CN110659638A - License plate recognition method and device, computer equipment and storage medium - Google Patents
License plate recognition method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110659638A CN110659638A CN201910904848.XA CN201910904848A CN110659638A CN 110659638 A CN110659638 A CN 110659638A CN 201910904848 A CN201910904848 A CN 201910904848A CN 110659638 A CN110659638 A CN 110659638A
- Authority
- CN
- China
- Prior art keywords
- license plate
- view
- image sample
- information
- target 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.)
- Granted
Links
Images
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
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a license plate recognition method and device, computer equipment and a storage medium, relates to the technical field of video monitoring, and aims to improve the license plate recognition precision. The main technical scheme of the invention is as follows: acquiring license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate; performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle; and inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
Description
Technical Field
The invention relates to the technical field of video monitoring, in particular to a license plate recognition method and device, computer equipment and a storage medium.
Background
With the development of economy in China, the holding quantity of vehicle markets is continuously increased, the problems of difficult parking, effective management of parking spaces and the like are further aggravated, and on-road parking is carried out at the same time under the historical background. The parking in the road is as the important link of intelligent transportation, indicates in both sides or one side of road safety red line within range, sets up a plurality of parking stall and supplies the incoming and outgoing vehicle to park temporarily, and this kind of parking stall has and sets up more nimble, the parking stall turnover rate is fast, occupation space is few, maintenance cost low grade advantage, can alleviate current "parking difficult" scheduling problem to a certain extent.
At present, after a vehicle stops at a parking position, if the license plate of the vehicle can be captured in the process of periodically capturing the vehicle by front-end camera equipment, the license plate of the vehicle is identified from a captured picture by a picture identification technology. However, in a parking scene, the resolution of the license plate of a vehicle parked in a near parking space is large enough to be accurately identified, while the resolution of the license plate of a vehicle parked in a far parking space is small, and the length and the width of the vehicle are only dozens of pixels, so that the precision of license plate identification is greatly reduced by comparing the license plate number.
Disclosure of Invention
The invention provides a license plate recognition method and device, computer equipment and a storage medium, which are used for improving the recognition precision of a license plate.
The embodiment of the invention provides a license plate identification method, which comprises the following steps:
acquiring license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate;
performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle;
and inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
The embodiment of the invention provides a license plate recognition device, which comprises:
the system comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring license plate information of a target vehicle through camera equipment, and the license plate information comprises four vertexes of a license plate;
the projection transformation module is used for performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle;
and the recognition module is used for inputting the license plate information of the positive visual angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, and the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the above-mentioned license plate recognition method when executing the computer program.
A computer-readable storage medium, in which a computer program is stored, which, when executed by a processor, implements the above-mentioned license plate recognition method.
According to the license plate recognition method and device, the computer equipment and the storage medium provided by the invention, firstly, the license plate information of a target vehicle is obtained through the camera equipment, and the license plate information comprises four vertexes of a license plate; then, performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle; and finally, inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample. Compared with the prior art that the license plate of the vehicle is recognized from a snapshot picture through an image recognition technology, the license plate recognition method has the advantages that after the license plate information of the target vehicle is obtained, four vertexes of the license plate information are subjected to projection transformation to obtain the license plate information of the front view angle of the license plate information of the front view angle, then the obtained license plate information of the front view angle is input into the license plate recognition model to obtain the license plate recognition result of the target vehicle, and the license plate recognition model is obtained through training according to a far-view license plate image sample and a near-view license plate image sample, so that the license plate recognition model can recognize the far-away vehicle, and the license plate recognition accuracy can be improved through the license plate recognition method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic diagram of an application environment of a license plate recognition method according to an embodiment of the present invention;
FIG. 2 is a flowchart of a license plate recognition method according to an embodiment of the invention;
FIG. 3 is a diagram illustrating the result of projective transformation performed on four vertices according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating a license plate recognition model training process according to an embodiment of the present invention;
FIG. 5 is a flowchart of a license plate recognition method according to an embodiment of the invention;
FIG. 6 is a schematic block diagram of a license plate recognition apparatus according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present 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.
The license plate recognition method provided by the application can be applied to the application environment shown in fig. 1, wherein the camera device is communicated with the server through a network. The method comprises the steps that a server obtains license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate; performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle; and inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample. The computer device may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, among others. The server may be implemented as a stand-alone server or as a server cluster consisting of a plurality of servers.
