MC-DTSVMs的双偏振气象雷达降水粒子分类方法

doi:10.19665/j.issn1001-2400.2020.04.018

摘要/Abstract

摘要：

为解决双偏振气象雷达偏振参量数据随机缺失情况下的降水粒子分类问题,提出一种矩阵填充-决策树支持向量机多分类器的双偏振气象雷达降水粒子联合分类方法。首先,根据矩阵填充算法对存在随机缺失的偏振参量数据进行重构; 然后,利用训练数据对决策树支持向量机多分类器进行学习;最后,使用学习好的决策树支持向量机多分类器实现对重构数据的降水粒子分类。通过对实测数据的处理及结果分析,证明了该方法能够有效地解决偏振参量数据随机缺失情况下的降水粒子分类问题。

关键词: 双偏振气象雷达, 降水粒子分类, 矩阵填充, 支持向量机, 随机缺失

Abstract:

In order to solve the problem of precipitation particle classification in the case of random missing of polarization parameter data of two-polarization meteorological radar, a method based on matrix completion(MC) and decision tree support vector machine multi-classifier(DTSVMs) is proposed. First, the polarization parameter data with random miss is reconstructed according to the matrix completion algorithm, and then the training data are used to learn the DTSVMs, and finally the precipitation particle classification of the reconstructed data is realized by using the DTSVMs with good learning. By processing the measured data and analyzing the results, it is proven that this method can effectively solve the precipitation particle classification problem in the case of random missing of polarization parameter data.

Key words: dual polarization weather radar, hydrometeor classification, matrix completion, support vector machine, random missing

中图分类号:

TN959.4

李海,尚金雷,孙婷逸,冯青,庄子波. MC-DTSVMs的双偏振气象雷达降水粒子分类方法[J]. 西安电子科技大学学报, 2020, 47(4): 132-140.

LI Hai,SHANG Jinlei,SUN Tingyi,FENG Qing,ZHUANG Zibo. Method for hydrometeor classification based on MC-DTSVMs[J]. Journal of Xidian University, 2020, 47(4): 132-140.

