向量距离中角度信息对时空Kriging的影响

删除或更新信息，请邮件至freekaoyan#163.com(#换成@)

清华大学辅仁网/2017-07-07

陈鼎新^1,2, 陆文凯¹, 刘代志²

1. 清华大学自动化系, 智能技术与系统国家重点实验室, 北京 100084;
2. 火箭军工程大学 907教研室, 西安 710025

Vector distance direction information for spatio-temporal Kriging

CHEN Dingxin^1,2, LU Wenkai¹, LIU Daizhi²

1. State Key Laboratory of Intelligent Technology and Systems, Department of Automation, Tsinghua University, Beijing 100084, China;
2. Staff Room 907, PLA Rocket Force University of Engineering, Xi'an 710025, China

摘要:

输出: BibTeX | EndNote (RIS)

摘要时空Kriging算法的核心, 是将变差函数的概念扩展到时空域, 变差函数的构建过程基础是计算时间切片和空间切片的向量距离。该文讨论了向量距离对构建时空变差函数的影响, 提出了空间距离加角度差异的向量距离模型。以地磁场观测数据作为对象, 分别用L1范数、L2范数和新距离模型对数据进行分析, 比较3种距离定义下的时空Kriging插值性能。结果表明: 加入了角度信息的向量距离, 能够更有效地表征数据, 提高时空Kriging的插值精度。

关键词 ：地磁场,时空Kriging,变差函数,向量角度

Abstract：The spatio-temporal Kriging method can be significantly improved by extending the variogram definition to the space-time domain. The key step in constructing the spatio-temporal variogram is to calculate the vector distances between the time slices and the space slices. This study analyzes the influence of the vector distance on the spatio-temporal variogram construction and presents a vector distance model that includes both the magnitude and the direction information. The algorithm was evaluated using magnetic field data with the evaluations based on the L1 norm and the L2 norm. The results show that the model with the additional direction information in the vector distance, more effectively represented the data characteristics which improved the spatio-temporal Kriging interpolation accuracy.

Key words：geomagnetic field spatio-temporal Kriging variogram vector direction

收稿日期: 2015-12-01 出版日期: 2016-05-19

ZTFLH:

P318.2

通讯作者:陆文凯, 教授, E-mail: lwkmf@tsinghua.edu.cnE-mail: lwkmf@tsinghua.edu.cn

引用本文:

陈鼎新, 陆文凯, 刘代志. 向量距离中角度信息对时空Kriging的影响[J]. 清华大学学报（自然科学版）, 2016, 65(5): 553-557.
CHEN Dingxin, LU Wenkai, LIU Daizhi. Vector distance direction information for spatio-temporal Kriging. Journal of Tsinghua University(Science and Technology), 2016, 65(5): 553-557.

链接本文:

http://jst.tsinghuajournals.com/CN/10.16511/j.cnki.qhdxxb.2016.25.015或 http://jst.tsinghuajournals.com/CN/Y2016/V65/I5/553

图表:

图１　台站分布图

表１　条件变差函数的拟合参数

表２　交叉验证结果的时间统计量

图２　不同距离定义下的条件变差函数拟合结果

图３　每个时间切片处的结果分析

参考文献:

[1] Iaco S D, Myers D E, Posa D. The linear coregionalization model and the product-sum space-time variogram[J].Mathematical Geology, 2003,35(35):25-38.
[2] Carla N, Amilcar S. Geostatistical space-time simulation model for air quality prediction[J].Environmetrics, 2005,16(4):393-404.
[3] Li S, Shu H, Xu Z. Spatial-temporal statistics and analysis of rainfall in Jilin Province[C]//International Workshop on Computer Science for Environmental Engineering and Ecoinformatics, 2011, Kunming, China. Berlin:Springer-Verlag, 2011:255-261.
[4] Tang Y. Comparison of semivariogram models for Kriging monthly rainfall in eastern China[J].Journal of Zhejiang University Science,2002,3(5):584-590.
[5] Tilmann G. Nonseparable, stationary covariance functions for space-time data[J].Journal of the American Statistical Association, 2002,97(458):590-600.
[6] Cichota R, Hurtado A L B, Lier Q D J V. Spatio-temporal variability of soil water tension in a tropical soil in Brazil[J].Geoderma,2006,133(s 3-4):231-243.
[7] Lark R M, Bellamy P H, Rawlins B G. Spatio-temporal variability of some metal concentrations in the soil of eastern England, and implications for soil monitoring[J].Geoderma,2006,133(s 3-4):363-379.
[8] Gething P W, Akinson P M, Noor A M. A local space-time Kriging approach applied to a national outpatient malaria data set[J].Computers & Geosciences, 2007,33(10-50):1337-1350.
[9] Noel C. Spatial prediction and ordinary Kriging[J].Mathematical Geology, 1988,20(4):405-421.
[10] De C L, Myers D, Posa D. Estimating and modeling space-time correlation structures[J].Stat Probab Lett, 2001,51(1):9-14.
[11] Iaco S D, Myers D E, Posa D. Nonseparable space-time covariance models:Some parametric families[J].Mathematical Geology,2002,34(1):23-42.
[12] Siders I V, Gabella, Erdin R, et al. Real-time radar-rain-gauge merging using spatiio-temporal co-Kriging with external drift in the alpine terrain of Switzerland[J].Journal of the Royal Meteorological Society, 2014,140(680):1097-1111.
[13] Kalivas J H. Overview of two-norm (L2) and one-norm (L1) Tikhonov regularization variants for full wavelength or sparse spectral multivariate calibration models or maintenance[J].Journal of Chemometrics, 2012,26(6):218-230.
[14] Jin L H, Li D H, Song Y M. Combining vector ordering and spatial information for color image interpolation[J].Image and Vision Computing, 2009,27(4):410-416.
[15] Astola J, Haavisto J, Neuvo Y. Vector median filters[J].Proceedings of the IEEE, 1990,78(4):678-689.
[16] Trahanias P E, Karakos D G, Venetsanopoulos A N. Directional processing of color images:Theory and experimental results[J].IEEE Transactions on Image Processing, 1996,5(6):868-880
[17] De C L, Myers D, Posa D. Estimating and modeling space-time correlation structures[J].Stat Probab Lett, 2001,51(1):9-14.
[18] Cesare L D, Myers D E, Posa D. Product-sum covariance for space-time modeling:An environmental application[J].Environmetrics, 2001,12(1):11-23.
[19] 李莎, 舒红, 董林. 基于时空变异函数的Kriging插值及实现[J]. 计算机工程与应用, 2011,47(23):25-38. LI Sha, SHU Hong, DONG Lin. Research and realization of Kriging interpolation based on spatial-temporal variogram[J]. Computer Engineering and Applications, 2011,47(23):25-26. (in Chinese)

No related articles found!