Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Evaluation of stitching-induced strain uncertainties in extended-field-of-view in situ X-ray tomography

Tambe, Indrajeet ; Laçaj, Endri LU ; Olsson, Pär A.T. LU ; Hallberg, Håkan LU orcid ; Engqvist, Jonas LU orcid ; Orbulov, Imre Norbert and Hektor, Johan LU (2026) In Materials Characterization 240.
Abstract

High-resolution synchrotron X-ray tomography combined with digital volume correlation (DVC) enables characterization of three-dimensional strain fields. When local tomography is used to investigate specimens larger than the field of view (FOV), the resulting subvolumes must be stitched together to form a single volume, thereby introducing strain uncertainties in DVC analyses. The present work quantifies such strain uncertainties obtained from extended-FOV synchrotron X-ray tomography of an Al99.5 metal matrix syntactic foam. To achieve this, zero-strain repeated scans were performed on the specimen. Two local DVC strategies, regular-grid DVC and discrete DVC, were applied to the stitched volumes to quantify the spatial distribution and... (More)

High-resolution synchrotron X-ray tomography combined with digital volume correlation (DVC) enables characterization of three-dimensional strain fields. When local tomography is used to investigate specimens larger than the field of view (FOV), the resulting subvolumes must be stitched together to form a single volume, thereby introducing strain uncertainties in DVC analyses. The present work quantifies such strain uncertainties obtained from extended-FOV synchrotron X-ray tomography of an Al99.5 metal matrix syntactic foam. To achieve this, zero-strain repeated scans were performed on the specimen. Two local DVC strategies, regular-grid DVC and discrete DVC, were applied to the stitched volumes to quantify the spatial distribution and magnitude of strain uncertainties. Both approaches confirm a consistent spatial hierarchy of strain uncertainties. Non-overlapping regions exhibit uncertainties of approximately 0.06%, while overlapping regions show uncertainties of approximately 0.17% (Frobenius norm of the Green–Lagrangian strain tensor). The spatial distribution of strain uncertainty in the stitched volume indicates that uncertainties in overlapping regions are two to four times higher compared to non-overlapping regions. The zero-strain stitching uncertainty quantification method demonstrated here is recommended as a standard pre-characterization step for any DVC study involving stitched tomographic volumes.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Digital volume correlation, Strain uncertainty quantification, Synchrotron X-ray tomography, Volume stitching
in
Materials Characterization
volume
240
article number
116852
publisher
Elsevier
external identifiers
  • scopus:105048471659
ISSN
1044-5803
DOI
10.1016/j.matchar.2026.116852
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2026 The Authors
id
309383e4-2395-4653-82da-fd320385bebf
date added to LUP
2026-09-07 10:10:58
date last changed
2026-09-08 11:46:31
@article{309383e4-2395-4653-82da-fd320385bebf,
  abstract     = {{<p>High-resolution synchrotron X-ray tomography combined with digital volume correlation (DVC) enables characterization of three-dimensional strain fields. When local tomography is used to investigate specimens larger than the field of view (FOV), the resulting subvolumes must be stitched together to form a single volume, thereby introducing strain uncertainties in DVC analyses. The present work quantifies such strain uncertainties obtained from extended-FOV synchrotron X-ray tomography of an Al99.5 metal matrix syntactic foam. To achieve this, zero-strain repeated scans were performed on the specimen. Two local DVC strategies, regular-grid DVC and discrete DVC, were applied to the stitched volumes to quantify the spatial distribution and magnitude of strain uncertainties. Both approaches confirm a consistent spatial hierarchy of strain uncertainties. Non-overlapping regions exhibit uncertainties of approximately 0.06%, while overlapping regions show uncertainties of approximately 0.17% (Frobenius norm of the Green–Lagrangian strain tensor). The spatial distribution of strain uncertainty in the stitched volume indicates that uncertainties in overlapping regions are two to four times higher compared to non-overlapping regions. The zero-strain stitching uncertainty quantification method demonstrated here is recommended as a standard pre-characterization step for any DVC study involving stitched tomographic volumes.</p>}},
  author       = {{Tambe, Indrajeet and Laçaj, Endri and Olsson, Pär A.T. and Hallberg, Håkan and Engqvist, Jonas and Orbulov, Imre Norbert and Hektor, Johan}},
  issn         = {{1044-5803}},
  keywords     = {{Digital volume correlation; Strain uncertainty quantification; Synchrotron X-ray tomography; Volume stitching}},
  language     = {{eng}},
  publisher    = {{Elsevier}},
  series       = {{Materials Characterization}},
  title        = {{Evaluation of stitching-induced strain uncertainties in extended-field-of-view in situ X-ray tomography}},
  url          = {{http://dx.doi.org/10.1016/j.matchar.2026.116852}},
  doi          = {{10.1016/j.matchar.2026.116852}},
  volume       = {{240}},
  year         = {{2026}},
}