图/表 12

图1

图2

表1

图3

图4

图5

表2

图6

表3

图7

表4

图8

参考文献 17

[1]	李柏, 古庆同, 李瑞义, 等. 新一代天气雷达灾害性天气监测能力分析及未来发展[J]. 气象, 2013,39(3):265-280.
	LI Bai, GU Qingtong, LI Ruiyi, et al. Analyses on Disastrous Weather Monitoring Capability of CINRAD and Future Development[J]. Meteor Mon, 2013,39(3):265-280.
[2]	GUO X, FU D, LI X, et al. Advances in Cloud Physics and Weather Modification in China[J]. Advances in Atmospheric Sciences, 2015,32(2):230-249. doi: 10.1007/s00376-014-0006-9
[3]	刘黎平, 胡志群, 吴翀. 双线偏振雷达和相控阵天气雷达技术的发展和应用[J]. 气象科技进展, 2016,6(3):28-33.
	LIU Liping, WU Zhiqun, WU Chong. Development and Application of Dual Linear Polarization Radar and Phased-array Radar[J]. Advances in Meteorological Science and Technology, 2016,6(3):28-33.
[4]	张培昌, 杜秉玉, 戴铁丕. 雷达气象学[M]. 北京: 气象出版社, 2001: 9-30.
[5]	DONOHO D L. Compressed Sensing[J]. IEEE Transactions on Information Theory, 2006,52(4):1289-1306. doi: 10.1109/TIT.2006.871582
[6]	CANDÈS E J, RECHT B. Exact Matrix Completion via Convex Optimization[J]. Foundations of Computational Mathematics, 2009,9(6):717-772. doi: 10.1007/s10208-009-9045-5
[7]	VIVAR G, ZWERGAL A, NAVAB N, et al. Gerome Vivar Email Author Andreas Zwergal Nassir Navab[C]// International Workshop on Graphs in Biomedical Image Analysis, 2018: 24-31.
[8]	VANIDEVI M, SELVAGANESAN N. Fast Iterative WSVT Algorithm in WNN Minimization Problem for Multiuser Massive MIMO Channel Estimation[J]. International Journal of Communication Systems, 2018,31(1):e3378. doi: 10.1002/dac.v31.1
[9]	GENES C, ESNAOLA I, PERLAZA S, et al. Recovery of Missing Data in Correlated Smart Grid Datasets[C]// Proceedings of the 2019 IEEE Data Science Workshop. Piscataway: IEEE, 2019: 263-269.
[10]	石绍应, 王小谟, 曹晨, 等. 规则数确定的自适应模糊分类器[J]. 西安电子科技大学学报, 2017,44(2):81-87.
	SHI Shaoying, WANG Xiaomo, CAO Chen, et al. Adaptive Fuzzy Classifier with a Fixed Number of Fuzzy Rules[J]. Journal of Xidian University, 2017,44(2):81-87.
[11]	韩冰, 贾中华, 高新波. 改进的主成分分析网络极光图像分类方法[J]. 西安电子科技大学学报, 2017,44(1):83-88.
	HAN Bing, JIA Zhonghua, GAO Xinbo. Improved PCA Net for Aurora Images Classification[J]. Journal of Xidian University, 2017,44(1):83-88.
[12]	MARZANO F S, SCARANARI D, MONTOPOLI M, et al. Supervised Classification and Estimation of Hydrometeors from C-band Dual-polarized Radars: a Bayesian Approach[J]. IEEE Transactions on Geoscience and Remote Sensing, 2008,46(1):85-98. doi: 10.1109/TGRS.2007.906476
[13]	李海, 任嘉伟, 尚金雷. 一种基于模糊神经网络-模糊C均值聚类的双偏振气象雷达降水粒子分类方法[J]. 电子与信息学报, 2019,41(4):809-815.
	LI Hai, REN Jiawei, SHANG Jinlei. Hydrometeor Classification Method in Dual-polarization Weather Radar Based on Fuzzy Neural Network-fuzzy C-means[J]. Journal of Electronics and Information Technology, 2019,41(4):809-815.
[14]	ROBERTO N, BALDINI L, ADIROSI E, et al. A Support Vector Machine Hydrometeor Classification Algorithm for Dual-polarization Radar[J]. Atmosphere, 2017,8(8):134. doi: 10.3390/atmos8080134
[15]	BUCHSBAUM S, KESHET Y, OZANA N, et al. Bio-sensor Based on Multiclass Support Vector Machine with a Reject Option[C]// Proceedings of the Progress in Biomedical Optics and Imaging. Bellingham: SPIE, 2019: 108710O.
[16]	王正海, 方臣, 何凤萍, 等. 基于决策树多分类支持向量机岩性波谱分类[J]. 中山大学学报(自然科学版), 2014,53(6):93-97.
	WANG Zhenghai, FANG Chen, HE Fengping, et al. Hyperspectral Rock Spectral Classification Based on the Decision Tree-support Vector Machine ( DT-SVMs)[J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2014,53(6):93-97.
[17]	杨通晓, 岳彩军. 基于支持向量机的双偏振雷达对流降水类型识别方法研究[J]. 暴雨灾害, 2019,38(4):297-302.
	YANG Tongxiao, YUE Caijun. Research on Hydrometeor Classification of Convective Weather Based on SVM by Dual Linear Polarization Radar[J]. Torrential Rain and Disasters, 2019,38(4):297-302.

SVM多分类算法	O-v-R	1-v-1	DDAG	DT
SVM数量/个	9	32	32	8
运算时间/s	167	360	360	136

参数名称	收缩阈值τ	迭代步长δ	终止门限ε	终止迭代次数K
反射率因子	273.2	1.0	5%	200
差分反射率	93.5	0.8	5%	200
零滞后互相关系数	5.2	0.3	5%	200

偏振参量	平均误差	误差比率/%
反射率因子	0.417dBZ	3.8
差分反射率	0.124dBZ	4.2
零滞后互相关系数	0.012	3.6

偏振参量	气象目标总数	气象目标识别率/%	识别正确个数	识别准确率/%
NOAA提供的分类结果	168087	100	168087	100
未补偿数据分类结果	113728	68	70680	62
补偿数据分类结果	174114	104	152284	91