MOASM: Anomalous Sub-trajectory Monitoring Of Moving Objects Over Trajectory Streams

Anomaly detection based on space-temporal trajectory streams is an important issue, especially in the field of military surveillance and safety monitoring. Due to the complexity and variability of application scenarios and the quality of trajectory data, this problem still faces many challenges. In this paper, we first give some novel definitions of anomalous trajectory point and anomalous sub-trajectory, and establish an efficient index structure GH-P based on GeoHash. Then, we propose a framework called MOASM for dynamic monitoring of anomalous sub-trajectories of moving objects. Finally we have experiments on the AIS trajectory data. The experiment results demonstrate that our method can detect the anomalous sub-trajectories of a moving object, which may represent the behavior of deviating from the channel and changing the course suddenly. Ideally, MOASM can support dynamic monitoring task for moving objects, which is significant for specific applications.