Non-linear 2D and 3D Registration Using Block-Matching and B-Splines

Heike Hufnagel; Xavier Pennec; Grégoire Malandain; Heinz Handels; Nicholas Ayache

doi:10.1007/3-540-26431-0_67

Non-linear 2D and 3D Registration Using Block-Matching and B-Splines

Heike Hufnagel, Xavier Pennec, Grégoire Malandain, Heinz Handels, Nicholas Ayache

Institute of Medical Informatics

INRIA Sophia Antipolis

Abstract

We developed a non-linear registration technique to align images that feature anatomical variabilities. The algorithm is based on a block-matching technique that identifies a sparse displacement vector field from the iconic features of two images. Subsequently, the displacement vectors are used as sampling points to estimate a parametric non-linear transformation that is represented by a tensor product of B-Splines. The B-Spline transformation estimation approximates the correspondences while minimizing the second order derivatives in the transformation function. The block-matching and the transformation estimation are then iterated in a multiscale framework to improve robustness and accuracy. Experiments on 2D histological slices and 3D MR images show qualitatively good results.

Original language	English
Title of host publication	Bildverarbeitung für die Medizin 2005
Number of pages	5
Place of Publication	Berlin, Heidelberg
Publisher	Springer Verlag
Publication date	01.12.2005
Pages	325-329
ISBN (Print)	978-3-540-25052-4
ISBN (Electronic)	978-3-540-26431-6
DOIs	https://doi.org/10.1007/3-540-26431-0_67
Publication status	Published - 01.12.2005
Event	Workshops Bildverarbeitung fur die Medizin 2005 - Heidelberg, Germany Duration: 13.03.2005 → 15.03.2005 Conference number: 97569

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/3-540-26431-0_67

Cite this

@inproceedings{c4e40268efc043e098274c18de8fb5f6,

title = "Non-linear 2D and 3D Registration Using Block-Matching and B-Splines",

abstract = "We developed a non-linear registration technique to align images that feature anatomical variabilities. The algorithm is based on a block-matching technique that identifies a sparse displacement vector field from the iconic features of two images. Subsequently, the displacement vectors are used as sampling points to estimate a parametric non-linear transformation that is represented by a tensor product of B-Splines. The B-Spline transformation estimation approximates the correspondences while minimizing the second order derivatives in the transformation function. The block-matching and the transformation estimation are then iterated in a multiscale framework to improve robustness and accuracy. Experiments on 2D histological slices and 3D MR images show qualitatively good results.",

author = "Heike Hufnagel and Xavier Pennec and Gr{\'e}goire Malandain and Heinz Handels and Nicholas Ayache",

year = "2005",

month = dec,

day = "1",

doi = "10.1007/3-540-26431-0_67",

language = "English",

isbn = "978-3-540-25052-4",

series = "Informatik aktuell",

publisher = "Springer Verlag",

pages = "325--329",

booktitle = "Bildverarbeitung f{\"u}r die Medizin 2005",

note = "Workshops Bildverarbeitung fur die Medizin 2005 , BVM 2005 ; Conference date: 13-03-2005 Through 15-03-2005",

}

Hufnagel, H, Pennec, X, Malandain, G, Handels, H & Ayache, N 2005, Non-linear 2D and 3D Registration Using Block-Matching and B-Splines. in Bildverarbeitung für die Medizin 2005 . Informatik aktuell, Springer Verlag, Berlin, Heidelberg, pp. 325-329, Workshops Bildverarbeitung fur die Medizin 2005 , Heidelberg, Baden-Württemberg, Germany, 13.03.05. https://doi.org/10.1007/3-540-26431-0_67

TY - GEN

T1 - Non-linear 2D and 3D Registration Using Block-Matching and B-Splines

AU - Hufnagel, Heike

AU - Pennec, Xavier

AU - Malandain, Grégoire

AU - Handels, Heinz

AU - Ayache, Nicholas

N1 - Conference code: 97569

PY - 2005/12/1

Y1 - 2005/12/1

N2 - We developed a non-linear registration technique to align images that feature anatomical variabilities. The algorithm is based on a block-matching technique that identifies a sparse displacement vector field from the iconic features of two images. Subsequently, the displacement vectors are used as sampling points to estimate a parametric non-linear transformation that is represented by a tensor product of B-Splines. The B-Spline transformation estimation approximates the correspondences while minimizing the second order derivatives in the transformation function. The block-matching and the transformation estimation are then iterated in a multiscale framework to improve robustness and accuracy. Experiments on 2D histological slices and 3D MR images show qualitatively good results.

AB - We developed a non-linear registration technique to align images that feature anatomical variabilities. The algorithm is based on a block-matching technique that identifies a sparse displacement vector field from the iconic features of two images. Subsequently, the displacement vectors are used as sampling points to estimate a parametric non-linear transformation that is represented by a tensor product of B-Splines. The B-Spline transformation estimation approximates the correspondences while minimizing the second order derivatives in the transformation function. The block-matching and the transformation estimation are then iterated in a multiscale framework to improve robustness and accuracy. Experiments on 2D histological slices and 3D MR images show qualitatively good results.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-84879915390&origin=inward&txGid=9244a21aebbcad8a49147bd8599e5f3e

U2 - 10.1007/3-540-26431-0_67

DO - 10.1007/3-540-26431-0_67

M3 - Conference contribution

SN - 978-3-540-25052-4

T3 - Informatik aktuell

SP - 325

EP - 329

BT - Bildverarbeitung für die Medizin 2005

PB - Springer Verlag

CY - Berlin, Heidelberg

T2 - Workshops Bildverarbeitung fur die Medizin 2005

Y2 - 13 March 2005 through 15 March 2005

ER -

Non-linear 2D and 3D Registration Using Block-Matching and B-Splines

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this