CN112926583A - License plate recognition method and license plate recognition system - Google Patents
License plate recognition method and license plate recognition system Download PDFInfo
- Publication number
- CN112926583A CN112926583A CN202110450373.9A CN202110450373A CN112926583A CN 112926583 A CN112926583 A CN 112926583A CN 202110450373 A CN202110450373 A CN 202110450373A CN 112926583 A CN112926583 A CN 112926583A
- Authority
- CN
- China
- Prior art keywords
- license plate
- image
- recognition
- coordinates
- vertex
- 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
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Character Discrimination (AREA)
- Image Analysis (AREA)
- Character Input (AREA)
Abstract
The invention provides a license plate recognition method and a license plate recognition system, which comprise the following steps: acquiring a first image containing a license plate; detecting the approximate position of the license plate through a first detection model, and cutting out a license plate image; positioning four vertex coordinates of the license plate in the license plate image through a second detection model; obtaining a transformation matrix of affine transformation by taking the obtained vertex coordinates as original coordinate points and taking four vertex coordinates of the license plate transformed into the expected plane view angle through affine transformation as transformation point coordinates; carrying out affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and finally, identifying the license plate information through a third detection model. According to the invention, after affine transformation processing is used for positioning based on the coordinates of the license plate, a license plate picture which is easier to recognize the license plate is input, the accuracy of license plate recognition is improved, and the problem that the license plate cannot be correctly recognized due to the fact that a license plate recognition system is too inclined in shooting angle is avoided.
Description
Technical Field
The invention relates to the technical field of image processing and recognition, in particular to a license plate recognition method and a license plate recognition system, and especially aims to improve the license plate recognition rate through the recognition processing of the invention when the shooting angle of a camera used for collecting license plate images is too large or the problems of inclination, angle deflection and the like of the shot images exist.
Background
The License Plate Recognition technology (VLPR) is an image processing technology that integrates front-end License Plate image acquisition with a background Recognition processing system, acquires a License Plate image of a Vehicle by using on-site image acquisition and shooting equipment, transmits the License Plate image to the background Recognition system for image processing, and recognizes a License Plate number by using visual Recognition processing. The license plate recognition technology has been widely applied to various scenes, such as license plate recognition and unmanned charging system of parking lot, Electronic Toll Collection (ETC) system of expressway, overspeed violation recognition system based on a camera for traffic police, road flow monitoring and management system, automatic license plate grade and release control of special areas or units (military districts, confidential units and the like), along with the further development of computer video image recognition technology, the accuracy and efficiency of license plate recognition technology are gradually improved, and the license plate recognition technology plays an important practical role in maintaining traffic safety and urban public security, preventing traffic jam, realizing traffic automation management and scientific efficiency management of residential areas and parking lots.
The existing license plate recognition technology VLPR mainly comprises a front end and a rear end, wherein the front end is responsible for image acquisition and needs to be deployed to a passage through which a vehicle passes, and aims at the direction of a vehicle entrance and well debugs the angle and the direction. The back end is responsible for image processing and processing of recognition results, wherein the image processing aims at recognizing license plates, the main recognition algorithm comprises license plate detection and license plate recognition, the license plate detection refers to positioning the position of the license plate from the image collected at the front end, and gray detection, edge detection or object detection algorithm (anchor or anchor-free algorithm) can be adopted, while the license plate recognition is the key, the existing algorithms are more, such as neural network recognition based on CNN and recognition based on image retrieval.
However, the existing license plate recognition result generally obtains license plate image recognition based on an image shot at a very ideal angle, and in a deployment process and a subsequent use process, after a camera angle is deviated or changed, a license plate image of a shot vehicle is deformed and is not shot for a license plate. Meanwhile, the traditional camera cannot be changed after being deployed, so that the system cannot correctly recognize the license plate due to the fact that the shooting angle is too inclined, the traditional license plate recognition model requires that the erected camera is very close to the license plate to be recognized and keeps right, but the camera is fixed, the vehicle is not fixed, and the traditional license plate recognition system cannot correctly recognize the license plate due to the fact that the shooting angle is too inclined.
