Solar Dynamic Observatory
The Solar Dynamics Observatory (SDO) is a NASA spacecraft providing continuous high-resolution images of the Sun, aiming to understand its dynamic processes and improve space weather forecasting. Current research leverages SDO's vast dataset with machine learning techniques, employing models like Long Short-Term Memory networks (LSTMs), convolutional neural networks (CNNs), and denoising diffusion probabilistic models (DDPMs) to predict solar events such as active region emergence and solar flares, and to improve image processing tasks like super-resolution and cloud removal. These advancements enhance our ability to analyze solar phenomena, leading to more accurate space weather predictions and a deeper understanding of the Sun's behavior, with implications for protecting space-based and terrestrial technologies.
Papers
A Quantum Fuzzy-based Approach for Real-Time Detection of Solar Coronal Holes
Sanmoy Bandyopadhyay, Suman Kundu
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