CN117115801A - License plate authenticity identification method, device, equipment and storage medium - Google Patents

License plate authenticity identification method, device, equipment and storage medium Download PDF

Info

Publication number
CN117115801A
CN117115801A CN202311308895.0A CN202311308895A CN117115801A CN 117115801 A CN117115801 A CN 117115801A CN 202311308895 A CN202311308895 A CN 202311308895A CN 117115801 A CN117115801 A CN 117115801A
Authority
CN
China
Prior art keywords
license plate
authenticity
image
identification result
true
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202311308895.0A
Other languages
Chinese (zh)
Inventor
陈昭俊
黄海波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Jieshun Science and Technology Industry Co Ltd
Original Assignee
Shenzhen Jieshun Science and Technology Industry Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Jieshun Science and Technology Industry Co Ltd filed Critical Shenzhen Jieshun Science and Technology Industry Co Ltd
Priority to CN202311308895.0A priority Critical patent/CN117115801A/en
Publication of CN117115801A publication Critical patent/CN117115801A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/625—License plates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/19—Recognition using electronic means
    • G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06V30/19147—Obtaining sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/19—Recognition using electronic means
    • G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06V30/19173—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a license plate authenticity identification method, device, equipment and storage medium, relating to the technical field of license plate identification, comprising the following steps: acquiring license plate images in vehicle videos acquired by a camera; performing target detection on the license plate image by using a license plate target detection model to obtain license plate position information and a first true and false recognition result; based on the license plate position information, classifying and identifying the license plate image by utilizing a license plate image classification model to obtain a second true and false identification result; and determining a target authenticity identification result based on the first authenticity identification result and the second authenticity identification result. According to the license plate identification method, the license plate image is identified and judged for multiple times by using the license plate target detection model and the license plate image classification model, so that the license plate authenticity result is obtained, the behavior that a vehicle owner evades payment by using fake license plates is effectively reduced, the loss of a parking lot is effectively reduced, the parking management order is greatly maintained, and the license plate identification method has the advantages of being low in cost, strong in robustness, high in identification effect accuracy and the like.

Description

License plate authenticity identification method, device, equipment and storage medium
Technical Field
The present invention relates to the field of license plate recognition technologies, and in particular, to a license plate authenticity recognition method, device, equipment and storage medium.
Background
In the parking field, the daily use of parking lots today has been kept away from license plate recognition technology, especially when the vehicle needs to be driven away from the field for payment. The current mainstream parking charging modes are: when a vehicle enters a parking lot, a license plate recognition device arranged in an exit/entrance area of the parking lot automatically recognizes license plate information of the vehicle, records the entrance time and the exit time of the vehicle, calculates parking cost and generates a parking bill.
However, some owners can use fake license plates to escape fee in order to escape fee, for example, fake license plates such as mobile phone license plates, paper license plates or fake license plates made of other materials are placed in front of a camera in the license plate recognition device, and the license plate recognition device cannot recognize the authenticity of the license plates, so that the license plate recognition device can open a barrier gate by mistake after recognizing the fake license plates, and the owners can escape fee by releasing the involved vehicles, thereby causing losses to a parking lot.
Disclosure of Invention
Based on the above, it is necessary to provide a license plate authenticity identification method, device, equipment and storage medium for effectively reducing the behavior of a vehicle owner for evading payment by using fake license plates, further effectively reducing the loss of a parking lot, and largely maintaining the parking management order.
A license plate authenticity identification method, comprising:
acquiring license plate images in vehicle videos acquired by a camera;
performing target detection on the license plate image by using a license plate target detection model to obtain license plate position information and a first true and false identification result;
based on the license plate position information, classifying and identifying the license plate image by utilizing a license plate image classification model to obtain a second true and false identification result;
and determining a target authenticity identification result based on the first authenticity identification result and the second authenticity identification result.
According to the license plate authenticity identification method provided by the invention, based on the license plate position information, the license plate image is classified and identified by utilizing a license plate image classification model to obtain a second license plate authenticity identification result, and the license plate authenticity identification method comprises the following steps:
obtaining at least one license plate amplified image based on the license plate position information;
classifying and identifying the at least one license plate amplified image by using the license plate image classification model to obtain a license plate authenticity result and a license plate authenticity score corresponding to the at least one license plate amplified image;
and determining the second authenticity identification result based on the license plate authenticity result corresponding to the at least one license plate amplification image, the license plate authenticity score and a preset scoring strategy.
According to the license plate authenticity identification method provided by the invention, the target authenticity identification result is determined based on the first authenticity identification result and the second authenticity identification result, and the method comprises the following steps:
determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result;
acquiring an image frame number of a license plate image corresponding to the initial false identification result, and judging whether the image frame number meets the preset detection image frame number or not;
if the image frame number meets the preset detection image frame number, determining the target authenticity identification result based on the initial authenticity identification result;
and if the image frame number does not meet the preset detection image frame number, returning to the step of acquiring the license plate image in the vehicle video acquired by the camera until the image frame number meets the preset detection image frame number.
According to the license plate authenticity identification method provided by the invention, the first authenticity identification result comprises license plate categories and first license plate authenticity scores; the second authenticity identification result comprises a license plate authenticity result and a second license plate authenticity score;
the determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result includes:
If the first license plate authenticity score is outside a preset detection score range, comparing the second license plate authenticity score with a preset classification score range based on the license plate authenticity result;
if the second license plate authenticity score is within the preset classification score range, determining that the initial authenticity identification result is the license plate type;
and if the second card authenticity score is outside the preset classification score range, determining the initial authenticity identification result based on the first card authenticity score and the second card authenticity score.
