| 53 | 0 | 1509 |
| 下载次数 | 被引频次 | 阅读次数 |
针对空间尺度转换中常见的区域边界拓扑失真与预测值不连续问题，本文提出一种基于边界距离加权的空间尺度平滑算法。该算法摒弃传统基于区域中心的权重分配思路，通过计算预测点至相邻区域边界的几何距离，构建连续的距离衰减函数，进而动态融合不同局部模型的预测结果，在保留区域空间特征的同时，能生成跨边界连续平滑的预测曲面。基于美国境内雪荷载数据的实验表明，本文算法在保持地理空间结构一致性方面优于移动窗口法与多尺度地理加权回归(MGWR)法；十折交叉验证显示，普通最小二乘回归(OLS)、广义可加模型(GAM)与Kriging模型的均方误差(MSE)分别降低80.5%、32.0%与18.5%;参数敏感性分析确定平滑距离阈值S为25 km时算法性能最优。该算法可为地理信息系统制图综合、遥感影像边缘优化及强空间异质性地理建模提供有效工具。
Abstract：To address the persistent challenges of topological distortion at regional boundaries and discontinuity of predicted values during spatial scale transformation, this paper proposes a novel spatial smoothing algorithm based on boundary-distance weighting.Departing from the traditional region-centroid-based weighting paradigms, the proposed algorithm quantifies cross-boundary spatial interaction intensity by computing the geometric distance from any prediction point to adjacent regional boundaries and constructs a continuous distance decay function to dynamically integrate predictions from multiple local models, each representing a distinct geographic subunit.The proposed algorithm preserves intra-regional spatial characteristics while generating smooth and continuous prediction surfaces across boundaries.Experiments conducted using snow load data within the contiguous United States demonstrate that the proposed algorithm surpasses the moving window method and multi-scale geographically weighted regression(MGWR) in maintaining the geospatial structural consistency.Computational efficiency is significantly enhanced through a multi-level optimization strategy integrating geometric library acceleration, polygon simplification, and parallel computing.Ten-fold cross-validation results reveal that the mean squared errors(MSE) of ordinary least squares(OLS) regression, generalized additive models(GAM),and Kriging models are reduced by 80.5%,32.0%,and 18.5%,respectively.Parameter sensitivity analysis further identifies that the algorithm achieves optimal performance when the smoothing distance threshold S is set to 25 km.This study provides an effective computational tool for cartographic generalization in geographic information systems, edge optimization in remote sensing imagery, and geographical modeling under strong spatial heterogeneity.
[1] 陆大道.关于地理学的“人-地系统”理论研究[J].地理研究，2002,21(2):135-145.
[2] 吴传钧.论地理学的研究核心：人地关系地域系统[J].经济地理，1991(3):1-6.
[3] 吴儒练，古雄，蔡俊兴，等.长三角地区绿色制造企业空间格局及驱动因素研究[J].地理与地理信息科学，2025,41(2):56-63.
[4] Wang Huiwen,Huang Ting,Wang Shan.A flexible spatial autoregressive modelling framework for mixed covariates of multiple data types[EB/OL].(2018-11-07)[2026-01-10].arXiv:1811.02809.
[5] 郭洁，董磊，邬伦，等.空间异质性建模方法[J].地理学报，2025,80(3):567-585.
[6] Gyanwali S,Karki S,Jang K M,et al.Implications for spatial non-stationarity and the neighborhood effect averaging problem (NEAP) in green inequality research:evidence from three states in the USA[J].Journal of Geographical Systems,2025,27(1):7-30.
[7] Shaw T A,Stevens B.The other climate crisis[J].Nature,2025,639(8056):877-887.
[8] Li Shuangcheng,Zhao Zhiqiang,Miaomiao Xie,et al.Investigating spatial non-stationary and scale-dependent relationships between urban surface temperature and environmental factors using geographically weighted regression[J].Environmental Modelling & Software,2010,25(12):1789-1800.
[9] Haas T C.Lognormal and moving window methods of estimating acid deposition[J].Journal of the American Statistical Association,1990,85(412):950-963.