In an embodiment, as shown in fig. 2, a license plate recognition method is provided, which is described by taking the server in fig. 1 as an example, and includes the following steps:
and S10, acquiring the license plate information of the target vehicle through the camera equipment.
The camera shooting equipment in the embodiment of the invention comprises gun camera shooting equipment and dome camera shooting equipment, wherein the gun camera shooting equipment can shoot nearby vehicles, and the dome camera shooting equipment can shoot distant vehicles and specific detailed contents of the vehicles. It should be noted that, when judging whether the vehicles are the same, the license plate of the vehicle has uniqueness, the license plate is the best identifier for the identity of the vehicle, and the license plate information is used as the judgment basis, so that the occurrence of the fracture can be greatly reduced. Therefore, in the embodiment of the present invention, the license plate information of the target vehicle, that is, the license plate picture of the target vehicle, which is captured by the camera device, includes four vertices of the license plate, that is, the license plate information includes four vertices of the license plate, and the four vertices of the license plate are four vertices of a license plate circumscribed rectangle.
And S20, performing projection transformation on the four vertexes of the license plate information to obtain the license plate information of the positive viewing angle.
It should be noted that the projective transformation is a special case of the perspective transformation: the perspective transformation of the coplanar points, namely projection transformation or coplanar point imaging, uses linear transformation realized by a homographic matrix (homographic matrix) to transform a group of coplanar points from one plane to another plane, the parallel relation and proportion among straight lines are kept unchanged, and the angle can be adjusted by the transformation matrix.
As shown in fig. 3, in the implementation of the present invention, four vertexes of the license plate information are subjected to projection transformation, that is, the license plate information directly photographed by the photographing apparatus is adjusted, so that the license plate information is adjusted to a front view angle, that is, four vertexes, which are photographed by the photographing apparatus in fig. 3 and are not the front view angle, are subjected to projection transformation, so that 4 vertexes of the license plate information are a planar rectangle, so that the license plate number of the target vehicle is identified according to the license plate information of the front view angle in the subsequent step.
And S30, inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle.
The license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample. In one embodiment provided by the invention, a far-view license plate image sample and a near-view license plate image sample are data in a real scene, the far-view license plate image sample comprises far-view license plate image information and a license plate number corresponding to the far-view license plate image information, the near-view license plate image sample comprises near-view license plate image information and a license plate number corresponding to the near-view license plate image information, a license plate recognition model is obtained by training the far-view license plate image sample and the near-view license plate image sample, and the license plate numbers of target vehicles in different scenes can be recognized through the license plate model.
In an embodiment provided by the present invention, as shown in fig. 4, before the license plate information of the front viewing angle is input into a license plate recognition model to obtain a license plate recognition result of the target vehicle, the method further includes:
s101, obtaining the long-range license plate image sample and the short-range license plate image sample.
And S102, performing projection transformation on the far-view license plate image sample and the near-view license plate image sample to respectively obtain a far-view license plate image sample at a positive viewing angle and a near-view license plate image sample at a positive viewing angle.
S103, performing model training on the long-range license plate image sample at the positive viewing angle and the short-range license plate image sample at the positive viewing angle to obtain the license plate recognition model.
In the embodiment of the invention, a far-view license plate image sample and a near-view license plate image sample are obtained in a real scene, then the far-view license plate image sample and the near-view license plate image sample are subjected to projection transformation to respectively obtain a far-view license plate image sample at a positive viewing angle and a near-view license plate image sample at a formal angle, and model training is carried out according to the far-view license plate image sample and the near-view license plate image sample at the formal angle to obtain a license plate recognition model. The license plate recognition model can be used for recognizing the license plate information of a target vehicle, and the license plate recognition model is obtained by training license plate image samples at different positions under different scenes, so that the license plate numbers of the vehicles under different scenes and different positions can be recognized through the license plate recognition model.
