Inferring Vector Magnetic Fields from Stokes Profiles of GST/NIRIS Using a Convolutional Neural Network
Inferring Vector Magnetic Fields from Stokes Profiles of GST/NIRIS Using a Convolutional Neural Network
复制标题
DOI：
10.3847/1538-4357/ab8818
复制
发表时间：
2020-05
期刊：
影响因子：
--
通讯作者：
Hao Liu;Yan Xu;Jiasheng Wang;J. Jing;Chang Liu;J. T. Wang;Haimin Wang
中科院分区：
文献类型：
--
作者：
Hao Liu;Yan Xu;Jiasheng Wang;J. Jing;Chang Liu;J. T. Wang;Haimin Wang
我们提出了一种新的机器学习方法的基础上卷积神经网络（CNN）和Milne-Eddington（ME）方法的斯托克斯反演。本研究中使用的斯托克斯测量是由大熊太阳天文台1.6米古德太阳望远镜（GST）上的近红外成像分光偏振仪（NIRIS）进行的。通过学习由基于物理的ME工具准备的训练数据中的潜在模式，所提出的CNN方法能够从GST/NIRIS的斯托克斯轮廓推断矢量磁场。实验结果表明，我们的CNN方法比广泛使用的ME方法产生更平滑，更干净的磁图。此外，CNN方法比ME方法快4至6倍，能够在几乎真实的时间内产生矢量磁场，这对空间天气预报至关重要。具体而言，CNN方法处理包括GST/NIRIS的斯托克斯轮廓的720 × 720像素的图像需要150 s。最后，CNN推断的结果与ME计算的结果高度相关，并且更接近ME的结果，平均而言，Pearson积矩相关系数（PPMCC）比其他机器学习算法（如多重支持向量回归和多层感知器（MLP））更接近1。特别是，根据我们的实验研究，CNN方法在PPMCC中平均比当前最好的机器学习方法（MLP）高出2.6%。因此，所提出的基于物理辅助的深度学习的CNN工具可以被认为是GST/NIRIS获得的高分辨率极化观测的斯托克斯反演的替代有效方法。
We propose a new machine-learning approach to Stokes inversion based on a convolutional neural network (CNN) and the Milne–Eddington (ME) method. The Stokes measurements used in this study were taken by the Near InfraRed Imaging Spectropolarimeter (NIRIS) on the 1.6 m Goode Solar Telescope (GST) at the Big Bear Solar Observatory. By learning the latent patterns in the training data prepared by the physics-based ME tool, the proposed CNN method is able to infer vector magnetic fields from the Stokes profiles of GST/NIRIS. Experimental results show that our CNN method produces smoother and cleaner magnetic maps than the widely used ME method. Furthermore, the CNN method is four to six times faster than the ME method and able to produce vector magnetic fields in nearly real time, which is essential to space weather forecasting. Specifically, it takes ∼50 s for the CNN method to process an image of 720 × 720 pixels comprising Stokes profiles of GST/NIRIS. Finally, the CNN-inferred results are highly correlated to the ME-calculated results and closer to the ME’s results with the Pearson product-moment correlation coefficient (PPMCC) being closer to 1, on average, than those from other machine-learning algorithms, such as multiple support vector regression and multilayer perceptrons (MLP). In particular, the CNN method outperforms the current best machine-learning method (MLP) by 2.6%, on average, in PPMCC according to our experimental study. Thus, the proposed physics-assisted deep learning–based CNN tool can be considered as an alternative, efficient method for Stokes inversion for high-resolution polarimetric observations obtained by GST/NIRIS.