RAN Yuan-bo,SUN Min,GAO Meng-qing,et al.Study on Hydrometeor Identification based on Deep Learning[J].Journal of Chengdu University of Information Technology,2017,(06):590-596.[doi:10.16836/j.cnki.jcuit.2017.06.003]
双偏振天气雷达水凝物识别研究
- Title:
- Study on Hydrometeor Identification based on Deep Learning
- 文章编号:
- 2096-1618(2017)06-0590-07
- Keywords:
- dual polarization doppler weather radar; hydrometeor classification; polarized parameter; deep learning; fuzzy logic; clustering
- 分类号:
- TN959.4
- 文献标志码:
- A
- 摘要:
- 对双偏振天气雷达回波的水凝物分类,是利用降水粒子对极化电磁波的散射特性对水凝物相态进行识别的过程。不同相态降水粒子由于在形状、大小和空间取向等方面存在差异,对一定极化状态的电磁波会产生不同的散射特性,导致与这些特性密切相关的雷达偏振参量也各不相同,综合利用这些偏振参量,可以有效地识别出各种水凝物的相态。提出一种利用深度学习和模糊逻辑算法进行联合判别的水凝物相态识别方法:首先,采用深度学习算法对降水粒子所对应的雷达回波产品数据进行特征提取,并利用Softmax分类器对提取到的特征进行分类,实现降水粒子所对应降水模式的识别; 其次,在已知降水模式的情况下,利用模糊逻辑算法,实现对降水粒子类型的最优判别; 最后,结合其他雷达产品对分类结果进行分析与比较,发现两者能够达到极好的吻合。这种采用深度学习方法对降水粒子进行初次聚类,再利用模糊逻辑算法实现精确聚类的方法,大大提高了水凝物识别的准确性。
- Abstract:
- Hydrometeor classification for Dual Polarization Doppler Weather Radar echo is a procedure that identifieshydrometeor types basing on the scattering properties of precipitation particles to polarized electromagnetic waves. The difference in shape, size or spatial orientation between different types of hydrometeor will produce different scattering characteristics for the electromagnetic waves in a certain polarization state, Moreover, the polarized parameters, which are calculated from the radar data and closely associated with these characteristics, are also different. The comprehensive use of these polarized parameters can effectively improve the identification accuracy of the phase of various hydrometeors. In this paper, a new identification method of the hydrometeor type basing on deep learning and fuzzy logic methods is proposed: Firstly, the deep learning(DL)approaches is used for extracting the features from the polarized products of the hydrometeors, and the Softmax classifier is applied to classify the pattern of precipitations about rain,snow and hail based on the features extracted by deep learning algorithm.Secondly, the adaptive fuzzy logic algorithm is adoptive for identify the precipitation particles in various precipitation patterns. Finally, the hydrometeor classifier has been applied to astratiform cloud precipitation process, and it is found that the identification result agrees with the other polarized products.The deep learning method is used for initial clustering, then using the fuzzy logic method for accurate clustering results, that greatly improves the accuracy of hydrometeor classification.
参考文献/References:
[1] 刘黎平,王致君,徐宝祥,等.我国双线偏振雷达探测理论及应用研究[J].高原气象,1997,16(1):99-104.
[2] Bringi V N,Thurai M,Hannesen R.Dual-polarization weather radar handbook[J].AMS-Gematronik GmbH,2007(2).
[3] Schuur T,Ryzhkov A,Heinselman P,et al.Observations and Classification of Echoes with the Polarimetric WSR-88D radar[J].2003(1).
[4] Zhang Z,Zheng G,Wei M,et al.Research on the system of advanced X-band Doppler weather radar with dual-linear polarization capability[C].Advanced Computational Intelligence(ICACI),2012 IEEE Fifth International Conference on.IEEE,2012:739-741.
[5] 汪月霞,林伙海,何建新,等.双偏振天气雷达降水粒子相态识别研究[C].中国气象学会年会,2013.
[6] Galati G,Pavan G.Estimation techniques for rainfall rate using differential phase shift in X-band weather radar[C].Geoscience and Remote Sensing Symposium Proceedings,1998.IGARSS'98.1998 IEEE International.IEEE,1998,1:138-140.
[7] Erhan D,Courville A,Bengio Y.Understanding representations learned in deep architectures[J].Department d'Informatique et Recherche Operationnelle,University of Montreal,QC,Canada,Tech.Rep,2010,1355.
[8] Lecun Y,Bengio Y,Hinton G.Deep learning[J].Nature,2015,521(7553):436-444.
[9] Schmidhuber J.Deep learning in neural networks:an overview[J].Neural Networks the Official Journal of the International Neural Network Society,2014,61:85-117.
[10] Sutskever I,Martens J,Dahl G,et al.On the importance of initialization and momentum in deep learning[C].International Conference on International Conference on Machine Learning.JMLR.org,2013:III-1139.
[11] Liu H,Chandrasekar V.Classification of hydrometeors based on polarimetric radar measurements:Development of fuzzy logic and neuro-fuzzy systems,and in situ verification[J].Journal of Atmospheric and Oceanic Technology,2000,17(2):140-164.
[12] Chandrasekar V,Keränen R,Lim S,et al.Recent advances in classification of observations from dual polarization weather radars[J].Atmospheric Research,2013,119:97-111.
[13] 刘亚男,肖辉,姚振东,等.X波段双极化雷达对云中水凝物粒子的相态识别[J].气候与环境研究,2012,17(6):925-936.
相似文献/References:
[1]杜 捷,王海江,孙召平.基于模糊逻辑的双偏振天气雷达地物杂波识别算法[J].成都信息工程大学学报,2022,37(02):159.[doi:10.16836/j.cnki.jcuit.2022.02.008]
DU Jie,WANG Haijiang,SUN Zaoping.Ground Clutter Recognition Algorithm of Dual Polarization Weather Radar based on Fuzzy Logic[J].Journal of Chengdu University of Information Technology,2022,37(06):159.[doi:10.16836/j.cnki.jcuit.2022.02.008]
备注/Memo
收稿日期:2017-09-11 基金项目:四川省教育厅基金资助项目(16ZA0209); 四川省科技厅基金资助项目(2016JY0106)