CN113569874A - License plate number re-identification method and device, computer equipment and storage medium - Google Patents
License plate number re-identification method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113569874A CN113569874A CN202111011764.7A CN202111011764A CN113569874A CN 113569874 A CN113569874 A CN 113569874A CN 202111011764 A CN202111011764 A CN 202111011764A CN 113569874 A CN113569874 A CN 113569874A
- Authority
- CN
- China
- Prior art keywords
- license plate
- image
- plate image
- database
- vehicle
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Databases & Information Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a license plate number re-identification method, a license plate number re-identification device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring a first license plate image of an approaching vehicle, and storing the first license plate image to a first database; acquiring a second license plate image of the vehicle on the scene; determining whether a target license plate image corresponding to the second license plate image exists in the first database in a characteristic extraction mode; and if the target license plate image exists in the first database, the vehicles leaving the parking lot are released. The invention utilizes computer vision to extract the characteristics of the license plate, realizes the matching of the license plate based on the re-recognition algorithm and can reduce a large amount of data labels. Compared with a mainstream character matching algorithm, the license plate heavy identification scheme is lighter, and the localized optimization can be rapidly carried out on a newly added service scene.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a license plate number re-recognition method, a license plate number re-recognition device, computer equipment and a storage medium.
Background
Most parking lots can automatically register license plates of vehicles entering and leaving, and especially for vehicles leaving, the vehicles can be released only on the basis of confirming that the vehicles have entering records, so that illegal persons can be prevented from stealing the vehicles from the parking lots and driving out, and the benefits of vehicle owners are guaranteed. At present, a license plate number of a vehicle is usually obtained by adopting a character recognition mode during license plate registration, and the specific steps comprise vehicle image acquisition, image denoising, character segmentation, character recognition and the like. The character recognition can comprise directly carrying out license plate character recognition on the image and carrying out recognition through a character-level character recognition model. On one hand, for the recognition of the characters of the license plate, because the license plate has a single-row license plate and a double-row license plate, the complexity of the site environment of a construction site is particularly considered, the recognition of the character level of the license plate is time-consuming and labor-consuming, and the accuracy is low; on the other hand, for training a character-level character recognition model, a large amount of license plate data needs to be marked in the training process of the character recognition module, a large amount of manpower and material resources are consumed, the training cost is high, and the recognition effect is not ideal.
Disclosure of Invention
The invention aims to provide an accurate and efficient non-character-level license plate matching scheme to solve the problems in the prior art.
In order to achieve the aim, the invention provides a license plate number re-identification method, which comprises the following steps:
acquiring a first license plate image of an approaching vehicle, and storing the first license plate image to a first database;
acquiring a second license plate image of the vehicle on the scene;
determining whether a target license plate image corresponding to the second license plate image exists in the first database in a characteristic extraction mode;
and if the target license plate image exists in the first database, the vehicles leaving the parking lot are released.
According to the license plate number weight recognition method provided by the invention, the step of acquiring the first license plate image of the approaching vehicle and storing the first license plate image in the first database comprises the following steps:
acquiring a first head image of the incoming vehicle;
respectively rotating the first head image according to a plurality of preset angles to obtain a plurality of first rotating images;
respectively extracting corresponding first license plate images from each first rotating image;
and adding a first classification identifier to all first license plate images corresponding to the same first head image and storing the first classification identifiers in the first database, wherein the first classification identifiers correspond to the incoming vehicles one to one.
According to the license plate number re-identification method provided by the invention, the step of acquiring the second license plate image of the vehicle coming out of the field comprises the following steps:
acquiring a second head image of the factory vehicle;
rotating the second head image according to a plurality of preset angles respectively to obtain a plurality of second rotated images;
and respectively extracting corresponding second license plate images from each second rotation image.
According to the license plate number re-identification method provided by the invention, the step of determining whether the target license plate image corresponding to the second license plate image exists in the first database comprises the following steps:
respectively calculating single similarity between each second license plate image and each first license plate image with the same first classification mark;
calculating a composite similarity between the second license plate image and the first license plate image based on all the single similarities;
and taking the first license plate image with the highest comprehensive similarity and exceeding a preset threshold value as the target license plate object.