In another embodiment provided by the invention, license plate image samples of vehicles at different positions are obtained in a real scene, for example, a distant view license plate image and a near view license plate image are obtained for a certain vehicle, the near view license plate image can be used as a label of the distant view license plate image, and then a license plate recognition model is trained to obtain the license plate recognition model. The license plate recognition model obtained in the way obtains the license plate information of the front view angle of the target vehicle, and then inputs the license plate information into the license plate recognition model to obtain the close-range license plate information of the target vehicle, namely, the license plate recognition result of the target vehicle can be obtained through the close-range license plate information.
For example, the camera device obtains a far-view license plate image and a near-view license plate image corresponding to different vehicles in a real scene, and then scales and fills the far-view license plate image and the near-view license plate image to a fixed size in an equal ratio, specifically scales and fills the far-view license plate image to 30 × 30 pixels, and scales and fills the near-view license plate image to 120 × 120 pixels. And then training the model according to the zoomed distant view license plate image and the zoomed near view license plate image to obtain a license plate recognition model. After a distant view license plate image of a target vehicle is acquired, the distant view license plate image is processed to obtain a distant view license plate image with the pixel of 30 multiplied by 30, then the processed distant view license plate image is input into a license plate recognition model, and license plate data with 120 multiplied by 120 pixels and subjected to network computing supersampling is output.
It should be noted that, because the license plate information obtained from the camera device has different distances from the camera device in different parking spaces, and the camera device itself may slightly shake during the operation process, the size of the license plate image collected in different parking spaces and different time periods, whether in a close view or a distant view, will be different, and the aspect ratio will be different. In order to meet the requirement of network training, the near-view license plate and the far-view license plate need to be scaled to be uniform in size.
Therefore, after obtaining the far-view license plate image sample at the positive viewing angle and the near-view license plate image sample at the positive viewing angle, the method further comprises: scaling and filling the forward-view long-range license plate image sample into a first preset size pixel in an equal proportion; and scaling the close-range license plate image sample of the formal corner in an equal proportion and filling the sample into a second preset size pixel, wherein the first preset size pixel is lower than the second preset size pixel. Whether the sample is a far-view license plate image sample or a near-view license plate image sample, the sample is scaled in an equal ratio and filled to a fixed size, and the difference is that the sample of the near-view license plate image is a preset multiple of the sample of the far-view license plate image, for example, the sample of the far-view license plate image is scaled and filled to 30 × 30 pixels, and the sample of the near-view license plate image is scaled and filled to 120 × 120 pixels. Specifically, when a license plate image sample is subjected to a scaling mode, bicubic interpolation needs to be performed while details of original data are kept as much as possible. For example, in order to facilitate the requirement of network training, the near-view license plate and the far-view license plate need to be scaled to a uniform size, and it is ensured that the width and height of the image of the near-view license plate are respectively 4 times of the fixed multiple of the width and height of the far-view license plate: the close-up image captured by the camera is about 4 times the long-range image.
Correspondingly, the performing model training on the far-view license plate image sample at the positive viewing angle and the near-view license plate image sample at the formal angle to obtain the license plate recognition model comprises: and performing model training on the long-range license plate image sample in the first preset size pixel and the short-range license plate image sample in the second preset size pixel to obtain the license plate recognition model.
In a specific application scenario provided in the embodiment of the present invention, the image capturing apparatus includes a variable-focus dome camera, and is capable of providing image data of the same license plate with different resolutions, where the license plate image data includes four detected vertices, projecting the license plate image data to a main view through the four detected vertices, scaling and filling a low-resolution long-range license plate image to an input size such as 30 × 30, scaling and filling a high-resolution short-range license plate image to an output size 120 × 120, and calculating L2 loss pixel by pixel and summing network output and target. The method comprises the steps of scaling and filling a long-range view license plate image or a short-range view license plate image to a fixed size in an equal ratio, wherein the difference is that the long-range view license plate image is scaled and filled to 30 multiplied by 30 pixels, and the short-range view license plate image is scaled and filled to 120 multiplied by 120 pixels. The scaling should preserve as much detail as possible of the original data.
The invention provides a license plate recognition method, which comprises the steps of firstly, obtaining license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate; then, performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle; and finally, inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample. Compared with the prior art that the license plate of the vehicle is recognized from a snapshot picture through an image recognition technology, the license plate recognition method has the advantages that after the license plate information of the target vehicle is obtained, four vertexes of the license plate information are subjected to projection transformation to obtain the license plate information of the front view angle of the license plate information of the front view angle, then the obtained license plate information of the front view angle is input into the license plate recognition model to obtain the license plate recognition result of the target vehicle, and the license plate recognition model is obtained through training according to a far-view license plate image sample and a near-view license plate image sample, so that the license plate recognition model can recognize the far-away vehicle, and the license plate recognition accuracy can be improved through the license plate recognition method.
