Image error correction in x-ray phase contrast imaging by convolutional neural networks

Jingjia Yang; Minghui Zhu; Jun Yang; Jianheng Huang; Yaohu Lei; Xin Liu; Ji Li; Jinchuan Guo

doi:10.1117/12.2654655

15 December 2022 Image error correction in x-ray phase contrast imaging by convolutional neural networks

Jingjia Yang, Minghui Zhu, Jun Yang, Jianheng Huang, Yaohu Lei, Xin Liu, Ji Li, Jinchuan Guo

Proceedings Volume 12478, Thirteenth International Conference on Information Optics and Photonics (CIOP 2022); 1247822 (2022) https://doi.org/10.1117/12.2654655
Event: Thirteenth International Conference on Information Optics and Photonics (CIOP 2022), 2022, Xi'an, China

Abstract

When there is mechanical drift in X-ray phase contrast imaging system, the position of the grating will produce random error, and the intensity of the obtained image will have a certain deviation, and the information retrieved by using the phase step method may be accompanied with Moire artifacts. In order to overcome this limitation, we introduce the convolutional neural network (CNN) to address it. The training data is downloaded from Kaggle, and the fringe graph with random deviation is combined as the network input, while the label is defined as the first-order difference image along the horizontal direction of the image. Both simulation and experiment show that CNN can not only retrieve the phase signal of the sample, but also remove some Moire artifacts with regular shape to improve the image quality. As a result, the utilization rate of X-ray in imaging system can be improved.

Citation Download Citation

Jingjia Yang, Minghui Zhu, Jun Yang, Jianheng Huang, Yaohu Lei, Xin Liu, Ji Li, and Jinchuan Guo "Image error correction in x-ray phase contrast imaging by convolutional neural networks", Proc. SPIE 12478, Thirteenth International Conference on Information Optics and Photonics (CIOP 2022), 1247822 (15 December 2022); https://doi.org/10.1117/12.2654655

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available