According to the license plate authenticity identification method provided by the invention, the initial authenticity identification result is determined based on the first license plate authenticity score and the second license plate authenticity score, and the method comprises the following steps:
calculating an average value of the true-false scores of the first cards and the true-false scores of the second cards to obtain true-false average scores;
if the true-false average score is smaller than the preset average threshold value, determining that the initial true-false recognition result is a true license plate result;
and if the true-false average score is larger than or equal to the preset average threshold value, determining the initial true-false identification result as the license plate category.
According to the license plate authenticity identification method provided by the invention, after the target authenticity identification result is determined based on the first authenticity identification result and the second authenticity identification result, the license plate authenticity identification method further comprises the following steps:
if the target true-false recognition result is a real license plate result, generating a release instruction, and issuing the release instruction to a channel gate to control the channel gate to open a release target vehicle;
and if the target authenticity identification result is a fake license plate result, generating an escape and payment warning, and pushing the escape and payment warning to a parking lot manager to remind the parking lot manager to collect the target vehicle.
According to the license plate authenticity identification method provided by the invention, before the license plate image in the vehicle video acquired by the camera is acquired, the license plate authenticity identification method further comprises the following steps:
acquiring a plurality of groups of training license plate image samples and training sample labels corresponding to the plurality of groups of training license plate image samples;
performing iterative training on an initial license plate target detection model according to each training license plate image sample and a training sample label corresponding to each training license plate image sample to obtain the license plate target detection model;
amplifying each training license plate image sample to obtain a plurality of groups of training amplified images and training sample labels corresponding to the plurality of groups of training amplified images;
And carrying out iterative training on an initial license plate image classification model according to each training license plate image sample, training sample labels corresponding to each training license plate image sample, each training amplification image and training sample labels corresponding to each training amplification image to obtain the license plate image classification model.
A license plate authenticity identification device, comprising:
the acquisition module is used for acquiring license plate images in the vehicle video acquired by the camera;
the detection module is used for carrying out target detection on the license plate image by utilizing a license plate target detection model to obtain license plate position information and a first true and false identification result;
the classification module is used for classifying and identifying the license plate images by utilizing a license plate image classification model based on the license plate position information to obtain a second true and false identification result;
and the determining module is used for determining the true and false identification result based on the first true and false identification result and the second true and false identification result.
An electronic device comprises a memory, a processor and computer readable instructions stored in the memory and capable of running on the processor, wherein the processor executes the computer readable instructions to realize the license plate authenticity identification method.
One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform a license plate authenticity identification method as described above.
According to the license plate authenticity identification method, the device, the equipment and the storage medium, the license plate images in the vehicle video acquired by the camera are acquired, the license plate images are subjected to target detection by utilizing the license plate target detection model, so that license plate position information and a first authenticity identification result are obtained, the license plate images are subjected to classification identification by utilizing the license plate image classification model based on the license plate position information, a second authenticity identification result is obtained, and then the target authenticity identification result is determined according to the first authenticity identification result and the second authenticity identification result, so that the behavior of a vehicle owner for evading payment by using fake license plates is effectively reduced, and further the field loss is effectively reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a license plate authenticity identification method provided by the invention;
FIG. 2 is a second flow chart of the license plate authenticity identification method provided by the invention;
FIG. 3 is a schematic diagram of a license plate authenticity identification device provided by the invention;
fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terminology used in the one or more embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of the invention. As used in one or more embodiments of the invention, the singular forms "a," "an," "the," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present invention refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that, although the terms first, second, etc. may be used in one or more embodiments of the invention to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first may also be referred to as a second, and similarly, a second may also be referred to as a first, without departing from the scope of one or more embodiments of the invention. The word "if" as used herein may be interpreted as "at … …" or "when … …", depending on the context.
Fig. 1 is a schematic flow chart of a license plate authenticity identification method provided by the invention. As shown in fig. 1, the license plate authenticity identification method includes:
s11, acquiring license plate images in vehicle videos acquired by a camera;
it should be noted that, the camera can be set up in parking area access & exit area, also can set up on the overpass, can also set up near the roadside barrier to gather the license plate video of in-road vehicle and off-road vehicle, and then carry out recognition analysis to the vehicle video.
It should be further noted that, when the vehicle passes the camera, the vehicle video refers to a video image collected by the camera, and the license plate image refers to one frame of image captured in the vehicle video.
Specifically, when a target vehicle drives through the camera, the camera shoots the target vehicle, and then a vehicle video is obtained, so that the vehicle video is intercepted, and a license plate image is obtained.
Step S12, carrying out target detection on the license plate image by utilizing a license plate target detection model to obtain license plate position information and a first true and false identification result;
it should be noted that, the license plate target detection model is obtained by performing iterative training for multiple times according to a plurality of training license plate image samples and training sample labels corresponding to the plurality of training license plate image samples.
It should be further noted that, the license plate position information refers to specific position information of the license plate in the license plate image, and may be represented by using a top left corner coordinate (x 1, y 1) and a bottom right corner coordinate (x 2, y 2) and the like. The first fake recognition result refers to a license plate fake recognition result output by the license plate target detection model and comprises a license plate category, a first license plate fake score and the like.
Specifically, the license plate image is input into a license plate target detection model to perform target detection, so that license plate position information and a first true and false recognition result output by the license plate target detection model are obtained.
Step S13, based on license plate position information, classifying and identifying license plate images by utilizing a license plate image classification model to obtain a second true and false identification result;
It should be noted that the license plate image classification model is obtained by performing multiple iterative training on a plurality of training license plate image samples, training sample labels corresponding to the plurality of training license plate image samples, a plurality of training amplification images and training sample labels corresponding to the plurality of training amplification images. The second authenticity identification result refers to a license plate authenticity identification result output by the license plate image classification model and comprises a license plate authenticity result, a second license plate authenticity score and the like.
