Tracing Halpha Fibrils through Bayesian Deep Learning

Abstract

We present a new deep learning method, dubbed FibrilNet, for tracing chromospheric fibrils in Halpha images of solar observations. Our method consists of a data pre-processing component that prepares training data from a threshold-based tool, a deep learning model implemented as a Bayesian convolutional neural network for probabilistic image segmentation with uncertainty quantification to predict fibrils, and a post-processing component containing a fibril-fitting algorithm to determine fibril orientations. The FibrilNet tool is applied to high-resolution Halpha images from an active region (AR 12665) collected by the 1.6 m Goode Solar Telescope (GST) equipped with high-order adaptive optics at the Big Bear Solar Observatory (BBSO). We quantitatively assess the FibrilNet tool, comparing its image segmentation algorithm and fibril-fitting algorithm with those employed by the threshold-based tool. Our experimental results and major findings are summarized as follows. First, the image segmentation results (i.e., detected fibrils) of the two tools are quite similar, demonstrating the good learning capability of FibrilNet. Second, FibrilNet finds more accurate and smoother fibril orientation angles than the threshold-based tool. Third, FibrilNet is faster than the threshold-based tool and the uncertainty maps produced by FibrilNet not only provide a quantitative way to measure the confidence on each detected fibril, but also help identify fibril structures that are not detected by the threshold-based tool but are inferred through machine learning. Finally, we apply FibrilNet to full-disk Halpha images from other solar observatories and additional high-resolution Halpha images collected by BBSO/GST, demonstrating the tool's usability in diverse datasets.

20 pages, 12 figures

References in corpus (20)

• Natural Scales in Geographical Patterns
• Variational Inference: A Review for Statisticians
• Deep learning of multi-element abundances from high-resolution spectroscopic data
• On-disk counterparts of type II spicules in the Ca II 854.2 nm and Halpha lines
• Predicting Solar Flares Using a Long Short-Term Memory Network
• Search for high velocities in the disk counterpart of type II spicules
• Deep Learning Identifies High-z Galaxies in a Central Blue Nugget Phase in a Characteristic Mass Range
• On fibrils and field lines: The nature of H$\alpha$ fibrils in the solar chromosphere
• Can We Improve the Preprocessing of Photospheric Vector Magnetograms by the Inclusion of Chromospheric Observations?
• Transverse Oscillations in Slender Ca II H Fibrils Observed with Sunrise/SuFI
• Inference of the chromospheric magnetic field orientation in the Ca II 8542 A line fibrils
• Using convolutional neural networks to predict galaxy metallicity from three-color images
• Detecting solar system objects with convolutional neural networks
• Tracing the Chromospheric and Coronal Magnetic Field with AIA, IRIS, IBIS, and ROSA Data
• Automated Spatiotemporal Analysis of Fibrils and Coronal Rain using the Rolling Hough Transform
• Morphological properties of slender Ca II H fibrils observed by SUNRISE II
• Force-Free Field Reconstructions Enhanced by Chromospheric Magnetic Field Data
• Dynamics of internetwork chromospheric fibrils: Basic properties and MHD kink waves
• Identifying and Tracking Solar Magnetic Flux Elements with Deep Learning
• Calibrating GONG Magnetograms with End-to-end Instrument Simulation I: Background, the GONG Instrument, and End-to-end Simulation

Cited by in corpus (2)

• Variational Inference as an alternative to MCMC for parameter estimation and model selection
• A Deep Learning Approach to Generating Photospheric Vector Magnetograms of Solar Active Regions for SOHO/MDI Using SDO/HMI and BBSO Data