留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

利用深度学习开展偏振雷达定量降水估测研究

皇甫江 胡志群 郑佳锋 朱永杰 尹晓燕 左园园

皇甫江,胡志群,郑佳锋,朱永杰,尹晓燕,左园园. 2022. 利用深度学习开展偏振雷达定量降水估测研究. 气象学报,80(4):565-577 doi: 10.11676/qxxb2022.046
引用本文: 皇甫江,胡志群,郑佳锋,朱永杰,尹晓燕,左园园. 2022. 利用深度学习开展偏振雷达定量降水估测研究. 气象学报,80(4):565-577 doi: 10.11676/qxxb2022.046
Huangfu Jiang, Hu Zhiqun, Zheng Jiafeng, Zhu Yongjie, Yin Xiaoyan, Zuo Yuanyuan. 2022. A study on polarization radar quantitative precipitation estimation using deep learning. Acta Meteorologica Sinica, 80(4):565-577 doi: 10.11676/qxxb2022.046
Citation: Huangfu Jiang, Hu Zhiqun, Zheng Jiafeng, Zhu Yongjie, Yin Xiaoyan, Zuo Yuanyuan. 2022. A study on polarization radar quantitative precipitation estimation using deep learning. Acta Meteorologica Sinica, 80(4):565-577 doi: 10.11676/qxxb2022.046

利用深度学习开展偏振雷达定量降水估测研究

doi: 10.11676/qxxb2022.046
基金项目: 国家重点研发计划项目(2019YFC1510304)、广东省重点领域研发计划项目(2020B1111200001)、中国气象局大气探测重点开放实验室开放课题(U2021Z05)、青年科学基金项目(42105141)
详细信息
    作者简介:

    皇甫江,主要从事雷达气象研究。E-mail:3200101052@stu.cuit.edu.cn

    通讯作者:

    胡志群,主要从事雷达气象研究。 E-mail:huzq@cma.gov.cn

  • 中图分类号: P415

A study on polarization radar quantitative precipitation estimation using deep learning

  • 摘要: 利用2018—2020年经偏振升级改造后的广州S波段双偏振雷达(CINRAD/SAD)82892个体扫的0.5°仰角数据,以及雷达100 km探测范围内1109个雨量站共计538560个分钟雨量数据,分别构建了单参量、三参量雷达定量降水估测(QPE)深度学习网络架构(Z-Rnet、KDP-Rnet、Pol-Rnet),并以KDP=0.5°/km为阈值分别训练得到大雨、小雨、总体等9个定量降水估测模型。在常用的均方误差作为损失函数的基础上,对不同降水强度采用不同权重提出了自定义损失函数,并利用比率偏差、相对偏差、均方差、平均绝对误差和平均相对误差作为评价指标对模型进行评估。通过对以积层混合云为主、以对流云为主和以层状云为主的3次降水过程的模型验证结果表明,利用深度学习训练的模型有较好的定量降水估测效果,区分雨强的小雨、大雨模型比不区分雨强的总体模型的效果要好。采用自定义损失函数模型效果更好,其均方差、平均绝对误差和平均相对误差分别较采用传统均方误差损失函数提升了8.62%、12.52%、16.34%。自定义损失函数中,采用ZH-ZDR-KDP三参量网络架构训练得到的定量降水估测模型效果最好,其均方差、平均绝对误差和平均相对误差分别较采用ZH的单参量Z-Rnet架构提升6.82%、8.43%、7.22%;较采用KDP的单参量KDP-Rnet架构提升12.33%、17.61%、17.26%。

     

  • 图 1  雨量计上空雷达0.5°仰角3×3个距离库数据示意

    Figure 1.  Schematic diagram of the 0.5° elevation angle 3×3 distance radar data above the rain gauge

    图 2  双偏振雷达资料处理技术路线

    Figure 2.  Flowchart of dual polarization radar data preprocessing

    图 3  自动雨量站分布 (图中间的黑色三角为雷达位置,黑色圆点为雷达探测100 km范围内雨量站)

    Figure 3.  Distribution of automatic rainfall stations (The black triangles in the center show the radar position,and the black dots indicate rainfall stations within 100 km of radar detection)

