U-net-based blocked artifacts removal method for dynamic computed tomography

Bo Wang; Zhiqiang Chen; Wim Dewulf; Ruben Pauwels; Zhiyang Yao; Qinhan Hou; Yongshun Xiao

doi:10.1364/AO.58.003748

U-net-based blocked artifacts removal method for dynamic computed tomography

Bo Wang, Zhiqiang Chen, Wim Dewulf, Ruben Pauwels, Zhiyang Yao, Qinhan Hou, Yongshun Xiao^*

^*Corresponding author for this work

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

6 Citations (Scopus)

Abstract

Airplane engines are vital aircraft components, so regular inspections of the engines are required to ensure their stable operation. A dynamic computed tomography (CT) system has been proposed by our group for in situ nondestructive testing of airplane engines, which takes advantage of the rotor's self-rotation. However, static parts of the engines cause blocked artifacts in the reconstructed image, leading to misinterpretations of the condition of engines. In this paper, in order to remove the artifacts produced by the projection of the static parts in CT reconstruction, two deep-learning-based methods are proposed, which use U-Net to perform correction in the projection domain. The projection of static parts can be estimated by a well-trained U-Net and subsequently can be subtracted from the projections of the engine. Finally, the rotor can be reconstructed from the corrected projections. The results shown in this paper indicate that the proposed methods are practical and effective for removing those blocked artifacts and recovering the details of rotating parts, which will, in turn, maximize the utilization of the dynamic CT system for in situ engine tests.

Original language	English
Journal	Applied Optics
Volume	58
Issue	14
Pages (from-to)	3748-3753
Number of pages	6
ISSN	1559-128X
DOIs	https://doi.org/10.1364/AO.58.003748
Publication status	Published - 2019
Externally published	Yes

Access to Document

10.1364/AO.58.003748

Cite this

@article{24378f7a61f04528860782eeff9ef15d,

title = "U-net-based blocked artifacts removal method for dynamic computed tomography",

abstract = "Airplane engines are vital aircraft components, so regular inspections of the engines are required to ensure their stable operation. A dynamic computed tomography (CT) system has been proposed by our group for in situ nondestructive testing of airplane engines, which takes advantage of the rotor's self-rotation. However, static parts of the engines cause blocked artifacts in the reconstructed image, leading to misinterpretations of the condition of engines. In this paper, in order to remove the artifacts produced by the projection of the static parts in CT reconstruction, two deep-learning-based methods are proposed, which use U-Net to perform correction in the projection domain. The projection of static parts can be estimated by a well-trained U-Net and subsequently can be subtracted from the projections of the engine. Finally, the rotor can be reconstructed from the corrected projections. The results shown in this paper indicate that the proposed methods are practical and effective for removing those blocked artifacts and recovering the details of rotating parts, which will, in turn, maximize the utilization of the dynamic CT system for in situ engine tests.",

author = "Bo Wang and Zhiqiang Chen and Wim Dewulf and Ruben Pauwels and Zhiyang Yao and Qinhan Hou and Yongshun Xiao",

note = "Funding Information: Funding. National Natural Science Foundation of China (NSFC) (51727813). Funding Information: National Natural Science Foundation of China (NSFC) (51727813). Publisher Copyright: {\textcopyright} 2019 Optical Society of America.",

year = "2019",

doi = "10.1364/AO.58.003748",

language = "English",

volume = "58",

pages = "3748--3753",

journal = "Applied Optics",

issn = "1559-128X",

publisher = "Optica Publishing Group",

number = "14",

}

TY - JOUR

T1 - U-net-based blocked artifacts removal method for dynamic computed tomography

AU - Wang, Bo

AU - Chen, Zhiqiang

AU - Dewulf, Wim

AU - Pauwels, Ruben

AU - Yao, Zhiyang

AU - Hou, Qinhan

AU - Xiao, Yongshun

N1 - Funding Information: Funding. National Natural Science Foundation of China (NSFC) (51727813). Funding Information: National Natural Science Foundation of China (NSFC) (51727813). Publisher Copyright: © 2019 Optical Society of America.

PY - 2019

Y1 - 2019

N2 - Airplane engines are vital aircraft components, so regular inspections of the engines are required to ensure their stable operation. A dynamic computed tomography (CT) system has been proposed by our group for in situ nondestructive testing of airplane engines, which takes advantage of the rotor's self-rotation. However, static parts of the engines cause blocked artifacts in the reconstructed image, leading to misinterpretations of the condition of engines. In this paper, in order to remove the artifacts produced by the projection of the static parts in CT reconstruction, two deep-learning-based methods are proposed, which use U-Net to perform correction in the projection domain. The projection of static parts can be estimated by a well-trained U-Net and subsequently can be subtracted from the projections of the engine. Finally, the rotor can be reconstructed from the corrected projections. The results shown in this paper indicate that the proposed methods are practical and effective for removing those blocked artifacts and recovering the details of rotating parts, which will, in turn, maximize the utilization of the dynamic CT system for in situ engine tests.

AB - Airplane engines are vital aircraft components, so regular inspections of the engines are required to ensure their stable operation. A dynamic computed tomography (CT) system has been proposed by our group for in situ nondestructive testing of airplane engines, which takes advantage of the rotor's self-rotation. However, static parts of the engines cause blocked artifacts in the reconstructed image, leading to misinterpretations of the condition of engines. In this paper, in order to remove the artifacts produced by the projection of the static parts in CT reconstruction, two deep-learning-based methods are proposed, which use U-Net to perform correction in the projection domain. The projection of static parts can be estimated by a well-trained U-Net and subsequently can be subtracted from the projections of the engine. Finally, the rotor can be reconstructed from the corrected projections. The results shown in this paper indicate that the proposed methods are practical and effective for removing those blocked artifacts and recovering the details of rotating parts, which will, in turn, maximize the utilization of the dynamic CT system for in situ engine tests.

UR - http://www.scopus.com/inward/record.url?scp=85065759119&partnerID=8YFLogxK

U2 - 10.1364/AO.58.003748

DO - 10.1364/AO.58.003748

M3 - Journal article

C2 - 31158188

AN - SCOPUS:85065759119

SN - 1559-128X

VL - 58

SP - 3748

EP - 3753

JO - Applied Optics

JF - Applied Optics

IS - 14

ER -

U-net-based blocked artifacts removal method for dynamic computed tomography

Abstract

Access to Document

Other files and links

Fingerprint

Cite this