Multi-dimensional trajectory analysis method and device

Abstract

The invention provides a multi-dimensional trajectory analysis method and device. The multi-dimensional trajectory analysis method comprises the steps: structuring multi-dimensional data and then storing the structured multi-dimensional data in a data warehouse; extracting a target trajectory and each associated trajectory from a data warehouse; dividing the target trajectory and the associated trajectories into trajectory segments; determining a space-time characteristic value of each track section; for an associated trajectory similar to the target trajectory, counting the number of trajectory segments with the same spatial-temporal characteristic values as the associated trajectory and the target trajectory, and taking the number as an associated value of the associated trajectory; forassociated trajectories different from the target trajectories, converting the target trajectories into feature trajectories similar to the associated trajectories, and counting the number of trajectory segments with the same spatial-temporal feature values as the associated trajectories to serve as associated values of the associated trajectories; sorting the associated trajectories according tothe associated values; and determining the associated track ranked at the top preset position as the track of the target associated object. By applying the embodiment of the invention, the efficiencyand accuracy of multi-dimensional trajectory analysis are improved.

Description

technical field [0001] The invention relates to the field of multidimensional trajectory analysis, in particular to a multidimensional trajectory analysis method and device. Background technique [0002] In order to strengthen the security protection of the city, it is usually necessary to analyze the trajectory of some targets to determine other objects traveling with the target. No. Find the car taken by the companions and so on. [0003] At present, the existing multi-dimensional trajectory analysis methods usually directly compare the trajectory of the target with the trajectory of other objects, and then determine whether the two are related. However, in the real environment, there are always problems such as data duplication and single-point data over-density in the trajectory data of each dimension, traffic congestion, reasonable distribution of collection equipment, network congestion and delay, front-end and back-end The degree of cohesion of the system, etc., wil...

AI Technical Summary

Problems solved by technology

Although this kind of situation also reflects the behavior trajectory of the car owner within this time frame, in terms of the analysis of the entire behavior trajectory of the target (car owner), the information entropy of the trajectory generated by the target in the time period of the traffic jam will be will be lower

At the same time, this situation will also cause the data density of the target to increase in the spatial range near the traffic jam location, thereby increasing the trajectory scale of the target during the correlation analysis, and ultimately increasing the difficulty of analysis and reducing the efficiency of analysis

[0004] In addition, due to differences in the acquisition process of different types of acquisition equipment, the data density of trajectories of different dimensions varies greatly

For example, the data collection density of sniffing devices such as WIFI and RFID is several times higher than that of bayonet devices such as vehicles and portraits. Difficulty, it is easy to cause problems such as low accuracy of correlation analysis and low efficiency of data analysis

Images