In the prior art, before character segmentation is performed on a license plate image, correction processing of the license plate image is realized through position correction, such as affine transformation or transmission transformation, but the correction processing aims at the whole license plate image, and a clear license plate recognition effect is still difficult to obtain after the correction processing of the affine transformation.
Prior art documents:
patent document 1: CN112434700A
Disclosure of Invention
The invention aims to provide a license plate recognition system and a license plate recognition method, which are used for inputting a license plate picture which is easier to recognize a license plate after affine transformation is used for positioning based on the coordinates of the license plate, improving the accuracy of license plate recognition and avoiding the problem that the license plate recognition system cannot correctly recognize the license plate because the shooting angle is too inclined.
The first aspect of the present invention provides a license plate recognition method, including:
acquiring a first image containing a license plate;
detecting coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of the license plate contained in the first image through a first detection model to obtain a license plate image;
positioning four vertex coordinates of the license plate in the license plate image through a second detection model;
obtaining a transformation matrix of affine transformation by taking the obtained vertex coordinates as original coordinate points and taking four vertex coordinates of the license plate transformed into the expected plane view angle through affine transformation as transformation point coordinates;
carrying out affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and
and taking the transformed second image as an input, and identifying the license plate information through a third detection model.
Preferably, the third detection model is configured to recognize the license plate information through character recognition.
Preferably, the first detection model is set as a target detector for detecting a license plate, the target detector is used for detecting the approximate position of the license plate in the first image, and the license plate image is cut out through coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of the license plate.
Preferably, the second detection model is a license plate image recognition model based on key point recognition, and the license plate image recognition model based on key point recognition is obtained by pre-training, namely a recognition model trained by constructing a training set by using four vertex coordinates of an labeled license plate;
the positioning of the four vertex coordinates of the license plate in the license plate image through the second detection model comprises the following steps:
and taking the license plate image as input, and positioning four vertex coordinates of the license plate in the license plate image by the license plate image recognition model based on the key point recognition.
Preferably, the obtaining of the transformation matrix is configured to be derived by the following formula:
wherein the content of the first and second substances,
the original coordinate point is represented by a coordinate of the original coordinate point,
the coordinates of the transformed points are represented,
representing a transformation matrix;
representing a new origin of coordinates;
representing the new basis vectors.
According to a second aspect of the present invention, there is provided a license plate recognition system, comprising:
a module for acquiring a first image containing a license plate;
a module for detecting coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of a license plate contained in the first image through a first detection model to obtain a license plate image;
a module for positioning the coordinates of four vertexes of the license plate in the license plate image through a second detection model;
a module for obtaining a transformation matrix of affine transformation by using the obtained vertex coordinates as original coordinate points and four vertex coordinates of the license plate transformed into the desired plane view angle through affine transformation as transformation point coordinates;
a module for performing affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and
and a module for identifying the license plate information through a third detection model by taking the transformed second image as input.
It should be understood that all combinations of the foregoing concepts and additional concepts described in greater detail below can be considered as part of the inventive subject matter of this disclosure unless such concepts are mutually inconsistent. In addition, all combinations of claimed subject matter are considered a part of the presently disclosed subject matter.
Drawings
The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description taken in conjunction with the accompanying drawings. Additional aspects of the present invention, such as features and/or advantages of exemplary embodiments, will be apparent from the description which follows, or may be learned by practice of specific embodiments in accordance with the teachings of the present invention.
The figures are not drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. Embodiments of various aspects of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:
fig. 1 is a flowchart illustrating a license plate recognition method according to a first embodiment of the present invention.
FIG. 2 is a flowchart of license plate detection according to a first embodiment of the present invention.
FIG. 3 is a flowchart of license plate location according to a first embodiment of the present invention.
