Local and global uncertainty in binary tomographic reconstruction

László G. Varga, L. Nyúl, Antal Nagy, Péter Balázs

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In binary tomography the goal is to reconstruct the inner structure of homogeneous objects from their projections. This is usually required from a low number of projections, which are also likely to be affected by noise and measurement errors. In general, the distorted and incomplete projection data holds insufficient information for the correct reconstruction of the original object. In this paper, we describe two methods for approximating the local uncertainty of the reconstructions, i.e., identifying how the information stored in the projections determine each part of the reconstructed image. These methods can measure the uncertainty of the reconstruction without any knowledge from the original object itself. Moreover, we provide a global uncertainty measure that can assess the information content of a projection set and predict the error to be expected in the reconstruction of a homogeneous object. We also give an experimental evaluation of our proposed methods, mention some of their possible applications, and describe how the uncertainty measure can be used to improve the performance of the DART reconstruction algorithm.

Original languageEnglish
Pages (from-to)52-62
Number of pages11
JournalComputer Vision and Image Understanding
Volume129
DOIs
Publication statusPublished - 2014

Fingerprint

Measurement errors
Tomography
Uncertainty

Keywords

  • Binary tomography
  • DART
  • Non-destructive testing
  • Optimization
  • Reconstruction
  • Uncertainty

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Local and global uncertainty in binary tomographic reconstruction. / Varga, László G.; Nyúl, L.; Nagy, Antal; Balázs, Péter.

In: Computer Vision and Image Understanding, Vol. 129, 2014, p. 52-62.

Research output: Contribution to journalArticle

Varga, László G. ; Nyúl, L. ; Nagy, Antal ; Balázs, Péter. / Local and global uncertainty in binary tomographic reconstruction. In: Computer Vision and Image Understanding. 2014 ; Vol. 129. pp. 52-62.
@article{490cdf278db1463cb81da500a0251e93,
title = "Local and global uncertainty in binary tomographic reconstruction",
abstract = "In binary tomography the goal is to reconstruct the inner structure of homogeneous objects from their projections. This is usually required from a low number of projections, which are also likely to be affected by noise and measurement errors. In general, the distorted and incomplete projection data holds insufficient information for the correct reconstruction of the original object. In this paper, we describe two methods for approximating the local uncertainty of the reconstructions, i.e., identifying how the information stored in the projections determine each part of the reconstructed image. These methods can measure the uncertainty of the reconstruction without any knowledge from the original object itself. Moreover, we provide a global uncertainty measure that can assess the information content of a projection set and predict the error to be expected in the reconstruction of a homogeneous object. We also give an experimental evaluation of our proposed methods, mention some of their possible applications, and describe how the uncertainty measure can be used to improve the performance of the DART reconstruction algorithm.",
keywords = "Binary tomography, DART, Non-destructive testing, Optimization, Reconstruction, Uncertainty",
author = "Varga, {L{\'a}szl{\'o} G.} and L. Ny{\'u}l and Antal Nagy and P{\'e}ter Bal{\'a}zs",
year = "2014",
doi = "10.1016/j.cviu.2014.05.006",
language = "English",
volume = "129",
pages = "52--62",
journal = "Computer Vision and Image Understanding",
issn = "1077-3142",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Local and global uncertainty in binary tomographic reconstruction

AU - Varga, László G.

AU - Nyúl, L.

AU - Nagy, Antal

AU - Balázs, Péter

PY - 2014

Y1 - 2014

N2 - In binary tomography the goal is to reconstruct the inner structure of homogeneous objects from their projections. This is usually required from a low number of projections, which are also likely to be affected by noise and measurement errors. In general, the distorted and incomplete projection data holds insufficient information for the correct reconstruction of the original object. In this paper, we describe two methods for approximating the local uncertainty of the reconstructions, i.e., identifying how the information stored in the projections determine each part of the reconstructed image. These methods can measure the uncertainty of the reconstruction without any knowledge from the original object itself. Moreover, we provide a global uncertainty measure that can assess the information content of a projection set and predict the error to be expected in the reconstruction of a homogeneous object. We also give an experimental evaluation of our proposed methods, mention some of their possible applications, and describe how the uncertainty measure can be used to improve the performance of the DART reconstruction algorithm.

AB - In binary tomography the goal is to reconstruct the inner structure of homogeneous objects from their projections. This is usually required from a low number of projections, which are also likely to be affected by noise and measurement errors. In general, the distorted and incomplete projection data holds insufficient information for the correct reconstruction of the original object. In this paper, we describe two methods for approximating the local uncertainty of the reconstructions, i.e., identifying how the information stored in the projections determine each part of the reconstructed image. These methods can measure the uncertainty of the reconstruction without any knowledge from the original object itself. Moreover, we provide a global uncertainty measure that can assess the information content of a projection set and predict the error to be expected in the reconstruction of a homogeneous object. We also give an experimental evaluation of our proposed methods, mention some of their possible applications, and describe how the uncertainty measure can be used to improve the performance of the DART reconstruction algorithm.

KW - Binary tomography

KW - DART

KW - Non-destructive testing

KW - Optimization

KW - Reconstruction

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=84908359926&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84908359926&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2014.05.006

DO - 10.1016/j.cviu.2014.05.006

M3 - Article

AN - SCOPUS:84908359926

VL - 129

SP - 52

EP - 62

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

ER -