As shown in fig. 5, in an embodiment provided by the present invention, after the license plate information of the front viewing angle is input into a license plate recognition model to obtain a license plate recognition result of the target vehicle, the method further includes:
and S40, inquiring whether the license plate identification result exists in a license plate database.
Various license plate numbers are stored in the license plate database, and the license plate numbers are legal and real license plate numbers. In the embodiment of the invention, after the license plate recognition result of the target vehicle is obtained through the license plate recognition model, the license plate number of the target vehicle is obtained, whether the license plate recognition of the target vehicle is correct or not is determined by inquiring whether the license plate recognition result exists in the license plate database, and if the license plate recognition result exists in the license plate database, the step S50A is skipped; if the license plate recognition result does not exist in the license plate database, the step S50B is skipped.
And S50A, if the license plate recognition result exists, preliminarily determining that the license plate recognition result is correct.
In the embodiment of the invention, if the license plate recognition result of the target vehicle exists in the license plate database, the license plate recognition result can be preliminarily determined to be correct, and then the step S60A is skipped to continuously verify whether the license plate recognition result is correct.
S50B, if the license plate recognition result does not exist, the license plate recognition result is determined to be wrong.
Step S50B is a parallel step of step S50A, and if the license plate database does not have the license plate recognition result of the target vehicle, it may be directly determined that the license plate recognition result is incorrect.
And S60A, identifying the license plate number from the license plate information of the target vehicle through an image identification technology.
It should be noted that the image recognition technology in the embodiment of the present invention may be an existing image recognition technology, and the purpose of using the image recognition technology is to recognize the license plate number of the target vehicle from the license plate information, so as to determine whether the recognition result of the target license plate number is correct through the license plate numbers recognized in different manners in the subsequent steps.
And S70A, matching the license plate recognition result with the recognized license plate number according to the corresponding position of the license plate number.
In the embodiment of the invention, after the license plate number identified by the image identification technology is acquired and the license plate identification result of the target vehicle is obtained from the license plate identification model, the license plate number identified by the image identification technology and the license plate identification result obtained from the license plate identification model are matched according to positions. For example, if the license plate number of the target vehicle is acquired as jing E71089 through the image recognition technology and the license plate number of the target vehicle is recognized as jing E71088 through the license plate recognition model, the two license plate numbers are subjected to corresponding position matching, and the number of matched positions can be determined to be 6.
And S80A, if the matching digit is larger than a preset value, determining that the identification result of the target vehicle is correct.
The preset value can be set according to actual requirements, the preset value is less than or equal to the total digits of the license plate number, and the larger the preset value is, the higher the matching digits of the license plate number are required to be; the smaller the preset value is set, the lower the number of matching bits of the license plate number is required. For example, the license plate number of the target vehicle is acquired as jing E71089 through the image recognition technology, the license plate number of the target vehicle recognized through the license plate recognition model is jing E71088, and if the preset value is set to be 5, the recognition result of the target vehicle is determined to be correct.
In one embodiment provided by the present invention, the method further comprises: and taking the correct recognition result of the target vehicle and the license plate information of the corresponding front view angle as a new training sample, and performing update training on the license plate recognition model. Therefore, the continuous updating of the vehicle identification model is realized, and the accuracy of the license plate number identification of the target vehicle is ensured.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In an embodiment, a license plate recognition device is provided, and the license plate recognition device corresponds to the license plate recognition method in the embodiment one to one. As shown in fig. 6, the license plate recognition device includes an acquisition module 10, a projective transformation module 20, and a recognition module 30. The functional modules are explained in detail as follows:
the system comprises an acquisition module 10, a storage module and a display module, wherein the acquisition module is used for acquiring license plate information of a target vehicle through camera equipment, and the license plate information comprises four vertexes of a license plate;
the projection transformation module 20 is configured to perform projection transformation on four vertexes of the license plate information to obtain license plate information at a positive viewing angle;
and the recognition module 30 is configured to input the license plate information at the positive viewing angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, where the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
Further, the apparatus further comprises:
the obtaining module 10 is further configured to obtain the distant view license plate image sample and the close view license plate image sample;
the projection transformation module 20 is further configured to perform projection transformation on the far-view license plate image sample and the near-view license plate image sample to obtain a far-view license plate image sample at a positive viewing angle and a near-view license plate image sample at a normal viewing angle respectively;
and the training module 40 is used for performing model training on the long-range license plate image sample at the positive viewing angle and the short-range license plate image sample at the positive viewing angle to obtain the license plate recognition model.