Fig. 4 is an example of an image captured by the license plate recognition system when the camera capture angle is too oblique.
Fig. 5 is an example of an image for recognition input to a model, which is obtained by performing affine transformation on an image after performing a transformation matrix obtained by positioning key points according to the result of a target detector in a license plate recognition process according to the first embodiment of the present invention with fig. 4 as an original image.
Fig. 6 is an example of an image obtained by performing affine transformation on the result of the target detector according to the prior art using fig. 4 as an original image and inputting the result into a model for recognition.
Fig. 7 is an example of an image obtained by processing a plurality of license plates according to the license plate recognition method of the first embodiment of the present invention and input to a second model for recognition.
Detailed Description
In order to better understand the technical content of the present invention, specific embodiments are described below with reference to the accompanying drawings.
In this disclosure, aspects of the present invention are described with reference to the accompanying drawings, in which a number of illustrative embodiments are shown. Embodiments of the present disclosure are not necessarily intended to include all aspects of the invention. It should be appreciated that the various concepts and embodiments described above, as well as those described in greater detail below, may be implemented in any of numerous ways, as the disclosed concepts and embodiments are not limited to any one implementation. In addition, some aspects of the present disclosure may be used alone, or in any suitable combination with other aspects of the present disclosure.
According to the embodiment of the first aspect of the invention, the license plate recognition method aims to solve the problem of how to improve the accuracy of license plate recognition under the condition that the shooting angle is not ideal when the image shot in a license plate recognition system (such as a recognition system of the entrance and exit position of a parking lot, a license plate recognition system of a highway toll station, a highway/urban road violation and/or a speed measurement system) is subjected to license plate recognition.
In an example of the present invention, an approximate position of a license plate in a collected image is detected, the image is cut to obtain a license plate image, then a key point positioning is performed on the license plate image, for example, a pre-trained key point recognition model is used for recognition, so that coordinates of four vertexes of the license plate can be determined, on the basis, a transformation matrix for performing transformation in the middle is calculated by the coordinates of the four vertexes of the license plate and coordinates expected to be transformed to a planar view angle, and then an affine transformation is performed on the whole collected image according to the transformation matrix, so that an input image more beneficial to license plate recognition is obtained and input into the license plate recognition model for character recognition, thereby obtaining license plate information output.
With reference to fig. 1-3, a specific implementation of the present invention includes:
s101: acquiring a first image containing a license plate, for example, acquiring an original image from a front-end acquisition end (camera) of a license plate recognition system through a data interface;
s102: detecting coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of the license plate contained in the first image through a first detection model to obtain a license plate image;
s103: positioning four vertex coordinates of the license plate in the license plate image through a second detection model;
s104: obtaining a transformation matrix of affine transformation by taking the obtained vertex coordinates as original coordinate points and taking four vertex coordinates of the license plate transformed into the expected plane view angle through affine transformation as transformation point coordinates;
s105: carrying out affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and
s106: and taking the transformed second image as an input, and identifying the license plate information through a third detection model.
Preferably, the aforementioned third detection model is configured to recognize the license plate information through character recognition, for example, a CNN network-based license plate recognition model, which may be generated through pre-training.
As an optional embodiment, in the training process of the license plate recognition model based on the CNN network, firstly, license plate data used for training needs to be constructed, and then the license plate data is trained to obtain the recognition model.
The construction of the license plate data requires the construction of a character set, for example, a province abbreviation, a capital one because of letters and numbers. And then generating Chinese and English character information, obtaining diversified license plate data information through data augmentation processing, and generating a training data set for training through a plurality of scene pictures as backgrounds.
In further embodiments, the third detection model may also be generated or trained based on other means, such as an Open CV trained SVM recognition model.
Preferably, the aforementioned first detection model is configured as an object detector for detecting the approximate position of the license plate in the first image, for example, by detecting with a conventional object detector, such as SSD, centrnet, etc., and cropping the license plate image by the coordinates of the top left vertex and the bottom right vertex of the minimum rectangular detection frame of the license plate.
