计算机科学 ›› 2022, Vol. 49 ›› Issue (8): 33-39.doi: 10.11896/jsjkx.210600161

• 数据库&大数据&数据科学* 上一篇    下一篇

基于时空注意力克里金的边坡形变数据插值方法

黎嵘繁, 钟婷, 吴劲, 周帆, 匡平   

  1. 电子科技大学信息与软件工程学院 成都 610054
  • 收稿日期:2021-06-21 修回日期:2021-10-15 发布日期:2022-08-02
  • 通讯作者: 周帆(fan.zhou@uestc.edu.cn)
  • 作者简介:(rongfanli1998@gmail.com)
  • 基金资助:
    国家自然科学基金(62072077);国家重点研发计划(2019YFB1406202);四川省科技计划项目专项资金(2020YFG0234)

Spatio-Temporal Attention-based Kriging for Land Deformation Data Interpolation

LI Rong-fan, ZHONG Ting, WU Jin, ZHOU Fan, KUANG Ping   

  1. School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610054,China
  • Received:2021-06-21 Revised:2021-10-15 Published:2022-08-02
  • About author:LI Rong-fan,born in 1998,postgra-duate.His main research interests include graph neural network and data mining.
    ZHOU Fan,born in 1981,Ph.D,asso-ciate professor,is a member of China Computer Federation.His main research interests include machine lear-ning,spatio-temporal data mining,data mining and knowledge discovery.
  • Supported by:
    National Natural Science Foundation of China(62072077),National Key R & D Program of China(2019YFB1406202) and Sichuan Science and Technology Program(2020YFG0234).

摘要: 山体滑坡每年都会对人们的生命财产安全造成重大损失,是常见的地质灾害之一。为了对山体滑坡进行防控,需要广泛地监测山体表面的沉降过程,但是由于恶劣气候和监测成本等难以克服的困难,山体沉降数据的收集呈现出局部数据不完整、数据采样不均衡和监测点动态变化等特点,使得山体滑坡的防控研究受到阻碍,给数据的采集和分析工作提出了新的要求。现有方法从空间角度对缺失进行补充,但忽略了时间维度的依赖关系。为了解决上述问题,对不完整的INSAR数据填充进行了研究,利用时空掩码矩阵对时空依赖关系进行建模,利用多头注意力对多层次的空间关系进行综合学习,并在克里金法(Kriging)的基础上提出了新的使用时空注意力的克里金插值法,实现了对复杂时空特征的深层理解。在真实数据集上的数据恢复实验验证了该算法可以有效地学习复杂的时空特征,并在3种不同的数据缺失情景下都取得了优于现存插值算法的表现。

关键词: 插值法, 克里金, 山体滑坡, 时空数据挖掘, 时空注意力

Abstract: Landslide is one of the most common geological hazards,it causes significant damage to people’s life and property everyyear.In order to prevent and control landslides,it is necessary to monitor the land surface extensively.However,insurmountable difficulties such as severe climate and high monitoring cost impede the collection of land surface data,resulting in incomplete local data,unbalanced data sampling and dynamic changes of monitoring points,which hinder the prevention and control research of landslide and put forward new demand to the data collection and analysis.Existing methods try to handle incomplete data from spatial perspective,which,however,ignore temporal dependencies that are important for data interpolation.To solve the above problems,the incomplete INSAR data filling is studied,the spatio-temporal dependence is modeled by using the spatio-temporal mask matrix,the multi-level spatial relationship is comprehensively studied by using multi-head attention,and a novel Kriging interpolation method using spatio-temporal attention is proposed on the basis of Kriging.It realizes the deep understanding of complex temporal and spatial features.Interpolation experiments on real-world INSAR datasets show that the proposed model is capable to learn sophisticated spatial and temporal features effectively,and achieves better performance than the state-of-the-art methods in three different data interpolation scenarios.