Specifically, based on license plate position information, the license plate image is subjected to expansion processing to obtain at least one license plate expansion image, and then the at least one license plate expansion image is input into a license plate image classification model to be subjected to classification recognition, so that a second true and false recognition result output by the license plate image classification model is obtained.
Step S14, determining a target authenticity identification result based on the first authenticity identification result and the second authenticity identification result.
It should be noted that the target false recognition result includes a true license plate result and a plurality of fake license plate results such as a mobile phone license plate, a paper license plate and a plastic license plate.
Specifically, based on the first authenticity identification result and the second authenticity identification result, an initial authenticity identification result is determined, then an image frame number of a license plate image corresponding to the initial authenticity identification result is obtained, whether the image frame number meets the preset detection image frame number is judged, and therefore a target authenticity identification result is determined based on the judging result.
According to the embodiment of the invention, the license plate image in the vehicle video acquired by the camera is acquired, the license plate image is further subjected to target detection by utilizing the license plate target detection model to obtain the license plate position information and the first fake identification result, the license plate image is classified and identified by utilizing the license plate image classification model based on the license plate position information to obtain the second fake identification result, and the target fake identification result is determined according to the first fake identification result and the second fake identification result, so that the behavior of a vehicle owner for escaping from payment by using fake license plates is effectively reduced, the loss of a vehicle field is further effectively reduced, the parking management order is largely maintained, and the method has the advantages of low cost, strong robustness, high identification effect accuracy and the like.
In one embodiment of the present invention, based on license plate position information, classification and identification are performed on license plate images by using a license plate image classification model to obtain a second true and false identification result, including:
obtaining at least one license plate amplified image based on license plate position information; classifying and identifying at least one license plate amplified image by utilizing a license plate image classification model to obtain a license plate authenticity result corresponding to the at least one license plate amplified image and a license plate authenticity score; and determining a second authenticity identification result based on the license plate authenticity result corresponding to the at least one license plate amplified image and a license plate authenticity score and a preset scoring strategy.
The license plate expansion image refers to a license plate expansion image obtained by expanding a license plate image. The license plate authenticity result comprises two license plate recognition results of a real license plate and a fake license plate, and the license plate authenticity score refers to the license plate authenticity score output by the license plate image classification model aiming at the license plate amplified image. The preset scoring strategy comprises a plurality of strategy methods such as taking the maximum value, taking the average value, weighting the average value and the like, so as to obtain the accurate license plate true-false scoring condition to the maximum extent.
Specifically, based on license plate position information, performing amplification processing on a license plate image to obtain at least one license plate amplification image, wherein the amplification processing comprises multiple amplification processing modes such as image random rotation and outward amplification, cutting, filling color conversion, image splicing, noise addition and the like, for example, the license plate image is outward amplified by 100%, the multiple amplification processing modes can be combined for use, and the license plate amplification image can be set according to actual conditions without limitation. In an embodiment, if the license plate position information is the upper left corner coordinates (x 1, y 1) and the lower right corner coordinates (x 2, y 2) of the license plate image position, the license plate position information is taken as the center position for outward expansion, so as to obtain a license plate expansion image with outward expansion of 100%, outward expansion of 300% and outward expansion of 500%.
Further, at least one vehicle extension image is input into a license plate image classification model for classification and identification, and a license plate authenticity result and a license plate authenticity score corresponding to at least one license plate extension image are obtained, for example, a license plate authenticity result and a license plate authenticity score corresponding to 100% of the extended license plate extension image are a fake license plate and 0.4, a license plate authenticity result and a license plate authenticity score corresponding to 300% of the extended license plate extension image are a fake license plate and 0.6, and a license plate authenticity result and a license plate authenticity score corresponding to 500% of the extended license plate extension image are a fake license plate and 0.8.
Further, according to a preset scoring strategy, a second authenticity identification result is determined based on a license plate authenticity result and a license plate authenticity score corresponding to at least one license plate amplification image, and if the preset scoring strategy is the same type license plate maximum value taking strategy, under the premise that the license plate authenticity results belong to fake license plates, the second authenticity identification result is determined to be fake license plates and 0.8 in the license plate authenticity scores of 0.4, 0.6 and 0.8 respectively.
In an embodiment, the license plate authenticity result and the license plate authenticity score corresponding to 100% of the license plate extension image are the fake license plate and 0.4, the license plate authenticity result and the license plate authenticity score corresponding to 300% of the license plate extension image are the real license plate and 0.6, the license plate authenticity result and the license plate authenticity score corresponding to 500% of the license plate extension image are the fake license plate and 0.7, if the preset score policy is the average value of the same kind of license plate authenticity results, the maximum value of the multiple license plate authenticity results is obtained, the average value of the license plate authenticity score of the same kind of fake license plate is obtained, the counterfeit license plate authenticity score is 0.55, and then the 0.6 of the real license plate is compared with the 0.55 of the fake license plate, and the second license plate is determined to be the real license plate and 0.6.
According to the embodiment of the invention, the license plate image classification model is utilized to carry out classification recognition on at least one license plate amplified image, so that the license plate authenticity result and the license plate authenticity score corresponding to the at least one license plate amplified image are obtained, and further, the second authenticity recognition result is determined based on the license plate authenticity result corresponding to the at least one license plate amplified image, the license plate authenticity score and the preset score strategy, so that the license plate amplified images are classified and recognized through the license plate image classification model, the authenticity recognition result output by the license plate target detection model is further determined, the authenticity condition of the license plate is effectively recognized, the accuracy of the authenticity recognition result is improved, and the license plate authenticity misjudgment condition is effectively reduced.
