Automatic Classification of Magnetic Chirality of Solar Filaments from H-Alpha Observations
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918145806041088 |
|---|---|
| author | Chalmers, Alexis Ahmadzadeh, Azim |
| author_facet | Chalmers, Alexis Ahmadzadeh, Azim |
| contents | In this study, we classify the magnetic chirality of solar filaments from H-Alpha observations using state-of-the-art image classification models. We establish the first reproducible baseline for solar filament chirality classification on the MAGFiLO dataset. The MAGFiLO dataset contains over 10,000 manually-annotated filaments from GONG H-Alpha observations, making it the largest dataset for filament detection and classification to date. Prior studies relied on much smaller datasets, which limited their generalizability and comparability. We fine-tuned several pre-trained, image classification architectures, including ResNet, WideResNet, ResNeXt, and ConvNeXt, and also applied data augmentation and per-class loss weights to optimize the models. Our best model, ConvNeXtBase, achieves a per-class accuracy of 0.69 for left chirality filaments and $0.73$ for right chirality filaments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_18214 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Automatic Classification of Magnetic Chirality of Solar Filaments from H-Alpha Observations Chalmers, Alexis Ahmadzadeh, Azim Solar and Stellar Astrophysics Artificial Intelligence In this study, we classify the magnetic chirality of solar filaments from H-Alpha observations using state-of-the-art image classification models. We establish the first reproducible baseline for solar filament chirality classification on the MAGFiLO dataset. The MAGFiLO dataset contains over 10,000 manually-annotated filaments from GONG H-Alpha observations, making it the largest dataset for filament detection and classification to date. Prior studies relied on much smaller datasets, which limited their generalizability and comparability. We fine-tuned several pre-trained, image classification architectures, including ResNet, WideResNet, ResNeXt, and ConvNeXt, and also applied data augmentation and per-class loss weights to optimize the models. Our best model, ConvNeXtBase, achieves a per-class accuracy of 0.69 for left chirality filaments and $0.73$ for right chirality filaments. |
| title | Automatic Classification of Magnetic Chirality of Solar Filaments from H-Alpha Observations |
| topic | Solar and Stellar Astrophysics Artificial Intelligence |
| url | https://arxiv.org/abs/2509.18214 |