Vision transformer for multi-domain phase retrieval in coherent diffraction imaging
Optics Express · 13.07.2026 · 34 (14) · 26196
DOI: 10.1364/oe.597254 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Express, online erschienen am 08.07.2026. Verfasst von Jialun Liu und 2 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: United States Department of Energy, U.S. National Science Foundation, Engineering and Physical Sciences Research Council.
Abstract (Originalsprache)
Bragg coherent diffraction imaging (BCDI) phase retrieval becomes difficult in the strong-phase regime, where a crystal contains distortions beyond half a lattice spacing. An important special case is the phase domain problem, where blocks of a crystal are displaced with sharp jumps at domain walls. This generates split Bragg peaks and dense fringe structures for which classical iterative solvers often stagnate or return different solutions from different initialisations. We introduce an unsupervised vision transformer (ViT) using Fourier attention to solve this block-phase, multi-domain phase retrieval problem directly from measured 2D Bragg diffraction intensities. Through token mixing and multiscale Fourier attention, Fourier ViT couples reciprocal-space information globally to improve convergence in the strong-phase regime. We have validated the approach on synthetic Voronoi multi-domain crystals with strong-phase contrast under realistic noise corruptions. We further tested it on experimental diffraction from weak-phase SrTiO 3 , Fe 3 O 4 across its phase transition, and strong-phase La 0.5 Ca 0.5 MnO 3 nanocrystals. For the cases studied here, we have achieved reciprocal-space mismatch ( χ 2 ) comparable to or lower than the compared baselines. Our method preserves domain-resolved phase reconstructions on synthetic data as the number of domains increases. On experimental data, the iterative method, Fourier ViT, and the complex convolutional neural network baseline reach similar χ 2 under the same fixed support.
Förderung
- United States Department of Energy
- U.S. National Science Foundation
- Engineering and Physical Sciences Research Council
- European Research Council
Zitieren
APA
Liu, J., Yang, D., & Robinson, I. (2026). Vision transformer for multi-domain phase retrieval in coherent diffraction imaging. Optics Express, 34(14), 26196. https://doi.org/10.1364/oe.597254
Vancouver
Liu J, Yang D, Robinson I. Vision transformer for multi-domain phase retrieval in coherent diffraction imaging. Optics Express. 2026;34(14):26196. doi:10.1364/oe.597254
BibTeX
@article{Liu2026,
author = {Liu, Jialun and Yang, David and Robinson, Ian},
title = {Vision transformer for multi-domain phase retrieval in coherent diffraction imaging},
journal = {Optics Express},
year = {2026},
volume = {34},
number = {14},
pages = {26196},
doi = {10.1364/oe.597254}
}
EUVIMED