Key words: Interpolation, Kriging, Landslide, Spatio-Temporal attention, Spatio-Temporal data mining

中图分类号: 

  • TP183
[1]BOZZANO F,CIPRIANI I,MAZZANTI P,et al.Displacement patterns of a landslide affected by human activities:insights from ground-based InSAR monitoring[J].Natural Hazards,2011,59(3):1377-1396.
[2]GAO W,DAI S,CHEN X.Landslide prediction based on a combination intelligent method using the GM and ENN:two cases of landslides in the Three Gorges Reservoir,China[J].Landslides,2020,17(1):111-126.
[3]HAJIMORADLOU A,ROBERTI G,POOLE D.Predicting Land-slides Using Locally Aligned Convolutional Neural Networks[J].arXiv:1911.04651,2019.
[4]HUANG R Q.Large-scale landslides and their sliding mechanisms in China since the 20th century [J].Chinese Journal of Rock Mechanics and Engineering,2007(3):433-454.
[5]ZHU A X,WANG R,QIAO J,et al.An expert knowledge-based approach to landslide susceptibility mapping using GIS and fuzzy logic[J].Geomorphology,2014,214:128-138.
[6]VAKHSHOORI V,ZARE M.Landslide susceptibility mapping by comparing weight of evidence,fuzzy logic,and frequency ratio methods[J].Geomatics,Natural Hazards and Risk,2016,7(5):1731-1752.
[7]ZHOU J,LU P,YANG Y.Reservoir landslides and its hazard effects for the hydropower station:a case study[C]//Workshop on World Landslide Forum.Cham:Springer,2017:699-706.
[8]GAN B R,YANG X G,ZHOU J W.GIS-based remote sensing analysis of the spatial-temporalevolution of landslides in a hydropower reservoir in southwest China[J].Geomatics,Natural Hazards and Risk,2019,10(1):2291-2312.
[9]CHEN W,XIE X,PENG J,et al.GIS-based landslide susceptibility modelling:a comparative assessment of kernel logistic regression,NaÏve-Bayes tree,and alternating decision tree models[J].Geomatics,Natural Hazards and Risk,2017,8(2):950-973.
[10]KALANTAR B,PRADHAN B,NAGHIBI S A,et al.Assess-ment of the effects of training data selection on the landslide susceptibility mapping:a comparison between support vector machine(SVM),logistic regression(LR) and artificial neural networks(ANN)[J].Geomatics,Natural Hazards and Risk,2018,9(1):49-69.
[11]HONG H,POURGHASEMI H R,POURTAGHI Z S.Land-slide susceptibility assessment in Lianhua County(China):a comparison between a random forest data mining technique and bivariate and multivariate statistical models[J].Geomorphology,2016,259:105-118.
[12]HONG H,PRADHAN B,JEBUR M N,et al.Spatial prediction of landslide hazard at the Luxi area (China) using support vector machines[J].Environmental Earth Sciences,2016,75(1):1-14.
[13]LEI T,ZHANG Y,LV Z,et al.Landslide inventory mappingfrom bitemporal images using deep convolutional neural networks[J].IEEE Geoscience and Remote Sensing Letters,2019,16(6):982-986.
[14]YU B,YIN H,ZHU Z.Spatio-temporal graph convolutional networks:a deep learning framework for traffic forecasting[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence.2018:3634-3640.
[15]LI Y,YU R,SHAHABI C,et al.Diffusion Convolutional Recurrent Neural Network:Data-Driven Traffic Forecasting[C]//International Conference on Learning Representations.2018:1-16.
[16]WANG X,MA Y,WANG Y,et al.Traffic flow prediction via spatial temporal graph neural network[C]//Proceedings of the Web Conference.2020:1082-1092.
[17]HAMILTON W L,YING R,LESKOVEC J.Inductive representation learning on large graphs[J].arXiv:1706.02216,2017.
[18]CHIANG W L,LIU X,SI S,et al.Cluster-gcn:An efficient algorithm for training deep and large graph convolutional networks[C]//Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining.2019:257-266.
