Learning the solar latent space: sigma-variational autoencoders for multiple channel solar imaging
Learning the solar latent space: sigma-variational autoencoders for multiple channel solar imaging
复制标题
学习太阳潜在空间：用于多通道太阳成像的西格玛变分自动编码器
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
M. Jah
中科院分区：
文献类型：
--
作者：
Edward J. E. Brown;S. Bonasera;Bernard Benson;Jorge A. Pérez;Giacomo Acciarini;Atılım Güne¸s Baydin;Christopher Bridges;Meng Jin;Eric Sutton;M. Jah
这项研究使用西格玛变分自动编码器来学习太阳图像的潜在空间，该图像使用美国宇航局太阳动力学观测站上的大气成像组件 (AIA) 和日震和磁成像仪 (HMI) 仪器拍摄的 12 个通道。该模型能够将大型图像数据集显着压缩至其原始大小的 0.19%，同时仍然能够熟练地重建原始图像。作为利用学习到的表示的下游任务，本研究展示了与现成的预训练 ResNet 特征提取器相比，使用学习到的潜在空间作为输入来改进 F30 太阳射电通量指数的预测。最后，开发的模型可用于通过从学习的潜在空间采样来生成逼真的合成太阳图像。
This study uses a sigma-variational autoencoder to learn a latent space of solar images using the 12 channels taken by Atmospheric Imaging Assembly (AIA) and the Helioseismic and Magnetic Imager (HMI) instruments on-board the NASA Solar Dynamics Observatory. The model is able to significantly compress the large image dataset to 0.19% of its original size while still proficiently reconstructing the original images. As a downstream task making use of the learned representation, this study demonstrates the of use the learned latent space as an input to improve the forecasts of the F30 solar radio flux index, compared to an off-the-shelf pretrained ResNet feature extractor. Finally, the developed models can be used to generate realistic synthetic solar images by sampling from the learned latent space.