Anomalous Trajectory Detection Algorithm Based On Trajectory Compression And Road Network Partition

With the burgeoning of car-hailing platforms,the crime rate of online car Hailing increases significantly,where vehicle trajectory often shows abnormal phenomenon.Aiming at these problems,from the point of view of road consumption,this paper deeply researches on the anomalous trajectory detection algorithm and incremental detection algorithm based on trajectory compression and road network partition.The main contents are as follows:(1)An anomalous trajectory detection algorithm based on road network partition is proposed.Firstly,we shift in focus of modeling from trajectories to roads,juggle the influence of time and distance,which effectively improves the accuracy of results.Secondly,trajectories are mapped on road network by map matching probability to improve the data quality.Then,according to road nodes,trajectories are divided into Trajectory segment sequence to support road consumption modeling.Finally,we propose and define the concept of consumption threshold matrix,in which data's range detected is expanded.(2)Based on the above research,an incremental anomalous trajectory detection algorithm based on trajectory compression and road network partition is proposed.Firstly,a compression method with high compression rate and small error is found by comparing different trajectory compression algorithms to compress trajectories,reduce the memory consumption and improve the efficiency of the algorithm.Then,the trajectory segments are stored in road section blocks to optimize the storage strategy and reduce the time needed to retrieve the trajectory data.Finally,the road model is updated in the process of new trajectory entry and detection to realize the detection of new abnormal trajectory.The effectiveness of RPATD and i-RPATD are verified by Shanghai trajectory dataset in this paper.Compared RPATD with i BOAT,TRAOD,TADSS and TPRO,RPATD has higher efficiency and accuracy.And compared i-RPATD with RPATD,it is verified that incremental road mode is meaningful.