SOT/HMI LOS Magnetogram Pairings Dataset for "A Diffusion-Based Deep Learning Approach for the Super-Resolution of SDO/HMI Magnetograms Using Hinode/SOT Data"

This dataset contains aligned pairs of Hinode/Solar Optical Telescope (SOT) and Solar Dynamics Observatory/Helioseismic and Magnetic Imager (SDO/HMI) line-of-sight (LOS) magnetograms used in the manuscript "A Diffusion-Based Deep Learning Approach for the Super-Resolution of SDO/HMI Magnetograms Using Hinode/SOT Data" by Congzhong Sun, Ruxue Wang, et al. The dataset consists of spatially aligned SOT/HMI magnetogram pairs collected at different observation times. These paired magnetograms are used for training, testing, and visualization experiments in the study of deep learning-based super-resolution reconstruction of SDO/HMI magnetograms. The dataset is organized into four archives: "hmi-train" contains low-resolution solar LOS magnetograms observed by SDO/HMI, which are used as input data for model training. "sot-train" contains the corresponding high-resolution solar LOS magnetograms observed by Hinode/SOT, which serve as the target data for supervised training of the super-resolution model. "hmi-test" contains low-resolution solar LOS magnetograms observed by SDO/HMI, which are used as input data for model evaluation. "sot-test" contains the corresponding high-resolution solar LOS magnetograms observed by Hinode/SOT, which serve as reference data for evaluating and visualizing the reconstructed results. To facilitate the matching and usage of paired data, each FITS file is named according to the observation timestamp of the corresponding HMI magnetogram. Therefore, files with the same filename in the paired archives (i.e., "hmi-train" with "sot-train", and "hmi-test" with "sot-test") represent corresponding SOT/HMI magnetogram samples.