Deep Learning Enhanced Quantum Holography with Undetected Photons
Author Information
Author(s): Fan Weiru, Qian Gewei, Wang Yutong, Xu Chen-Ran, Chen Ziyang, Liu Xun, Li Wei, Liu Xu, Liu Feng, Xu Xingqi, Wang Da-Wei, Yakovlev Vladislav V.
Primary Institution: Zhejiang University
Hypothesis
Can deep learning improve the performance of quantum holography with undetected photons?
Conclusion
The study demonstrates that deep learning can significantly enhance the imaging capabilities of quantum holography by improving spatial resolution and reducing noise.
Supporting Evidence
- Deep learning can extract images from single-shot holograms.
- DL-QHUP improves spatial resolution and noise resilience.
- The methodology allows for high-speed imaging.
- DL-QHUP can be applied in various fields including biomedical imaging.
- Training with diverse datasets enhances generalization.
Takeaway
This study shows that using deep learning can help create clearer images from special light patterns, making it easier to see details in things like medical images or materials.
Methodology
The study developed a convolution-based deep learning model to extract images from single-shot holograms, improving noise resilience and spatial resolution.
Limitations
The performance may degrade if there is a mismatch between training and testing datasets.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website