
Àrees temàtiques de la UPC::Ciències de la visió
Descripción
Sumario:© Sociedad Española de Óptica. A new unsupervised method for segmentation of objects of diverse nature with the common feature of connectivity (e.g. branching trees or net-shaped objects) is proposed. A preferred application to the vasculature segmentation of retinal images has been illustrated using images from DRIVE database. In the pre-processing stage, the method overcomes the common problem of non-uniform illumination of eye fundus images. The method follows with an iterative algorithm that starts with a seed and adds, at each step, a new vessel segment connected to the previously segmented part. The result preserves the connectivity as a distinct feature of the retinal vessel tree. The segmentation performance is evaluated through common signal detection metrics: sensitivity, specificity and accuracy.