基于信息熵的空间插值影响因素分析

Analysis of Influencing Factors of Spatial Interpolation Based on Information Entropy

空间数据作为一种地理现象的离散度量广泛应用于各类地理空间分析。然而,空间数据经常由于各种原因出现数据的缺失,存在不完备的特性,需要利用空间插值方法对其进行插值估计。现有国内外对于空间插值方法的研究较多,但是对于空间插值影响因素的分析研究则相对较少。本文首先对国内外相关空间插值研究工作进行总结和回顾,进而探讨不同采样密度、采样分布和采样粒度对于空间插值精度的影响。在此基础上,提出一种基于信息熵的空间插值影响因素分析方法。该方法针对采样数据的不同采样密度和采样分布,对空间数据的可插值性进行比较分析,进而探讨插值粒度,即插值网格分辨率对于空间数据可插值性的影响。最后,采用一组实际数据验证了使用信息熵对空间数据进行可插值性研究的可行性和适用性。

Spatial data plays a key role for measuring geographical phenomena in human life.As a discrete geographic phenomena measure,spatial data is widespread in people's lives.However,the spatial data is often caused by unexpected observation station and the data is missing.Because of the imperfection of spatial data,spatial interpolation method is needed to estimate the missing data.There are a lot of researches for spatial interpolation methods,but the interpolation analysis of spatial data is less.In this paper,we review the research work of spatial interpolation,and then discuss the effect of different sampling density,sampling distribution and sample size on the results of spatial interpolation,and propose a new method based on information entropy.In this method,the sampling density and sampling distribution of the sample data are first analyzed,and the interpolation of spatial data is analyzed.In the cases of the same data sampling density and sampling distribution,the effect of interpolation granularity is discussed.Finally,the feasibility and applicability of using the information entropy is verified by using a set of spatial data.