The embodiment of the invention relates to an efficient anomaly recognition method on sparse trajectory data. The method comprises the following steps: constructing a hierarchical pattern tree by using a first algorithm based on all event patterns; and based on the hierarchical mode tree, calculating whether the track mode of the traffic track T is matched with any event mode by using a second algorithm. According to the embodiment of the invention, the event mode behind the event is compressed through the hierarchical mode tree method. Upon checking whether an upcoming traffic trajectory is related to an event pattern, the calculation may be stopped in advance according to a hierarchical pattern tree. The MTTD time of FDM is reduced through construction of a hierarchical mode tree and a rapid matching mode, and the calculation cost is reduced. The accelerated FDM calculation efficiency is very high, a large amount of trajectory data can be effectively processed, and instant response can be made in real time.
本公开实施例是关于一种在稀疏轨迹数据上的高效异常识别方法。该方法包括:基于所有事件模式,利用第一算法构建分层模式树;基于所述分层模式树,利用第二算法计算交通轨迹T的轨迹模式是否与任一事件模式匹配。本公开实施例通过分层模式树方法压缩事件背后的事件模式。在检查即将到来的交通轨迹是否与事件模式相关时,可以根据分层模式树提前停止计算。通过构建分层模式树和快速匹配的方式减少FDM的MTTD时间,减少计算的成本。加速后的FDM的计算效率很高,能够有效处理大量轨迹数据,能够对实时做出即时响应。
Efficient anomaly identification method on sparse trajectory data
一种在稀疏轨迹数据上的高效异常识别方法
2024-06-21
Patent
Electronic Resource
Chinese
Trajectory anomaly detection system and online trajectory anomaly detection method
European Patent Office | 2024
|High-speed target sparse flight trajectory data prediction method and system
European Patent Office | 2025
|