The article discusses advancements in astronomical image restoration using a Transformer-based deep learning model. It highlights the model's effectiveness in enhancing resolution and reducing noise for moderate noise levels, while acknowledging limitations faced when noise levels become excessive. Specific datasets such as HST, GalSim, and JWST are utilized for testing, demonstrating the model's capabilities and the challenges in restoring vital image parameters. The study meticulously analyzes performance metrics and visual outcomes, pointing to areas needing further improvement, particularly under high noise conditions.
Although our Transformer-based deep learning model provides state-of-the-art performance in both resolution enhancement and noise reduction for moderate noise levels, restoration becomes impossible when the noise level exceeds a threshold.
The model excels in enhancing image resolution and reducing noise effectively; however, severe noise degrades the quality, posing significant challenges in image restoration.
Collection
[
|
...
]