Further, the apparatus further comprises:
a scaling and filling module 50, configured to scale and fill the front-view license plate image sample at the front viewing angle into a first preset size pixel; scaling and filling a sample of the close-range license plate image of the formal corner into a second preset size pixel in an equal proportion, wherein the first preset size pixel is lower than the second preset size pixel;
the training module 40 is specifically configured to perform model training on the long-range license plate image samples in the first preset-size pixels and the short-range license plate image samples in the second preset-size pixels to obtain the license plate recognition model.
Further, the apparatus further comprises:
the query module 60 is used for querying whether the license plate recognition result exists in a license plate database;
the determining module 70 is configured to preliminarily determine that the license plate recognition result is correct if the license plate recognition result exists; and if the license plate recognition result does not exist, determining that the license plate recognition result is wrong.
Further, the apparatus further comprises:
the recognition module 30 is further configured to recognize a license plate number from the license plate information of the target vehicle through an image recognition technology;
a matching module 80 for matching the license plate recognition result and the recognized license plate number according to the corresponding position of the license plate number;
the determining module 70 is configured to determine that the identification result of the target vehicle is correct if the number of matching bits is greater than a preset value.
Further, the training module 40 is further configured to take the correct recognition result of the target vehicle and the license plate information of the front view angle corresponding to the correct recognition result as a new training sample, and perform update training on the license plate recognition model.
For specific limitations of the license plate recognition device, reference may be made to the above limitations of the license plate recognition method, which are not described herein again. All or part of the modules in the license plate recognition device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a license plate recognition method.
In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
acquiring license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate;
performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle;
and inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate;
performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle;
and inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile 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), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present 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 solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (14)
1. A method for recognizing a license plate, the method comprising:
acquiring license plate information of a target vehicle through camera equipment, wherein the license plate information comprises four vertexes of a license plate;
performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle;
and inputting the license plate information of the front view angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, wherein the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
2. The method for recognizing a license plate of claim 1, wherein before the license plate information at the front viewing angle is input into a license plate recognition model to obtain a license plate recognition result of the target vehicle, the method further comprises:
acquiring the far-view license plate image sample and the near-view license plate image sample;
performing projection transformation on the far-view license plate image sample and the near-view license plate image sample to respectively obtain a far-view license plate image sample at a positive viewing angle and a near-view license plate image sample at a formal angle;
and performing model training on the long-range license plate image sample at the positive viewing angle and the short-range license plate image sample at the formal angle to obtain the license plate recognition model.
3. The method for recognizing a license plate of claim 2, wherein after obtaining the forward-view far-view license plate image sample and the normal-view near-view license plate image sample, the method further comprises:
scaling and filling the forward-view long-range license plate image sample into a first preset size pixel in an equal proportion;
scaling and filling a sample of the close-range license plate image of the formal corner into a second preset size pixel in an equal proportion, wherein the first preset size pixel is lower than the second preset size pixel;
the model training of the far-view license plate image sample at the positive viewing angle and the near-view license plate image sample at the formal angle to obtain the license plate recognition model comprises the following steps:
and performing model training on the long-range license plate image sample in the first preset size pixel and the short-range license plate image sample in the second preset size pixel to obtain the license plate recognition model.
4. The method for recognizing a license plate of claim 1, wherein after the license plate information of the front viewing angle is input into a license plate recognition model to obtain a license plate recognition result of the target vehicle, the method further comprises:
inquiring whether the license plate identification result exists in a license plate database;
if the license plate recognition result exists, preliminarily determining that the license plate recognition result is correct;
and if the license plate recognition result does not exist, determining that the license plate recognition result is wrong.