In one embodiment of the present invention, determining a target authenticity identification result based on a first authenticity identification result and a second authenticity identification result includes:
determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result; acquiring an image frame number of a license plate image corresponding to the initial false identification result, and judging whether the image frame number meets the preset detection image frame number or not; if the image frame number meets the preset detection image frame number, determining a target authenticity identification result based on the initial authenticity identification result; if the image frame number does not meet the preset detection image frame number, returning to the step of acquiring license plate images in the vehicle video acquired by the camera until the image frame number meets the preset detection image frame number.
It should be noted that, the image frame number is the sequential number of the multi-frame license plate image cut out from the vehicle video, for example, if the license plate image is the second frame license plate image cut out from the vehicle video, the image frame number of the license plate image is 2. The image frame number may be set according to practical situations, such as roman numerals, arabic numerals, etc., and is not limited herein.
It should be further noted that the preset detection image frame number refers to a preset license plate image frame number which is to be intercepted in the vehicle video and used for detection, so as to perform multiple model detection on the license plate image, further perform comprehensive analysis on multiple initial authenticity identification results output by the model, and finally determine the target authenticity identification result.
Specifically, based on the first and second authenticity identification results, an initial authenticity identification result is determined, and then an image frame number of a license plate image corresponding to the initial authenticity identification result is obtained, and whether the image frame number meets a preset detection image frame number is judged, if the image frame number meets the preset detection image frame number, a target authenticity identification result is determined based on the initial authenticity identification result, for example, if the preset detection image frame number is three license plate images in a continuous detection vehicle video and the image frame number of the license plate image corresponding to the initial authenticity identification result is a third frame, the image frame number meets the preset detection image frame number is determined, and the initial authenticity identification result of the first two frames and a preset voting mechanism are combined, and the target authenticity identification result is determined, for example, if the initial authenticity identification result corresponding to the three license plate images is a mobile phone license plate, a paper license plate and a paper license plate, the target authenticity identification result is determined to be a paper license plate.
If the image frame number does not meet the preset detection image frame number, returning to the step of acquiring license plate images in the vehicle video acquired by the camera until the image frame number meets the preset detection image frame number. For example, if the preset detected image frame number is five license plate images in the vehicle video, and the image frame number of the license plate image corresponding to the initial true-false recognition result is the fourth frame, then the initial true-false recognition result is judged to not meet the preset detected image frame number, and the step of acquiring the license plate image in the vehicle video acquired by the camera is returned to be executed until the image frame number is judged to be the fifth frame.
According to the embodiment of the invention, the initial authenticity identification result is determined based on the first authenticity identification result and the second authenticity identification result, so that the image frame number of the license plate image corresponding to the initial authenticity identification result is obtained, whether the image frame number meets the preset detection image frame number is judged, and the target authenticity identification result is determined based on the judgment result, so that the accuracy of license plate authenticity identification is effectively improved, misjudgment is reduced, and the management efficiency of a parking lot is prevented from being reduced.
In one embodiment of the present invention, the first genuine-fake recognition result includes a license plate category and a first license plate genuine-fake score; the second authenticity identification result comprises a license plate authenticity result and a second license plate authenticity score; determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result, comprising:
If the first vehicle license plate authenticity score is outside the preset detection score range, comparing the second vehicle license plate authenticity score with the preset classification score range based on a license plate authenticity result; if the true-false score of the second license plate is within the preset classification score range, determining that the initial true-false recognition result is the license plate type; if the second card authenticity score is outside the preset classification score range, determining an initial authenticity identification result based on the first card authenticity score and the second card authenticity score.
The license plate category includes a plurality of license plate authenticity categories such as a real license plate, a mobile phone license plate, a paper license plate and the like, and the first license plate authenticity score refers to a license plate authenticity score output by a license plate target detection model. The second license plate true-false score refers to a license plate true-false score output by the license plate image classification model.
The preset detection score range refers to a preset threshold value for comparing with the true-false score of the first license plate output by the license plate target detection model. The preset classification score range refers to a preset threshold value which is compared with the authenticity score of the second vehicle license plate output by the license plate image classification model.
Specifically, if the first vehicle license plate authenticity score is located outside the preset detection score range, the second vehicle license plate authenticity score is compared with the preset classification score range based on the license plate authenticity result. If the true-false score of the second license plate is within the preset classification score range, determining the initial true-false recognition result as the license plate type. Additionally, if the second card authenticity score is outside the preset classification score range, determining an initial authenticity identification result based on the first card authenticity score and the second card authenticity score.
For example, if the license plate category is a mobile phone license plate, the first license plate authenticity score is 0.4, the preset detection score range is 0.5-1.0, the license plate authenticity result is a fake license plate, the second license plate authenticity score is 0.7, and the preset classification score range is 0.6-1.0, and because the first license plate authenticity score is 0.4 outside the preset detection score range, the second license plate authenticity score is 0.7 compared with the preset classification score range, the initial authenticity recognition result is determined to be the license plate category, namely, the initial authenticity recognition result is determined to be the mobile phone license plate.
Preferably, referring to fig. 2, if the first license plate authenticity score is within the preset detection score range, the initial authenticity identification result is determined to be the license plate category. For example, if the license plate type is a paper license plate, the first license plate is true and false, the preset detection score range is 0.5-1.0, and the initial true and false recognition result is a paper license plate because the true and false score of the first license plate is within the preset detection score range.
