Brownian Warps for Non-Rigid Registration

Mads Nielsen; Peter Johansen; Andrew D. Jackson; Benny Elley Lautrup; Søren Hauberg

doi:10.1007/s10851-008-0083-4

Brownian Warps for Non-Rigid Registration

Mads Nielsen, Peter Johansen, Andrew D. Jackson, Benny Elley Lautrup, Søren Hauberg

6 Citationer (Scopus)

Abstract

A Brownian motion model in the group of diffeomorphisms has been introduced as inducing a least committed prior on warps. This prior is source-destination symmetric, fulfills a natural semi-group property for warps, and with probability 1 creates invertible warps. Using this as a least committed prior, we formulate a Partial Differential Equation for obtaining the maximally likely warp given matching constraints derived from the images. We solve for the free boundary conditions, and the bias toward smaller areas in the finite domain setting. Furthermore, we demonstrate the technique on 2D images, and show that the obtained warps are also in practice source-destination symmetric and in an example on X-ray spine registration provides extrapolations from landmark point superior to those of spline solutions.
Udgivelsesdato: July

Originalsprog	Engelsk
Tidsskrift	Journal of Mathematical Imaging and Vision
Vol/bind	31
Udgave nummer	2-3
Sider (fra-til)	221-231
Antal sider	11
ISSN	0924-9907
DOI	https://doi.org/10.1007/s10851-008-0083-4
Status	Udgivet - 2008

Emneord

Det Natur- og Biovidenskabelige Fakultet

Adgang til dokumentet

10.1007/s10851-008-0083-4

Citationsformater

@article{953af430cc7d11dd9473000ea68e967b,

title = "Brownian Warps for Non-Rigid Registration",

abstract = "A Brownian motion model in the group of diffeomorphisms has been introduced as inducing a least committed prior on warps. This prior is source-destination symmetric, fulfills a natural semi-group property for warps, and with probability 1 creates invertible warps. Using this as a least committed prior, we formulate a Partial Differential Equation for obtaining the maximally likely warp given matching constraints derived from the images. We solve for the free boundary conditions, and the bias toward smaller areas in the finite domain setting. Furthermore, we demonstrate the technique on 2D images, and show that the obtained warps are also in practice source-destination symmetric and in an example on X-ray spine registration provides extrapolations from landmark point superior to those of spline solutions. Udgivelsesdato: July",

keywords = "Faculty of Science, DIKU, Image Group, Non-rigid registration, Brownian motion, Central limit theorem, Invariance",

author = "Mads Nielsen and Peter Johansen and Jackson, {Andrew D.} and Lautrup, {Benny Elley} and S{\o}ren Hauberg",

year = "2008",

doi = "10.1007/s10851-008-0083-4",

language = "English",

volume = "31",

pages = "221--231",

journal = "Journal of Mathematical Imaging and Vision",

issn = "0924-9907",

publisher = "Springer",

number = "2-3",

}

TY - JOUR

T1 - Brownian Warps for Non-Rigid Registration

AU - Nielsen, Mads

AU - Johansen, Peter

AU - Jackson, Andrew D.

AU - Lautrup, Benny Elley

AU - Hauberg, Søren

PY - 2008

Y1 - 2008

N2 - A Brownian motion model in the group of diffeomorphisms has been introduced as inducing a least committed prior on warps. This prior is source-destination symmetric, fulfills a natural semi-group property for warps, and with probability 1 creates invertible warps. Using this as a least committed prior, we formulate a Partial Differential Equation for obtaining the maximally likely warp given matching constraints derived from the images. We solve for the free boundary conditions, and the bias toward smaller areas in the finite domain setting. Furthermore, we demonstrate the technique on 2D images, and show that the obtained warps are also in practice source-destination symmetric and in an example on X-ray spine registration provides extrapolations from landmark point superior to those of spline solutions. Udgivelsesdato: July

AB - A Brownian motion model in the group of diffeomorphisms has been introduced as inducing a least committed prior on warps. This prior is source-destination symmetric, fulfills a natural semi-group property for warps, and with probability 1 creates invertible warps. Using this as a least committed prior, we formulate a Partial Differential Equation for obtaining the maximally likely warp given matching constraints derived from the images. We solve for the free boundary conditions, and the bias toward smaller areas in the finite domain setting. Furthermore, we demonstrate the technique on 2D images, and show that the obtained warps are also in practice source-destination symmetric and in an example on X-ray spine registration provides extrapolations from landmark point superior to those of spline solutions. Udgivelsesdato: July

KW - Faculty of Science

KW - DIKU

KW - Image Group

KW - Non-rigid registration

KW - Brownian motion

KW - Central limit theorem

KW - Invariance

U2 - 10.1007/s10851-008-0083-4

DO - 10.1007/s10851-008-0083-4

M3 - Journal article

SN - 0924-9907

VL - 31

SP - 221

EP - 231

JO - Journal of Mathematical Imaging and Vision

JF - Journal of Mathematical Imaging and Vision

IS - 2-3

ER -