In the specific implementation process, a license plate data set can be used for training a target detector capable of detecting the license plate, an image is input during actual use, and coordinates of the top left vertex and the bottom right vertex of the minimum rectangular frame of the license plate in the image can be output.
Preferably, the second detection model is a license plate image recognition model based on key point recognition. The license plate image recognition model based on key point recognition can be obtained by adopting pre-training, namely, a recognition model trained by constructing a training set by using four vertex coordinates of a labeled license plate;
the positioning of the four vertex coordinates of the license plate in the license plate image through the second detection model comprises the following steps:
and taking the license plate image as input, and positioning four vertex coordinates of the license plate in the license plate image by the license plate image recognition model based on the key point recognition.
In an actual shooting scene, due to reasons such as an angle, coordinates of a license plate in an image are not basically a marked rectangle, so that more detailed coordinate information (namely position information of four vertexes) of the license plate needs to be acquired through license plate positioning. The license plate positioning can be trained by adopting a reactive facial landmark detector (PFLD) model with key point positioning, and a detection model is trained by marking coordinates of four vertexes of the license plate and then forming a training set. Therefore, after the approximate position of the license plate is detected based on the target detector, the coordinates of the four vertexes of the license plate can be further obtained through license plate positioning.
Then, with reference to the flow shown in fig. 1, we obtain a transformation matrix of affine transformation by using the obtained vertex coordinates as original coordinate points and four vertex coordinates of the license plate that are affine-transformed into the desired planar view angle as transformation point coordinates.
Preferably, the obtaining of the transformation matrix is configured to be derived by the following formula:
wherein the content of the first and second substances,
the original coordinate point is represented by a coordinate of the original coordinate point,
the coordinates of the transformed points are represented,
representing a transformation matrix;
representing a new origin of coordinates;
representing the new basis vectors.
Based on this, the first image (i.e. the originally obtained image, as in fig. 4) is further affine transformed according to the obtained transformation matrix, obtaining a transformed second image, as in fig. 5.
Finally, the converted second image (for example, fig. 5) is used as an input, and the license plate information is recognized by the third detection model.
As shown in fig. 6, the image is an example of an image obtained by performing affine transformation on the result of target detection in the prior art and then inputting the result to the second model for recognition, using fig. 4 as the original image. In the conventional license plate recognition logic of license plate positioning, image correction (e.g., affine change or transmission transformation) and license plate character recognition, affine transformation correction is directly performed on the result of license plate positioning (i.e., an image containing the approximate position of the license plate directly cut out from an original shot image), and in the process of pulling to a planar view angle, if only the approximate position of the license plate obtained before affine change and the position information of the edge of the license plate are inaccurate, the quality difference of the pictures obtained after affine change is large, and the problem of blurring is very easy to occur, for example, as shown in fig. 6, which is not favorable for recognition.
Referring to fig. 5, in the license plate recognition process of the present invention, a transformation matrix obtained by positioning the key points is first performed according to the result of the target detector, and then an image example for input to the second model for recognition is obtained by performing affine transformation on the image.
Therefore, aiming at each originally obtained first image with the license plate, namely the image containing the license plate shot by a camera, the license plate is firstly detected by a target detector, the approximate position of the license plate is identified and the license plate image is obtained by cutting, then on the basis, the license plate is positioned by a detection model identified by key points, four vertexes of the license plate, namely the exact position of the license plate, are determined, and a transformation matrix to a plane visual angle is determined, namely each originally obtained first image corresponds to one transformation matrix, and then the corresponding original first image is subjected to pertinence affine transformation, so that the pertinence of each image correction is ensured, an ideal license plate image is obtained and is used as the input of a license plate identification model, and the accuracy of license plate identification is improved.