According to the embodiment of the invention, when the true and false score of the first vehicle license plate is out of the preset detection score range, the true and false score of the second vehicle license plate is compared with the preset classification score range based on the true and false result of the license plate, so that the initial true and false recognition result is accurately judged by setting a plurality of preset thresholds, and further the result judgment under a plurality of scenes is realized, thereby greatly improving the accuracy of the true and false recognition result and reducing the possibility of false judgment of the license plate.
In one embodiment of the present invention, determining an initial authenticity identification result based on the first card authenticity score and the second card authenticity score comprises:
calculating an average value of the true-false score of the first card and the true-false score of the second card to obtain a true-false average score; if the true-false average score is smaller than the preset average threshold value, determining that the initial true-false identification result is a true license plate result; if the true-false average score is larger than or equal to the preset average threshold value, determining the initial true-false identification result as the license plate category.
The true-false average score is an average value of the true-false score of the first card and the true-false score of the second card, and the preset average threshold value is a preset threshold value compared with the true-false average score.
Specifically, an average value is calculated for the true-false score of the first card and the true-false score of the second card to obtain a true-false average score, and then the true-false average score is compared with a preset average threshold value. If the true-false average score is smaller than the preset average threshold value, determining that the initial true-false identification result is a true license plate result; if the true-false average score is larger than or equal to the preset average threshold value, determining the initial true-false identification result as the license plate category.
In an embodiment, if the license plate type is a mobile phone license plate, the first license plate is true and false, the preset detection score range is 0.5-1.0, the license plate true and false result is a fake license plate, the second license plate true and false score is 0.7, the preset classification score range is 0.8-1.0, and the preset average threshold value is 0.5. And under the premise that the true-false score of the first vehicle card is outside the preset detection score range and the true-false score of the second vehicle card is outside the preset classification score range, averaging the true-false score of the first vehicle card and the true-false score of the second vehicle card to obtain a true-false average score of 0.5, namely, the true-false average score is equal to a preset average threshold value of 0.5, and determining that the initial true-false recognition result is the mobile phone license plate.
In another embodiment, if the license plate type is a paper license plate, the first license plate score is 0.2, and the preset detection score range is 0.5-1.0. The fake license plate is obtained through fake license plate, the fake score of the second license plate is 0.4, and the preset classification score range is 0.6-1.0. The preset average threshold is 0.5. Under the premise that the true-false score of the first vehicle card is out of the preset detection score range and the true-false score of the second vehicle card is out of the preset classification score range, the true-false score of the first vehicle card and the true-false score of the second vehicle card are averaged to obtain a true-false average score of 0.3, namely, the true-false average score is smaller than a preset average threshold value of 0.5, and the initial true-false recognition result is determined to be the true license plate.
According to the embodiment of the invention, the true-false average score is obtained by calculating the average value of the true-false score of the first vehicle card and the true-false score of the second vehicle card, and then the initial true-false identification result is determined based on the comparison result, so that the flow of the initial true-false identification result is further refined, and the accuracy of the true-false identification result is further effectively improved.
In one embodiment of the present invention, after determining the target authenticity identification result based on the first authenticity identification result and the second authenticity identification result, the method further includes:
if the target true-false identification result is a true license plate result, generating a release instruction, and issuing the release instruction to the channel gate to control the channel gate to open a release target vehicle; if the target authenticity identification result is a fake license plate result, an escape warning is generated, and the escape warning is pushed to a parking lot manager to remind the parking lot manager to collect the target vehicle.
The escape warning includes a result of authenticity of the target license plate, vehicle information of the target vehicle, and the like.
Specifically, if the target true-false identification result is a real license plate result, a release instruction is generated, and then the release instruction is issued to the channel gate to control the channel gate to open and release the target vehicle. Optionally, before releasing the target vehicle, a parking fee for the target vehicle is acquired, and a fee is charged to the owner of the target vehicle based on the parking fee.
In addition, if the target false recognition result is a fake license plate result such as a mobile phone license plate, an escape warning is generated, and the escape warning is pushed to a parking lot manager to remind the parking lot manager to collect the target vehicle, wherein pushing can be performed in a mode such as a short message, an APP and a parking lot display device. In addition, can also show the warning of collecting fee to the display device of parking area exit to remind the car owner to collect fee, avoid causing the condition of collecting fee to collect fee.
According to the embodiment of the invention, when the target true and false identification result is the real license plate result, a release instruction is generated and issued to the channel gate to control the channel gate to open and release the target vehicle; in addition, when the target authenticity identification result is a fake license plate result, an escape warning is generated and pushed to a parking lot manager, and then the parking lot manager is reminded of carrying out additional payment on the target vehicle, so that the income loss of the parking lot is effectively reduced, the management order of the parking lot is maintained, the vehicle owners can be effectively warned, and the escape behavior is prevented from becoming a normal state.
In one embodiment of the present invention, before acquiring the license plate image in the vehicle video acquired by the camera, the method further includes:
Acquiring a plurality of groups of training license plate image samples and training sample labels corresponding to the plurality of groups of training license plate image samples; according to each training license plate image sample and the training sample label corresponding to each training license plate image sample, carrying out iterative training on the initial license plate target detection model to obtain a license plate target detection model; amplifying each training license plate image sample to obtain a plurality of groups of training amplified images and training sample labels corresponding to the plurality of groups of training amplified images; and carrying out iterative training on the initial license plate image classification model according to each training license plate image sample, training sample labels corresponding to each training license plate image sample, each training amplification image and training sample labels corresponding to each training amplification image to obtain a license plate image classification model.
