Data Processing and Algorithms

Solar radio burst automatic recognition

With hough transform on the dynamic spectrum, we can detect the line-like solar radio bursts: type III solar radio bursts.

Detail of the algorithm is described in the paper: AutoTypeIII

Source code is available at github

Machine learning to generate Microware imaging from EUV imaging

This tool can generate microwave radio imaging with EUV radio imaging data [Zhang et al 2020]. The concept map is shown as follows:

The motivation of this work is to fill the gaps in the radio interferometric data, which usually have poor data continuity. After training, the model can generate microwave radio imagery data with a correlation level of >99% between the observation data and generated data. The generated data can serve as a reference for the radio interferometric data, and also be used in active region analysis.