TY - GEN
T1 - Deep regression for imaging solar magnetograms using pyramid generative adversarial networks
AU - Alshehhi, Rasha
N1 - Funding Information:
The author thanks team members of the SDO mission and STEREO mission and acknowledge effort devoted to develop open-source python packages: Sunpy and Keras. This work is supported by the New York University, Abu Dhabi Kawader Research Program. Computational resources are provided by HPC center in the New York university, Abu Dhabi.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/6
Y1 - 2020/6
N2 - Monitoring a large active region in the farside of the Sun is important for space weather forecasting. However, direct imaging of the farside is currently not available and usually physicists rely on seismic holography to infer farside magnetograms. On other hand, mapping between holography and magnetic images is non-trivial. In this work, Generative Adversarial Network (GAN) is used; which consists of a pyramid of modified pixel2pixel architectures to capture internal distributions at different scales with higher quality. Generative model is trained and evaluated using frontside of Solar Dynamic Observatory (SDO): Atmospheric Imaging Assembly (AIA) and Helioseismic and Magnetic Imager (HMI) magnetograms. Farside solar magnetograms from Extreme UltraViolet Imager (EUVI) farside data is also generated. The generative model successfully generates frontside solar magnetograms and outperforms state-of-The art method. It also help to monitor the magnetic changes from farside to frontside using generated solar magnetograms.
AB - Monitoring a large active region in the farside of the Sun is important for space weather forecasting. However, direct imaging of the farside is currently not available and usually physicists rely on seismic holography to infer farside magnetograms. On other hand, mapping between holography and magnetic images is non-trivial. In this work, Generative Adversarial Network (GAN) is used; which consists of a pyramid of modified pixel2pixel architectures to capture internal distributions at different scales with higher quality. Generative model is trained and evaluated using frontside of Solar Dynamic Observatory (SDO): Atmospheric Imaging Assembly (AIA) and Helioseismic and Magnetic Imager (HMI) magnetograms. Farside solar magnetograms from Extreme UltraViolet Imager (EUVI) farside data is also generated. The generative model successfully generates frontside solar magnetograms and outperforms state-of-The art method. It also help to monitor the magnetic changes from farside to frontside using generated solar magnetograms.
UR - http://www.scopus.com/inward/record.url?scp=85090130369&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85090130369&partnerID=8YFLogxK
U2 - 10.1109/CVPRW50498.2020.00110
DO - 10.1109/CVPRW50498.2020.00110
M3 - Conference contribution
AN - SCOPUS:85090130369
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 807
EP - 815
BT - Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020
PB - IEEE Computer Society
T2 - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020
Y2 - 14 June 2020 through 19 June 2020
ER -