5. The method for recognizing the license plate of claim 4, wherein after the preliminary determination that the license plate recognition result is correct, the method further comprises:
identifying a license plate number from the license plate information of the target vehicle through an image identification technology;
matching the license plate recognition result with the recognized license plate number according to the corresponding position of the license plate number;
and if the matching digit is larger than a preset value, determining that the identification result of the target vehicle is correct.
6. The method for recognizing a license plate of claim 4 or 5, further comprising:
and taking the correct recognition result of the target vehicle and the license plate information of the corresponding front view angle as a new training sample, and performing update training on the license plate recognition model.
7. A device for recognizing a license plate, the device comprising:
the system comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring license plate information of a target vehicle through camera equipment, and the license plate information comprises four vertexes of a license plate;
the projection transformation module is used for performing projection transformation on four vertexes of the license plate information to obtain license plate information of a positive visual angle;
and the recognition module is used for inputting the license plate information of the positive visual angle into a license plate recognition model to obtain a license plate recognition result of the target vehicle, and the license plate recognition model is obtained by training according to a far-view license plate image sample and a near-view license plate image sample.
8. The device for recognizing license plates of claim 7, further comprising:
the acquisition module is further used for acquiring the far-view license plate image sample and the near-view license plate image sample;
the projection transformation module is also used for performing projection transformation on the far-view license plate image sample and the near-view license plate image sample to respectively obtain a far-view license plate image sample at a positive viewing angle and a near-view license plate image sample at a positive viewing angle;
and the training module is used for carrying out model training on the long-range license plate image sample at the positive visual angle and the short-range license plate image sample at the formal angle to obtain the license plate recognition model.
9. The device for recognizing license plates of claim 8, further comprising:
the scaling and filling module is used for scaling and filling the positive-view distant view license plate image sample into a first preset size pixel in an equal proportion; scaling and filling a sample of the close-range license plate image of the formal corner into a second preset size pixel in an equal proportion, wherein the first preset size pixel is lower than the second preset size pixel;
the training module is specifically configured to perform model training on the long-range license plate image samples in the first preset-size pixels and the short-range license plate image samples in the second preset-size pixels to obtain the license plate recognition model.
10. The device for recognizing license plates of claim 7, further comprising:
the query module is used for querying whether the license plate identification result exists in a license plate database;
the determining module is used for preliminarily determining that the license plate recognition result is correct if the license plate recognition result exists; and if the license plate recognition result does not exist, determining that the license plate recognition result is wrong.
11. The device for recognizing license plates of claim 10, further comprising:
the identification module is also used for identifying the license plate number from the license plate information of the target vehicle through an image identification technology;
the matching module is used for matching the license plate recognition result with the recognized license plate number according to the corresponding position of the license plate number;
the determining module is used for determining that the identification result of the target vehicle is correct if the matching digit is larger than a preset value.
12. The license plate recognition device of claim 10 or 11, wherein the training module is further configured to update and train the license plate recognition model by using a correct recognition result of the target vehicle and license plate information of a corresponding front view angle as a new training sample.
13. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for recognizing a license plate according to any one of claims 1 to 6 when executing the computer program.
14. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, implements a method for recognizing a license plate according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910904848.XA CN110659638B (en) | 2019-09-24 | 2019-09-24 | License plate recognition method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910904848.XA CN110659638B (en) | 2019-09-24 | 2019-09-24 | License plate recognition method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110659638A true CN110659638A (en) | 2020-01-07 |
| CN110659638B CN110659638B (en) | 2023-04-07 |
Family
ID=69038852
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910904848.XA Active CN110659638B (en) | 2019-09-24 | 2019-09-24 | License plate recognition method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659638B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111797713A (en) * | 2020-06-16 | 2020-10-20 | 浙江大华技术股份有限公司 | License plate recognition method and photographing device |
| CN113807302A (en) * | 2021-09-26 | 2021-12-17 | 重庆紫光华山智安科技有限公司 | License plate recognition algorithm accuracy rate testing method and device, electronic equipment and readable storage medium |
| KR20220091995A (en) * | 2020-12-24 | 2022-07-01 | 인천대학교 산학협력단 | Electronic device capable of rectangularizing an inclined license plate image through machine learning-based vertex extraction and homography matrix calculation, and the operating method thereof |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073691A (en) * | 2010-12-23 | 2011-05-25 | 杭州海康威视数字技术股份有限公司 | Vehicle monitoring method and system thereof |
| CN102610119A (en) * | 2012-03-22 | 2012-07-25 | 广州杰赛科技股份有限公司 | Reverse car locating method and reverse car locating system |
| CN103065138A (en) * | 2012-12-06 | 2013-04-24 | 新疆公众信息产业股份有限公司 | Recognition method of license plate number of motor vehicle |
| CN103065135A (en) * | 2013-01-25 | 2013-04-24 | 上海理工大学 | License number matching algorithm based on digital image processing |
| CN105069765A (en) * | 2015-07-22 | 2015-11-18 | 广东迅通科技股份有限公司 | Fuzzy license plate reconstruction method based on characteristic learning |
| CN106169076A (en) * | 2016-07-22 | 2016-11-30 | 中山大学 | A kind of angle license plate image storehouse based on perspective transform building method |
| CN106203433A (en) * | 2016-07-13 | 2016-12-07 | 西安电子科技大学 | In a kind of vehicle monitoring image, car plate position automatically extracts and the method for perspective correction |
| CN106251637A (en) * | 2016-09-13 | 2016-12-21 | 杭州东方通信软件技术有限公司 | A kind of intelligent parking lot licence plate recognition method |
| CN106570436A (en) * | 2016-11-14 | 2017-04-19 | 三峡大学 | Substation equipment meter reading system and meter reading method based on two-dimensional code |
| CN107316495A (en) * | 2017-08-16 | 2017-11-03 | 武汉理工大学 | Residential parking space shared system and management method |
| CN107680382A (en) * | 2017-02-24 | 2018-02-09 | 上海享泊信息科技有限公司 | Licence plate recognition method, module, cloud server and managing system of car parking |
| CN108805807A (en) * | 2017-04-26 | 2018-11-13 | 伟诠电子股份有限公司 | Splicing method and system for ring scene images |
| CN108830213A (en) * | 2018-06-12 | 2018-11-16 | 北京理工大学 | Car plate detection and recognition methods and device based on deep learning |
| CN108985137A (en) * | 2017-06-02 | 2018-12-11 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method, apparatus and system |
| CN109145900A (en) * | 2018-07-30 | 2019-01-04 | 中国科学技术大学苏州研究院 | A kind of licence plate recognition method based on deep learning |
| CN109690624A (en) * | 2016-09-15 | 2019-04-26 | 高通股份有限公司 | Automatic scene calibration method for video analysis |
-
2019
- 2019-09-24 CN CN201910904848.XA patent/CN110659638B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073691A (en) * | 2010-12-23 | 2011-05-25 | 杭州海康威视数字技术股份有限公司 | Vehicle monitoring method and system thereof |
| CN102610119A (en) * | 2012-03-22 | 2012-07-25 | 广州杰赛科技股份有限公司 | Reverse car locating method and reverse car locating system |
| CN103065138A (en) * | 2012-12-06 | 2013-04-24 | 新疆公众信息产业股份有限公司 | Recognition method of license plate number of motor vehicle |
| CN103065135A (en) * | 2013-01-25 | 2013-04-24 | 上海理工大学 | License number matching algorithm based on digital image processing |
| CN105069765A (en) * | 2015-07-22 | 2015-11-18 | 广东迅通科技股份有限公司 | Fuzzy license plate reconstruction method based on characteristic learning |
| CN106203433A (en) * | 2016-07-13 | 2016-12-07 | 西安电子科技大学 | In a kind of vehicle monitoring image, car plate position automatically extracts and the method for perspective correction |
| CN106169076A (en) * | 2016-07-22 | 2016-11-30 | 中山大学 | A kind of angle license plate image storehouse based on perspective transform building method |
| CN106251637A (en) * | 2016-09-13 | 2016-12-21 | 杭州东方通信软件技术有限公司 | A kind of intelligent parking lot licence plate recognition method |
