地理要素的分布式空间近似查询处理方法  

Distributed approximate spatial query processing method of geographical features

作　　者：仇阿根[1,2] 刘纪平[2] 张志然[2] 董珍珍[2] 马磊[3] 钱新林[2] 

机构地区：[1]武汉大学,武汉430072 [2]中国测绘科学研究院,北京100830 [3]北京市测绘设计研究院,北京100038

出　　处：《测绘科学》2017年第7期57-64,共8页Science of Surveying and Mapping

基　　金：测绘地理信息公益性行业科研专项项目(201512032);测绘地理信息公益性行业科研专项项目(201512027);中国测绘科学研究院基本科研业务费项目(7771614);国家重点研发计划项目(2016YFC0803108)

摘　　要：针对传统空间查询无法满足地理数据交互式可视化对处理时间要求的问题,以窗口查询为例,提出了一种空间近似查询处理方法。该方法包括预处理和查询两步:在预处理阶段,利用分布化的线简化算法对空间对象进行顾及误差的预处理采样,将采样过程及误差值用树型结构保存;在查询阶段,以豪斯多夫距离定义数据可视化的误差,进行误差可知的顶点即时采样与截取,从而实现针对可视化应用的高效的空间近似查询处理。在Hadoop集群上利用77GB的OpenStreetMap数据集进行了实验,证实了本方法的效力与效率。Interactive visualization of global scale fine detailed geographic data calls for high perform- ance spatial query processing especially the windowing queries. This paper studies the approximate spatial query processing, which is dedicated for data visualization, to tackle the performance problem. Taking the geographic features as the target dataset and hausdorff distance of original and simplified feature as visual- ization error, it presents the distributed approximate spatial query processing techniques which comprise two stages. In the first stage, features are preprocessed using Ramer-Douglas-Peuker algorithm to sample points considering visualization error, then the order during the points sampling and the errors are pre- served in binary trees. In the second stage, points of features are retrieved by binary trees＇ breadth-first- searching with error priority queue and spatial constraint. Extensive experiments on real-world data show the effectiveness and efficiency of the introduced method.

关 键 词：窗口查询 顶点采样 线简化算法 分布式内存计算 二叉树层次结构 误差优先广度遍历 

分 类 号：P208[天文地球—地图制图学与地理信息工程] TP392[天文地球—测绘科学与技术]