According to the license plate number re-identification method provided by the invention, the single similarity calculation step comprises the following steps:
respectively extracting a first feature in the first license plate image and a second feature in the second license plate image by using a feature extraction model;
and calculating the Euclidean distance between the first feature and the second feature, and converting the Euclidean distance into the single similarity.
According to the license plate number re-identification method provided by the invention, the step of calculating the comprehensive similarity between the second license plate image and the first license plate image based on all the single similarities comprises the following steps:
and weighting and summing all the single similarities to obtain the comprehensive similarity.
According to the license plate number re-identification method provided by the invention, the training step of the feature extraction model comprises the following steps:
obtaining a plurality of sample license plate images, wherein each sample license plate image is labeled with a corresponding original feature code in advance;
taking the sample license plate image as input data, taking sample characteristics extracted from the sample license plate image as output data to train a neural network model, and enabling sample characteristic codes obtained after the sample characteristics are coded to approach the original characteristic codes;
and taking the trained neural network model as the feature extraction model.
In order to achieve the above object, the present invention further provides a license plate re-recognition apparatus, including:
the system comprises an approaching image acquisition module, a first database and a second database, wherein the approaching image acquisition module is suitable for acquiring a first license plate image of an approaching vehicle and storing the first license plate image to the first database;
the departure image acquisition module is suitable for acquiring a second license plate image of a departure vehicle;
the target determining module is used for determining whether a target license plate image corresponding to the second license plate image exists in the first database in a characteristic extraction mode;
and the releasing module is suitable for releasing the vehicles on the spot if the target license plate image exists in the first database.
To achieve the above object, the present invention further provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when executing the computer program.
To achieve the above object, the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the above method.
Compared with the prior art, the invention has the following beneficial effects:
(1) the feature extraction is carried out on the license plate by using the computer vision, the matching of the license plate is realized based on the re-recognition algorithm, and a large amount of data marking can be reduced.
(2) Due to the fact that the data label of the re-identification algorithm is low in obtaining cost and the model is convenient to optimize, compared with a mainstream matching algorithm, the license plate matching method is lighter in weight, and the new service scene can be quickly optimized locally.
Drawings
FIG. 1 is a flowchart of a first embodiment of a license plate number re-identification method of the present invention;
FIG. 2 is a schematic flow chart of a first embodiment of the present invention for obtaining an image of a first license plate;
FIG. 3 is a schematic diagram illustrating a rotation of a vehicle head image according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart of a first embodiment of the present invention for obtaining a second license plate image;
FIG. 5 is a schematic flow chart of a first embodiment of the present invention for determining a license plate image of a target;
FIG. 6 is a schematic flowchart of comparing based on feature vectors according to an embodiment of the present invention;
FIG. 7 is a schematic flow chart of training a feature extraction model according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a program module of a first embodiment of the license plate number re-identification apparatus of the present invention;
fig. 9 is a schematic hardware structure diagram of a license plate number re-identification device according to a first embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Referring to fig. 1, the present embodiment provides a method for recognizing a license plate number, which includes the following steps:
s100, acquiring a first license plate image of an approaching vehicle, and storing the first license plate image to a first database.
The method is used for acquiring the license plate number information of all vehicles entering the field through the image. A first head image of an approaching vehicle can be shot through an image acquisition device arranged at a first fixed position, and a first license plate image is cut out of the first head image. The position of the license plate number can be determined from the first head image through the existing arbitrary image segmentation algorithm and cut. For example, an infrared distance meter can be installed near the image acquisition equipment, and when the distance meter measures that the incoming vehicles are at a preset distance, the image acquisition equipment is started to shoot a first locomotive image, so that the proportion of locomotives in the shot first locomotive images of all the incoming vehicles is approximately equal. Further, the display range of the first license plate image in the first license plate image can be calculated according to the installation position of the license plate on the vehicle head, and therefore the first license plate image can be obtained by cutting according to the display range. In addition, a license plate extraction model trained by a machine can be used for directly outputting the cut first license plate image according to the input first head image. In short, any method capable of extracting the first license plate image including the license plate number from the vehicle head image in the prior art is within the protection scope of the embodiment.
The step stores the extracted first license plate image in a first database for searching and comparing when the vehicle leaves. For example, a category label, such as a truck type, a car type, a bus type, etc., may be further added to the first license plate image stored in the first database, thereby further narrowing the search range.
And S200, acquiring a second license plate image of the vehicle on the scene.
