顾及地理语义的运动轨迹相似性度量模型  

Trajectory similarity measurement model considering geographic semantics

作　　者：周慧君 罗世佳 蒋和平 张晶[1,2,3,4] ZHOU Huijun;LUO Shijia;JIANG Heping;ZHANG Jing(Beijing Laboratory of Water Resources Security,Capital Normal University,Beijing 100048,China;Key Laboratory of 3-Dimensional Information Acquisition and Application,Capital Normal University,Beijing 100048,China;National Experimental Teaching Center of Urban Environmental Processes and Digital Simulation,Capital Normal University,Beijing 100048,China;College of Resource Environment and Tourism,Capital Normal University,Beijing 100048,China)

机构地区：[1]首都师范大学水资源安全北京实验室,北京100048 [2]首都师范大学三维信息获取与应用教育部重点实验室,北京100048 [3]首都师范大学城市环境过程虚拟仿真国家级实验教学中心,北京100048 [4]首都师范大学资源环境与旅游学院,北京100048

出　　处：《测绘通报》2023年第3期67-73,149,共8页Bulletin of Surveying and Mapping

基　　金：国家自然科学基金(42071376)。

摘　　要：移动对象的运动行为往往受到所处地理环境的影响,针对现有轨迹相似性分析无法考虑地理环境因素的现状,本文提出了一种顾及地理语义的多重运动特征轨迹相似性度量方法。该方法首先以扩展到时间层面的数据立方体为基础,对轨迹序列的地理语义和运动特征进行符号化;然后利用多个特征值计算得到的加权编辑距离作为衡量轨迹相似性的标准;最后将该相似性度量方法与谱聚类结合,利用真实数据验证了该度量模型的有效性与优势。The motion behaviour of moving objects is often influenced by the geographical environment in which they are located.In response to the current situation that existing trajectory similarity analysis often fails to take into account the geographical environment,a multiple motion feature trajectory similarity measure that takes into account geographical semantics is proposed.The method is based on a data cube that extends to the temporal level,and symbolises the geographic semantics and motion features of the trajectory sequence.Finally,this similarity measure is combined with spectral clustering to explore the motion behaviour and patterns of moving objects,and the validity and advantages of the model are verified using real data.

关 键 词：地理环境 地理语义 运动轨迹 运动特征 时空立方体 

分 类 号：P208[天文地球—地图制图学与地理信息工程]