Single-shot grating-based X-ray phase contrast imaging via generative adversarial network

•The spotlights of the submitted manuscript are as follows:.•A highly efficient U-net based generative adversarial network (GAN) for solving the phase signal extraction problem with single-shot intensity pattern in grating-based XPCI system is proposed. The innovative single-shot XPCI (SS-XPCI) netw...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Optics and lasers in engineering 2022-05, Vol.152, p.106960, Article 106960
Hauptverfasser: Xu, Yueshu, Tao, Siwei, Bian, Yinxu, Bai, Ling, Tian, Zonghan, Hao, Xiang, Kuang, Cuifang, Liu, Xu
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•The spotlights of the submitted manuscript are as follows:.•A highly efficient U-net based generative adversarial network (GAN) for solving the phase signal extraction problem with single-shot intensity pattern in grating-based XPCI system is proposed. The innovative single-shot XPCI (SS-XPCI) network is effective to improve the spatial resolution and image contrast of the reconstructed images.•The simulated and experimental results demonstrate that our method is able to obtain results with high resolution and image contrast which is competitive with the conventional PS approach, while being less time-consuming and low-dose.•Even not being trained by the “noisy” datasets, the SS-XPCI network is capable of achieving near-optimal results as the noise level increases. If the network is trained with a predefined noise, the reconstruction performance could be further improved. Compared with the fourier analysis method, the SS-XPCI is not sensitive to the distance between the source grating G0 and the π-shift phase grating G1. Talbot-Lau interferometry obtains X-ray differential phase contrast (DPC) signals of object by subtracting multiple moiré patterns acquired by phase-stepping (PS) procedure. Due to the need of multiple intensity measurements, the long measuring time is inevitable in the conventional DPC imaging, giving rise to X-ray dose and fluctuations. In this paper, we propose a single-shot X-ray phase contrast imaging (XPCI) method based on deep learning. Specifically, in hardware, we propose to replace the analysis absorption grating with the high-resolution X-ray detector system to avoid the illumination flux loss caused by the analysis grating. In software, we employ a U-net based generative adversarial network (GAN) for solving the phase reconstruction problem with single-shot intensity pattern. By examining the performance on a variety of simulated and experimental datasets, we demonstrate that our approach, in spite of only using single intensity pattern, is able to obtain results with high resolution and image contrast which is competitive with the conventional PS approach, while being less time-consuming and low-dose.
ISSN:0143-8166
1873-0302
DOI:10.1016/j.optlaseng.2022.106960