EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

The MS-RadarFormer: A Transformer-Based Multi-Scale Deep Learning Model for Radar Echo Extrapolation

Huantong Geng, Fangli Wu, Xiaoran Zhuang, Liangchao Geng, Boyang Xie, Zhanpeng Shi

Remote Sensing · 10.01.2024 · 16 (2) · 274

DOI: 10.3390/rs16020274 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 10.01.2024. Verfasst von Huantong Geng und 5 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Open Grants of China Meteorological Administration Radar Meteorology Key Laboratory, China Meteorological Administration Innovation and Development Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As a spatial–temporal sequence prediction task, radar echo extrapolation aims to predict radar echoes’ future movement and intensity changes based on historical radar observations. Two urgent issues still need to be addressed in deep learning radar echo extrapolation models. First, the predicted radar echo sequences often exhibit echo-blurring phenomena. Second, over time, the output echo intensities from the model gradually weaken. In this paper, we propose a novel model called the MS-RadarFormer, a Transformer-based multi-scale deep learning model for radar echo extrapolation, to mitigate the two above issues. We introduce a multi-scale design in the encoder–decoder structure and a Spatial–Temporal Attention block to improve the precision of radar echoes and establish long-term dependencies of radar echo features. The model uses a non-autoregressive approach for echo prediction, avoiding accumulation errors during the recursive generation of future echoes. Compared to the baseline, our model shows an average improvement of 15.8% in the critical success index (CSI), an average decrease of 8.3% in the false alarm rate (FAR), and an average improvement of 16.2% in the Heidke skill score (HSS).

Förderung

  • National Natural Science Foundation of China
  • Open Grants of China Meteorological Administration Radar Meteorology Key Laboratory
  • China Meteorological Administration Innovation and Development Program
  • China Meteorological Administration Key Innovation Team

Zitieren

APA

Geng, H., Wu, F., Zhuang, X., Geng, L., Xie, B., & Shi, Z. (2024). The MS-RadarFormer: A Transformer-Based Multi-Scale Deep Learning Model for Radar Echo Extrapolation. Remote Sensing, 16(2), 274. https://doi.org/10.3390/rs16020274

Vancouver

Geng H, Wu F, Zhuang X, Geng L, Xie B, Shi Z. The MS-RadarFormer: A Transformer-Based Multi-Scale Deep Learning Model for Radar Echo Extrapolation. Remote Sensing. 2024;16(2):274. doi:10.3390/rs16020274

BibTeX
@article{Geng2024,
  author = {Geng, Huantong and Wu, Fangli and Zhuang, Xiaoran and Geng, Liangchao and Xie, Boyang and Shi, Zhanpeng},
  title = {The MS-RadarFormer: A Transformer-Based Multi-Scale Deep Learning Model for Radar Echo Extrapolation},
  journal = {Remote Sensing},
  year = {2024},
  volume = {16},
  number = {2},
  pages = {274},
  doi = {10.3390/rs16020274}
}
Kostenlos starten →