Fast Solar Image Classification Using Deep Learning and its Importance for Automation in Solar Physics

Abstract

The volume of data being collected in solar physics has exponentially increased over the past decade and with the introduction of the $\text{Daniel K. Inouye Solar Telescope}$ (DKIST) we will be entering the age of petabyte solar data. Automated feature detection will be an invaluable tool for post-processing of solar images to create catalogues of data ready for researchers to use. We propose a deep learning model to accomplish this; a deep convolutional neural network is adept at feature extraction and processing images quickly. We train our network using data from $\text{Hinode/Solar Optical Telescope}$ (SOT) H$\alpha$ images of a small subset of solar features with different geometries: filaments, prominences, flare ribbons, sunspots and the quiet Sun ($\text{i.e.}$ the absence of any of the other four features). We achieve near perfect performance on classifying unseen images from SOT ($\approx$99.9\%) in 4.66 seconds. We also for the first time explore transfer learning in a solar context. Transfer learning uses pre-trained deep neural networks to help train new deep learning models $\text{i.e.}$ it teaches a new model. We show that our network is robust to changes in resolution by degrading images from SOT resolution ($\approx$0.33${}^{\prime \prime }$ at $\lambda$=6563Å) to $\text{Solar Dynamics Observatory/Atmospheric Imaging Assembly}$ (SDO/AIA) resolution ($\approx$1.2${}^{\prime \prime }$) without a change in performance of our network. However, we also observe where the network fails to generalise to sunspots from SDO/AIA bands 1600/1700Å due to small-scale brightenings around the sunspots and prominences in SDO/AIA 304Å due to coronal emission.

19 pages, 9 figures, accepted for publication in Solar Physics

References in corpus (13)

• The Solar Optical Telescope for the Hinode Mission: An Overview
• An Observational Overview of Solar Flares
• Solar Flare Prediction Model with Three Machine-Learning Algorithms Using Ultraviolet Brightening and Vector Magnetogram
• Forecasting Solar Flares Using Magnetogram-based Predictors and Machine Learning
• Predicting Solar Flares Using SDO/HMI Vector Magnetic Data Product and Random Forest Algorithm
• A New Tool for CME Arrival Time Prediction Using Machine Learning Algorithms: CAT-PUMA
• DeepVel: deep learning for the estimation of horizontal velocities at the solar surface
• Identifying typical Mg II flare spectra using machine learning
• Enhancing SDO/HMI images using deep learning
• Real-time multiframe blind deconvolution of solar images
• The FIRST Classifier: Compact and Extended Radio Galaxy Classification using Deep Convolutional Neural Networks
• RADYNVERSION: Learning to Invert a Solar Flare Atmosphere with Invertible Neural Networks
• Automated Detection Methods for Solar Activities and an Application for Statistic Analysis of Solar Filament

Cited by in corpus (11)

• Multi-channel coronal hole detection with convolutional neural networks
• Unsupervised classification of simulated magnetospheric regions
• Instrument-To-Instrument translation: Instrumental advances drive restoration of solar observation series via deep learning
• Using Multiple Instance Learning for Explainable Solar Flare Prediction
• A Machine Learning Approach to Correcting Atmospheric Seeing in Solar Flare Observations
• Analysing AIA Flare Observations using Convolutional Neural Networks
• Automatic Detection of Occulted Hard X-ray Flares Using Deep-Learning Methods
• Solar activity classification based on Mg II spectra: towards classification on compressed data
• Automatic detection of Ellerman bombs using Deep Learning
• Siberian radioheliograph image classification using ensemble of CLIP, EfficientNet and CatBoost models
• A Robust Deep Learning Framework for Prominence Detection through Composite Feature Representations