Vehicle-track-based detection method of abnormal wandering behaviors in region

Abstract

The invention provides a vehicle-track-based detection method of abnormal wandering behaviors in a region. The method includes the following steps that track data is preprocessed, wherein recording data of all intelligent monitoring and recording devices in a city is collected on a daily basis; a vehicle-track data form Vehicle Trajectory is generated; after being preprocessed, the track data is read in a database table, and a vehicle-track data table is generated; according to a mode of the abnormal wandering behaviors in the region and track information of the obtained data, attributive characters of vehicle tracks are extracted; regarding fives vehicle-track attributive characters, two threshold values are set, and threshold value setting methods include a statistical-data-based threshold value setting method and an expert-experience-based threshold value setting method; general information of abnormally wandering vehicles is extracted. According to the method, by processing the track data, the purpose is achieved that the abnormally wandering vehicles of which the tracks are unclosed in the region are automatically identified; meanwhile, when the method is used for showing the vehicle tracks, the data processing is more visual, convenient and efficient.

Description

technical field [0001] The invention relates to the technical field of abnormal behavior pattern recognition, in particular to a method for detecting abnormal behaviors of wandering in areas based on vehicle tracks. Background technique [0002] With the acceleration of the informatization process, a large number of road intelligent monitoring equipment (electronic checkpoints for short) are deployed on all levels of urban roads, gradually forming a networked vehicle monitoring pattern, providing data for abnormal behavior detection based on vehicle trajectories Base. At the same time, the rapid improvement of computer performance and the rapid development of distributed processing technology also provide technical support for the analysis and processing of massive data. [0003] At present, most of the research on abnormal vehicle behavior detection is through video image processing, pattern recognition and other technologies to directly identify the abnormal driving behav...

AI Technical Summary

Problems solved by technology

[0006] However, the trajectory of a vehicle wandering for a certain target does not necessarily form a closed curve, and it is likely to be a trajectory that is chaotic and accompanied by a certain period of time.

Moreover, although the number of electronic checkpoints is large, it has not reached the level of coverage of the whole city without dead ends. Most of the abnormal vehicles found by this method are vehicles detouring in a large circular area, which is not the same as that found in the field of public security. Abnormal behaviors of wandering have certain differences, and abnormal behaviors that endanger public safety cannot be effectively detected

Images