    图 4  雨量数据预处理技术路线

    Figure 4.  Flowchart of rainfall data preprocessing

    图 5  单参量网络架构流程

    Figure 5.  Flowchart of single-moment network

    图 6  多参量网络架构流程

    Figure 6.  Flowchart of three-moment network

    图 7  采用均方误差作为损失函数时,大雨模型(a1、b1、c1)、小雨模型(a2、b2、c2)和总模型(a3、b3、c3)中分别采用KDP (a1、a2、a3)、ZH (b1、b2、b3) 和ZH-ZDR-KDP (c1、c2、c3) 作为输入因子的散点 (色阶为解释方差)

    Figure 7.  Scatter plots of heavy rain (a1,b1,c1),light rain (a2,b2,c2),all rain (a3,b3,c3) data models fitted with traditional MSE as loss function and with KDP (a1,a2,a3),ZH (b1,b2,b3) and ZH-ZDR-KDP (c1,c2,c3) as input factors,respectively (The color code indicates the explanatory variance)

    图 8  同图7,但为采用自定义损失函数

    Figure 8.  Same as Fig. 7 but with self-defined loss function

    图 9  2019年6月11日12时36分0.5°仰角雷达PPI

    Figure 9.  Radar PPI of 0.5° elevation at 12:36 UTC 11 June 2019

    图 10  2020年6月9日10时48分0.5°仰角雷达PPI

    Figure 10.  Radar PPI of 0.5° elevation at 10:48 UTC 9 June 2020

    图 11  2018年9月16日06时30分0.5°仰角雷达回波PPI

    Figure 11.  Radar PPI of 0.5° elevation at 06:30 UTC 16 September 2018

    表  1  各参数标准化的最大、最小值设置

    Table  1.   Maximum and minimum values of standardized parameters

    参数MinMax
    R0.1 mm/h250 mm/h
    ZH10 dB70 dB
    ZDR0 dB7 dB
    KDP0°/km7°/km
    下载: 导出CSV

    表  2  各模型训练集数据量

    Table  2.   The amount in the training dataset of each model

    模型/数据训练数据数量(组)
    大雨模型71490
    小雨模型471200
    总模型620690
    合计1163380
    下载: 导出CSV

    表  3  自定义损失函数权重系数

    Table  3.   Weight coefficients of self-defined loss function

    模型参数雨强区间(mm/h)权重
    大雨模型KDP(0,10.0)、[10.0, 25.0)、[25.0,45.0)、[45.0,60.0)、[60.0,100.0)、[100.0, 250)8.0、 5.0、 1.1、 3.0、 6.0、 10.0
    ZH(0,10.0)、[10.0, 25.0)、[25.0,45.0)、[45.0,60.0)、[60.0,100.0)、[100.0, 250)8.0、 5.0、 1.1、 3.0、 6.0、 9.0
    ZH-ZDR-KDP(0,10.0)、[10.0, 25.0)、[25.0,45.0)、[45.0,60.0)、[60.0,100.0)、[100.0, 250)10.0、 5.0、 1.1、 3.0、 5.0、 10.0
    小雨模型KDP(0,3.0)、[3.0,9.0)、 [9.0,15.0)、 [15.0,30.0)、[30.0, 50.0)、 [50.0,250)5.0、 1.5、 1.1、 2.0、 5.0、 8.0
    ZH(0,3.0)、[3.0,9.0)、 [9.0,15.0)、 [15.0,30.0)、[30.0, 50.0)、 [50.0,250)8.0、 1.5、 1.1、 2.0、 5.0、 8.0
    ZH-ZDR-KDP(0,3.0)、[3.0,9.0)、 [9.0,15.0)、 [15.0,30.0)、[30.0, 50.0)、 [50.0,250)8.0、 1.5、 1.1、 2.0、 5.0、 8.0
    总模型 KDP(0,3.0)、[3.0,9.0)、 [9.0,15.0)、 [15.0,30.0)、[30.0, 50.0)、 [50.0,250)8.0、 1.5、 1.1、 2.0、 5.0、 8.0
    ZH(0,3.0)、[3.0,9.0)、 [9.0,15.0)、 [15.0,30.0)、[30.0, 50.0)、 [50.0,250)8.0、 1.5、 1.1、 2.0、 5.0、 8.0
    ZH-ZDR-KDP(0,3.0)、[3.0,9.0)、 [9.0,15.0)、 [15.0,30.0)、[30.0, 50.0)、 [50.0,250)8.0、 1.5、 1.1、 2.0、 5.0、 8.0
    下载: 导出CSV

