2004 International Conference on Image Processing, 2004. ICIP '04.
DOI: 10.1109/icip.2004.1421346
|View full text |Cite
|
Sign up to set email alerts
|

Fuzzy non-rigid motion estimation robust to rotation

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
3
0
1

Publication Types

Select...
2

Relationship

1
1

Authors

Journals

citations
Cited by 2 publications
(4 citation statements)
references
References 4 publications
0
3
0
1
Order By: Relevance
“…The main issue of this work is the constraint-forcing procedure on second stage, because this is the point where model restrictions must be applied. More details about the motion estimation algorithm can be found in [2,3,4]. The paper is organized as follows: section 2 presents the basic ideas to achieve a better image matching by using multiple models; section 3 presents comparative results to prove the validity of this approach and illustrate the estimation improvement; finally, section 4 contains the conclusions.…”
Section: Regularizationmentioning
confidence: 99%
See 2 more Smart Citations
“…The main issue of this work is the constraint-forcing procedure on second stage, because this is the point where model restrictions must be applied. More details about the motion estimation algorithm can be found in [2,3,4]. The paper is organized as follows: section 2 presents the basic ideas to achieve a better image matching by using multiple models; section 3 presents comparative results to prove the validity of this approach and illustrate the estimation improvement; finally, section 4 contains the conclusions.…”
Section: Regularizationmentioning
confidence: 99%
“…For each pixel in I(x, y) a Similarity Map (SM) ρ c (x, y) of (2N + 1)-pixel width is computed according to [2,3,4], which contains the similarity between a block of I(x, y) centered at the pixel c ≡ (c x , c y ) and a set of equal size blocks of J(x, y) in a (2N + 1) search length.…”
Section: Parametric Modelsmentioning
confidence: 99%
See 1 more Smart Citation
“…La aparición de nuevas tecnologías en adquisición de imágenes, en particular en el ámbito médico, demanda técnicas más avanzadas de registro de imagen para automatizar tareas de diagnóstico y tratamiento médico, en concreto, técnicas de estimación de movimiento no rígido y fusión de imágenes procedentes de diferentes fuentes. En los trabajos de Morales [169,170,171] y la Tesis Doctoral [168] se aborda el problema de la estimación de movimiento no rígido entre imágenes, con la propuesta de un método novel de registro de imagen que consta de dos etapas bien definidas: en la primera etapa se establece una correspondencia espacial difusa entre ambas imágenes, y en la segunda se obtiene, a partir del resultado anterior, los vectores de movimiento con la correspondencia espacial explícita mediante un método iterativo que implementa la regularización Tikhonov. El uso de regularización Tikhonov resulta de asumir que la correspondencia espacial a estimar es un mapa continuo y de carácter suave.…”
Section: Estimación De Movimiento No Rígido Mediante Modelos Deformablesunclassified