Automatic wide-eld registration and mosaicking of noisy OCTA images using template matching and dierential evolution

Moya García, Alejandro. (2020). Automatic wide-eld registration and mosaicking of noisy OCTA images using template matching and dierential evolution Master Thesis, Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia Artificial

Ficheros (Some files may be inaccessible until you login with your e-spacio credentials)
Nombre Descripción Tipo MIME Size
Moya_Garcia_Alejandro_TFM.pdf Moya_Garcia_Alejandro_TFM.pdf application/pdf 24.88MB

Título Automatic wide-eld registration and mosaicking of noisy OCTA images using template matching and dierential evolution
Autor(es) Moya García, Alejandro
Abstract Optical Coherence Tomography Angiography (OCTA) is a novel non-invasive ophthalmological technique used to perform a detailed examination of the eye fundus vascularity. However, each of the images obtained by this technique only cover a small retinal area. Thus, ophthalmologists have to take complementary images of the eye fundus from dierent angles in order to obtain a complete visualization of patients' eye fundus. In particular, each set of images must be manually registered by a clinician, being a tedious and time-consuming process. In this work, we propose an approach based on template matching and dierential evolution to automatically register a set of OCTA images characterized by containing noise and artifacts. The proposed method is divided into three main steps. First, a preprocessing step used to extract the main vascular network is applied on every image. Then, an algorithm based on dierential evolution is run on every 2-combination of OCTA images in order to nd the best overlap between them. Finally, a greedy algorithm iteratively selects the best pairs of images (according to their tness) to create the complete mosaic. The proposed method was evaluated via the registration of several sets of OCTA images with the purpose of building their associated mosaics. Results show that our approach is robust and able to achieve a good approximation to the optimal mosaic.
Notas adicionales Trabajo de Fin de Máster Universitario en Investigación en Inteligencia Artificial. UNED
Materia(s) Ingeniería Informática
Palabra clave OCTA
retinal imaging
template matching
differential evolution
mosaicking
Editor(es) Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia Artificial
Director/Tutor Carmona Suárez, Enrique J.
Fecha 2020-06
Formato application/pdf
Identificador bibliuned:master-ETSInformatica-IIA-Amoya
http://e-spacio.uned.es/fez/view/bibliuned:master-ETSInformatica-IIA-Amoya
Idioma eng
Versión de la publicación acceptedVersion
Nivel de acceso y licencia http://creativecommons.org/licenses/by-nc-nd/4.0
info:eu-repo/semantics/openAccess
Tipo de recurso master Thesis
Tipo de acceso Acceso abierto

 
Versiones
Versión Tipo de filtro
Contador de citas: Google Scholar Search Google Scholar
Estadísticas de acceso: 430 Visitas, 197 Descargas  -  Estadísticas en detalle
Creado: Fri, 24 Sep 2021, 19:32:44 CET