    表  4  采用均方误差作为损失函数的评估结果

    Table  4.   Model evaluation results by using MSE as the loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP1.0130.01312.599.370.286
    ZH1.0550.05212.1838.930.272
    ZH-ZDR-KDP1.0090.00911.5788.6870.265
    小雨模型KDP1.1080.0975.8263.7440.372
    ZH1.0000.0005.0813.4640.344
    ZH-ZDR-KDP1.0320.0315.1453.4530.343
    总模型 KDP0.9030.1077.1315.0360.432
    ZH0.9390.0656.3894.2610.366
    ZH-ZDR-KDP1.0650.0616.3593.8950.334
    下载: 导出CSV

    表  5  采用自定义损失函数的评估结果

    Table  5.   Models evaluation results by using the self-defined loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP0.9660.03612.1599.11110.278
    ZH0.9850.01511.8638.8550.27
    ZH-ZDR-KDP1.0340.03311.3788.4570.258
    小雨模型KDP0.9120.0975.6183.8730.385
    ZH0.9850.0164.8673.1960.318
    ZH-ZDR-KDP1.0190.0184.7913.110.309
    总模型 KDP1.0480.0466.5814.0340.346
    ZH1.0440.0426.1693.7470.322
    ZH-ZDR-KDP1.0360.0345.813.5490.305
    下载: 导出CSV

    表  6  采用均方误差作为损失函数的评估结果

    Table  6.   Evaluation results by using MSE as the loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP1.0060.00610.5518.0920.257
    ZH0.9960.0049.6857.1210.226
    ZH-ZDR-KDP0.9160.0929.8547.6610.244
    小雨模型KDP1.0890.0825.6413.7620.364
    ZH0.9330.0724.9823.7420.362
    ZH-ZDR-KDP0.9330.0724.9143.7030.358
    总模型 KDP0.8560.1686.4324.9770.466
    ZH0.8450.1845.6324.2350.396
    ZH-ZDR-KDP0.9310.0745.2803.7340.349
    下载: 导出CSV

    表  7  采用自定义损失函数的评估结果

    Table  7.   Evaluation results by using self-defined loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP0.9780.0239.8727.5780.241
    ZH0.9340.0709.5427.0320.224
    ZH-ZDR-KDP0.9270.0789.2857.1170.226
    小雨模型KDP0.8900.1245.1793.7690.364
    ZH0.8970.1144.6563.3140.320
    ZH-ZDR-KDP0.8950.1174.5333.2200.311
    总模型 KDP1.0010.0015.4143.3880.317
    ZH0.9390.0655.0843.1490.295
    ZH-ZDR-KDP0.9030.1074.9073.0670.287
    下载: 导出CSV

    表  8  采用均方误差作为损失函数的评估结果

    Table  8.   Evaluation results by using MSE as the loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP0.9200.08812.1779.2340.308
    ZH1.0310.03011.3508.3110.278
    ZH-ZDR-KDP0.9270.07910.7277.9440.265
    小雨模型KDP1.1300.1155.3723.5700.360
    ZH1.0690.0655.0163.5680.360
    ZH-ZDR-KDP1.0720.0674.7323.3100.334
    总模型 KDP0.8670.1536.2104.6730.445
    ZH0.9590.0435.8524.4020.385
    ZH-ZDR-KDP1.0560.0535.3433.2830.313
    下载: 导出CSV

    表  9  采用自定义损失函数的评估结果

    Table  9.   Evaluation results by using self-defined loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP0.8710.14812.3779.4830.317
    ZH0.9710.03011.0798.1430.272
    ZH-ZDR-KDP0.9310.07410.5597.9240.265
    小雨模型KDP0.9390.0654.8583.5110.354
    ZH1.0830.0764.4712.8790.290
    ZH-ZDR-KDP1.0700.0654.1592.6940.272
    总模型 KDP1.0210.0215.6593.3900.323
    ZH1.1300.1155.5653.2690.311
    ZH-ZDR-KDP1.0610.0584.9322.8450.271
    下载: 导出CSV