The license plate image samples are license plate images in a pre-collected vehicle video and comprise real license plate samples and fake license plate samples, wherein the real license plate samples can be obtained through cameras erected in roads and nearby outside the roads; the fake license plate sample mainly comprises two parts: one part of license plate images for escaping by using various fake license plates by a plurality of vehicle owners, and the other part of license plate images are obtained by simulating attack means possibly used by the vehicle owners, for example, recording vehicle videos by using fake license plates such as mobile phone license plates, paper license plates and license plates made of other materials, so as to obtain fake license plate samples.
Additionally, to ensure robustness of the model, multiple scenes need to be collected, including various real license plate samples and counterfeit license plate samples in different scenes such as daytime, night-time, rainy days, and foggy days. The training sample label comprises license plate marking positions, license plate categories, license plate authenticity results and the like. Optionally, the license plate marking position can be marked by using marking software LabelImg to mark 4 points on the upper, lower, left and right sides of a license plate in a license plate image to form a rectangular frame so as to mark a specific license plate position in the license plate image.
It should be further noted that, the training expansion image is an expansion image obtained after the training license plate image sample is expanded, where the expansion modes include multiple expansion modes such as image overturning, image random rotation and scaling, clipping, filling color conversion, image stitching and noise adding, for example, expanding the training license plate image sample by 100%, and the like, which is not limited herein.
Furthermore, before model training, preprocessing, such as cutting into a uniform size, processing into a uniform format, and normalizing, needs to be performed on each training license plate image sample and each training extension image.
Specifically, the model training step of the license plate target detection model is as follows: and inputting each training license plate image sample into the initial license plate target detection model to obtain a predicted value output by the initial license plate target detection model, wherein a YOLO series target detection algorithm can be used to calculate a model loss value based on the predicted value and training sample labels corresponding to each training license plate image sample by using a loss function, and in the embodiment, the loss function can be set according to actual requirements without specific limitation. After the model loss value is obtained through calculation, the training process is finished, model parameters in the initial license plate target detection model are updated by using an error back propagation algorithm and the like, and then the next training is carried out. In the training process, judging whether the updated initial license plate target detection model meets the preset training ending condition, if so, taking the updated initial license plate target detection model as a license plate target detection model, and if not, continuing training the model, wherein the preset training ending condition comprises loss convergence, reaching the maximum iteration number threshold and the like.
Further, the model training step of the license plate image classification model is as follows: amplifying each training license plate image sample to obtain a plurality of groups of training amplified images and training sample labels corresponding to the plurality of groups of training amplified images, inputting each training license plate image sample and each training amplified image into an initial license plate image classification model to obtain a predicted value output by the initial license plate image classification model, wherein a VGG image classification algorithm can be used, and further model loss values are obtained by calculating a loss function based on the predicted values, the training sample labels corresponding to each training license plate image sample and the training sample labels corresponding to each training amplified image, and in the embodiment, the loss function can be set according to actual requirements without specific limitation. After the model loss value is obtained through calculation, the training process is finished, and then the model parameters in the initial license plate image classification model are updated by using algorithms such as an error back propagation algorithm and the like, and then the next training is carried out. In the training process, judging whether the updated initial license plate image classification model meets the preset training ending condition, if so, taking the updated initial license plate image classification model as a license plate image classification model, and if not, continuing training the model, wherein the preset training ending condition comprises loss convergence, reaching the maximum iteration number threshold and the like.
According to the embodiment of the invention, the initial license plate target detection model and the initial license plate image classification model are subjected to repeated iterative training by utilizing the plurality of training license plate image samples and the plurality of training amplification images, so that the license plate target detection model and the license plate image classification model are obtained, the robustness of the model is improved, and the accuracy of the model output result is effectively improved.
In an embodiment, the main flow of the license plate authenticity identification method can refer to fig. 2.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In an embodiment, a license plate authenticity identification device is provided, and the license plate authenticity identification device corresponds to the license plate authenticity identification method in the embodiment one by one. As shown in fig. 3, the license plate authenticity identification device includes an acquisition module 31, a detection module 32, a classification module 33, and a determination module 34. The functional modules are described in detail as follows:
the acquisition module 31 is used for acquiring license plate images in the vehicle video acquired by the camera;
The detection module 32 is configured to perform target detection on the license plate image by using a license plate target detection model, so as to obtain license plate position information and a first true and false identification result;
the classification module 33 is configured to perform classification and identification on the license plate image by using the license plate image classification model based on the license plate position information, so as to obtain a second true and false identification result;
the determining module 34 is configured to determine the authenticity identification result based on the first authenticity identification result and the second authenticity identification result.
The license plate authenticity identification device is also used for:
obtaining at least one license plate amplified image based on license plate position information;
classifying and identifying at least one license plate amplified image by utilizing a license plate image classification model to obtain a license plate authenticity result corresponding to the at least one license plate amplified image and a license plate authenticity score;
and determining a second authenticity identification result based on the license plate authenticity result corresponding to the at least one license plate amplified image and a license plate authenticity score and a preset scoring strategy.
The license plate authenticity identification device is also used for:
determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result;
acquiring an image frame number of a license plate image corresponding to the initial false identification result, and judging whether the image frame number meets the preset detection image frame number or not;
If the image frame number meets the preset detection image frame number, determining a target authenticity identification result based on the initial authenticity identification result;
if the image frame number does not meet the preset detection image frame number, returning to the step of acquiring license plate images in the vehicle video acquired by the camera until the image frame number meets the preset detection image frame number.
The license plate authenticity identification device is also used for:
if the first vehicle license plate authenticity score is outside the preset detection score range, comparing the second vehicle license plate authenticity score with the preset classification score range based on a license plate authenticity result;
if the true-false score of the second license plate is within the preset classification score range, determining that the initial true-false recognition result is the license plate type;
if the second card authenticity score is outside the preset classification score range, determining an initial authenticity identification result based on the first card authenticity score and the second card authenticity score.