The method is used for acquiring the license plate number information of all vehicles on the spot through the image. A second head image of the field vehicle can be captured by an image capture device disposed at a second fixed location and a second license plate image can be cut from the second head image. The position of the second license plate number can be determined from the second head image through any existing image segmentation algorithm and cut. For example, an infrared distance meter can be installed near the image acquisition device, and when the distance meter measures that the outgoing vehicle is at a preset distance, the image acquisition device is started to shoot a second vehicle head image, so that the proportion of the vehicle heads in the shot vehicle head images of all outgoing vehicles is approximately equal. Further, the display range of the second license plate image in the second license plate image can be calculated according to the installation position of the license plate on the vehicle head, and therefore cutting is conducted according to the display range to obtain the second license plate image. In addition, the license plate extraction model can be trained through a machine, and the license plate extraction model directly outputs the cut second license plate image according to the input second head image. In short, any method capable of extracting the first license plate image including the license plate number from the second head image in the prior art is within the protection scope of the present embodiment.
And S300, determining whether a target license plate image corresponding to the second license plate image exists in the first database in a characteristic extraction mode.
The manner of feature extraction may include using a machine-trained feature extraction model. For example, a second feature related to the second license plate number is extracted from the second license plate image through the feature extraction model, a first feature related to the first license plate number is extracted from the first license plate image through the feature extraction model, and a target first feature identical to the second feature of the license plate number of the vehicle to be discharged is determined by comparing the similarity of the first feature and the second feature. It can be understood that the target license plate image corresponding to the first feature of the target is an image captured when the vehicle to be departed enters the parking lot.
S400, if the target license plate image exists in the first database, the vehicles leaving the parking lot are released.
The presence of the target license plate image in the first database means that the vehicle to be departed has an entrance record, in which case the vehicle is determined to be normally entering and exiting, and the vehicle is allowed to drive away. The query result interface of the first database can be electrically connected with the controller of the gate or the railing, and once the target license plate image is queried from the first database, the controller is activated to open the gate or the railing, so that automatic passing is realized.
FIG. 2 illustrates a schematic flow chart of an embodiment of the present invention for obtaining a first license plate image. As shown in fig. 2, step S100 includes:
and S110, acquiring a first head image of the incoming vehicle.
And S120, rotating the first head image according to a plurality of preset angles respectively to obtain a plurality of first rotating images. In order to obtain more reference images, the method increases the consideration of actual conditions such as uneven road, vehicle body swing and the like, and improves the subsequent comparison accuracy. Fig. 3 is a schematic diagram illustrating a rotation of a vehicle head image according to an embodiment of the present invention. As shown in fig. 3, the first image is a first head image originally shot, and is equivalent to a first rotation image obtained by rotating at a preset angle of 0 degrees; the second image is a first rotating image obtained by rotating at a preset angle of-15 degrees; the second image is a first rotated image obtained by rotating the second image by a preset angle of 15 degrees. Those skilled in the art understand that the preset angle may be adjusted according to actual situations, and the embodiment does not limit this.
And S130, respectively extracting corresponding first license plate images from each first rotating image. The step of extracting the first license plate image is described in step S100, and is not described herein again.
And S140, adding first classification marks to all first license plate images corresponding to the same first head image and storing the first classification marks in the first database, wherein the first classification marks correspond to the incoming vehicles one by one.
The first classification mark in this step can be used for identifying the entering and exiting vehicles. For example, a unique radio frequency identification tag is installed on a vehicle body for an internal vehicle, and when the vehicle enters the field, the radio frequency identification device scans the identification tag on the vehicle body to confirm that the vehicle is an internal self-owned vehicle.
FIG. 4 shows a schematic flow diagram of a method of obtaining a second card image according to an embodiment of the invention. As shown in fig. 4, step S200 includes:
and S210, acquiring a second head image of the factory vehicle.
And S220, rotating the second head image according to a plurality of preset angles respectively to obtain a plurality of second rotated images.
In order to obtain more reference images, the method increases the consideration of actual conditions such as uneven road, vehicle body swing and the like, and improves the subsequent comparison accuracy. Those skilled in the art understand that the preset angle may be adjusted according to actual situations, and the embodiment does not limit this.
And S230, respectively extracting corresponding second license plate images from each second rotation image. The step of extracting the first license plate image is described in step S100, and is not described herein again.
