The increasing pervasiveness of object tracking technologies leads to huge volumes of spatiotemporal data collected in the form of trajectory streams. The discovery of useful group patterns from moving objects' movement behaviours in trajectory streams is critical for real-time applications ranging from transportation management to military surveillance. Motivated by this, we first propose a novel pattern, called evolving group, which models the unusual group events of moving objects that travel together within density connected clusters in evolving streaming trajectories. Our theoretical analysis and empirical study on the Osaka Pedestrian data and Beijing Taxi data demonstrate its effectiveness in capturing the development, evolution, and trend of group events of moving objects in streaming context. Moreover, we propose a discovery method that efficiently supports online detection of evolving groups over massive-scale trajectory streams using a sliding window. It contains three phases along with a set of novel optimization techniques designed to minimize the computation costs. Furthermore, to scale to huge workloads over evolving streams, we extend our discovery method to a parallel framework by using a sector-based partition. Our comprehensive empirical study demonstrates that our online discovery framework is effective and efficient on real-world high-volume trajectory streams.
翻译:物体跟踪技术日益普及,导致以轨迹流的形式收集了大量零星时空数据。发现轨道流中移动物体移动行为的有效群体模式对于从运输管理到军事监视等实时应用至关重要。我们首先提出一个新的模式,即不断变化的群体,该模式在不断演变的轨迹中模拟在密度相连的集群内移动物体的异常群体事件,在不断演变的轨迹中进行移动。我们对大阪派斯特里亚数据和北京出租车数据的理论分析和经验研究表明,它在捕捉流动物体群落事件的发展、演变和趋势方面是有效的。此外,我们提出一个发现方法,高效地支持利用滑动窗口对大规模轨迹流中演变中的群群进行在线检测。它包含三个阶段,以及一套旨在尽量减少计算成本的新型优化技术。此外,为了在不断演变的溪流中缩小巨大的工作量,我们利用基于部门的分隔,将我们的发现方法扩展为平行框架。我们的全面经验研究表明,我们的在线发现框架在现实世界高容量轨道流中是有效和高效的。