Super-Resolution of SOHO/MDI Magnetograms of Solar Active Regions Using SDO/HMI Data and an Attention-Aided Convolutional Neural Network
Super-Resolution of SOHO/MDI Magnetograms of Solar Active Regions Using SDO/HMI Data and an Attention-Aided Convolutional Neural Network Chunhui Xu, Jason T. L. Wang, Haimin Wang, Haodi Jiang, Qin Li, Yasser Abduallah, Yan Xu arXiv:2403.18302v1 Announce Type: new Abstract: Image super-resolution has been an important subject in image processing and recognition. Here, we present an attention-aided convolutional neural network (CNN) for solar image super-resolution. Our method, named SolarCNN, aims to enhance the quality of line-of-sight (LOS) magnetograms of solar active regions (ARs) collected by the Michelson Doppler Imager (MDI) on board the Solar and Heliospheric Observatory (SOHO). The ground-truth labels used for training SolarCNN are theRead More →