arrow
返回

SFTformer: A Spatial-Frequency-Temporal Correlation-Decoupling Transformer for Radar Echo Extrapolation

delete2024-01-01
delete5
delete
OA
AI
L
Liangyu Xu
W
Wanxuan Lu
H
Hongfeng Yu *
F
Fanglong Yao
X
Xian Sun
K
Kun Fu
DOI:10.1109/TGRS.2024.3367857delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Extrapolating future weather radar echoes from past observations is a complex task vital for precipitation nowcasting. The spatial morphology and temporal evolution of radar echoes exhibit a certain degree of correlation, yet they also possess independent characteristics. Existing methods learn unified spatial and temporal representations in a highly coupled feature space, emphasizing the correlation between spatial and temporal features but neglecting the explicit modeling of their independent characteristics, which may result in mutual interference between them. To effectively model the spatiotemporal dynamics of radar echoes, we propose a spatial-frequency-temporal correlation-decoupling transformer (SFTformer). The model leverages stacked multiple SFT-Blocks to not only mine the correlation of the spatiotemporal dynamics of echo cells but also avoid the mutual interference between the temporal modeling and the spatial morphology refinement by decoupling them. Furthermore, inspired by the practice that weather forecast experts effectively review historical echo evolution to make accurate predictions, SFTfomer incorporates a joint training paradigm for historical echo sequence reconstruction and future echo sequence prediction. Experimental results on the HKO-7 dataset and ChinaNorth-2021 dataset demonstrate the superior performance of SFTfomer in short-term (1 h), mid-term (2 h), and long-term (3 h) precipitation nowcasting.
Keyword:
Radar
Spatiotemporal phenomena
Extrapolation
Transformers
Predictive models
Correlation
Morphology
Correlation decoupling
prediction-reconstruction joint training paradigm
SFT-Block

期刊

IEEE Transactions on Geoscience and Remote Sensing 封面图
IEEE Transactions on Geoscience and Remote Sensing
IF:
8.6
论文数:
2.1W
被引数:
10.7W

机构

C
chinese academy of sciences
学者数:
56.7W
论文数: 45.0W
被引数: 704
引用论文

引用论文

Fatigue and damage accumulation of veneer porcelain pressed on Y-TZP
err2010-04-01
err0
PREAI
errEstevam A. Bonfante; Paulo G. Coelho; Petra C. Guess; Van P. Thompson; Nelson R.F.A. Silva
err分享
err收藏
err分享
err收藏
Development of physical, mechanical, antibacterial and cell growth properties of poly(glycerol sebacate urethane) (PGSU) with helping of curcumin and hydroxyapatite nanoparticles
err2021-01-01
err0
errOAAI
errVafa Fakhri; Aliakbar Jafari; Mir Alireza Shafiei; Mohammad Vahid Ehteshamfar; Shima Khalighiyan; Hadi Hosseini; Vahabodin Goodarzi; Frederik R. Wurm; Mehrdad Moosazadeh Moghaddam; Hossein Ali Khonakdar
err分享
err收藏
Heart rate control during cycling exercise using Takagi-Sugeno models
err2011-01-01
err0
errOAAI
errSami MOHAMMAD; Thierry Marie GUERRA; Jean Marie GROBOIS; Bernard HECQUET
err分享
err收藏
学者 查看更多内容