关于点实体密度的插值算法探究.pdfVIP

关于点实体密度的插值算法探究 邹丹 （福州大学福建省空间信息工程研究中心，空间数据挖掘与信息共享教育部重点实验室，福州，350002 ） 摘要：居民地是人口分布的载体，了解居民地的分布特征对确定人口分布有重要意义。密度 估计法是研究居民地等点实体分布的主要的密度计算法，而对于区域内点实体密度的衡量， 主要有局部平均法和核密度估计法这两种方法，前者未考虑到权重问题，因此结果不能准确 反映区域内点的实体分布，而核密度估计法采用点实体和计算密度处点的距离的四次函数作 为权重，并且能够消除边缘效应，是一种实用的密度估计法。本文介绍了核密度估计方法， 并且基于 AO8.3 与 VC++6.0 进行编程实现，采用点数据进行验证，并与 ArcGIS 软件中的密 度计算结果进行了对比分析。 关键词：点实体 ，核密度估计，边缘效应 Research on Interpolation Algorithm of Point Entity Density Zou Dan (Spatial Information Research Center, Fujian Province; Key Laboratory of Spatial Data Mining and Information Sharing, Ministry of Education, Fuzhou University, Fuzhou 350002, China) Abstract: Resident is the carrier of population distribution, to understand the distribution characteristics of the residents is significant to determine the population. Density estimation method is a main density calculation method to study the distribution of point entities like residents, and for measuring density of point entities within the region, mainly are local average method and the kernel density estimation these two methods, the former does not take into account the weight issue, so the results can not accurately reflect the physical distribution of points within the region, while the kernel density estimation applies a four-point function of distance between point entity and the place where density is calculated for the weight ,and it can eliminate the edge effect, is a practical density estimation. This article describes the kernel density estimation methods, and is based on AO8.3 with VC6.0 be programmed, using point data validation, and are compared and analyzed w ith the density calculation results under ArcGIS. Keywords: point entity , Kernel Density Estimation ,edge effect 1 引言 点实体的密度计算是地理信息系统领域中的基本问题，它是对居民点密度计算、建筑物 密度计算等地学应用问题进行归