首页|基于参考数据的OSM道路数据质量评估模型

基于参考数据的OSM道路数据质量评估模型

OSM Road Data Quality Evaluation Model Based on Reference Data

扫码查看
随着计算机技术的发展与互联网的普及,作为新型地理信息模式的志愿地理信息(Vol-unteered Geographic Information,VGI)受到越来越多学者的关注,逐渐成为专业地理信息的重要补充.由于松散的制图标准和非专业数据编辑,具有代表性的OpenStreetMap(OSM)在许多项目中的应用受到了阻碍,其可靠性和适用性成为研究人员最感兴趣的话题.提出了一种基于参考数据的OSM道路数据质量评价模型,从完整性、主题准确性和位置精度进行分析,并将该模型作为评估OSM数据适用性的方法.以郑州市部分区域的OSM道路数据为例,展开OSM数据质量评估研究,得出质量元素空间分布上的差异与联系.结果表明,与权威数据相比,OSM数据集的几何完整性达到 97%,高等级道路分类精度高达 98%,名称准确性超过88%,两个数据集之间的平均重叠百分比超过 90%,名称完整性的最低结果不足 40%,总体质量比较高.
With the development of computer technology and the popularity of Internet,volunteered geographic information(VGI)as a new geographic information model has received more and more at-tention from scholars and gradually becomes an important supplement to professional geographic in-formation.The representative OpenStreetMap(OSM)has hindered the application of OSM in many projects due to loose mapping standards and non-professional data editing,and its reliability and ap-plicability have become the most interesting topics for researchers.This paper presents an OSM road data quality evaluation model based on reference data,which is studied in terms of completeness,subject accuracy and location accuracy,as a method to assess the applicability of OSM data.In this study,the OSM data quality evaluation study is carried out with OSM road data in some areas of Zhengzhou City,and the differences and associations in the spatial distribution of quality elements are derived.The results show that,compared with authoritative data,the OSM dataset has 97%length completeness,98%classification accuracy of high-grade roads,over 88%name accuracy,while the average overlap percentage between the two datasets is over 90%,with the lowest result being less than 40%for name completeness.The overall quality is relatively high.

volunteer geographic informationOpenStreetMapspatial data quality elementsquali-ty assessment

闫旭强、刘新贵、李慧敏、李帅

展开 >

信息工程大学,河南 郑州 450001

志愿者地理信息 OpenStreetMap道路数据 空间数据质量元素 质量评估

国家自然科学基金资助项目

42101455

2024

信息工程大学学报
中国人民解放军信息工程大学科研部

信息工程大学学报

影响因子:0.276
ISSN:1671-0673
年,卷(期):2024.25(1)
  • 14