Dataset of Solar Active Regions in the Solar Full-disk Magnetograms

Accurate identification of these regions can help forecast the impact of solar activities on Earth's environment. This dataset utilizes solar full-disk magnetograms observed by the Helioseismic and Magnetic Imager onboard the Solar Dynamic Observatory (2010-2019), combined with NOAA AR numbers provided by the SolarMonitor website.

The active regions are annotated using an image processing-based recognition method along with manual labeling. The dataset consists of 6,975 solar full-disk magnetograms, taken every 12 hours, with a total numbe of 26531 annotations consisting of the active-region informationfor each magnetogram.This dataset serves as a benchmark resource for developing deep learning solar active region detection models, and aims to enhance predictive capabilities for severe space weather phenomena through physics-informed training samples.