Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bangun, Arya, Töllner, Maximilian, Zhao, Xuan, Kübel, Christian, Scharr, Hanno
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.07633
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914147949608960
author Bangun, Arya
Töllner, Maximilian
Zhao, Xuan
Kübel, Christian
Scharr, Hanno
author_facet Bangun, Arya
Töllner, Maximilian
Zhao, Xuan
Kübel, Christian
Scharr, Hanno
contents We introduce FlowTIE, a neural-network-based framework for phase reconstruction from 4D-Scanning Transmission Electron Microscopy (STEM) data, which integrates the Transport of Intensity Equation (TIE) with a flow-based representation of the phase gradient. This formulation allows the model to bridge data-driven learning with physics-based priors, improving robustness under dynamical scattering conditions for thick specimen. The validation on simulated datasets of crystalline materials, benchmarking to classical TIE and gradient-based optimization methods are presented. The results demonstrate that FlowTIE improves phase reconstruction accuracy, fast, and can be integrated with a thick specimen model, namely multislice method.
format Preprint
id arxiv_https___arxiv_org_abs_2511_07633
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FlowTIE: Flow-based Transport of Intensity Equation for Phase Gradient Estimation from 4D-STEM Data
Bangun, Arya
Töllner, Maximilian
Zhao, Xuan
Kübel, Christian
Scharr, Hanno
Machine Learning
We introduce FlowTIE, a neural-network-based framework for phase reconstruction from 4D-Scanning Transmission Electron Microscopy (STEM) data, which integrates the Transport of Intensity Equation (TIE) with a flow-based representation of the phase gradient. This formulation allows the model to bridge data-driven learning with physics-based priors, improving robustness under dynamical scattering conditions for thick specimen. The validation on simulated datasets of crystalline materials, benchmarking to classical TIE and gradient-based optimization methods are presented. The results demonstrate that FlowTIE improves phase reconstruction accuracy, fast, and can be integrated with a thick specimen model, namely multislice method.
title FlowTIE: Flow-based Transport of Intensity Equation for Phase Gradient Estimation from 4D-STEM Data
topic Machine Learning
url https://arxiv.org/abs/2511.07633