Radar Echo Extrapolation Algorithm Integrating Spatiotemporal Reconstruction Unit and Transformer

NAVIGATE

Table of Contents

STATISTICS

Viewed31

Downloads23

Radar Echo Extrapolation Algorithm Integrating Spatiotemporal Reconstruction Unit and Transformer

[HTML] PDF下载 (23)

[1]FANG Wei,WANG Haoxi.Radar Echo Extrapolation Algorithm Integrating Spatiotemporal Reconstruction Unit and Transformer[J].Journal of Zhengzhou University (Engineering Science),2025,46(05):137-144.[doi:10.13705/j.issn.1671-6833.2025.02.007]

Copy

Journal of Zhengzhou University (Engineering Science)[ISSN 1671-6833/CN 41-1339/T] Volume: 46 Number of periods: 2025 05 Page number: 137-144 Column: Public date: 2025-08-10

Title:: Radar Echo Extrapolation Algorithm Integrating Spatiotemporal Reconstruction Unit and Transformer

Author(s):: FANG Wei¹; 2; 3; 4; WANG Haoxi¹; 1. School of Computer Science, Nanjing University of Information Science & Technology, Nanjing 210044, China; 2.School of Software, Nanjing University of Information Science & Technology, Nanjing 210044, China; 3.China Meteorological Administration Basin Heavy Rainfall Key Laboratory/Hubei Key Laboratory for Heavy Rain Monitoring and Warning Research, Wuhan Research Institute of Heavy Rain, China Meteorological Administration, Wuhan 430205,China;4.State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081, China

Keywords:: deep learning; radar echo extrapolation; Transformer; spatiotemporal reconstruction unit; global average pooling

CLC:: TP399

DOI:: 10.13705/j.issn.1671-6833.2025.02.007

Abstract:: Addressing the limitations of existing deep learning-based radar echo extrapolation algorithms in spatialtemporal feature extraction and long-term dependency modeling, a novel SRU-Former model that integrated spatialtemporal reconstruction unit (SRU) and Transformer was proposed for radar echo extrapolation. Firstly, a newly designed SRU was introduced into the model′s encoder and decoder to extract fine-grained spatiotemporal features from radar images via separation, transformation, and reconstruction. Secondly, a variant architecture model of Transformer, Poolformer, was introduced between the encoder and decoder, using global average pooling to replace the self-attention mechanism, thereby assisting the model in modeling highly dynamic radar sequences. Finally, SRU-Former was trained and tested on two meteorological radar datasets from Jiangsu Province and Shanghai City, respectively, and compared with current mainstream deep learning models. In the 2-hour extrapolation task, SRUFormer achieved optimal values in four metrics: CSI, FAR, MSE, and SSIM. Specifically, CSI was improved by 0.020 on the Jiangsu Province dataset and by 0.048 on the Shanghai City dataset. Experimental results showed that SRU-Former effectively improved model prediction accuracy, with more precise capture of strong echo regions and clearer detail textures in later extrapolation stages.