[19]ZENG H,ZHOU H,SRIVASTAVA A,et al.GraphSAINT:Graph Sampling Based Inductive Learning Method[C]//International Conference on Learning Representations.2019:1-19.
[20]VASWANI A,SHAZEER N,PARMAR N,et al.Attention is All you Need[C]//Conference on Neural Information Proces-sing Systems.2017:6000-6010.
[21]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks[J].arXiv:1710.10903,2017.
[22]DONG J,ZHANG L,LIAO M,et al.Improved correction of seasonal tropospheric delay in InSAR observations for landslide deformation monitoring[J].Remote Sensing of Environment,2019,233:111370.
[23]CARLÀ T,INTRIERI E,RASPINI F,et al.Perspectives on the prediction of catastrophic slope failures from satellite InSAR[J].Scientific Reports,2019,9(1):1-9.
[24]LE N D,ZIDEK J V.Statistical Analysis of EnvironmentalSpace-Time Processes[J].Journal of the American Statistical Association,2007,102:1477-1477.
[25]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks[J].arXiv:1609.02907,2016.
[26]WU Z,PAN S,CHEN F,et al.A comprehensive survey ongraph neural networks[J].IEEE Transactions on Neural Networks and Learning Systems,2021,32(1):4-24.
[27]SONG S J,LAN C L,XING J L,et al.An end-to-end spatio-temporal attention model for human action recognition from skeleton data[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2017:4263-4270.
[28]ZHAO Z,YANG Q,CAI D,et al.Video Question Answering via Hierarchical Spatio-Temporal Attention Networks[C]//International Joint Conference on Artificial Intelligence.2017:3518-3524.
[29]CRESSIE N.The origins of kriging[J].Mathematical Geology,1990,22(3):239-252.
[30]TAPOGLOU E,KARATZAS G P,TRICHAKIS I C,et al.Aspatio-temporal hybrid neural network-Kriging model for groundwater level simulation[J].Journal of Hydrology,2014,519:3193-3203.
[31]FRANCHI G,YAO A,KOLB A.Supervised deep Kriging for single-image super-resolution[C]//German Conference on Pattern Recognition.Cham:Springer,2018:638-649.
[32]WU Y,ZHUANG D,LABBE A,et al.Inductive graph neural networks for spatiotemporal kriging[J].arXiv:2006.07527,2020.
[33]ZHOU T,SHAN H,BANERJEE A,et al.Kernelized probabilistic matrix factorization:Exploiting graphs and side information[C]//Proceedings of the 2012 SIAM International Confe-rence on Data mining.Society for Industrial and Applied Mathematics,2012:403-414.
[1] 高越, 傅湘玲, 欧阳天雄, 陈松龄, 闫晨巍.
基于时空自适应图卷积神经网络的脑电信号情绪识别
EEG Emotion Recognition Based on Spatiotemporal Self-Adaptive Graph ConvolutionalNeural Network
计算机科学, 2022, 49(4): 30-36. https://doi.org/10.11896/jsjkx.210900200
[2] 刘亚臣, 黄雪莹.
卫星监测时空大数据蠕变特征提取及预警算法
Research on Creep Feature Extraction and Early Warning Algorithm Based on Satellite MonitoringSpatial-Temporal Big Data
计算机科学, 2021, 48(11A): 258-264. https://doi.org/10.11896/jsjkx.201000071
[3] 刘浪, 李梁, 但远宏.
用于视频修复的连贯语义时空注意力网络
Coherent Semantic Spatial-Temporal Attention Network for Video Inpainting
计算机科学, 2021, 48(10): 239-245. https://doi.org/10.11896/jsjkx.200600130
[4] 游兰, 韩雪薇, 何正伟, 肖丝雨, 何渡, 潘筱萌.
基于改进Seq2Seq的短时AIS轨迹序列预测模型
Improved Sequence-to-Sequence Model for Short-term Vessel Trajectory Prediction Using AIS Data Streams
计算机科学, 2020, 47(9): 169-174. https://doi.org/10.11896/jsjkx.190800060
[5] 刘长赟,杨宇迪,周丽华,赵丽红.
带有时间标签的流行社交位置发现
Discovering Popular Social Location with Time Label
计算机科学, 2019, 46(7): 186-194. https://doi.org/10.11896/j.issn.1002-137X.2019.07.029
[6] 张敬芝 高强 耿桦 潘金贵.
统计自然语言处理中的线性插值平滑技术

计算机科学, 2007, 34(6): 223-225.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!