Track anomaly detection method, device, equipment and medium

Abstract

The present invention relates to the field of artificial intelligence, and provides an artificial intelligence-based trajectory abnormality detection method, device, equipment and medium, which can determine the probability of each trajectory segment according to current trajectory data and historical trajectory data, and each trajectory data includes two-way intersections and The time passing through the two-way intersection increases the information of the time dimension, which makes up for the deficiency of simply using spatial information, thereby effectively improving the recall rate of vehicle track anomaly detection, and improving the one-way intersection information into two-way intersection information, effectively The false detection caused by the opposite direction of travel is reduced, thereby improving the accuracy of vehicle trajectory abnormality detection, and then it can automatically and accurately determine whether the current trajectory data is abnormal to assist vehicle detection. The present invention also relates to block chain technology, and abnormal detection results are stored in the block chain. The present invention can be applied in intelligent traffic scenarios, thereby promoting the construction of intelligent cities.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based trajectory abnormality detection method, device, equipment and medium. Background technique [0002] Vehicle trajectory anomaly detection can be applied indirectly or directly to many scenarios in the traffic field (such as drunk driving, taxi detours, suspected car escape, etc.), which plays a very important role in intelligent transportation systems and attracts more and more people. More and more researchers have paid attention to it, and many vehicle trajectory anomaly detection algorithms have emerged. [0003] At present, from the comprehensive point of view of timeliness and accuracy, the detection algorithm based on intersection information is relatively effective, and has become one of the more popular algorithms for vehicle trajectory anomaly detection. [0004] However, because the traditional vehicle trajectory...

AI Technical Summary

Problems solved by technology

[0004] However, since the traditional vehicle trajectory anomaly detection algorithm based on intersection information only uses one-way intersection information in space, the accuracy and recall are limited to a certain extent, and there is still a lot of room for improvement.

Images