| CN109690624A (en) * | 2016-09-15 | 2019-04-26 | 高通股份有限公司 | Automatic scene calibration method for video analysis |
| CN106570436A (en) * | 2016-11-14 | 2017-04-19 | 三峡大学 | Substation equipment meter reading system and meter reading method based on two-dimensional code |
| CN107680382A (en) * | 2017-02-24 | 2018-02-09 | 上海享泊信息科技有限公司 | Licence plate recognition method, module, cloud server and managing system of car parking |
| CN108805807A (en) * | 2017-04-26 | 2018-11-13 | 伟诠电子股份有限公司 | Splicing method and system for ring scene images |
| CN108985137A (en) * | 2017-06-02 | 2018-12-11 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method, apparatus and system |
| CN107316495A (en) * | 2017-08-16 | 2017-11-03 | 武汉理工大学 | Residential parking space shared system and management method |
| CN108830213A (en) * | 2018-06-12 | 2018-11-16 | 北京理工大学 | Car plate detection and recognition methods and device based on deep learning |
| CN109145900A (en) * | 2018-07-30 | 2019-01-04 | 中国科学技术大学苏州研究院 | A kind of licence plate recognition method based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| MENG DONG ET.AL: ""A CNN-Based Approach for Automatic License Plate Recognition in the Wild"", 《BRITISH MACHINE VISION CONFERENCE 2017》 * |
| 徐辉: "车辆牌照自动识别技术研究与应用", 《企业导报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111797713A (en) * | 2020-06-16 | 2020-10-20 | 浙江大华技术股份有限公司 | License plate recognition method and photographing device |
| KR20220091995A (en) * | 2020-12-24 | 2022-07-01 | 인천대학교 산학협력단 | Electronic device capable of rectangularizing an inclined license plate image through machine learning-based vertex extraction and homography matrix calculation, and the operating method thereof |
| KR102557697B1 (en) * | 2020-12-24 | 2023-07-19 | 인천대학교 산학협력단 | Electronic device capable of rectangularizing an inclined license plate image through machine learning-based vertex extraction and homography matrix calculation, and the operating method thereof |
| CN113807302A (en) * | 2021-09-26 | 2021-12-17 | 重庆紫光华山智安科技有限公司 | License plate recognition algorithm accuracy rate testing method and device, electronic equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110659638B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110659638B (en) | License plate recognition method and device, computer equipment and storage medium | |
| US20230345135A1 (en) | Method, apparatus, and device for processing images, and storage medium | |
| CN110738219A (en) | Method and device for extracting lines in image, storage medium and electronic device | |
| CN111191653A (en) | License plate recognition method and device, computer equipment and storage medium | |
| CN110620881A (en) | License plate exposure compensation method and device, computer equipment and storage medium | |
| CN111161202A (en) | Vehicle behavior information acquisition method and device, computer equipment and storage medium | |
| CN110705433A (en) | Bridge deformation monitoring method, device and equipment based on visual perception | |
| CN111652937B (en) | Vehicle-mounted camera calibration method and device | |
| CN113963072B (en) | Binocular camera calibration method and device, computer equipment and storage medium | |
| CN112634628B (en) | Vehicle speed determination method, terminal and storage medium | |
| CN111445513B (en) | Plant canopy volume acquisition method and device based on depth image, computer equipment and storage medium | |
| CN113284190A (en) | Calibration method, calibration device, calibration equipment, storage medium and product | |
| CN112102378A (en) | Image registration method and device, terminal equipment and computer readable storage medium | |
| CN116468914A (en) | Page comparison method and device, storage medium and electronic equipment | |
| CN114757846A (en) | Image correction method and device, storage medium and electronic device | |
| CN113593297B (en) | Parking space state detection method and device | |
| CN109919998B (en) | Satellite attitude determination method and device and terminal equipment | |
| CN109376653B (en) | Method, apparatus, device and medium for locating vehicle | |
| CN110599801A (en) | Parking management method and device, computer equipment and storage medium | |
| CN114066930A (en) | Planar target tracking method and device, terminal equipment and storage medium | |
| CN114581310A (en) | Chromosome straightening method, chromosome straightening device, computer equipment and storage medium | |
| CN112200004A (en) | Training method and device of image detection model and terminal equipment | |
| CN117437646A (en) | License plate recognition method, license plate recognition device, computer equipment and storage medium | |
| CN113780286A (en) | Object recognition method and device, storage medium and electronic device | |
| CN113223076B (en) | Coordinate system calibration method, device and storage medium for vehicle and vehicle-mounted camera |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |