Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 49 min ago
Peer-reviewedCC BY 4.0The Astrophysical Journal Supplement Series

Data-driven Deep SwinTransUnet Enables the Prediction of 4K Future Solar Extreme Ultraviolet Images

  • Hao Li1,2,3,4,5
  • Tao Shu1,2
  • Zhou Chen1,2
5 affiliations
  1. 1Nanchang University
  2. 2National Center for Space Weather
  3. 3Institute of Automation
  4. 4University of Chinese Academy of Sciences
  5. 5China National Center for Bioinformation
Open at sourcePDFDOICite

Abstract

Solar extreme ultraviolet (EUV) images are crucial for understanding the structure and dynamics of the solar corona and predicting various space weather hazards. However, there is a significant research gap in forecasting future 4K solar EUV images, which are vital as high-quality inputs to support downstream tasks for more accurate prediction of spaceborne hazards like magnetic storms. Current mainstream approaches to space weather forecasting typically focus on predicting one or several parameters of the Sun from existing EUV images, rather than generating holistic predictions of full-disk solar images and remaining restricted to resolutions. To address the prediction of future solar EUV images, this study combines the advantages of the Swin transformer and TransUNet, proposing a new method: Data-driven Deep SwinTransUnet Enables the Prediction of 4K Future Solar Extreme Ultraviolet Images (Swin-4KEUV). Considering the strict hardware requirements of 4K images, each 4K image is segmented into nine overlapping 2K segments (with a 1024 pixel overlap) during processing. During the training process, Swin-4KEUV learns the weighted relationships among existing solar EUV images of different regions to forecast future images at 12 and 24 minute lead times. These results demonstrate the substantial potential of Swin-4KEUV for 4K solar EUV image prediction and its capability to serve as an advanced data foundation for downstream space weather applications. Furthermore, the extended forecasting capacity of the model is explored in our discussion section, evaluating multistep prediction results up to eight frames into the future with a total lead time of 96 minutes.

Related

  1. Solar extreme-ultraviolet waves and other associated solar activities
    Apoorv Dashora, Ramesh Chandra, Pooja Devi
    arXivPreprintarXiv licenceFull textSolar & stellar physicsarXiv
  2. Kinematic Relationship Between Solar Extreme Ultraviolet Waves and Type II Metric Radio Bursts
    Ramesh Chandra, Apoorv Dashora, P. F. Chen +1
    arXivPreprintCC BY 4.0Full textSolar & stellar physicsarXiv
  3. Improving Solar Extreme Ultraviolet Irradiance Modeling with Differential Emission Measure Informed Spectra
    Courtney L. Peck, Shah M. Bahauddin, Daniel B. Seaton +3
    The Astrophysical JournalPeer-reviewedCC BY 4.0Full textSolar & stellar physicsCrossref
  4. Multidomain Solar Extreme-ultraviolet Image Translation and Its Application to Differential Emission Measure Analysis
    Mingyu Jeon, Daeil Kim, Junmu Youn +1
    The Astrophysical Journal Supplement SeriesPeer-reviewedCC BY 4.0Full textSolar & stellar physicsCrossref
  5. Extreme Stellar Death and Galaxy Feedback at z = 2: A Rest-frame Ultraviolet Characterization of the Strongly Lensed Superluminous Supernova 2025wny
    Willem B. Hoogendam, David O. Jones, Christopher J. Storfer +26
    arXivPreprintCC BY 4.0Full textGalaxiesHigh-energy astrophysicsarXiv

Cite

BibTeXRIS
BibTeX
@article{li2026datadriven,
  title = {{Data-driven Deep SwinTransUnet Enables the Prediction of 4K Future Solar Extreme Ultraviolet Images}},
  author = {Hao Li and Tao Shu and Zhou Chen},
  journal = {The Astrophysical Journal Supplement Series},
  year = {2026},
  month = oct,
  doi = {10.3847/1538-4365/aea219},
  url = {https://iopscience.iop.org/article/10.3847/1538-4365/aea219},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Data-driven Deep SwinTransUnet Enables the Prediction of 4K Future Solar Extreme Ultraviolet Images
AU  - Hao Li
AU  - Tao Shu
AU  - Zhou Chen
PY  - 2026
DA  - 2026/10/14
JO  - The Astrophysical Journal Supplement Series
DO  - 10.3847/1538-4365/aea219
UR  - https://iopscience.iop.org/article/10.3847/1538-4365/aea219
AB  - Solar extreme ultraviolet (EUV) images are crucial for understanding the structure and dynamics of the solar corona and predicting various space weather hazards. However, there is a significant research gap in forecasting future 4K solar EUV images, which are vital as high-quality inputs to support downstream tasks for more accurate prediction of spaceborne hazards like magnetic storms. Current mainstream approaches to space weather forecasting typically focus on predicting one or several parameters of the Sun from existing EUV images, rather than generating holistic predictions of full-disk solar images and remaining restricted to resolutions. To address the prediction of future solar EUV images, this study combines the advantages of the Swin transformer and TransUNet, proposing a new method: Data-driven Deep SwinTransUnet Enables the Prediction of 4K Future Solar Extreme Ultraviolet Images (Swin-4KEUV). Considering the strict hardware requirements of 4K images, each 4K image is segmented into nine overlapping 2K segments (with a 1024 pixel overlap) during processing. During the training process, Swin-4KEUV learns the weighted relationships among existing solar EUV images of different regions to forecast future images at 12 and 24 minute lead times. These results demonstrate the substantial potential of Swin-4KEUV for 4K solar EUV image prediction and its capability to serve as an advanced data foundation for downstream space weather applications. Furthermore, the extended forecasting capacity of the model is explored in our discussion section, evaluating multistep prediction results up to eight frames into the future with a total lead time of 96 minutes.
SN  - 0067-0049
SN  - 1538-4365
ER  -