The license plate authenticity identification device is also used for:
calculating an average value of the true-false score of the first card and the true-false score of the second card to obtain a true-false average score;
if the true-false average score is smaller than the preset average threshold value, determining that the initial true-false identification result is a true license plate result;
If the true-false average score is larger than or equal to the preset average threshold value, determining the initial true-false identification result as the license plate category.
The license plate authenticity identification device is also used for:
if the target true-false identification result is a true license plate result, generating a release instruction, and issuing the release instruction to the channel gate to control the channel gate to open a release target vehicle;
if the target authenticity identification result is a fake license plate result, an escape warning is generated, and the escape warning is pushed to a parking lot manager to remind the parking lot manager to collect the target vehicle.
The license plate authenticity identification device is also used for:
acquiring a plurality of groups of training license plate image samples and training sample labels corresponding to the plurality of groups of training license plate image samples;
according to each training license plate image sample and the training sample label corresponding to each training license plate image sample, carrying out iterative training on the initial license plate target detection model to obtain a license plate target detection model;
amplifying each training license plate image sample to obtain a plurality of groups of training amplified images and training sample labels corresponding to the plurality of groups of training amplified images;
and carrying out iterative training on the initial license plate image classification model according to each training license plate image sample, training sample labels corresponding to each training license plate image sample, each training amplification image and training sample labels corresponding to each training amplification image to obtain a license plate image classification model.
The specific limitation of the license plate authenticity identification device can be referred to the limitation of the license plate authenticity identification method hereinabove, and the description thereof will not be repeated here. All or part of each module in the license plate authenticity identification device can be realized by software, hardware and a combination thereof. The above modules may be embedded in hardware or independent of a processor in the electronic device, or may be stored in software in a memory in the electronic device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, an electronic device is provided, which may be a server. The electronic device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic device includes a readable storage medium, an internal memory. The readable storage medium stores an operating system, computer readable instructions, and a database. The internal memory provides an environment for the execution of an operating system and computer-readable instructions in a readable storage medium. The database of the electronic equipment is used for storing data related to the license plate authenticity identification method. The network interface of the electronic device is used for communicating with an external terminal through a network connection. The computer readable instructions when executed by the processor implement a license plate authenticity identification method. The readable storage medium provided by the present embodiment includes a nonvolatile readable storage medium and a volatile readable storage medium.
In one embodiment, an electronic device is provided, which may be a terminal device, and an internal structure diagram thereof may be as shown in fig. 4. The electronic device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic device includes a readable storage medium. The readable storage medium stores computer readable instructions. The network interface of the electronic device is used for communicating with an external terminal through a network connection. The computer readable instructions when executed by the processor implement a license plate authenticity identification method. The readable storage medium provided by the present embodiment includes a nonvolatile readable storage medium and a volatile readable storage medium.
In one embodiment, an electronic device is provided that includes a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, when executing the computer readable instructions, implementing the steps of the license plate authenticity identification method as described above.
In one embodiment, a readable storage medium is provided, the readable storage medium storing computer readable instructions that when executed by a processor implement the license plate authenticity identification method steps described above. Those skilled in the art will appreciate that implementing all or part of the above described embodiment methods may be accomplished by instructing the associated hardware by computer readable instructions, which may be stored on a non-volatile readable storage medium or a volatile readable storage medium, which when executed may comprise the above described embodiment methods. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.

Claims (10)

1. The license plate authenticity identification method is characterized by comprising the following steps of:
acquiring license plate images in vehicle videos acquired by a camera;
performing target detection on the license plate image by using a license plate target detection model to obtain license plate position information and a first true and false identification result;
Based on the license plate position information, classifying and identifying the license plate image by utilizing a license plate image classification model to obtain a second true and false identification result;
and determining a target authenticity identification result based on the first authenticity identification result and the second authenticity identification result.
2. The license plate authenticity identification method according to claim 1, wherein the step of classifying and identifying the license plate image based on the license plate position information by using a license plate image classification model to obtain a second license plate authenticity identification result comprises the steps of:
obtaining at least one license plate amplified image based on the license plate position information;
classifying and identifying the at least one license plate amplified image by using the license plate image classification model to obtain a license plate authenticity result and a license plate authenticity score corresponding to the at least one license plate amplified image;
and determining the second authenticity identification result based on the license plate authenticity result corresponding to the at least one license plate amplification image, the license plate authenticity score and a preset scoring strategy.
3. The license plate authenticity identification method according to claim 1, wherein the determining a target authenticity identification result based on the first authenticity identification result and the second authenticity identification result comprises:
Determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result;
acquiring an image frame number of a license plate image corresponding to the initial false identification result, and judging whether the image frame number meets the preset detection image frame number or not;
if the image frame number meets the preset detection image frame number, determining the target authenticity identification result based on the initial authenticity identification result;
and if the image frame number does not meet the preset detection image frame number, returning to the step of acquiring the license plate image in the vehicle video acquired by the camera until the image frame number meets the preset detection image frame number.
4. The license plate authenticity identification method according to claim 3, wherein the first authenticity identification result comprises a license plate category and a first license plate authenticity score; the second authenticity identification result comprises a license plate authenticity result and a second license plate authenticity score;
the determining an initial authenticity identification result based on the first authenticity identification result and the second authenticity identification result includes:
if the first license plate authenticity score is outside a preset detection score range, comparing the second license plate authenticity score with a preset classification score range based on the license plate authenticity result;
If the second license plate authenticity score is within the preset classification score range, determining that the initial authenticity identification result is the license plate type;
and if the second card authenticity score is outside the preset classification score range, determining the initial authenticity identification result based on the first card authenticity score and the second card authenticity score.
