Method and device for searching fake plate suspicion vehicle

Abstract

The invention relates to the field of intelligent transportation, and discloses a method and device for searching a fake plate suspicion vehicle. According to the method for searching the fake plate suspicion vehicle, with respect to two adjacent passage records in snapshot time, if at least one middle pass point needing to be snapshot exists in all routes between pass points of the two passage records without detection records, a vehicle is judged to be the fake plate suspicion vehicle, and therefore a novel method convenient and reliable to use is provided for the judgment of the fake plate suspicion vehicle. A vehicle object-oriented storage and search method is adopted for vehicle historical data in the system, all vehicles with different license plates in the system are used as the maximum record number, a unique index is established through the license plates, the tiny text message with the memory content of the passage records of each vehicle being within a single-vehicle object record is increased, storage resources are saved, the searching speed of the historical vehicle passing record is increased, the problems that a plurality of database records need to be searched in fake plate suspicion vehicle comparison, so that resource consumption is large and real-time implementation can not be achieved are solved, and real-time effective detection can be achieved.

Description

technical field [0001] The invention relates to the field of intelligent transportation, in particular to the detection technology of suspected vehicles with false license plates. Background technique [0002] With the increase in the number of motor vehicles in recent years, there are more and more criminal activities related to vehicles. The license plate, as the only ID card of the vehicle, is the key clue to check the crimes related to vehicles, but because of this, many criminals Molecules use the method of forging and hanging other people's license plates, that is, decks, to evade responsibility. For example, some car owners use other license plates to drive illegally for "convenient" driving without being restricted by electronic police. Even if caught by the electronic police, they will not be found on their own heads, and they will get away with it. Driven by economic interests, many car owners do not go through registration after buying a car, and directly use oth...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem of license plate inspection, there are currently many solutions, which can be roughly divided into two categories. One is the method of complete video detection, which uses the spatiotemporal characteristics and registration information of the vehicle to identify license plate vehicles, but the problem lies in video detection and identification of vehicles. On the one hand, the method of color, license plate and car logo cannot be 100% accurate. Even if there is an error of 5%, based on the data of tens of millions of records per day, there will be hundreds of thousands of false positives. Therefore, only the license plate is used directly. The comparison methods such as vehicle color and vehicle logo are not feasible in actual operation due to too many false positives; on the other hand, the spatio-temporal information of the vehicle is used. These solutions often use real-time detection results to the central relational database retrieval method, because Vehicle databases are often in the hundreds of millions and are constantly accumulating. Retrieval and comparison can only be theoretically feasible, and the effect of real-time detection and alarm cannot be guaranteed at all.

In the Chinese patent with the application number 200910099475.X, the method of storing and caching can solve the problem of quick comparison, but because it is aimed at the principle of travel time between two card points for real-time license plate matching and comparison, the judgment of travel Therefore, in the actual situation where there are more and more card points, the required cache storage and the consumed computing resources increase exponentially, and finally it can only be used for the limited number of card points. The plate detection of driving vehicles cannot solve the problem of large data volume plate car comparison caused by the actual needs of cross-regional plate vehicle detection in principle, and it is impossible to realize the post-event plate analysis and alarm function from historical vehicle records , so there are serious limitations in the application

Images