Deep generative priors for robust and efficient electron ptychography
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908643832627200 |
|---|---|
| author | McCray, Arthur R. C. Ribet, Stephanie M. Varnavides, Georgios Ophus, Colin |
| author_facet | McCray, Arthur R. C. Ribet, Stephanie M. Varnavides, Georgios Ophus, Colin |
| contents | Electron ptychography enables dose-efficient atomic-resolution imaging, but conventional reconstruction algorithms suffer from noise sensitivity, slow convergence, and extensive manual hyperparameter tuning for regularization, especially in three-dimensional multislice reconstructions. We introduce a deep generative prior (DGP) framework for electron ptychography that uses the implicit regularization of convolutional neural networks to address these challenges. Two DGPs parameterize the complex-valued sample and probe within an automatic-differentiation mixed-state multislice forward model. Compared to pixel-based reconstructions, DGPs offer four key advantages: (i) greater noise robustness and improved information limits at low dose; (ii) markedly faster convergence, especially at low spatial frequencies; (iii) improved depth regularization; and (iv) minimal user-specified regularization. The DGP framework promotes spatial coherence and suppresses high-frequency noise without extensive tuning, and a pre-training strategy stabilizes reconstructions. Our results establish DGP-enabled ptychography as a robust approach that reduces expertise barriers and computational cost, delivering robust, high-resolution imaging across diverse materials and biological systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07795 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Deep generative priors for robust and efficient electron ptychography McCray, Arthur R. C. Ribet, Stephanie M. Varnavides, Georgios Ophus, Colin Image and Video Processing Materials Science Electron ptychography enables dose-efficient atomic-resolution imaging, but conventional reconstruction algorithms suffer from noise sensitivity, slow convergence, and extensive manual hyperparameter tuning for regularization, especially in three-dimensional multislice reconstructions. We introduce a deep generative prior (DGP) framework for electron ptychography that uses the implicit regularization of convolutional neural networks to address these challenges. Two DGPs parameterize the complex-valued sample and probe within an automatic-differentiation mixed-state multislice forward model. Compared to pixel-based reconstructions, DGPs offer four key advantages: (i) greater noise robustness and improved information limits at low dose; (ii) markedly faster convergence, especially at low spatial frequencies; (iii) improved depth regularization; and (iv) minimal user-specified regularization. The DGP framework promotes spatial coherence and suppresses high-frequency noise without extensive tuning, and a pre-training strategy stabilizes reconstructions. Our results establish DGP-enabled ptychography as a robust approach that reduces expertise barriers and computational cost, delivering robust, high-resolution imaging across diverse materials and biological systems. |
| title | Deep generative priors for robust and efficient electron ptychography |
| topic | Image and Video Processing Materials Science |
| url | https://arxiv.org/abs/2511.07795 |