FIG. 5 is a schematic flow chart illustrating a process for determining a license plate image of a target vehicle according to an embodiment of the invention. As shown in fig. 5, step S300 includes:
and S310, respectively calculating single similarity between each second license plate image and each first license plate image with the same first classification mark.
As described above, after the rotation by the preset angle, a plurality of second license plate images corresponding to the same outgoing vehicle can be obtained. The second license plate images have the same license plate number but different image angles. Similarly, a plurality of first license plate images corresponding to the same incoming vehicle may also be obtained.
In one example, assume that the first license plate image includes FA1、FA2And FA3The second license plate image includes FB1、FB2And FB3. Through the feature extraction model, the feature vector T can be extracted from the first license plate image respectivelyA1、TA2、TA3And extracting a feature vector T from the second license plate imageB1、TB2、TB3. In this embodiment, calculating the similarity between each second license plate image and each first license plate image may be implemented by calculating the similarity of the corresponding feature vectors.
FIG. 6 shows a schematic flowchart of comparing based on feature vectors according to an embodiment of the present invention. By feature vector TB1、TB2、TB3And a feature vector TA1、TA2、TA3Respectively, a plurality of single similarities, such as T, can be obtainedB1And TA1Comparing to obtain a single similarity S11，TB1And TA2Comparing to obtain a single similarity S12，TB1And TA3Comparing to obtain a single similarity S13… … and so on, a single similarity S can be obtained21、S22、S23、S31、S32、S33. Each single similarity can be calculated by the Euclidean distance between two feature vectors, and the Euclidean distance is converted into a percentage value as the single similarity after being subjected to corresponding normalization processing. It can be understood that the smaller the euclidean distance, the higher the single degree of similarity.
And S320, calculating the comprehensive similarity between the second license plate image and the first license plate image based on all the single similarities.
A weighted average of all the single similarities may be taken as the overall similarity. The weight of each single similarity can be set according to the actual road condition, for example, the actual road condition is relatively flat, and the weight of the single similarity between license plate images with a preset angle of 0 degree is set to be the maximum; setting the weight of single similarity between license plate images with a preset angle of-15 degrees as the maximum when the actual road condition sinks to the left; and setting the weight of the single similarity between the license plate images with the preset angle of 15 degrees to be maximum when the actual road condition sinks to the right, and the like. By calculating the comprehensive similarity, the relationship between the first license plate image and the second license plate image can be reflected more comprehensively and accurately, so that the license plate re-identification efficiency is improved.
And S330, taking the first license plate image with the highest comprehensive similarity and exceeding a preset threshold as the target license plate object.
The feature extraction model in this embodiment may be formed by machine learning training using an existing artificial neural network. FIG. 7 is a schematic flow chart diagram of training a feature extraction model according to an embodiment of the present invention, including the following steps:
s710, obtaining a plurality of sample license plate images, wherein each sample license plate image is labeled with a corresponding original feature code in advance. The original feature codes can be codes of standard feature vectors extracted according to known license plate number numbers in sample license plate images. The standard feature vector may be calculated in advance based on the color, brightness, and other features.
And S720, taking the sample license plate image as input data, taking the sample characteristics extracted from the sample license plate image as output data to train a neural network model, so that the sample characteristic codes obtained after the sample characteristics are coded approach to the original characteristic codes. The neural network model in the present embodiment includes, but is not limited to, a convolutional neural network, a residual error shrinkage network, a multi-layer perceptron, a boltzmann machine, and other network models.
And S730, taking the trained neural network model as the feature extraction model.
Through the steps, the feature extraction model of the embodiment can accurately extract the feature vector related to the license plate number from the picture of the vehicle head, so that effective guarantee is provided for license plate re-identification.
With continued reference to fig. 8, a license plate number re-identification apparatus is shown, in this embodiment, the license plate number re-identification apparatus 80 may include or be divided into one or more program modules, and the one or more program modules are stored in a storage medium and executed by one or more processors to implement the license plate number re-identification method. The program module referred to in the present invention means a series of computer program instruction segments capable of performing a specific function, and is more suitable than the program itself for describing the execution process of the license plate number re-recognition apparatus 80 in the storage medium. The following description will specifically describe the functions of the program modules of the present embodiment:
the approach image acquisition module 81 is adapted to acquire a first license plate image of an approach vehicle and store the first license plate image in a first database;
a departure image acquisition module 82 adapted to acquire a second license plate image of a departure vehicle;
a target determination module 83 adapted to determine whether a target license plate image corresponding to the second license plate image exists in the first database;
a passing module 84 adapted to pass the outgoing vehicle if the target license plate image exists in the first database.
