EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automatic Equatorial Ionospheric Irregularities Detection From GOLD Images Using Deep Learning

Long Tang, Yumei Li, Huijian Zhao, Xueyan Chen, Qiong Ding, Fang Zou, Xiaohong Zhang

Space Weather · 2026-11 · 24 (11)

DOI: 10.1029/2026sw005273 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Space Weather, online erschienen am 06.11.2026. Verfasst von Long Tang und 6 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Guangzhou Municipal Science and Technology Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate detection of Equatorial Ionization Anomaly (EIA) and Equatorial Plasma Bubble (EPB) is critical for understanding ionospheric dynamic characteristics and mitigating their adverse impacts on satellite‐based communication and navigation systems. This study presents a novel method to automatically detect equatorial ionospheric irregularities (including EIA and EPB) using deep learning models based on the Global‐scale Observations of the Limb and Disk (GOLD) images. Specifically, GOLD 135.6 nm nightglow data from 2021 to 2023 are utilized to train and evaluate the performance of three deep learning architectures: U‐shaped Convolutional Neural Network (U‐Net), DeepLab Version 3 Plus (DeepLabV3+), and Feature Pyramid Network (FPN). The results indicate that all three models achieve the detection rate of 100% in identifying equatorial ionospheric irregularities. In overall comparison, the U‐Net model outperforms the others, achieving the highest coverage rate of 95.78%. To assess the general capability of the models, EIA and EPB events during 2024 are also investigated. The results show that all three models can still achieved the detection rate of 100% and the U‐Net model maintains the best coverage rate of 95.96%. In addition, all three models have achieved superior performance across common evaluation metrics in terms of accuracy, precision, recall, and F1‐score.

Förderung

  • National Natural Science Foundation of China
  • Guangzhou Municipal Science and Technology Project

Zitieren

APA

Tang, L., Li, Y., Zhao, H., Chen, X., Ding, Q., Zou, F., & Zhang, X. (2026). Automatic Equatorial Ionospheric Irregularities Detection From GOLD Images Using Deep Learning. Space Weather, 24(11). https://doi.org/10.1029/2026sw005273

Vancouver

Tang L, Li Y, Zhao H, Chen X, Ding Q, Zou F, et al. Automatic Equatorial Ionospheric Irregularities Detection From GOLD Images Using Deep Learning. Space Weather. 2026;24(11). doi:10.1029/2026sw005273

BibTeX
@article{Tang2026,
  author = {Tang, Long and Li, Yumei and Zhao, Huijian and Chen, Xueyan and Ding, Qiong and Zou, Fang and Zhang, Xiaohong},
  title = {Automatic Equatorial Ionospheric Irregularities Detection From GOLD Images Using Deep Learning},
  journal = {Space Weather},
  year = {2026},
  volume = {24},
  number = {11},
  doi = {10.1029/2026sw005273}
}
Kostenlos starten →