Idioma:

Exploring the Limits of Synthetic Creation of Solar EUV Images via Image-to-image Translation

Salvatelli, Valentina ; dos Santos, Luiz F. G. ; Bose, Souvik ; Neuberg, Brad ; Cheung, Mark C. M. ; Janvier, Miho ; Jin, Meng ; Gal, Yarin ; Güneş Baydin, Atilim

The Astrophysical journal, 2022-10, Vol.937 (2), p.100 [Periódico revisado por pares]

Philadelphia: The American Astronomical Society

Texto completo disponível

Citações Citado por

Enviar para

Título:
Exploring the Limits of Synthetic Creation of Solar EUV Images via Image-to-image Translation
Autor: Salvatelli, Valentina ; dos Santos, Luiz F. G. ; Bose, Souvik ; Neuberg, Brad ; Cheung, Mark C. M. ; Janvier, Miho ; Jin, Meng ; Gal, Yarin ; Güneş Baydin, Atilim
Assuntos: Astrophysics ; Channels ; Coders ; Convolutional neural networks ; Corona ; Deep learning ; Deep space ; GPU computing ; Image quality ; Machine learning ; Mission planning ; Neural networks ; Permutations ; Physics ; Solar active regions ; Solar activity ; Solar and Stellar Astrophysics ; Solar corona ; Solar EUV ; Solar extreme ultraviolet emission ; Solar observatories ; Solar surface ; Solar telescopes ; Space missions
É parte de: The Astrophysical journal, 2022-10, Vol.937 (2), p.100
Notas: The Sun and the Heliosphere
AAS37846
NFR/262622
Descrição: Abstract The Solar Dynamics Observatory (SDO), a NASA multispectral decade-long mission that has been daily producing terabytes of observational data from the Sun, has been recently used as a use case to demonstrate the potential of machine-learning methodologies and to pave the way for future deep space mission planning. In particular, the idea of using image-to-image translation to virtually produce extreme ultraviolet channels has been proposed in several recent studies, as a way to both enhance missions with fewer available channels and to alleviate the challenges due to the low downlink rate in deep space. This paper investigates the potential and the limitations of such a deep learning approach by focusing on the permutation of four channels and an encoder–decoder based architecture, with particular attention to how morphological traits and brightness of the solar surface affect the neural network predictions. In this work we want to answer the question: can synthetic images of the solar corona produced via image-to-image translation be used for scientific studies of the Sun? The analysis highlights that the neural network produces high-quality images over 3 orders of magnitude in count rate (pixel intensity) and can generally reproduce the covariance across channels within a 1% error. However, the model performance drastically diminishes in correspondence to extremely high energetic events like flares, and we argue that the reason is related to the rareness of such events posing a challenge to model training.
Editor: Philadelphia: The American Astronomical Society
Idioma: Inglês;Norueguês

Links

View record in HAL

Voltar para lista de resultados

Realização: Logos de Redes Sociais:

Exploring the Limits of Synthetic Creation of Solar EUV Images via Image-to-image Translation

Salvatelli, Valentina ; dos Santos, Luiz F. G. ; Bose, Souvik ; Neuberg, Brad ; Cheung, Mark C. M. ; Janvier, Miho ; Jin, Meng ; Gal, Yarin ; Güneş Baydin, Atilim

Philadelphia: The American Astronomical Society

Buscando em bases de dados remotas. Favor aguardar.