EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module
In the field of weather forecasting, improving the accuracy of nowcasting is a highly researched topic, and radar echo extrapolation technology plays a crucial role in this process. Aiming to address the limitations of existing deep learning methods in radar echo extrapolation, this paper proposes a...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/6/1103 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850280283078656000 |
|---|---|
| author | Guangxin He Wei Wu Jing Han Jingjia Luo Lei Lei |
| author_facet | Guangxin He Wei Wu Jing Han Jingjia Luo Lei Lei |
| author_sort | Guangxin He |
| collection | DOAJ |
| description | In the field of weather forecasting, improving the accuracy of nowcasting is a highly researched topic, and radar echo extrapolation technology plays a crucial role in this process. Aiming to address the limitations of existing deep learning methods in radar echo extrapolation, this paper proposes a spatio-temporal long short-term memory (LSTM) network model that integrates an attention mechanism and the full-dimensional dynamic convolution technique. The multi-scale spatial and temporal features of radar images can be fully extracted by an efficient multi-scale attention module to enhance the model’s ability to perceive global and local information. The full-dimensional dynamic convolutional module introduces the dynamic attention mechanism in the spatial position and input and output channels of the convolutional kernel, adaptively adjusts the weight of the convolutional kernel, and improves the flexibility and efficiency of feature extraction. Combined with the network constructed by the above modules, the accuracy and time dependence of the model for predicting the strong echo region are significantly improved. Our experiments based on Jiangsu meteorological radar data show that the model achieved excellent results in terms of the Critical Success Index (CSI) and Heidke Skill Score (HSS), which show its efficiency and stability in predicting radar echo, especially under the condition of a high 35 dBZ threshold, and its prediction performance improved significantly. It provides an effective solution for fine short-term impending precipitation forecasting. |
| format | Article |
| id | doaj-art-18aac820ffc444d980ffe699fcab8a1a |
| institution | OA Journals |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-18aac820ffc444d980ffe699fcab8a1a2025-08-20T01:48:49ZengMDPI AGRemote Sensing2072-42922025-03-01176110310.3390/rs17061103EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution ModuleGuangxin He0Wei Wu1Jing Han2Jingjia Luo3Lei Lei4Key Laboratory of Meteorological Disaster, Ministry of Education (KLME), International Joint Research Laboratory on Climate and Environment Change (ILCEC), Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaKey Laboratory of Meteorological Disaster, Ministry of Education (KLME), International Joint Research Laboratory on Climate and Environment Change (ILCEC), Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaHainan Institute of Meteorological Sciences, Haikou 570203, ChinaKey Laboratory of Meteorological Disaster, Ministry of Education (KLME), International Joint Research Laboratory on Climate and Environment Change (ILCEC), Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaBeijing Meteorological Observatory, Beijing 100097, ChinaIn the field of weather forecasting, improving the accuracy of nowcasting is a highly researched topic, and radar echo extrapolation technology plays a crucial role in this process. Aiming to address the limitations of existing deep learning methods in radar echo extrapolation, this paper proposes a spatio-temporal long short-term memory (LSTM) network model that integrates an attention mechanism and the full-dimensional dynamic convolution technique. The multi-scale spatial and temporal features of radar images can be fully extracted by an efficient multi-scale attention module to enhance the model’s ability to perceive global and local information. The full-dimensional dynamic convolutional module introduces the dynamic attention mechanism in the spatial position and input and output channels of the convolutional kernel, adaptively adjusts the weight of the convolutional kernel, and improves the flexibility and efficiency of feature extraction. Combined with the network constructed by the above modules, the accuracy and time dependence of the model for predicting the strong echo region are significantly improved. Our experiments based on Jiangsu meteorological radar data show that the model achieved excellent results in terms of the Critical Success Index (CSI) and Heidke Skill Score (HSS), which show its efficiency and stability in predicting radar echo, especially under the condition of a high 35 dBZ threshold, and its prediction performance improved significantly. It provides an effective solution for fine short-term impending precipitation forecasting.https://www.mdpi.com/2072-4292/17/6/1103radar echo extrapolationdeep learninglong short-term memory networksshort-term weather forecast |
| spellingShingle | Guangxin He Wei Wu Jing Han Jingjia Luo Lei Lei EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module Remote Sensing radar echo extrapolation deep learning long short-term memory networks short-term weather forecast |
| title | EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module |
| title_full | EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module |
| title_fullStr | EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module |
| title_full_unstemmed | EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module |
| title_short | EOST-LSTM: Long Short-Term Memory Model Combined with Attention Module and Full-Dimensional Dynamic Convolution Module |
| title_sort | eost lstm long short term memory model combined with attention module and full dimensional dynamic convolution module |
| topic | radar echo extrapolation deep learning long short-term memory networks short-term weather forecast |
| url | https://www.mdpi.com/2072-4292/17/6/1103 |
| work_keys_str_mv | AT guangxinhe eostlstmlongshorttermmemorymodelcombinedwithattentionmoduleandfulldimensionaldynamicconvolutionmodule AT weiwu eostlstmlongshorttermmemorymodelcombinedwithattentionmoduleandfulldimensionaldynamicconvolutionmodule AT jinghan eostlstmlongshorttermmemorymodelcombinedwithattentionmoduleandfulldimensionaldynamicconvolutionmodule AT jingjialuo eostlstmlongshorttermmemorymodelcombinedwithattentionmoduleandfulldimensionaldynamicconvolutionmodule AT leilei eostlstmlongshorttermmemorymodelcombinedwithattentionmoduleandfulldimensionaldynamicconvolutionmodule |