• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.3.2016.tde-24052016-150250
Document
Author
Full name
Guillermo Angel Perez López
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2015
Supervisor
Committee
Kim, Hae Yong (President)
Ferreira, Ademar
Hernandez, Emilio Del Moral
Hirata, Nina Sumiko Tomita
Kurashima, Celso Setsuo
Title in Portuguese
Casamento de modelos baseado em projeções radiais e circulares invariante a pontos de vista.
Keywords in Portuguese
Asift
Casamento de modelos
Ciratefi
Forapro
Imagem digital
Invariância a escala
Invariância afim
Mudança de iluminação
Padrões repetitivos
Pontos-chaves
Sift
Simulação de pontos de vista
Transformação afim
Abstract in Portuguese
Este trabalho aborda o problema de casamento entre duas imagens. Casamento de imagens pode ser do tipo casamento de modelos (template matching) ou casamento de pontos-chaves (keypoint matching). Estes algoritmos localizam uma região da primeira imagem numa segunda imagem. Nosso grupo desenvolveu dois algoritmos de casamento de modelos invariante por rotação, escala e translação denominados Ciratefi (Circula, radial and template matchings filter) e Forapro (Fourier coefficients of radial and circular projection). As características positivas destes algoritmos são a invariância a mudanças de brilho/contraste e robustez a padrões repetitivos. Na primeira parte desta tese, tornamos Ciratefi invariante a transformações afins, obtendo Aciratefi (Affine-ciratefi). Construímos um banco de imagens para comparar este algoritmo com Asift (Affine-scale invariant feature transform) e Aforapro (Affine-forapro). Asift é considerado atualmente o melhor algoritmo de casamento de imagens invariante afim, e Aforapro foi proposto em nossa dissertação de mestrado. Nossos resultados sugerem que Aciratefi supera Asift na presença combinada de padrões repetitivos, mudanças de brilho/contraste e mudanças de pontos de vista. Na segunda parte desta tese, construímos um algoritmo para filtrar casamentos de pontos-chaves, baseado num conceito que denominamos de coerência geométrica. Aplicamos esta filtragem no bem-conhecido algoritmo Sift (scale invariant feature transform), base do Asift. Avaliamos a nossa proposta no banco de imagens de Mikolajczyk. As taxas de erro obtidas são significativamente menores que as do Sift original.
Title in English
Viewpoint invariant template matching based in radial and circular proejction.
Keywords in English
Affine invariant
Asift,
Ciratefi
Forapro
Illumination changes
Keypoints
Repetitive patterns
Scale invariant
Sift
Template matching
Viewpoints simulation
Abstract in English
This work deals with image matching. Image matchings can be modeled as template matching or keypoints matching. These algorithms search a region of the first image in a second image. Our group has developed two template matching algorithms invariant by rotation, scale and translation called Ciratefi (circular, radial and template matching filter) and Forapro (Fourier coefficients of radial and circular projection). The positive characteristics of Ciratefi and Forapro are: the invariance to brightness/contrast changes and robustness to repetitive patterns. In the first part of this work, we make Ciratefi invariant to affine transformations, getting Aciratefi (Affine-ciratefi). We have built a dataset to compare Aciratefi with Asift (Affine-scale invariant feature transform) and Aforapro (Affine-forapro). Asift is currently considered the best affine invariant image matching algorithm, and Aforapro was proposed in our master's thesis. Our results suggest that Aciratefi overcome Asift in the combined presence of repetitive patterns, brightness/contrast and viewpoints changes. In the second part of this work, we filter keypoints matchings based on a concept that we call geometric coherence. We apply this filtering in the well-known algorithm Sift (scale invariant feature transform), the basis of Asift. We evaluate our proposal in the Mikolajczyk images database. The error rates obtained are significantly lower than those of the original Sift.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2016-06-06
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.