Spatial interpolation based on spatial auto-regressive neural network
ZENG Jindi1,2, ZHANG Feng1,2, WU Sensen1,2, DU Zhenhong1,2, LIU Renyi1,2
1.Zhejiang Provincial Key Lab of GIS, Zhejiang University, Hangzhou 310028, China 2.Department of Geographic Information Science, Zhejiang University, Hangzhou 310027, China
Abstract：The calculation of the spatial autocorrelation weight based on spatial distance is the core of the classical spatial interpolation method. However, due to the complex nonlinear relationship between spatial distance and the autocorrelation weight, traditional spatial interpolation methods such as inverse distance weighting (IDW) and the Kriging method have limitations on the accurate weight calculation. In this paper,based on the strong nonlinear fitting ability of neural network, we establish a spatial auto-regressive neural network (SARNN) model by combining neural network and spatial autoregressive method, and realize the accurate calculation of spatial autocorrelation weights. In order to verify the validity and feasibility of the SARNN model, we use two types of simulation data and marine environment data for cross-validation, and compare the accuracy with that of the IDW and Kriging. The results show that the performance SARNN is significantly better than IDW and Kriging, regarding all the statistical indicators such as R2, RMSE, MAE and MAPE. At the same time, SARNN predicts the mutation data and extremum more accurately and improves the problems in traditional interpolation method such as low spatial smooth transition, “bull's eye” and sawtooth phenomenon. Therefore, SARNN provides a new idea of spatial interpolation and has a wider application potential.
1 FOTHERINGHAM A S, ROGERSON P A. The SAGE Handbook of Spatial Analysis[M]. London：Sage Publication, 2008. 2 TOBLER W R. A computer movie simulating urban growth in the Detroit region[J]. Economic Geography， 1970, 46(2):234-240. DOI: 10.2307/143141 3 GOODCHILD M F, ANSELIN L, DEICHMANN U. A framework for the areal interpolation of socioeconomic data[J]. Environment & Planning A, 1993, 25(3): 383-397. DOI: 10.1068/a250383 4 GOOVAERTS P. Geostatistical approaches for incorporating elevation into the spatial interpolation of rainfall[J]. Journal of Hydrology, 2000, 228(1): 113-129. DOI: 10.1016/S0022-1694(00)00144-X 5 ZHUANG L. Spatial interpolation methods of daily weather data in Northeast China[J]. Quarterly Journal of Applied Meteorology,2003, 14(5) : 605?615. DOI: 10.3969/j.issn.1001-7313.2003.05.011 6 CRESSIE N. The origins of Kriging[J]. Mathematical Geology, 1990, 22(3): 239-252. DOI: 10.1007/bf00889887 7 JIN L, HEAP A D. A review of comparative studies of spatial interpolation methods in environmental sciences: Performance and impact factors[J]. Ecological Informatics,2011, 6(3): 228-241. 8 ZHANG S, ZHANG J, YIN L, et al. A mathematical spatial interpolation method for the estimation of convective rainfall distribution over small watersheds[J]. Environmental Engineering Research,2016, 21(3): 226-232. 9 QIAO P, LI P, CHENG Y, et al. Comparison of common spatial interpolation methods for analyzing pollutant spatial distributions at contaminated sites[J]. Environmental Geochemistry and Health, 2019, 41(6): 2709-2730. DOI: 10.1007/s10653-019-00328-0 10 丁卉，余志，徐伟嘉，等. 3种区域空气质量空间插值方法对比研究[J]. 安全与环境学报， 2016, 16(3): 309-315. DOI: 10.13637/j.issn.1009-6094.2016.03.061 DING H, YU Z, XU W J, et al. Comparative study of the three spatial interpolation methods for the regional air quality evaluation[J]. Journal of Safety and Environment， 2016, 16(3): 309-315. DOI: 10.13637/j.issn.1009-6094.2016.03.061. 