Cargando…

Deep-learning-based ring artifact correction for tomographic reconstruction

X-ray tomography has been widely used in various research fields thanks to its capability of observing 3D structures with high resolution non-destructively. However, due to the nonlinearity and inconsistency of detector pixels, ring artifacts usually appear in tomographic reconstruction, which may c...

Descripción completa

Detalles Bibliográficos
Autores principales: Fu, Tianyu, Wang, Yan, Zhang, Kai, Zhang, Jin, Wang, Shanfeng, Huang, Wanxia, Wang, Yaling, Yao, Chunxia, Zhou, Chenpeng, Yuan, Qingxi
Formato: Online Artículo Texto
Lenguaje:English
Publicado: International Union of Crystallography 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10161896/
https://www.ncbi.nlm.nih.gov/pubmed/36897392
http://dx.doi.org/10.1107/S1600577523000917
_version_ 1785037591227138048
author Fu, Tianyu
Wang, Yan
Zhang, Kai
Zhang, Jin
Wang, Shanfeng
Huang, Wanxia
Wang, Yaling
Yao, Chunxia
Zhou, Chenpeng
Yuan, Qingxi
author_facet Fu, Tianyu
Wang, Yan
Zhang, Kai
Zhang, Jin
Wang, Shanfeng
Huang, Wanxia
Wang, Yaling
Yao, Chunxia
Zhou, Chenpeng
Yuan, Qingxi
author_sort Fu, Tianyu
collection PubMed
description X-ray tomography has been widely used in various research fields thanks to its capability of observing 3D structures with high resolution non-destructively. However, due to the nonlinearity and inconsistency of detector pixels, ring artifacts usually appear in tomographic reconstruction, which may compromise image quality and cause nonuniform bias. This study proposes a new ring artifact correction method based on the residual neural network (ResNet) for X-ray tomography. The artifact correction network uses complementary information of each wavelet coefficient and a residual mechanism of the residual block to obtain high-precision artifacts through low operation costs. In addition, a regularization term is used to accurately extract stripe artifacts in sinograms, so that the network can better preserve image details while accurately separating artifacts. When applied to simulation and experimental data, the proposed method shows a good suppression of ring artifacts. To solve the problem of insufficient training data, ResNet is trained through the transfer learning strategy, which brings advantages of robustness, versatility and low computing cost.
format Online
Article
Text
id pubmed-10161896
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher International Union of Crystallography
record_format MEDLINE/PubMed
spelling pubmed-101618962023-05-06 Deep-learning-based ring artifact correction for tomographic reconstruction Fu, Tianyu Wang, Yan Zhang, Kai Zhang, Jin Wang, Shanfeng Huang, Wanxia Wang, Yaling Yao, Chunxia Zhou, Chenpeng Yuan, Qingxi J Synchrotron Radiat Research Papers X-ray tomography has been widely used in various research fields thanks to its capability of observing 3D structures with high resolution non-destructively. However, due to the nonlinearity and inconsistency of detector pixels, ring artifacts usually appear in tomographic reconstruction, which may compromise image quality and cause nonuniform bias. This study proposes a new ring artifact correction method based on the residual neural network (ResNet) for X-ray tomography. The artifact correction network uses complementary information of each wavelet coefficient and a residual mechanism of the residual block to obtain high-precision artifacts through low operation costs. In addition, a regularization term is used to accurately extract stripe artifacts in sinograms, so that the network can better preserve image details while accurately separating artifacts. When applied to simulation and experimental data, the proposed method shows a good suppression of ring artifacts. To solve the problem of insufficient training data, ResNet is trained through the transfer learning strategy, which brings advantages of robustness, versatility and low computing cost. International Union of Crystallography 2023-03-10 /pmc/articles/PMC10161896/ /pubmed/36897392 http://dx.doi.org/10.1107/S1600577523000917 Text en © Tianyu Fu et al. 2023 https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.
spellingShingle Research Papers
Fu, Tianyu
Wang, Yan
Zhang, Kai
Zhang, Jin
Wang, Shanfeng
Huang, Wanxia
Wang, Yaling
Yao, Chunxia
Zhou, Chenpeng
Yuan, Qingxi
Deep-learning-based ring artifact correction for tomographic reconstruction
title Deep-learning-based ring artifact correction for tomographic reconstruction
title_full Deep-learning-based ring artifact correction for tomographic reconstruction
title_fullStr Deep-learning-based ring artifact correction for tomographic reconstruction
title_full_unstemmed Deep-learning-based ring artifact correction for tomographic reconstruction
title_short Deep-learning-based ring artifact correction for tomographic reconstruction
title_sort deep-learning-based ring artifact correction for tomographic reconstruction
topic Research Papers
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10161896/
https://www.ncbi.nlm.nih.gov/pubmed/36897392
http://dx.doi.org/10.1107/S1600577523000917
work_keys_str_mv AT futianyu deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT wangyan deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT zhangkai deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT zhangjin deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT wangshanfeng deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT huangwanxia deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT wangyaling deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT yaochunxia deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT zhouchenpeng deeplearningbasedringartifactcorrectionfortomographicreconstruction
AT yuanqingxi deeplearningbasedringartifactcorrectionfortomographicreconstruction