From Inpainting to Active Contours

Francois Bernard Lauze; Mads Nielsen

doi:10.1007/s11263-007-0088-2

From Inpainting to Active Contours

Francois Bernard Lauze, Mads Nielsen

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › peer review

3 Citationer (Scopus)

Abstract

Udgivelsesdato: August

Originalsprog	Engelsk
Tidsskrift	International Journal of Computer Vision
Vol/bind	79
Udgave nummer	1
Sider (fra-til)	31-43
Antal sider	14
ISSN	0920-5691
DOI	https://doi.org/10.1007/s11263-007-0088-2
Status	Udgivet - 2008

Emneord

Det Natur- og Biovidenskabelige Fakultet

Adgang til dokumentet

10.1007/s11263-007-0088-2

Citationsformater

@article{224aa3a096ca11dd86a6000ea68e967b,

title = "From Inpainting to Active Contours",

abstract = "Abstract Background subtraction is an elementary method for detection of foreground objects and their segmentations. Obviously it requires an observation image as well as a background one. In this work we attempt to remove the last requirement by reconstructing the background from the observation image and a guess on the location of the object to be segmented via variational inpainting method. A numerical evaluation of this reconstruction provides a {"}disocclusion measure{"} and the correct foreground segmentation region is expected to maximize this measure. This formulation is in fact an optimal control problem, where controls are shapes/regions and states are the corresponding inpaintings. Optimization of the disocclusion measure leads formally to a coupled contour evolution equation, an inpainting equation (the state equation) as well as a linear PDE depending on the inpainting (the adjoint state equation). The contour evolution is implemented in the framework of level sets. Finally, the proposed method is validated on various examples. We focus among others in the segmentation of calcified plaques observed in radiographs from human lumbar aortic regions. Keywords Segmentation - Inpainting - Active contours - Disocclusion - Adjoint methods - Variational methods ",

keywords = "Faculty of Science, Segmentation, Inpainting, Active contours, Disocclusion, Ajoint methods, Variational methods",

author = "Lauze, {Francois Bernard} and Mads Nielsen",

year = "2008",

doi = "10.1007/s11263-007-0088-2",

language = "English",

volume = "79",

pages = "31--43",

journal = "International Journal of Computer Vision",

issn = "0920-5691",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - From Inpainting to Active Contours

AU - Lauze, Francois Bernard

AU - Nielsen, Mads

PY - 2008

Y1 - 2008

N2 - Abstract Background subtraction is an elementary method for detection of foreground objects and their segmentations. Obviously it requires an observation image as well as a background one. In this work we attempt to remove the last requirement by reconstructing the background from the observation image and a guess on the location of the object to be segmented via variational inpainting method. A numerical evaluation of this reconstruction provides a "disocclusion measure" and the correct foreground segmentation region is expected to maximize this measure. This formulation is in fact an optimal control problem, where controls are shapes/regions and states are the corresponding inpaintings. Optimization of the disocclusion measure leads formally to a coupled contour evolution equation, an inpainting equation (the state equation) as well as a linear PDE depending on the inpainting (the adjoint state equation). The contour evolution is implemented in the framework of level sets. Finally, the proposed method is validated on various examples. We focus among others in the segmentation of calcified plaques observed in radiographs from human lumbar aortic regions. Keywords Segmentation - Inpainting - Active contours - Disocclusion - Adjoint methods - Variational methods

AB - Abstract Background subtraction is an elementary method for detection of foreground objects and their segmentations. Obviously it requires an observation image as well as a background one. In this work we attempt to remove the last requirement by reconstructing the background from the observation image and a guess on the location of the object to be segmented via variational inpainting method. A numerical evaluation of this reconstruction provides a "disocclusion measure" and the correct foreground segmentation region is expected to maximize this measure. This formulation is in fact an optimal control problem, where controls are shapes/regions and states are the corresponding inpaintings. Optimization of the disocclusion measure leads formally to a coupled contour evolution equation, an inpainting equation (the state equation) as well as a linear PDE depending on the inpainting (the adjoint state equation). The contour evolution is implemented in the framework of level sets. Finally, the proposed method is validated on various examples. We focus among others in the segmentation of calcified plaques observed in radiographs from human lumbar aortic regions. Keywords Segmentation - Inpainting - Active contours - Disocclusion - Adjoint methods - Variational methods

KW - Faculty of Science

KW - Segmentation

KW - Inpainting

KW - Active contours

KW - Disocclusion

KW - Ajoint methods

KW - Variational methods

U2 - 10.1007/s11263-007-0088-2

DO - 10.1007/s11263-007-0088-2

M3 - Journal article

VL - 79

SP - 31

EP - 43

JO - International Journal of Computer Vision

JF - International Journal of Computer Vision

SN - 0920-5691

IS - 1

ER -