Super-resolution of SDO/HMI Magnetograms Using Novel Deep Learning Methods

Rahman, Sumiaya; Moon, Yong‐Jae; Park, Eunsu; Siddique, Ashraf; Cho, Il-Hyun; Lim, Daye

doi:10.3847/2041-8213/ab9d79

Cited by 23 publications

(20 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using LOS magnetograms from GONG and other instruments that are explicitly cross-calibrated with HMI LOS magnetograms may significantly improve accuracy of the corresponding vector field instruments. In addition, deep-learning-based techniques for improving the resolution of magnetograms, namely, superresolution, are being successfully developed (Rahman et al 2020;Munoz-Jaramillo et al 2022). Using super-resolved LOS magnetograms as input to the CNN promises to yield more accurate CNN estimates of the vector field features.…”

Section: Discussionmentioning

confidence: 99%

Deep-learning Reconstruction of Sunspot Vector Magnetic Fields for Forecasting Solar Storms

Dhuri

Bhattacharjee

Hanasoge

et al. 2022

ApJ

View full text Add to dashboard Cite

Solar magnetic activity produces extreme solar flares and coronal mass ejections, which pose grave threats to electronic infrastructure and can significantly disrupt economic activity. It is therefore important to appreciate the triggers of explosive solar activity and develop reliable space weather forecasting. Photospheric vector magnetic field data capture sunspot magnetic field complexity and can therefore improve the quality of space weather prediction. However, state-of-the-art vector field observations are consistently only available from Solar Dynamics Observatory/Helioseismic and Magnetic Imager (HMI) since 2010, with most other current and past missions and observational facilities, such as Global Oscillations Network Group (GONG), only recording line-of-sight (LOS) fields. Here, using an inception-based convolutional neural network (CNN), we reconstruct HMI sunspot vector field features from LOS magnetograms of HMI and GONG with high fidelity (∼90% correlation) and sustained flare forecasting accuracy. We rebuild vector field features during the 2003 Halloween storms, for which only LOS field observations are available, and the CNN-estimated electric current helicity accurately captures the observed rotation of the associated sunspot prior to the extreme flares, showing a striking increase. Our study thus paves the way for reconstructing three solar cycles worth of vector field data from past LOS measurements, which are of great utility in improving space weather forecasting models and gaining new insights about solar activity.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep-learning Reconstruction of Sunspot Vector Magnetic Fields for Forecasting Solar Storms

Dhuri

Bhattacharjee

Hanasoge

et al. 2022

ApJ

View full text Add to dashboard Cite

show abstract

“…This method is especially suitable for sparse wavelength sampling. Rahman et al (2020) applied a residual attention model and a progressive generative adversarial network model to enhance the magnetograms of SDO/HMI. The magnetograms generated by the models are almost consistent with the corresponding target magnetograms.…”

Section: Related Workmentioning

confidence: 99%

“…The Helioseismic and Magnetic Imager (HMI) on the launched Solar Dynamics Observatory (SDO) can acquire full-disk images of the Sun with a pixel size of 0 5 and provide uninterrupted observations of the Sun (Schou et al 2012). But its spatial resolution is not sufficient to observe the small-scale structure of the Sun (Rahman et al 2020). GST can provide HMI with more detailed information about the solar images Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.…”

Section: Introductionmentioning

confidence: 99%

Improving the Spatial Resolution of Solar Images Based on an Improved Conditional Denoising Diffusion Probability Model

Song

et al. 2022

ApJS

View full text Add to dashboard Cite

The quality of solar images plays an important role in the analysis of small events in solar physics. Therefore, the improvement of image resolution based on super-resolution (SR) reconstruction technology has aroused the interest of many researchers. In this paper, an improved conditional denoising diffusion probability model (ICDDPM) based on the Markov chain is proposed for the SR reconstruction of solar images. This method reconstructs high-resolution (HR) images from low-resolution images by learning a reverse process that adds noise to HR images. To verify the effectiveness of the method, images from the Goode Solar Telescope at the Big Bear Solar Observatory and the Helioseismic and Magnetic Imager (HMI) on the Solar Dynamics Observatory are used to train a network, and the spatial resolution of reconstructed images is 4 times that of the original HMI images. The experimental results show that the performance based on ICDDPM is better than the previous work in subject judgment and object evaluation indexes. The reconstructed images of this method have higher subjective vision quality and better consistency with the HMI images. And the structural similarity and rms index results are also higher than the compared method, demonstrating the success of the resolution improvement using ICDDPM.

show abstract

“…The total data size from SDO to date is approximately 5 PB. This big data set obtained by SDO enables various studies and the application of deep learning techniques in solar physics (Armstrong and Fletcher, 2019;Jeong et al, 2020;Kim et al, 2019;Kucuk, Aydin, and Angryk, 2017;Park et al, 2018Park et al, , 2020Rahman et al, 2020). We use SDO/AIA and HMI data to train the solar event auto-detection models.…”

Section: Solar Dynamics Observatory (Sdo)mentioning

confidence: 99%

Solar Event Detection Using Deep-Learning-Based Object Detection Methods

et al. 2021

View full text Add to dashboard Cite

Research on the detection of solar events has been conducted over many years. Recently, deep learning and data-driven approaches have been applied to solar event recognition. In this study, we present solar event detection using deep-learning-based object detection methods for real-time space weather monitoring. First, we construct a new object detection dataset using imaging data obtained by the Solar Dynamics Observatory with bounding boxes as labels for three representative features: coronal holes, sunspots, and prominences. Second, we train two representative object detection models: the Single Shot MultiBox Detector (SSD) and the Faster Region-based Convolutional Neural Network (R-CNN) using the new dataset. The results show that both models perform similarly well for coronal hole and sunspot detection. For prominence detection, the SSD and Faster R-CNN exhibited relatively low performance. This study demonstrates that deep-learning-based object detection can successfully detect multiple types of solar events, and it may be extended to detect other solar events. In addition, we provide the dataset for further achievements of object detection studies in solar physics.

show abstract

Super-resolution of SDO/HMI Magnetograms Using Novel Deep Learning Methods

Cited by 23 publications

References 17 publications

Deep-learning Reconstruction of Sunspot Vector Magnetic Fields for Forecasting Solar Storms

Deep-learning Reconstruction of Sunspot Vector Magnetic Fields for Forecasting Solar Storms

Improving the Spatial Resolution of Solar Images Based on an Improved Conditional Denoising Diffusion Probability Model

Solar Event Detection Using Deep-Learning-Based Object Detection Methods

Contact Info

Product

Resources

About