Products

GONG H-Alpha Anomaly Dataset is a dataset of manually classified anomalous patterns in GONG H-Alpha solar observations from the years 2011 to 2021. The dataset contains 1000 anomalous images and 1800 non-anomalous images. Anomalous images exhibit anomalies such as shadow, objects obstructing the solar disk, an off-center solar disk, overly-bright, and the absence of normal texture. This dataset can be used for supervised or unsupervised algorithms to detect anomalous H-Alpha observations.

This Python package is designed for detecting anomalous H-Alpha observations of the Sun. Using H-Alpha Anomalyzer, bad-quality observations can be filtered out of millions of images very quickly, enhancing the training phase of Deep Neural Networks (DNNs). Our region-based probabilistic method offers explainability by assigning anomaly likelihoods to each cell of a given observation.

The project is a simple Python wrapper that makes use of NVSS's conversion software which in turn is a modernized version of Bill Cotton's implementation.

The credit goes exclusively to the developers associated to the projects listed above, not the developers of this software. The sole intention of creating this project is to customize it to serve as a component of the Operational MLEcoFi.

[https://bitbucket.org/dataresearchlab/mleco-fits2jpegconvertor/src/master/]

This web application facilitates search through H-Alpha images of the GONG network. This product is also a front end for our future products related to filaments. 

[https://dmlab.cs.gsu.edu/MLEco/GONGHAlphaViewer/]

The Manually Annotated GONG Filaments in H-ALpha Observations (MAGFiLO) is the largest collection of its kind. It contains 10,244 annotated filaments from 1,593 observations captured by the Global Oscillation Network Group (GONG), spanning the years 2011 through 2022. Each annotation details one filament’s segmentation, minimum bounding box, spine, and magnetic field chirality. With a total of over one thousand person-hours of annotation, and a double-blind review process, we ensured high-quality ground-truth data.