    表  10  采用均方误差作为损失函数的评估结果

    Table  10.   Evaluation results by using MSE as the loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP1.0950.08715.17810.9470.285
    ZH1.1180.15815.90311.2000.291
    ZH-ZDR-KDP1.0710.06613.3508.9130.232
    小雨模型KDP1.1500.1305.4163.2390.331
    ZH1.0290.0285.1113.4110.349
    ZH-ZDR-KDP1.0170.0174.8953.2580.333
    总模型 KDP0.9160.0916.5724.4490.421
    ZH0.9670.0346.5384.0520.383
    ZH-ZDR-KDP1.0600.0576.0883.3650.318
    下载: 导出CSV

    表  11  采用自定义损失函数的评估结果

    Table  11.   Evaluation results by using self-defined loss function

    模型参数BIASRBIASRMSE(mm)MAE(mm)MRE
    大雨模型KDP1.0150.01513.4579.9630.259
    ZH1.1120.10114.07610.1720.264
    ZH-ZDR-KDP1.0660.06212.0237.8680.205
    小雨模型KDP0.9540.0494.7463.0830.315
    ZH1.0340.0334.6062.8110.288
    ZH-ZDR-KDP1.0150.0154.3452.6770.274
    总模型 KDP1.1020.0925.7133.1900.302
    ZH1.1390.1225.9913.2830.310
    ZH-ZDR-KDP1.0730.0685.1462.8520.270
    下载: 导出CSV
  • [1] 陈超,胡志群,胡胜等. 2019. CINRAD-SA双偏振雷达资料在降水估测中的应用初探. 气象,45(1):113-125 doi: 10.7519/j.issn.1000-0526.2019.01.010

    Chen C,Hu Z Q,Hu S,et al. 2019. Preliminary application of CINRAD-SA dual polarization radar data in rainfall estimation. Meteor Mon,45(1):113-125 (in Chinese) doi: 10.7519/j.issn.1000-0526.2019.01.010
    [2] 郭瀚阳,陈明轩,韩雷等. 2019. 基于深度学习的强对流高分辨率临近预报试验. 气象学报,77(4):715-727 doi: 10.11676/qxxb2019.036

    Guo H Y,Chen M X,Han L,et al. 2019. High resolution nowcasting experiment of severe convections based on deep learning. Acta Meteor Sinica,77(4):715-727 (in Chinese) doi: 10.11676/qxxb2019.036
    [3] 胡志群,刘黎平,楚荣忠等. 2008. X波段双线偏振雷达不同衰减订正方法对比及其对降水估测影响研究. 气象学报,66(2):251-261 doi: 10.3321/j.issn:0577-6619.2008.02.011

    Hu Z Q,Liu L P,Chu R Z,et al. 2008. Comparison of different attenuation correction methods and their effects on estimated rainfall using X-band dual linear polarimetric radar. Acta Meteor Sinica,66(2):251-261 (in Chinese) doi: 10.3321/j.issn:0577-6619.2008.02.011
    [4] 黄小玉,陈媛,熊毅等. 2009. 漂移克里金方法在雷达和雨量计联合估测降水中的应用. 气象学报,67(2):288-297 doi: 10.3321/j.issn:0577-6619.2009.02.012

    Huang X Y,Chen Y,Xiong Y,et al. 2009. Merging radar and rain gauge data using Kriging with external drift (KED) for quantitative precipitation estimation. Acta Meteor Sinica,67(2):288-297 (in Chinese) doi: 10.3321/j.issn:0577-6619.2009.02.012
    [5] 黄兴友,马玉蓉,胡苏蔓. 2021. 基于深度学习的天气雷达回波序列外推及效果分析. 气象学报,79(5):817-827 doi: 10.11676/qxxb2021.041

    Huang X Y,Ma Y R,Hu S M. 2021. Extrapolation and effect analysis of weather radar echo sequence based on deep learning. Acta Meteor Sinica,79(5):817-827 (in Chinese) doi: 10.11676/qxxb2021.041
    [6] 蒋薇,刘芸芸,陈鹏等. 2021. 利用深度神经网络和先兆信号的江苏夏季降水客观预测方法. 气象学报,79(6):1035-1048 doi: 10.11676/qxxb2021.057

