A Deep Learning Model with Axial Attention for Radar Echo Extrapolation

Radar echo extrapolation is an important approach in precipitation nowcasting which utilizes historical radar echo images to predict future echo images. In this paper, we introduce the self-attention mechanism into Trajectory Gated Recurrent Unit (TrajGRU) model. Under the sequence-to-sequence frame...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu-Mei Xie, Ying-Liang Zhao, Shu-Yan Huang
Format: Article
Language:English
Published: Taylor & Francis Group 2024-12-01
Series:Applied Artificial Intelligence
Online Access:https://www.tandfonline.com/doi/10.1080/08839514.2024.2311003
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850255154000953344
author Yu-Mei Xie
Ying-Liang Zhao
Shu-Yan Huang
author_facet Yu-Mei Xie
Ying-Liang Zhao
Shu-Yan Huang
author_sort Yu-Mei Xie
collection DOAJ
description Radar echo extrapolation is an important approach in precipitation nowcasting which utilizes historical radar echo images to predict future echo images. In this paper, we introduce the self-attention mechanism into Trajectory Gated Recurrent Unit (TrajGRU) model. Under the sequence-to-sequence framework, we have developed a novel convolutional recurrent neural network called Self-attention Trajectory Gated Recurrent Unit (SA-TrajGRU), which incorporates the self-attention mechanism. The SA-TrajGRU model which combines spatiotemporal variant structure in TrajGRU and self-attention module is simple and effective. We evaluate our approach on the Moving MNIST-2 dataset and the CIKM AnalytiCup 2017 radar echo dataset. The experimental results show that the performance of the proposed SA-TrajGRU model is comparable to other convolutional recurrent neural network models. HSS and CSI scores of the SA-TrajGRU model are higher than scores of other models under the radar echo threshold of 25 dBZ, indicating that the SA-TrajGRU model has the most accurate prediction results under this threshold.
format Article
id doaj-art-ef6bc77075a9415c8e8aa0d948765dda
institution OA Journals
issn 0883-9514
1087-6545
language English
publishDate 2024-12-01
publisher Taylor & Francis Group
record_format Article
series Applied Artificial Intelligence
spelling doaj-art-ef6bc77075a9415c8e8aa0d948765dda2025-08-20T01:56:56ZengTaylor & Francis GroupApplied Artificial Intelligence0883-95141087-65452024-12-0138110.1080/08839514.2024.2311003A Deep Learning Model with Axial Attention for Radar Echo ExtrapolationYu-Mei Xie0Ying-Liang Zhao1Shu-Yan Huang2College of Electronics and Information Science, Fujian Jiangxia University, Fuzhou, ChinaSchool of Computer Science and Technology, Xi’an Jiaotong University, Xi’an, ChinaCollege of Electronics and Information Science, Fujian Jiangxia University, Fuzhou, ChinaRadar echo extrapolation is an important approach in precipitation nowcasting which utilizes historical radar echo images to predict future echo images. In this paper, we introduce the self-attention mechanism into Trajectory Gated Recurrent Unit (TrajGRU) model. Under the sequence-to-sequence framework, we have developed a novel convolutional recurrent neural network called Self-attention Trajectory Gated Recurrent Unit (SA-TrajGRU), which incorporates the self-attention mechanism. The SA-TrajGRU model which combines spatiotemporal variant structure in TrajGRU and self-attention module is simple and effective. We evaluate our approach on the Moving MNIST-2 dataset and the CIKM AnalytiCup 2017 radar echo dataset. The experimental results show that the performance of the proposed SA-TrajGRU model is comparable to other convolutional recurrent neural network models. HSS and CSI scores of the SA-TrajGRU model are higher than scores of other models under the radar echo threshold of 25 dBZ, indicating that the SA-TrajGRU model has the most accurate prediction results under this threshold.https://www.tandfonline.com/doi/10.1080/08839514.2024.2311003
spellingShingle Yu-Mei Xie
Ying-Liang Zhao
Shu-Yan Huang
A Deep Learning Model with Axial Attention for Radar Echo Extrapolation
Applied Artificial Intelligence
title A Deep Learning Model with Axial Attention for Radar Echo Extrapolation
title_full A Deep Learning Model with Axial Attention for Radar Echo Extrapolation
title_fullStr A Deep Learning Model with Axial Attention for Radar Echo Extrapolation
title_full_unstemmed A Deep Learning Model with Axial Attention for Radar Echo Extrapolation
title_short A Deep Learning Model with Axial Attention for Radar Echo Extrapolation
title_sort deep learning model with axial attention for radar echo extrapolation
url https://www.tandfonline.com/doi/10.1080/08839514.2024.2311003
work_keys_str_mv AT yumeixie adeeplearningmodelwithaxialattentionforradarechoextrapolation
AT yingliangzhao adeeplearningmodelwithaxialattentionforradarechoextrapolation
AT shuyanhuang adeeplearningmodelwithaxialattentionforradarechoextrapolation
AT yumeixie deeplearningmodelwithaxialattentionforradarechoextrapolation
AT yingliangzhao deeplearningmodelwithaxialattentionforradarechoextrapolation
AT shuyanhuang deeplearningmodelwithaxialattentionforradarechoextrapolation