The license plate number re-recognition device utilizes computer vision to extract the features of the license plate, realizes the matching of the license plate based on the re-recognition algorithm, and can reduce a large amount of data labels. Because the data label of the re-identification algorithm is low in obtaining cost and the model is convenient to optimize, the license plate number re-identification device provided by the embodiment is lighter than a mainstream matching algorithm, and can quickly perform localized optimization on a newly added service scene.
The embodiment also provides a computer device, such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server or a rack server (including an independent server or a server cluster composed of a plurality of servers) capable of executing programs, and the like. The computer device 90 of the present embodiment includes at least, but is not limited to: a memory 91 and a processor 92 communicatively connected to each other via a system bus, as shown in fig. 9. It should be noted that fig. 9 only shows a computer device 90 having components 91-92, but it should be understood that not all of the shown components are required to be implemented, and that more or fewer components may be implemented instead.
In this embodiment, the memory 91 (i.e., a readable storage medium) includes a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the storage 91 may be an internal storage unit of the computer device 90, such as a hard disk or a memory of the computer device 90. In other embodiments, the memory 91 may be an external storage device of the computer device 90, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like, provided on the computer device 90. Of course, the memory 91 may also include both internal and external memory units of the computer device 90. In this embodiment, the memory 91 is generally used for storing an operating system and various application software installed on the computer device 90, such as the program code of the license plate number re-identification apparatus 80 in the first embodiment. Further, the memory 91 can also be used to temporarily store various types of data that have been output or are to be output.
Processor 92 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 92 is typically used to control the overall operation of the computer device 90. In this embodiment, the processor 92 is configured to operate the program codes stored in the memory 91 or process data, for example, operate the license plate number re-identification device 80, so as to implement the license plate number re-identification method of the first embodiment.
The present embodiment also provides a computer-readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application mall, etc., on which a computer program is stored, which when executed by a processor implements corresponding functions. The computer-readable storage medium of this embodiment is used for storing a license plate number re-identification device 80, and when being executed by a processor, the license plate number re-identification method of the first embodiment is implemented.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example" or "some examples" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A method for re-identifying license plate numbers is characterized by comprising the following steps:
acquiring a first license plate image of an approaching vehicle, and storing the first license plate image to a first database;
acquiring a second license plate image of the vehicle on the scene;
determining whether a target license plate image corresponding to the second license plate image exists in the first database in a characteristic extraction mode;
and if the target license plate image exists in the first database, the vehicles leaving the parking lot are released.
2. The license plate number re-identification method according to claim 1, wherein the step of acquiring a first license plate image of an approaching vehicle and storing the first license plate image to a first database comprises:
acquiring a first head image of the incoming vehicle;
respectively rotating the first head image according to a plurality of preset angles to obtain a plurality of first rotating images;
respectively extracting corresponding first license plate images from each first rotating image;
and adding a first classification identifier to all first license plate images corresponding to the same first head image and storing the first classification identifiers in the first database, wherein the first classification identifiers correspond to the incoming vehicles one to one.
3. The license plate number re-recognition method as claimed in claim 2, wherein the step of obtaining a second license plate image of the departing vehicle comprises:
acquiring a second head image of the factory vehicle;
rotating the second head image according to a plurality of preset angles respectively to obtain a plurality of second rotated images;
and respectively extracting corresponding second license plate images from each second rotation image.
4. The license plate number re-recognition method of claim 3, wherein the step of determining whether a target license plate image corresponding to the second license plate image exists in the first database comprises:
respectively calculating single similarity between each second license plate image and each first license plate image with the same first classification mark;
calculating a composite similarity between the second license plate image and the first license plate image based on all the single similarities;
and taking the first license plate image with the highest comprehensive similarity and exceeding a preset threshold value as the target license plate object.
5. The method of claim 4, wherein the step of calculating the single similarity comprises:
respectively extracting a first feature in the first license plate image and a second feature in the second license plate image by using a feature extraction model;
and calculating the Euclidean distance between the first feature and the second feature, and converting the Euclidean distance into the single similarity.