Based on fig. 4, the image shown in fig. 5 is more easily recognizable to the naked eye than the image shown in fig. 6, as seen in conjunction with fig. 5 and 6.
Fig. 7 is a diagram illustrating an example of license plate processing and recognition in the first embodiment of the present invention. The image before transformation is a license plate image detected by a license plate detector (target detector), license plate recognition is often inaccurate when the angle is too large, and the image after transformation is an image after positioning through a license plate and performing affine transformation in the method based on the first embodiment of the invention. Therefore, the output of the license plate images processed by the method has high pertinence due to the affine transformation of each image, is realized after the key points of each license plate are positioned, has high consistency on the processing quality of the images, and is better beneficial to realizing the accurate recognition of the license plates.
According to a second aspect of the present invention, there is provided a license plate recognition system, comprising:
a module for acquiring a first image containing a license plate;
a module for detecting coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of a license plate contained in the first image through a first detection model to obtain a license plate image;
a module for positioning the coordinates of four vertexes of the license plate in the license plate image through a second detection model;
a module for obtaining a transformation matrix of affine transformation by using the obtained vertex coordinates as original coordinate points and four vertex coordinates of the license plate transformed into the desired plane view angle through affine transformation as transformation point coordinates;
a module for performing affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and
and a module for identifying the license plate information through a second detection model by taking the transformed second image as an input.
Wherein the third detection model is configured to recognize license plate information through character recognition.
The first detection model is set as a target detector for realizing license plate detection, the target detector is used for detecting the approximate position of a license plate in the first image, and a license plate image is cut out through coordinates of the upper left vertex and the lower right vertex of the minimum rectangular detection frame of the license plate.
The second detection model is a license plate image recognition model based on key point recognition, and the license plate image recognition model based on key point recognition is obtained by pre-training, namely, a recognition model trained by constructing a training set by using four vertex coordinates of a labeled license plate.
Wherein the obtaining of the transformation matrix is configured to be obtained by the manner of the above embodiment.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.
Claims (10)
1. A license plate recognition method is characterized by comprising the following steps:
acquiring a first image containing a license plate;
detecting coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of the license plate contained in the first image through a first detection model to obtain a license plate image;
positioning four vertex coordinates of the license plate in the license plate image through a second detection model;
obtaining a transformation matrix of affine transformation by taking the obtained vertex coordinates as original coordinate points and taking four vertex coordinates of the license plate transformed into the expected plane view angle through affine transformation as transformation point coordinates;
carrying out affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and
and taking the transformed second image as an input, and identifying the license plate information through a third detection model.
2. The license plate recognition method of claim 1, wherein the third detection model is configured to recognize license plate information through character recognition.
3. The license plate recognition method of claim 1, wherein the first detection model is configured as a target detector for detecting a license plate, the target detector is configured to detect an approximate position of the license plate in the first image, and the license plate image is cropped according to coordinates of a top left vertex and a bottom right vertex of a minimum rectangular detection frame of the license plate.
4. The license plate recognition method of claim 1, wherein the second detection model is a license plate image recognition model based on key point recognition, and the license plate image recognition model based on key point recognition is a recognition model obtained by pre-training, namely, a training set is constructed by marking four vertex coordinates of a license plate;
the positioning of the four vertex coordinates of the license plate in the license plate image through the second detection model comprises the following steps:
and taking the license plate image as input, and positioning four vertex coordinates of the license plate in the license plate image by the license plate image recognition model based on the key point recognition.
5. The license plate recognition method of any one of claims 1-4, wherein the obtaining of the transformation matrix is configured to be derived by the following formula:
wherein the content of the first and second substances,
the original coordinate point is represented by a coordinate of the original coordinate point,
the coordinates of the transformed points are represented,
representing a transformation matrix;
representing a new origin of coordinates;
representing the new basis vectors.
