一种基于时序的层次轨迹聚类算法  

A Hierarchical Trajectory Clustering Algorithm Based on Time Series

作　　者：冷泳林[1] 鲁富宇[2] LENG Yong-lin a LU Fu-yu b(a. College of Information Science and Technolog b. Office of Academic Affairs, Bohai University, Jinzhou 121000, China)

机构地区：[1]渤海大学信息科学与技术学院,辽宁锦州121000 [2]渤海大学教务处,辽宁锦州121000

出　　处：《重庆理工大学学报（自然科学）》2017年第3期123-127,157,共6页Journal of Chongqing University of Technology：Natural Science

基　　金：辽宁省社科基金资助项目(L14AGL002;L13AGL002);辽宁省教育科学规划十三五项目(JG16DB009);辽宁省社科联2015年度合作项目(lslgslhl-020)

摘　　要：聚类相似的运动轨迹,获取对象主要运动特征是轨迹路径聚类的目标之一。本文针对轨迹路径数据量大、传统整体轨迹聚类算法效率低等问题,提出了一种基于时序的层次轨迹聚类算法(hierarchical trajectory clustering algorithm based on time series,HTCTS)。算法首先将完整的轨迹数据按一定的时间间隔进行分割,然后对分割的子路径分别聚类,最后在对聚类子集进行二次聚类,生成最终的聚类结果。实验结果表明:HTCTS算法在聚类效率和聚类质量上高于整体轨迹聚类算法。Clustering movement trajectories to get the motion feature of object are one of the goals of the trajectory clustering. Aiming at the large scale trajectory data,to address the low efficiency of clustering,this paper proposes a hierarchical trajectory clustering algorithm based on time series（ HTCTS）. The algorithm first divides trajectory data by time interval,and then respectively clusters the sub paths. Finally,for all cluster subset,HTCTS executes cluster algorithm again to produce the final clustering results. The experimental results show that HTCTS algorithm in clustering efficiency and quality is superior to the trajectory clustering algorithms which cluster the whole dataset directly.

关 键 词：轨迹聚类 轨迹路径 相似性 AP算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]