6. The license plate number re-recognition method of claim 4, wherein the step of calculating a comprehensive similarity between the second license plate image and the first license plate image based on all the single similarities comprises:
and weighting and summing all the single similarities to obtain the comprehensive similarity.
7. The method of claim 5, wherein the training step of the feature extraction model comprises:
obtaining a plurality of sample license plate images, wherein each sample license plate image is labeled with a corresponding original feature code in advance;
taking the sample license plate image as input data, taking sample characteristics extracted from the sample license plate image as output data to train a neural network model, and enabling sample characteristic codes obtained after the sample characteristics are coded to approach the original characteristic codes;
and taking the trained neural network model as the feature extraction model.
8. A license plate re-recognition device, comprising:
the system comprises an approaching image acquisition module, a first database and a second database, wherein the approaching image acquisition module is suitable for acquiring a first license plate image of an approaching vehicle and storing the first license plate image to the first database;
the departure image acquisition module is suitable for acquiring a second license plate image of a departure vehicle;
the target determining module is used for determining whether a target license plate image corresponding to the second license plate image exists in the first database in a characteristic extraction mode;
and the releasing module is suitable for releasing the vehicles on the spot if the target license plate image exists in the first database.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented by the processor when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111011764.7A CN113569874A (en) | 2021-08-31 | 2021-08-31 | License plate number re-identification method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111011764.7A CN113569874A (en) | 2021-08-31 | 2021-08-31 | License plate number re-identification method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113569874A true CN113569874A (en) | 2021-10-29 |
Family
ID=78173216
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111011764.7A Pending CN113569874A (en) | 2021-08-31 | 2021-08-31 | License plate number re-identification method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113569874A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023178930A1 (en) * | 2022-03-23 | 2023-09-28 | 北京京东乾石科技有限公司 | Image recognition method and apparatus, training method and apparatus, system, and storage medium |
-
2021
- 2021-08-31 CN CN202111011764.7A patent/CN113569874A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023178930A1 (en) * | 2022-03-23 | 2023-09-28 | 北京京东乾石科技有限公司 | Image recognition method and apparatus, training method and apparatus, system, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9082038B2 (en) | Dram c adjustment of automatic license plate recognition processing based on vehicle class information | |
| Marzougui et al. | A lane tracking method based on progressive probabilistic Hough transform | |
| WO2020042489A1 (en) | Authentication method and apparatus for illegal parking case, and computer device | |
| CN108073928B (en) | License plate recognition method and device | |
| CN109886222B (en) | Face recognition method, neural network training method, device and electronic equipment | |
| US9400936B2 (en) | Methods and systems for vehicle tag number recognition | |
| US9792301B2 (en) | Multi-query privacy-preserving parking management system and method | |
| JP2015079497A (en) | Delayed vehicle identification for privacy protection | |
| CN111369801B (en) | Vehicle identification method, device, equipment and storage medium | |
| Salma et al. | Development of ANPR framework for Pakistani vehicle number plates using object detection and OCR | |
| US20210089790A1 (en) | Bidirectional statistical consolidation of multi-modal grouped observations | |
| CN113055823A (en) | Method and device for sharing bicycle based on roadside parking management | |
| CN115810134A (en) | Image acquisition quality inspection method, system and device for preventing car insurance from cheating | |
| CN112861567B (en) | Vehicle type classification method and device | |
| KR20220029929A (en) | Vehicle management assistant method and vehicle management assistant apparatus | |
| CN113569874A (en) | License plate number re-identification method and device, computer equipment and storage medium | |
| US20210192264A1 (en) | Vehicle damage detection method based on image analysis, electronic device and storage medium | |
| Al Nasim et al. | An automated approach for the recognition of bengali license plates | |
| CN113221894A (en) | License plate number identification method and device of vehicle, electronic equipment and storage medium | |
| Ng et al. | Detection and recognition of malaysian special license plate based on sift features | |
| Rahman et al. | A real-time end-to-end Bangladeshi license plate detection and recognition system for all situations including challenging environmental scenarios | |
| US20230342937A1 (en) | Vehicle image analysis | |
| CN110689481A (en) | Vehicle type identification method and device | |
| CN112906495B (en) | Target detection method and device, electronic equipment and storage medium | |
| CN111126336B (en) | Sample collection method, device and equipment |
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 |