[10] Matthew J H,Christensen W F,Terres M A.Nonstationary gaussian process models using spatial hierarchical clustering from finite differences[J].Technometrics,2017,59(1):93-101.
[11] Lu Binbin,Hu Yigong,Murakami D,et al.High-performance solutions of geographically weighted regression in R[J].Geo-Spatial Information Science,2022,25(4):536-549.
[12] Fortheringham A S,Oshan T M,Li Ziqi.Multiscale geographically weighted regression:theory and practice[M].1st ed.Boca Raton:CRC Press,2023.
[13] Zhang Yonglin,Li Shanlin,Dong Rencai,et al.Quantifying physical and psychological perceptions of urban scenes using deep learning[J].Land Use Policy,2021,111:105762.DOI:10.1016/j.landusepol.2021.105762.
[14] Forman R T T.Land mosaic:the ecology of landscapes and regions[M].Cambridge:Cambridge University Press,1995.
[15] Bail C A,Argyle L P,Brown T W,et al.Exposure to opposing views on social media can increase political polarization[J].Proceedings of the National Academy of Sciences of the United States of America,2018,115(37):9216-9221.
[16] 程建燕，康建荣，陈炳乾，等.一种新的土地利用图斑边界一致化改正算法[J].地理与地理信息科学，2019,35(6):15-19.
[17] Wang Zhichao,Chen Xinhai,Yan Junjun,et al.An intelligent mesh-smoothing method with graph neural networks[J].Frontiers of Information Technology & Electronic Engineering,2025,26(3):367-384.
[18] Lavreniuk M,Kussul N,Shelestov A,et al.Delineate anything flow:fast,country-level field boundary detection from any source[EB/OL].(2025-11-17)[2026-01-10].arXiv:2511.13417.
[19] Rocca L,Iuricich F,Puppo E.Disambiguating flat spots in discrete scalar fields[J].Graphical Models,2025,141:101299.DOI:10.1016/j.gmod.2025.101299.
[20] Liu Shuaiqi,Shao Tingting,Liu Siyuan,et al.An asymptotic multiscale symmetric fusion network for hyperspectral and multispectral image fusion[J].IEEE Transactions on Geoscience and Remote Sensing,2025,63:5503016.DOI:10.1109/TGRS.2025.3525840.
[21] O′Sullivan D.Geographically weighted regression:the analysis of spatially varying relationships[J].Geographical Analysis,2003,35(3):272-275.
[22] Richards J,Wadsworth J L.Spatial deformation for nonstationary extremal dependence[J].Environmetrics,2021,32(5):e2671.DOI:10.1002/env.2671.
[23] Tippmann S.Programming tools:adventures with R[J].Nature,2015,517(7532):109-110.
[24] Commission for Environment Cooperation.Ecological regions of North America:toward a common perspective[R].Montreal:Commission for Environmental Cooperation,1997.
[25] Li Limin,Wen Zongzhou,Wang Zhongsheng.Outlier detection and correction during the process of groundwater lever monitoring base on Pauta criterion with self-learning and smooth processing[C]//Theory,Methodology,Tools and Applications for Modeling and Simulation of Complex Systems.Singapore:Springer,2016:497-503.
[26] Hewage U H W A,Sinha R,Naeem M.Privacy-preserving data (stream) mining techniques and their impact on data mining accuracy:a systematic literature review[J].Artificial Intelligence Review,2023,56(9):10427-10464.
[27] Bean B,Maguire M,Sun Yan.Comparing design ground snow load prediction in Utah and Idaho[J].Journal of Cold Regions Engineering,2019,33(3):04019010.DOI:10.1061/(ASCE)CR.1943-5495.0000190.
基本信息:
中图分类号:P208
引用信息:
[1]饶加旺,李晶,马荣华,等.基于边界距离加权的空间尺度平滑算法及其在跨尺度地理预测中的应用[J].地理与地理信息科学,2026,42(04):9-16.
基金信息:
国家自然科学基金项目(42361144002); 江苏省自然资源科技创新项目(JSZRKJ202409)
2026-07-21
2026-07-21
2026-07-21