    Jiang W,Liu Y Y,Chen P,et al. 2021. Prediction of summer precipitation in Jiangsu province based on precursory factors:A deep neural network approach. Acta Meteor Sinica,79(6):1035-1048 (in Chinese) doi: 10.11676/qxxb2021.057
    [7] 李建通,张培昌. 1996. 最优插值法用于天气雷达测定区域降水量. 台湾海峡,15(3):255-259

    Li J T,Zhang P C. 1996. Optimum interpolation method used for measuring regional precipition with weather radar. J Oceanogr Taiwan Strait,15(3):255-259 (in Chinese)
    [8] 林炳干,张培昌,顾松山. 1997. 天气雷达测定区域降水量方法的改进与比较. 南京气象学院学报,20(3):67-73

    Lin B G,Zhang P C,Gu S S. 1997. Improvement of weather radar measured regional rainfall with comparison to other techniques. J Nanjing Instit Meteor,20(3):67-73 (in Chinese)
    [9] 刘黎平,钱永甫,王致君. 1996. 用双线偏振雷达研究云内粒子相态及尺度的空间分布. 气象学报,54(5):590-599 doi: 10.3321/j.issn:0577-6619.1996.05.008

    Liu L P,Qian Y F,Wang Z J. 1996. The study of spacial distribution of phase and size of hydrometeorsin cloud by dual linear polarization radar. Acta Meteor Sinica,54(5):590-599 (in Chinese) doi: 10.3321/j.issn:0577-6619.1996.05.008
    [10] 任萍,陈明轩,曹伟华等. 2020. 基于机器学习的复杂地形下短期数值天气预报误差分析与订正. 气象学报,78(6):1002-1020 doi: 10.11676/qxxb2020.060

    Ren P,Chen M X,Cao W H,et al. 2020. Error analysis and correction of short-term numerical weather prediction under complex terrain based on machine learning. Acta Meteor Sinica,78(6):1002-1020 (in Chinese) doi: 10.11676/qxxb2020.060
    [11] 邵月红,张万昌,刘永和等. 2009. BP神经网络在多普勒雷达降水量的估测中的应用. 高原气象,28(4):846-853

    Shao Y H,Zhang W C,Liu Y H,et al. 2009. Application of back-propagation neural network in precipitation estimation with doppler radar. Plateau Meteor,28(4):846-853 (in Chinese)
    [12] 王超,吴翀,刘黎平. 2019. X波段双线偏振雷达数据质量分析及控制方法. 高原气象,38(3):636-649

    Wang C,Wu C,Liu L P. 2019. Data quality analysis and control method of X-band dual polarization radar. Plateau Meteor,38(3):636-649 (in Chinese)
    [13] 张培昌. 1988. 雷达气象学. 北京: 气象出版社. Zhang P C. 1988. Radar Meteorology. Beijing: China Meteorological Press (in Chinese)
    [14] 张培昌,戴铁丕,王登炎等. 1992a. 最优化法求Z-I关系及其在测定降水量中的精度. 气象科学,12(3):333-338

    Zhang P C,Dai T P,Wang D Y,et al. 1992a. Derivation of the Z-I relationship by optimization and the accuracy in the quantitative rainfall measurement. Scientia Meteor Sinica,12(3):333-338 (in Chinese)
    [15] 张培昌,戴铁丕,傅德胜等. 1992b. 用变分方法校准数字化天气雷达测定区域降水量基本原理和精度. 大气科学,16(2):248-256 doi: 10.3878/j.issn.1006-9895.1992.02.13

    Zhang P C,Dai T P,Fu D S,et al. 1992b. Principle and accuracy of adjusting the area precipitation from digital weather radar through variational method. Chinese J Atmos Sci,16(2):248-256 (in Chinese) doi: 10.3878/j.issn.1006-9895.1992.02.13
    [16] 张扬. 2019. 业务双偏振雷达网与自动站联合定量降水估测方法及效果分析研究[D]. 北京: 中国气象科学研究院. Zhang Y. 2019. Study on the quantitative precipitation estimation algorithm utilized with the operational dual-polarization radar network and automatic stations and its effect analysis[D]. Beijing: Chinese Academy of Meteorological Sciences (in Chinese)
    [17] 郑佳锋,刘黎平,曾正茂等. 2016. Ka波段毫米波云雷达数据质量控制方法. 红外与毫米波学报,35(6):748-757

