Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar
Quantitative precipitation estimation (QPE) plays an important role in meteorology and hydrology. Currently, multichannel Doppler radar image is used for QPE based on some traditional methods like the <i>Z</i> − <i>R</i> relationship, which struggles to capture the complicate...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/4/1111 |
| _version_ | 1827755726991261696 |
|---|---|
| author | Weishu Chen Wenjun Hua Mengshu Ge Fei Su Na Liu Yujia Liu Anyuan Xiong |
| author_facet | Weishu Chen Wenjun Hua Mengshu Ge Fei Su Na Liu Yujia Liu Anyuan Xiong |
| author_sort | Weishu Chen |
| collection | DOAJ |
| description | Quantitative precipitation estimation (QPE) plays an important role in meteorology and hydrology. Currently, multichannel Doppler radar image is used for QPE based on some traditional methods like the <i>Z</i> − <i>R</i> relationship, which struggles to capture the complicated non-linear spatial relationship. Encouraged by the great success of using Deep Learning (DL) segmentation networks in medical science and remoting sensing, a UNet-based network named Reweighted Regression Encoder–Decoder Net (RRED-Net) is proposed for QPE in this paper, which can learn more complex non-linear information from the training data. Firstly, wavelet transform (WT) is introduced to alleviate the noise in radar images. Secondly, a wider receptive field is obtained by taking advantage of attention mechanisms. Moreover, a new Regression Focal Loss is proposed to handle the imbalance problem caused by the extreme long-tailed distribution in precipitation. Finally, an efficient feature selection strategy is designed to avoid exhaustion experiments. Extensive experiments on 465 real processes data demonstrate that the superiority of our proposed RRED-Net not only in the threat score (TS) in the severe precipitation (from 17.6% to 39.6%, ≥20 mm/h) but also the root mean square error (RMSE) comparing to the traditional Z-R relationship-based method (from 2.93 mm/h to 2.58 mm/h, ≥20 mm/h), baseline models and other DL segmentation models. |
| first_indexed | 2024-03-11T08:12:02Z |
| format | Article |
| id | doaj.art-a4ba2606148f4bdaa0f03d168861b80b |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T08:12:02Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a4ba2606148f4bdaa0f03d168861b80b2023-11-16T23:03:51ZengMDPI AGRemote Sensing2072-42922023-02-01154111110.3390/rs15041111Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler RadarWeishu Chen0Wenjun Hua1Mengshu Ge2Fei Su3Na Liu4Yujia Liu5Anyuan Xiong6Key Laboratory of Interactive Technology and Experience System, Ministry of Culture and Tourism, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaKey Laboratory of Interactive Technology and Experience System, Ministry of Culture and Tourism, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaKey Laboratory of Interactive Technology and Experience System, Ministry of Culture and Tourism, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaKey Laboratory of Interactive Technology and Experience System, Ministry of Culture and Tourism, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaNational Meteorological Information Center, Beijing 100081, ChinaNational Meteorological Information Center, Beijing 100081, ChinaNational Meteorological Information Center, Beijing 100081, ChinaQuantitative precipitation estimation (QPE) plays an important role in meteorology and hydrology. Currently, multichannel Doppler radar image is used for QPE based on some traditional methods like the <i>Z</i> − <i>R</i> relationship, which struggles to capture the complicated non-linear spatial relationship. Encouraged by the great success of using Deep Learning (DL) segmentation networks in medical science and remoting sensing, a UNet-based network named Reweighted Regression Encoder–Decoder Net (RRED-Net) is proposed for QPE in this paper, which can learn more complex non-linear information from the training data. Firstly, wavelet transform (WT) is introduced to alleviate the noise in radar images. Secondly, a wider receptive field is obtained by taking advantage of attention mechanisms. Moreover, a new Regression Focal Loss is proposed to handle the imbalance problem caused by the extreme long-tailed distribution in precipitation. Finally, an efficient feature selection strategy is designed to avoid exhaustion experiments. Extensive experiments on 465 real processes data demonstrate that the superiority of our proposed RRED-Net not only in the threat score (TS) in the severe precipitation (from 17.6% to 39.6%, ≥20 mm/h) but also the root mean square error (RMSE) comparing to the traditional Z-R relationship-based method (from 2.93 mm/h to 2.58 mm/h, ≥20 mm/h), baseline models and other DL segmentation models.https://www.mdpi.com/2072-4292/15/4/1111QPEdeep learningprecipitationDoppler radarwaveletlong-tailed distribution |
| spellingShingle | Weishu Chen Wenjun Hua Mengshu Ge Fei Su Na Liu Yujia Liu Anyuan Xiong Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar Remote Sensing QPE deep learning precipitation Doppler radar wavelet long-tailed distribution |
| title | Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar |
| title_full | Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar |
| title_fullStr | Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar |
| title_full_unstemmed | Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar |
| title_short | Severe Precipitation Recognition Using Attention-UNet of Multichannel Doppler Radar |
| title_sort | severe precipitation recognition using attention unet of multichannel doppler radar |
| topic | QPE deep learning precipitation Doppler radar wavelet long-tailed distribution |
| url | https://www.mdpi.com/2072-4292/15/4/1111 |
| work_keys_str_mv | AT weishuchen severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar AT wenjunhua severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar AT mengshuge severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar AT feisu severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar AT naliu severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar AT yujialiu severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar AT anyuanxiong severeprecipitationrecognitionusingattentionunetofmultichanneldopplerradar |