Unsupervised Deep Image Prior for Sparse-View and Limited-Angle Electron Tomography
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917536543539200 |
|---|---|
| author | Brosset, Serge Bueno, Daniel del Pozo David, Thomas Guetaz, Laure Ciuciu, Philippe Saghi, Zineb |
| author_facet | Brosset, Serge Bueno, Daniel del Pozo David, Thomas Guetaz, Laure Ciuciu, Philippe Saghi, Zineb |
| contents | Electron tomography (ET) plays an important role in the three-dimensional (3D) characterization of nanomaterials. However, under limited-angle and sparse-view conditions, conventional algorithms produce degraded reconstructions, which compromise the quality and interpretability of resulting 3D data. In this paper, we present deep image prior (DIP), an unsupervised deep learning (DL) approach, for highly degraded tomography acquisitions and demonstrate, using simulated data, that its performance is comparable to that of supervised approaches requiring training datasets, even for tilt ranges as limited as 60° and tilt increments of 10°. We then apply it to experimental data and show that it enables reliable 3D quantification under both sparse-view and limited-angle conditions, highlighting its potential for a wide range of materials and acquisition modalities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_27139 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Unsupervised Deep Image Prior for Sparse-View and Limited-Angle Electron Tomography Brosset, Serge Bueno, Daniel del Pozo David, Thomas Guetaz, Laure Ciuciu, Philippe Saghi, Zineb Image and Video Processing Computer Vision and Pattern Recognition Instrumentation and Detectors Electron tomography (ET) plays an important role in the three-dimensional (3D) characterization of nanomaterials. However, under limited-angle and sparse-view conditions, conventional algorithms produce degraded reconstructions, which compromise the quality and interpretability of resulting 3D data. In this paper, we present deep image prior (DIP), an unsupervised deep learning (DL) approach, for highly degraded tomography acquisitions and demonstrate, using simulated data, that its performance is comparable to that of supervised approaches requiring training datasets, even for tilt ranges as limited as 60° and tilt increments of 10°. We then apply it to experimental data and show that it enables reliable 3D quantification under both sparse-view and limited-angle conditions, highlighting its potential for a wide range of materials and acquisition modalities. |
| title | Unsupervised Deep Image Prior for Sparse-View and Limited-Angle Electron Tomography |
| topic | Image and Video Processing Computer Vision and Pattern Recognition Instrumentation and Detectors |
| url | https://arxiv.org/abs/2605.27139 |