References:: [1]薛丰昌, 章超钦, 王文硕, 等. 基于自注意力和门控循环神经网络的雷达回波外推算法研究[J]. 气象学报, 2024, 82(1): 127-135.
XUE F C, ZHANG C Q, WANG W S, et al. Improving radar echo extrapolation algorithms based on selfattention and gated recurrent neural networks[J]. Acta Meteorologica Sinica, 2024, 82(1): 127-135.
[2]黄旋旋, 朱科锋, 赵坤. 改进后TREC外推方法在台风临近降雨预报中的应用[J]. 气象科学, 2017, 37 (5): 610-618.
HUANG X X, ZHU K F, ZHAO K. The improved TREC algorithm and its application to the precipitation now-casting for landfall typhoon[J]. Journal of the Meteorological Sciences, 2017, 37(5): 610-618.
[3]DEL MORAL A, RIGO T, LLASAT M C. A radar-based centroid tracking algorithm for severe weather surveillance: identifying split/merge processes in convective systems[J]. Atmospheric Research, 2018, 213: 110-120.
[4]吴剑坤, 陈明轩. 基于雷达回波区域跟踪算法的临近预报技术进展[J]. 气象科技, 2018, 46(5): 899-909.
WU J K, CHEN M X. Advances in nowcasting techniques based on radar echo area tracking algorithm[J]. Meteorological Science and Technology, 2018, 46(5): 899-909.
[5]SHI X J, CHEN Z R, WANG H, et al. Convolutional LSTM network: a machine learning approach for precipitation nowcasting[J]. Advances in Neural Information Processing Systems, 2015, 2015: 802-810.
[6]WANG Y B, WU H X, ZHANG J J, et al. PredRNN: recurrent neural networks for predictive learning using spatiotemporal LSTMs[C]∥ Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates,2017: 879-888.
[7]WANG Y B, GAO Z F, LONG M S, et al. PredRNN++: towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning[EB/OL]. (2018-04-17) [2024-10-16]. https:∥doi. org/10. 48550/arXiv. 1804.06300.
[8]方巍, 庞林, 易伟楠. 基于深度时空融合网络的雷达回波外推模型[J]. 电子学报, 2023, 51(9): 2526-2538.
FANG W, PANG L, YI W N. Radar echo extrapolation model based on deep spatio-temporal fusion neural network[J]. Acta Electronica Sinica, 2023, 51(9): 25262538.
[9]刘颖, 杨鹏飞, 张立军, 等. 前馈神经网络和循环神经网络的鲁棒性验证综述[J]. 软件学报, 2023, 34 (7): 3134-3166. .
LIU Y, YANG P F, ZHANG L J, et al. Survey on robustness verification of feedforward neural networks and recurrent neural networks[J]. Journal of Software, 2023, 34(7): 3134-3166.
[10] RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation [C]∥Lecture Notes in Computer Science. Cham: Springer, 2015: 234-241.
[11]王善昊, 胡志群, 王福增, 等. 基于ConvLSTM融合RMAPS-NOW数据的雷达回波外推研究[J]. 气象学报, 2024, 82(4): 554-567.
WANG S H, HU Z Q, WANG F Z, et al. Extrapolation of radar echo based on ConvLSTM with fusion of RMAPSNOW data[J]. Acta Meteorologica Sinica, 2024, 82 (4): 554-567.
[12] TREBING K, STAN `CZYK T, MEHRKANOON S. SmaAt-UNet: precipitation nowcasting using a small attention-UNet architecture[J]. Pattern Recognition Letters, 2021, 145: 178-186.
[13] GAO Z Y, TAN C, WU L R, et al. SimVP: simpler yet better video prediction[C]∥2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2022: 3160-3170.
[14]李卫军, 张新勇, 高庾潇, 等. 基于门控时空注意力的视频帧预测模型[J]. 郑州大学学报(工学版), 2024, 45(1): 70-77, 121.
LI W J, ZHANG X Y, GAO Y X, et al. Video frame prediction model based on gated spatio-temporal attention [J]. Journal of Zhengzhou University (Engineering Science), 2024, 45(1): 70-77, 121.
[15] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL].(2017-07-12)[2024-1016].https:∥doi.org/10.48550/arXiv.1706.03762.
[16]方巍, 齐媚涵. 基于深度学习的高时空分辨率降水临近预报方法[J]. 地球科学与环境学报, 2023, 45 (3): 706-718.
FANG W, QI M H. Precipitation nowcasting method with high spatio-temporal resolution based on deep learning [J]. Journal of Earth Sciences and Environment, 2023, 45(3): 706-718.
[17] LIU Z, LIN Y T, CAO Y, et al. Swin Transformer: hierarchical vision Transformer using shifted windows[C]∥ 2021 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2021: 9992-10002.
[18] BI K F, XIE L X, ZHANG H H, et al. Pangu-weather: a 3D high-resolution model for fast and accurate global weather forecast[EB/OL]. (2022-11-03)[2024-1016]. https:∥doi.org/10.48550/arXiv.2211.02556.
[19] DALMAZ O, YURT M, CUKUR T. ResViT: residual vision transformers for multimodal medical image synthesis [J]. IEEE Transactions on Medical Imaging, 2022, 41(10): 2598-2614.
[20] YU W H, LUO M, ZHOU P, et al. MetaFormer is actually what you need for vision[C]∥2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE, 2022: 10809-10819.
[21]魏明军, 王镆涵, 刘亚志, 等. 基于特征融合和混合注意力的小目标检测[J]. 郑州大学学报(工学版), 2024, 45(3): 72-79.
WEI M J, WANG M H, LIU Y Z, et al. Small object detection based on feature fusion and mixed attention[J]. Journal of Zhengzhou University (Engineering Science), 2024, 45(3): 72-79.
[22] LI J F, WEN Y, HE L H. SCConv: spatial and channel reconstruction convolution for feature redundancy[C]∥ 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway: IEEE, 2023: 6153-6162.
[23] SMITH L N. Cyclical learning rates for training neural networks[C]∥2017 IEEE Winter Conference on Applications of Computer Vision. Piscataway: IEEE, 2017: 464-472.
[24] LIU Z, MAO H Z, WU C Y, et al. A ConvNet for the 2020s[C]∥2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 11966-11976.
[25] LI K C, WANG Y L, ZHANG J H, et al. UniFormer: unifying convolution and self-attention for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(10): 12581-1260.

Similar References:

Memo

Last Update: 2025-09-19