11 ARIF M , HUSSAIN I , HUSSAIN J , et al. GIS-based inverse distance weighting spatial interpolation technique for fluoride distribution in south west part of Nagaur district, Rajasthan[J]. Cogent Environmental Science,2015, 1(1):1038944. DOI: 10.1080/23311843.2015.1038944 12 BRUNSDON C, FOTHERINGHAM A S, CHARLTON M. Spatial nonstationarity and autoregressive models[J]. Environment & Planning A, 1998, 30(6): 957-973. DOI: 10.1068/a300957 13 WU B, LI R, HUANG B. A geographically and temporally weighted autoregressive model with application to housing prices[J]. International Journal of Geographical Information Science, 2014, 28(5): 1186-1204. DOI: 10.1080/13658816.2013.878463 14 李章义，陈媛. Kriging插值中不同变异函数对城市环境场强重构的影响[J]. 中国无线电, 2017(7): 51-53. DOI: 10.3969/j.issn.1672-7797.2017.07.034 LI Z Y, CHEN Y. The influence of different variogram in Kriging on the reconstruction of urban environmental field strength[J]. China Radio, 2017(7):51-53. DOI: 10.3969/j.issn.1672-7797.2017.07.034 15 SCHMIDHUBER J. Deep learning in neural networks: An overview[J]. Neural Networks,2015, 61: 85-117. DOI: 10.1016/j.neunet.2014.09.003 16 李纯斌，刘永峰，吴静，等. 基于BP神经网络和支持向量机的降水量空间插值对比研究——以甘肃省为例[J]. 草原与草坪, 2018, 38(4): 12-19. DOI: 10.3969/j.issn.1009-5500.2018.04.002 LI C B, LIU Y F, WU J, et al. Comparative study on spatial interpolation based on BP neural network and support vector machine-Taking Gansu province as an example [J]. Grassland and Turf,2018, 38(4): 12-19. DOI: 10.3969/j.issn.1009-5500.2018.04.002. 17 邱云翔，张潇潇，刘国东. 粒子群算法优化BP在降雨空间插值中的应用[J]. 长江科学院院报， 2017, 34(12): 28-32. DOI: 10.11988/ckyyb.20160837 QIU Y X, ZHANG X X, LIU G D. Application of PSO–BP in rainfall spatial interpolation[J]. Journal of Yangtze River Scientific Research Institute，2017, 34(12): 28-32. DOI: 10.11988/ckyyb.20160837 18 SOARES S A F , NETO G S , ROISENBERG M. Improving the incremental Gaussian mixture neural network model for spatial interpolation and geostatistical simulation[C]// 2016 International Joint Conference on Neural Networks (IJCNN). New York: IEEE, 2016. DOI: 10.1109/IJCNN.2016.7727511 19 ZHU D, CHENG X, ZHANG F, et al. Spatial interpolation using conditional generative adversarial neural networks[J]. International Journal of Geographical Information Science，2020, 34(4): 735-758. DOI: 10.1080/13658816.2019.1599122. 20 SRIVASTAVA N, HINTON G, KRIZHEVSKY A, et al. Dropout: A simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research,2014, 15(1): 1929-1958. 21 HINTON G E, SRIVASTAVA N, KRIZHEVSKY A, et al. Improving neural networks by preventing co?adaptation of feature detectors[J]. Computer Science，2012, 3(4): 212-223. DOI: 10.9774/GLEAF.978-1-909493-38-4_2 22 NAIR V, HINTON G E. Rectified linear units improve restricted boltzmann machines[C]//The 27th International Conference on Machine Learning (ICML-10). Haifa: ICML,2010: 807-814. 23 HE K, ZHANG X, REN S, et al. Delving deep into rectifiers: Surpassing human-level performance on imageNet classification[C]// 2015 IEEE International Conference on Computer Vision(ICCV).Washington,DC:IEEE Computer Society,2015: 1026-1034. DOI: 10.1109/ICCV.2015.123 24 IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[J].arXiv:1502.03167.[2015-03-02].http://arXir.org/pdf/1502.03167.pdf. 25 HARRIS P. A simulation study on specifying a regression model for spatial data: Choosing between autocorrelation and heterogeneity effects[J]. Geographical Analysis,2019, 51(2): 151-181. DOI: 10.1111/gean.12163
2020, Vol. 47 