    Zheng J F,Liu L P,Zeng Z M,et al. 2016. Ka-bard millimeter wave cloud radar data quality control. J Infrared Millim Waves,35(6):748-757 (in Chinese)
    [18] 郑玉,魏鸣,李南等. 2020. 基于循环神经网络改进雷达定量估测强降水. 中国科技论文,15(5):585-592 doi: 10.3969/j.issn.2095-2783.2020.05.014

    Zheng Y,Wei M,Li N,et al. 2020. Improved radar heavy precipitation estimation based on RNN. China Sciencepaper,15(5):585-592 (in Chinese) doi: 10.3969/j.issn.2095-2783.2020.05.014
    [19] Bringi V N,Huang G J,Chandrasekar V,et al. 2002. A methodology for estimating the parameters of a gamma raindrop size distribution model from polarimetric radar data:Application to a squall-line event from the TRMM/brazil campaign. J Atmos Ocean Technol,19(5):633-645 doi: 10.1175/1520-0426(2002)019<0633:AMFETP>2.0.CO;2
    [20] Chumchean S,Sharma A,Seed A. 2006. An integrated approach to error correction for real-time radar-rainfall estimation. J Atmos Ocean Technol,23(1):67-79 doi: 10.1175/JTECH1832.1
    [21] Fulton R A,Breidenbach J P,Seo D J,et al. 1998. The WSR-88D rainfall algorithm. Wea Forecast,13(2):377-395 doi: 10.1175/1520-0434(1998)013<0377:TWRA>2.0.CO;2
    [22] Gorgucci E,Scarchilli G,Chandrasekar V. 2000. Practical aspects of radar rainfall estimation using specific differential propagation phase. J Appl Meteor,39(6):945-955 doi: 10.1175/1520-0450(2000)039<0945:PAORRE>2.0.CO;2
    [23] Haberlie A M,Ashley W S. 2019. A radar-based climatology of mesoscale convective systems in the United States. J Climate,32(5):1591-1606 doi: 10.1175/JCLI-D-18-0559.1
    [24] Lee G W. 2006. Sources of errors in rainfall measurements by polarimetric radar:Variability of drop size distributions,observational noise,and variation of relationships between R and polarimetric parameters. J Atmos Ocean Technol,23(8):1005-1028 doi: 10.1175/JTECH1899.1
    [25] Reichstein M,Camps-Valls G,Stevens B,et al. 2019. Deep learning and process understanding for data-driven earth system science. Nature,566(7743):195-204 doi: 10.1038/s41586-019-0912-1
    [26] Seliga T A,Bringi V N. 1976. Potential use of radar differential reflectivity measurements at orthogonal polarizations for measuring precipitation. J Appl Meteor,15(1):69-76 doi: 10.1175/1520-0450(1976)015<0069:PUORDR>2.0.CO;2
    [27] Wang G L,Liu L P,Ding Y Y. 2012. Improvement of radar quantitative precipitation estimation based on real-time adjustments to Z-R relationships and inverse distance weighting correction schemes. Adv Atmos Sci,29(3):575-584 doi: 10.1007/s00376-011-1139-8
    [28] Wilson J W,Brandes E A. 1979. Radar measurement of rainfall:A summary. Bull Amer Meteor Soc,60(9):1048-1060 doi: 10.1175/1520-0477(1979)060<1048:RMORS>2.0.CO;2
    [29] Yin X Y,Hu Z Q,Zheng J F,et al. 2021. Study on radar echo-filling in an occlusion area by a deep learning algorithm. Remote Sens,13(9):1779 doi: 10.3390/rs13091779
  • 加载中
图(11) / 表(11)
计量
  • 文章访问数:  272
  • HTML全文浏览量:  29
  • PDF下载量:  127
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-02-28
  • 录用日期:  2022-06-10
  • 修回日期:  2022-03-29
  • 网络出版日期:  2022-04-01

目录

    /

    返回文章
    返回