Mostrar el registro sencillo del ítem

dc.contributor.advisorLópez-Rubio, Ezequiel 
dc.contributor.advisorRoé Vellvé, Nuria
dc.contributor.authorThurnhofer Hemsi, Karl Khader
dc.contributor.otherLenguajes y Ciencias de la Computaciónes_ES
dc.date.accessioned2021-05-12T09:16:33Z
dc.date.available2021-05-12T09:16:33Z
dc.date.created2020-10-01
dc.date.issued2021-05-12
dc.date.submitted2021-03-24
dc.identifier.urihttps://hdl.handle.net/10630/21791
dc.descriptionThe third part is exclusively dedicated to the super-resolution of Magnetic Resonance Images. In one of these works, an algorithm based on the random shifting technique is developed. Besides, we studied noise removal and resolution enhancement simultaneously. To end, the cost function of deep networks has been modified by different combinations of norms in order to improve their training. Finally, the general conclusions of the research are presented and discussed, as well as the possible future research lines that are able to make use of the results obtained in this Ph.D. thesis.es_ES
dc.description.abstractThis Ph.D. thesis is about image processing by computational intelligence techniques. Firstly, a general overview of this book is carried out, where the motivation, the hypothesis, the objectives, and the methodology employed are described. The use and analysis of different mathematical norms will be our goal. After that, state of the art focused on the applications of the image processing proposals is presented. In addition, the fundamentals of the image modalities, with particular attention to magnetic resonance, and the learning techniques used in this research, mainly based on neural networks, are summarized. To end up, the mathematical framework on which this work is based on, 𝓁ₚ-norms, is defined. Three different parts associated with image processing techniques follow. The first non-introductory part of this book collects the developments which are about image segmentation. Two of them are applications for video surveillance tasks and try to model the background of a scenario using a specific camera. The other work is centered on the medical field, where the goal of segmenting diabetic wounds of a very heterogeneous dataset is addressed. The second part is focused on the optimization and implementation of new models for curve and surface fitting in two and three dimensions, respectively. The first work presents a parabola fitting algorithm based on the measurement of the distances of the interior and exterior points to the focus and the directrix. The second work changes to an ellipse shape, and it ensembles the information of multiple fitting methods. Last, the ellipsoid problem is addressed in a similar way to the parabola.es_ES
dc.language.isoenges_ES
dc.publisherUMA Editoriales_ES
dc.rightsinfo:eu-repo/semantics/openAccesses_ES
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectProcesado de imágenes - Técnicas digitales - Tesis doctoraleses_ES
dc.subjectRedes neuronales (Informática) - Tesis doctoraleses_ES
dc.subject.otherInteligencia Artificiales_ES
dc.titleRobust computational intelligence techniques for visual information processinges_ES
dc.typeinfo:eu-repo/semantics/doctoralThesises_ES
dc.centroE.T.S.I. Informáticaes_ES
dc.rights.ccAttribution-NonCommercial-NoDerivatives 4.0 Internacional*


Ficheros en el ítem

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem

Attribution-NonCommercial-NoDerivatives 4.0 Internacional
Excepto si se señala otra cosa, la licencia del ítem se describe como Attribution-NonCommercial-NoDerivatives 4.0 Internacional