Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2403.07786 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915342205321216 |
|---|---|
| author | Liu, Ronald B. Liu, Zhe Wolf, Max G. A. Purohit, Krishna P. Fritz, Gregor Feng, Yi Hansen, Carsten G. Bagnaninchi, Pierre O. Solvas, Xavier Casadevall i Yang, Yunjie |
| author_facet | Liu, Ronald B. Liu, Zhe Wolf, Max G. A. Purohit, Krishna P. Fritz, Gregor Feng, Yi Hansen, Carsten G. Bagnaninchi, Pierre O. Solvas, Xavier Casadevall i Yang, Yunjie |
| contents | Advancements in high-throughput biomedical applications require real-time, large field-of-view (FOV) imaging. While current 2D lens-free imaging (LFI) systems improve FOV, they are often hindered by time-consuming multi-position measurements, extensive data pre-processing, and strict optical parameterization, limiting their application to static, thin samples. To overcome these limitations, we introduce GenLFI, combining a generative unsupervised physics-informed neural network (PINN) with a large FOV LFI setup for straightforward holographic image reconstruction, without multi-measurement. GenLFI enables real-time 2D imaging for 3D samples, such as droplet-based microfluidics and 3D cell models, in dynamic complex optical fields. Unlike previous methods, our approach decouples the reconstruction algorithm from optical setup parameters, enabling a large FOV limited only by hardware. We demonstrate a real-time FOV exceeding 550 mm$^2$, over 20 times larger than current real-time LFI systems. This framework unlocks the potential of LFI systems, providing a robust tool for advancing automated high-throughput biomedical applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_07786 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Physics-informed generative real-time lens-free imaging Liu, Ronald B. Liu, Zhe Wolf, Max G. A. Purohit, Krishna P. Fritz, Gregor Feng, Yi Hansen, Carsten G. Bagnaninchi, Pierre O. Solvas, Xavier Casadevall i Yang, Yunjie Optics Computer Vision and Pattern Recognition Advancements in high-throughput biomedical applications require real-time, large field-of-view (FOV) imaging. While current 2D lens-free imaging (LFI) systems improve FOV, they are often hindered by time-consuming multi-position measurements, extensive data pre-processing, and strict optical parameterization, limiting their application to static, thin samples. To overcome these limitations, we introduce GenLFI, combining a generative unsupervised physics-informed neural network (PINN) with a large FOV LFI setup for straightforward holographic image reconstruction, without multi-measurement. GenLFI enables real-time 2D imaging for 3D samples, such as droplet-based microfluidics and 3D cell models, in dynamic complex optical fields. Unlike previous methods, our approach decouples the reconstruction algorithm from optical setup parameters, enabling a large FOV limited only by hardware. We demonstrate a real-time FOV exceeding 550 mm$^2$, over 20 times larger than current real-time LFI systems. This framework unlocks the potential of LFI systems, providing a robust tool for advancing automated high-throughput biomedical applications. |
| title | Physics-informed generative real-time lens-free imaging |
| topic | Optics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.07786 |