Estimation of linear deformations of 2D and 3D fuzzy objects

Attila Tanács, Joakim Lindblad, Nataša Sladoje, Z. Kato

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Registration is a fundamental task in image processing, it is used to determine geometric correspondences between images taken at different times and/or from different viewpoints. Here we propose a general framework in n-dimensions to solve binary shape/object matching problems without the need of establishing additional point or other type of correspondences. The approach is based on generating and solving polynomial systems of equations. We also propose an extension which, provided that a suitable segmentation method can produce a fuzzy border representation, further increases the registration precision. Via numerous synthetic and real test we examine the different solution techniques of the polynomial systems of equations. We take into account a direct analytical, an iterative least-squares, and a combined method. Iterative and combined approaches produce the most precise results. Comparison is made against competing methods for rigid-body problems. Our method is orders of magnitude faster and is able to recover alignment regardless of the magnitude of the deformation compared to the narrow capture range of others. The applicability of the proposed methods is demonstrated on real X-ray images of hip replacement implants and 3D CT volumes of the pelvic area. Since the images must be parsed through only once, our approach is especially suitable for solving registration problems of large images.

Original languageEnglish
Pages (from-to)1387-1399
Number of pages13
JournalPattern Recognition
Volume48
Issue number4
DOIs
Publication statusPublished - Apr 1 2015

Fingerprint

Polynomials
Image processing
X rays

Keywords

  • Affine transformation
  • Fuzzy representation
  • Registration
  • System of polynomial equations

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Estimation of linear deformations of 2D and 3D fuzzy objects. / Tanács, Attila; Lindblad, Joakim; Sladoje, Nataša; Kato, Z.

In: Pattern Recognition, Vol. 48, No. 4, 01.04.2015, p. 1387-1399.

Research output: Contribution to journalArticle

Tanács, Attila ; Lindblad, Joakim ; Sladoje, Nataša ; Kato, Z. / Estimation of linear deformations of 2D and 3D fuzzy objects. In: Pattern Recognition. 2015 ; Vol. 48, No. 4. pp. 1387-1399.
@article{1263427c9dcd4e13a93b478d3acd297a,
title = "Estimation of linear deformations of 2D and 3D fuzzy objects",
abstract = "Registration is a fundamental task in image processing, it is used to determine geometric correspondences between images taken at different times and/or from different viewpoints. Here we propose a general framework in n-dimensions to solve binary shape/object matching problems without the need of establishing additional point or other type of correspondences. The approach is based on generating and solving polynomial systems of equations. We also propose an extension which, provided that a suitable segmentation method can produce a fuzzy border representation, further increases the registration precision. Via numerous synthetic and real test we examine the different solution techniques of the polynomial systems of equations. We take into account a direct analytical, an iterative least-squares, and a combined method. Iterative and combined approaches produce the most precise results. Comparison is made against competing methods for rigid-body problems. Our method is orders of magnitude faster and is able to recover alignment regardless of the magnitude of the deformation compared to the narrow capture range of others. The applicability of the proposed methods is demonstrated on real X-ray images of hip replacement implants and 3D CT volumes of the pelvic area. Since the images must be parsed through only once, our approach is especially suitable for solving registration problems of large images.",
keywords = "Affine transformation, Fuzzy representation, Registration, System of polynomial equations",
author = "Attila Tan{\'a}cs and Joakim Lindblad and Nataša Sladoje and Z. Kato",
year = "2015",
month = "4",
day = "1",
doi = "10.1016/j.patcog.2014.10.006",
language = "English",
volume = "48",
pages = "1387--1399",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "4",

}

TY - JOUR

T1 - Estimation of linear deformations of 2D and 3D fuzzy objects

AU - Tanács, Attila

AU - Lindblad, Joakim

AU - Sladoje, Nataša

AU - Kato, Z.

PY - 2015/4/1

Y1 - 2015/4/1

N2 - Registration is a fundamental task in image processing, it is used to determine geometric correspondences between images taken at different times and/or from different viewpoints. Here we propose a general framework in n-dimensions to solve binary shape/object matching problems without the need of establishing additional point or other type of correspondences. The approach is based on generating and solving polynomial systems of equations. We also propose an extension which, provided that a suitable segmentation method can produce a fuzzy border representation, further increases the registration precision. Via numerous synthetic and real test we examine the different solution techniques of the polynomial systems of equations. We take into account a direct analytical, an iterative least-squares, and a combined method. Iterative and combined approaches produce the most precise results. Comparison is made against competing methods for rigid-body problems. Our method is orders of magnitude faster and is able to recover alignment regardless of the magnitude of the deformation compared to the narrow capture range of others. The applicability of the proposed methods is demonstrated on real X-ray images of hip replacement implants and 3D CT volumes of the pelvic area. Since the images must be parsed through only once, our approach is especially suitable for solving registration problems of large images.

AB - Registration is a fundamental task in image processing, it is used to determine geometric correspondences between images taken at different times and/or from different viewpoints. Here we propose a general framework in n-dimensions to solve binary shape/object matching problems without the need of establishing additional point or other type of correspondences. The approach is based on generating and solving polynomial systems of equations. We also propose an extension which, provided that a suitable segmentation method can produce a fuzzy border representation, further increases the registration precision. Via numerous synthetic and real test we examine the different solution techniques of the polynomial systems of equations. We take into account a direct analytical, an iterative least-squares, and a combined method. Iterative and combined approaches produce the most precise results. Comparison is made against competing methods for rigid-body problems. Our method is orders of magnitude faster and is able to recover alignment regardless of the magnitude of the deformation compared to the narrow capture range of others. The applicability of the proposed methods is demonstrated on real X-ray images of hip replacement implants and 3D CT volumes of the pelvic area. Since the images must be parsed through only once, our approach is especially suitable for solving registration problems of large images.

KW - Affine transformation

KW - Fuzzy representation

KW - Registration

KW - System of polynomial equations

UR - http://www.scopus.com/inward/record.url?scp=84920694446&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920694446&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2014.10.006

DO - 10.1016/j.patcog.2014.10.006

M3 - Article

VL - 48

SP - 1387

EP - 1399

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 4

ER -