分级统计地图的模型评价与应用  

Model Evaluation and Application for Choroplethic Maps

作　　者：乔俊军[1] 胡冯伟 

机构地区：[1]武汉大学测绘学院,湖北武汉430079 [2]浙江省测绘科学技术研究院,浙江杭州310012

出　　处：《测绘地理信息》2017年第6期105-109,共5页Journal of Geomatics

基　　金：国家科技支撑计划资助项目(2008BAH31B02-3)

摘　　要：首先,从点纹、线纹、混合纹、底色4个方面归纳了分级统计地图的面状符号类型;然后,以统计数据的利用程度为视角对分级模型进行了分类,同时,给出了动态聚类初始聚类中心的计算方法,并提出了一种用级内数据集中度和级间均值离散度来评价分级精度的方法;最后,在CorelDRAW环境下进行了二次开发,实现了分级统计地图从模型分级、评价、选择到制图单元的点纹、线纹、底色填充等一系列过程的自动化,对提高统计地图制图水平和效率具有现实意义。This paper summarizes the area symbol types of choroplethic map from the four aspects which includes dots,lined,mixed grain,background.The classification models are categorized in the perspective of the usage degree of statistical data.A calculation method of initial cluster center of dynamic clustering is presented.Meanwhile,it proposes a method of applying the inter-level data within the mean concentration and dispersion level to evaluate the classification accuracy.Finally,under the CorelDRAW environment,it realizes the process automation of choroplethic map from the model classification,evaluation,selection to mapping unit of some points,lines,background filling.It is of practical significance to the improvement of the cartography level and statistics efficiency.

关 键 词：分级统计地图 分级模型 精度评价 

分 类 号：P283.7[天文地球—地图制图学与地理信息工程]