5. The license plate authenticity identification method according to claim 4, wherein the determining the initial license plate authenticity identification result based on the first license plate authenticity score and the second license plate authenticity score comprises:
calculating an average value of the true-false scores of the first cards and the true-false scores of the second cards to obtain true-false average scores;
if the true-false average score is smaller than the preset average threshold value, determining that the initial true-false recognition result is a true license plate result;
and if the true-false average score is larger than or equal to the preset average threshold value, determining the initial true-false identification result as the license plate category.
6. The license plate authentication method according to claim 1, wherein after determining a target authentication result based on the first authentication result and the second authentication result, further comprising:
If the target true-false recognition result is a real license plate result, generating a release instruction, and issuing the release instruction to a channel gate to control the channel gate to open a release target vehicle;
and if the target authenticity identification result is a fake license plate result, generating an escape and payment warning, and pushing the escape and payment warning to a parking lot manager to remind the parking lot manager to collect the target vehicle.
7. The license plate authenticity identification method according to claim 1, wherein before the license plate image in the vehicle video acquired by the camera is acquired, the method further comprises:
acquiring a plurality of groups of training license plate image samples and training sample labels corresponding to the plurality of groups of training license plate image samples;
performing iterative training on an initial license plate target detection model according to each training license plate image sample and a training sample label corresponding to each training license plate image sample to obtain the license plate target detection model;
amplifying each training license plate image sample to obtain a plurality of groups of training amplified images and training sample labels corresponding to the plurality of groups of training amplified images;
and carrying out iterative training on an initial license plate image classification model according to each training license plate image sample, training sample labels corresponding to each training license plate image sample, each training amplification image and training sample labels corresponding to each training amplification image to obtain the license plate image classification model.
8. A license plate authenticity identification device, comprising:
the acquisition module is used for acquiring license plate images in the vehicle video acquired by the camera;
the detection module is used for carrying out target detection on the license plate image by utilizing a license plate target detection model to obtain license plate position information and a first true and false identification result;
the classification module is used for classifying and identifying the license plate images by utilizing a license plate image classification model based on the license plate position information to obtain a second true and false identification result;
and the determining module is used for determining the true and false identification result based on the first true and false identification result and the second true and false identification result.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the license plate authenticity identification method according to any one of claims 1 to 7 when executing the program.
10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the license plate authenticity identification method according to any one of claims 1 to 7.
CN202311308895.0A 2023-10-10 2023-10-10 License plate authenticity identification method, device, equipment and storage medium Pending CN117115801A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311308895.0A CN117115801A (en) 2023-10-10 2023-10-10 License plate authenticity identification method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311308895.0A CN117115801A (en) 2023-10-10 2023-10-10 License plate authenticity identification method, device, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN117115801A true CN117115801A (en) 2023-11-24

Family

ID=88804149

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311308895.0A Pending CN117115801A (en) 2023-10-10 2023-10-10 License plate authenticity identification method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117115801A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117456481A (en) * 2023-12-22 2024-01-26 华南师范大学 Anti-fake license plate recognition method, system and terminal

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117456481A (en) * 2023-12-22 2024-01-26 华南师范大学 Anti-fake license plate recognition method, system and terminal
CN117456481B (en) * 2023-12-22 2024-05-14 华南师范大学 Anti-fake license plate recognition method, system and terminal

Similar Documents

Publication Publication Date Title
CN106600977B (en) Multi-feature recognition-based illegal parking detection method and system
US11455805B2 (en) Method and apparatus for detecting parking space usage condition, electronic device, and storage medium
CN111563494B (en) Behavior identification method and device based on target detection and computer equipment
CN110533923B (en) Parking management method and device, computer equipment and storage medium
CN103258213B (en) A kind of for the dynamic vehicle model recognizing method in intelligent transportation system
CN109190488B (en) Front vehicle door opening detection method and device based on deep learning YOLOv3 algorithm
CN110929589B (en) Method, apparatus, computer apparatus and storage medium for identifying vehicle characteristics
CN110706261A (en) Vehicle violation detection method and device, computer equipment and storage medium
CN109241938B (en) Road congestion detection method and terminal
CN113033604A (en) Vehicle detection method, system and storage medium based on SF-YOLOv4 network model
KR102089298B1 (en) System and method for recognizing multinational license plate through generalized character sequence detection
CN117115801A (en) License plate authenticity identification method, device, equipment and storage medium
CN111079751B (en) Method and device for identifying authenticity of license plate, computer equipment and storage medium
CN111753592B (en) Traffic sign recognition method, device, computer equipment and storage medium
CN111444798A (en) Method and device for identifying driving behavior of electric bicycle and computer equipment
CN110473426A (en) Parking management method, device, computer equipment and storage medium
CN115691148A (en) Intelligent charging auxiliary method, equipment and medium based on expressway
CN111091041A (en) Vehicle law violation judging method and device, computer equipment and storage medium
CN114219829A (en) Vehicle tracking method, computer equipment and storage device
CN111178359A (en) License plate number recognition method, device and equipment and computer storage medium
CN113192217B (en) Fee evasion detection method, fee evasion detection device, computer equipment and medium
de Almeida et al. Vehicle occurrence-based parking space detection
CN113283303A (en) License plate recognition method and device
CN117576674A (en) License plate recognition method, device, equipment and medium
CN116543189B (en) Target detection method, device, equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination