Type II and Type III Solar Radio Burst Classification Using Transfer Learning

The Sun periodically emits intense bursts of radio emission called solar radio bursts (SRBs). These bursts can disrupt radio communications and be indicative of large solar events that can disrupt technological infrastructure on Earth and in space. The risks posed by these events highlight the need for automated SRB classification, providing the potential to improve event detection and real-time monitoring, thereby enhancing techniques used to research space weather and related events. Using data recorded by the e-Callisto network, a dataset containing images of radio spectra was created. This dataset consists of three classes; Empty spectrograms, spectrograms containing Type II SRBs, and spectrograms containing Type III SRBs. This dataset was used to fine-tune several popular pre-trained deep learning models to classify Type II and Type III SRBs, including; VGGnet-19, MobileNet, ResNet-152, DenseNet-201 and YOLOv8. The results obtained from testing the models on the testing set ranged between an F1-score of 87% and 92%. The best performing model, YOLOv8, demonstrates that utilising pre-trained models for event classification can provide automated approaches to classying SRBs and provide a practical solution to the limited amount of data samples available for Type II SRBs.