6. A license plate recognition system, comprising:
a module for acquiring a first image containing a license plate;
a module for detecting coordinates of an upper left vertex and a lower right vertex of a minimum rectangular detection frame of a license plate contained in the first image through a first detection model to obtain a license plate image;
a module for positioning the coordinates of four vertexes of the license plate in the license plate image through a second detection model;
a module for obtaining a transformation matrix of affine transformation by using the obtained vertex coordinates as original coordinate points and four vertex coordinates of the license plate transformed into the desired plane view angle through affine transformation as transformation point coordinates;
a module for performing affine transformation on the first image according to the transformation matrix to obtain a transformed second image; and
and a module for identifying the license plate information through a third detection model by taking the transformed second image as input.
7. The license plate recognition system of claim 1, wherein the third detection model is configured to recognize license plate information through character recognition.
8. The license plate recognition system of claim 1, wherein the first detection model is configured as a target detector for license plate detection, the target detector is configured to detect an approximate position of a license plate in the first image, and the license plate image is cropped by coordinates of upper left vertex and lower right vertex of a minimum rectangular detection frame of the license plate.
9. The license plate recognition system of claim 1, wherein the second detection model is a license plate image recognition model based on key point recognition, and the license plate image recognition model based on key point recognition is a recognition model obtained by pre-training, namely, training a training set by constructing four vertex coordinates of a labeled license plate.
10. The license plate recognition system of any one of claims 6-9, wherein the transformation matrix is obtained by:
wherein the content of the first and second substances,
the original coordinate point is represented by a coordinate of the original coordinate point,
the coordinates of the transformed points are represented,
representing a transformation matrix;
representing a new origin of coordinates;
representing the new basis vectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110450373.9A CN112926583B (en) | 2021-04-25 | 2021-04-25 | License plate recognition method and license plate recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110450373.9A CN112926583B (en) | 2021-04-25 | 2021-04-25 | License plate recognition method and license plate recognition system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112926583A true CN112926583A (en) | 2021-06-08 |
| CN112926583B CN112926583B (en) | 2022-08-16 |
Family
ID=76174678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110450373.9A Active CN112926583B (en) | 2021-04-25 | 2021-04-25 | License plate recognition method and license plate recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112926583B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113642577A (en) * | 2021-10-14 | 2021-11-12 | 深圳市爱深盈通信息技术有限公司 | Low-contrast license plate recognition method, system, equipment and storage medium |
| CN114049622A (en) * | 2021-10-29 | 2022-02-15 | 深圳市爱深盈通信息技术有限公司 | License plate recognition method and system |
| CN114882489A (en) * | 2022-07-07 | 2022-08-09 | 浙江智慧视频安防创新中心有限公司 | Method, device, equipment and medium for horizontally correcting rotary license plate |
| CN116612459A (en) * | 2023-07-18 | 2023-08-18 | 小米汽车科技有限公司 | Target detection method, target detection device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414507A (en) * | 2019-07-11 | 2019-11-05 | 和昌未来科技（深圳）有限公司 | Licence plate recognition method, device, computer equipment and storage medium |
| JP2019207456A (en) * | 2018-05-28 | 2019-12-05 | 日本電信電話株式会社 | Geometric transformation matrix estimation device, geometric transformation matrix estimation method, and program |
| CN110969135A (en) * | 2019-12-05 | 2020-04-07 | 中南大学 | Vehicle logo recognition method in natural scene |
-
2021
- 2021-04-25 CN CN202110450373.9A patent/CN112926583B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2019207456A (en) * | 2018-05-28 | 2019-12-05 | 日本電信電話株式会社 | Geometric transformation matrix estimation device, geometric transformation matrix estimation method, and program |
| CN110414507A (en) * | 2019-07-11 | 2019-11-05 | 和昌未来科技（深圳）有限公司 | Licence plate recognition method, device, computer equipment and storage medium |
| CN110969135A (en) * | 2019-12-05 | 2020-04-07 | 中南大学 | Vehicle logo recognition method in natural scene |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113642577A (en) * | 2021-10-14 | 2021-11-12 | 深圳市爱深盈通信息技术有限公司 | Low-contrast license plate recognition method, system, equipment and storage medium |
| CN114049622A (en) * | 2021-10-29 | 2022-02-15 | 深圳市爱深盈通信息技术有限公司 | License plate recognition method and system |
| CN114882489A (en) * | 2022-07-07 | 2022-08-09 | 浙江智慧视频安防创新中心有限公司 | Method, device, equipment and medium for horizontally correcting rotary license plate |
| CN114882489B (en) * | 2022-07-07 | 2022-12-16 | 浙江智慧视频安防创新中心有限公司 | Method, device, equipment and medium for horizontally correcting rotating license plate |
| CN116612459A (en) * | 2023-07-18 | 2023-08-18 | 小米汽车科技有限公司 | Target detection method, target detection device, electronic equipment and storage medium |
| CN116612459B (en) * | 2023-07-18 | 2023-11-17 | 小米汽车科技有限公司 | Target detection method, target detection device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112926583B (en) | 2022-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112926583B (en) | License plate recognition method and license plate recognition system | |
| CN105702048B (en) | Highway front truck illegal road occupation identifying system based on automobile data recorder and method | |
| CN110619279B (en) | Road traffic sign instance segmentation method based on tracking | |
| Saha et al. | License Plate localization from vehicle images: An edge based multi-stage approach | |
| CN101937508A (en) | License plate localization and identification method based on high-definition image | |
| KR101788225B1 (en) | Method and System for Recognition/Tracking Construction Equipment and Workers Using Construction-Site-Customized Image Processing | |
| Soomro et al. | Vehicle number recognition system for automatic toll tax collection | |
| CN103646544B (en) | Based on the vehicle behavioural analysis recognition methods of The Cloud Terrace and camera apparatus | |
| CN102867418A (en) | Method and device for judging license plate identification accuracy | |
| CN104463232A (en) | Density crowd counting method based on HOG characteristic and color histogram characteristic | |
| CN107358236A (en) | A kind of number-plate number identifying system and method based on camera device | |
| Ko et al. | License plate surveillance system using weighted template matching | |
| CN111091041A (en) | Vehicle law violation judging method and device, computer equipment and storage medium | |
| Devi et al. | An Efficient Hybrid Technique for Automatic License Plate Recognitions | |
| Liu et al. | Automatic pedestrian crossing detection and impairment analysis based on mobile mapping system | |
| Moghassemi et al. | Iranian License Plate Recognition using connected component and clustering techniques | |
| Yim et al. | Integrated plate recognition and speed detection for intelligent transportation systems | |
| CN114140674B (en) | Electronic evidence availability identification method combined with image processing and data mining technology | |
| KR100869139B1 (en) | System for recognizing regisration number plate of vechicle using image input signal through mobile phone and method thereof | |
| Saha et al. | License plate localization using vertical edge map and hough transform based technique | |
| KR100538526B1 (en) | Automatic police enforcement system of a illegal stopping and parking vehicles | |
| KR20070044607A (en) | System for recognizing regisration number plate of vechicle using image input signal through mobile phone | |
| Lajish et al. | Mobile phone based vehicle license plate recognition for road policing | |
| KR20080049472A (en) | Information detecting system using photographing apparatus load in vehicle and artificial neural network | |
| Bold et al. | Smart license plate recognition using optical character recognition based on the multicopter |
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 | ||
| CP01 | Change in the name or title of a patent holder |
Address after: No.568 longmian Avenue, gaoxinyuan, Jiangning District, Nanjing City, Jiangsu Province, 211000 Patentee after: Xiaoshi Technology (Jiangsu) Co.,Ltd. Address before: No.568 longmian Avenue, gaoxinyuan, Jiangning District, Nanjing City, Jiangsu Province, 211000 Patentee before: NANJING ZHENSHI INTELLIGENT TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |