A detection method for license plates based on probability aggregation of vehicle hotspots

Abstract

A vehicle detection method based on probability aggregation of vehicle hotspot areas, comprising the following steps: S1. Extraction of checkpoint data; S2. Cleaning of checkpoint data; S3. Calculation of flow probability between checkpoints; S4. Abnormal calculation behavior Vehicle list: According to the bayonet pair of each vehicle extracted in S2, calculate the bayonet pair probability, arrange in descending order according to the number of times the probability is lower than the set threshold Ti, and select the first N lines as the abnormal behavior vehicle list according to needs; S5. Calculate the checkpoint points in the hotspot area of ​​the vehicle; S6. Calculate the probability of the vehicle passing between the checkpoint points in the two hotspot areas; S7. According to the probability between the checkpoints obtained in S6, aggregate all the checkpoints; S8. According to the vehicle checkpoints Aggregation calculates vehicle deck probability Z. The invention overcomes random bayonet number plate recognition errors, utilizes the relatively fixed characteristics of vehicle activity areas, screens out vehicles with a relatively high degree of suspicion of decking, greatly reduces the scope of suspicion, and has good practicability.

Description

technical field [0001] The invention relates to the field of intelligent transportation, in particular to a detection method for vehicles with license plates. Background technique [0002] Cars with license plates, commonly known as clone cars, refer to vehicles that go on the road through forged or illegally obtained license plates and driving licenses of other vehicles. With the development of the economy, the number of vehicles is increasing, and the license plate vehicles are also increasing. False license plates will disrupt the public security authorities' control over public safety. Since license plates do not have legal procedures and insurance, once a traffic accident occurs, the driver can easily escape. The set of license plates also provides criminal tools for criminal activities, which greatly increases the difficulty of solving crimes. Cars with license plates will evade various taxes and fees, causing a large loss of national taxes and fees, and disrupting t...

AI Technical Summary

Problems solved by technology

[0020] In order to overcome the shortcomings of the existing detection methods for vehicles with false checkpoints due to random bayonet number plate recognition errors, excessive suspect ranges, and poor practicability, the present invention provides a method for overcoming random bayonet number plate recognition errors, using vehicle Due to the relatively fixed characteristics of the activity area, it can screen out vehicles with a high degree of suspicion of decking, greatly reduce the scope of suspicion, and have good practicability. A decked vehicle detection method based on the probability aggregation of vehicle activity hotspots

Images