← Back to all news
arXiv Computer Vision By Sina Norouzi Kandalan, Haodi Jiang, Jason T. L. Wang, Qin Li

Super-Resolution of Solar Magnetograms via Adaptive Stratified Ensemble Learning with Uncertainty Estimation

Read the original on arXiv Computer Vision →

The paper presents a method for single‑image super‑resolution of solar magnetograms, converting low‑resolution SOHO/MDI data into high‑resolution SDO/HMI line‑of‑sight images. It uses a modified RRDBNet architecture initialized with ESRGAN weights and introduces an adaptive stratified specialist ensemble (SSE) that trains three specialist networks on different image complexity strata, guided by a lightweight router and uncertainty estimation. Experiments show the ensemble outperforms related approaches, improving reconstruction quality across heterogeneous space‑based instruments.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Related stories

arXiv Machine Learning
Jun 10

An adaptive framework for the axisymmetric pulsar magnetosphere using physics-informed Kolmogorov-Arnold networks

arXiv:2606. 10686v1 Announce Type: cross Abstract: The pulsar magnetosphere has only recently been addressed using Physics-Informed Neural Networks (PINNs), by deploying a domain-decomposition approach and treating the separatrix and equatorial current sheet as infinitesimally thin discontinuities.

By Spyros Rigas, Ioannis Contopoulos, Georgios Alexandridis, Antonios Nathanail
More like this →
arXiv AI
Sep 17

Physics-Informed Neural Networks for Fast Multilayer Spectral Inversion of H{\alpha} 6562.8 A and Ca II 8542.1 A Spectra

The paper presents a physics-informed neural network (PINN) that accelerates multilayer spectral inversion (MLSI) of solar chromospheric lines Hα 6562.8 Å and Ca II 8542.1 Å. The PINN predicts MLSI parameters from observed line profiles and uses a differentiable forward model to synthesize spectra, trained in two stages—first with spectral reconstruction loss, then fine‑tuned with conventional MLSI results on a single reference image. Applied to Fast Imaging Solar Spectrograph data, the method reproduces key spatial structures and achieves a 12–60× speedup, processing a raster in 5–15 s versus 3–5 min for traditional MLSI.

By Ziyang Zhang, Qin Li, Vasyl B. Yurchyshyn, Kangwoo Yi, Haimin Wang, Wenda Cao, Bo Shen
fine-tuningsafety
More like this →