CN110956169A - License plate recognition method and device and electronic equipment - Google Patents
License plate recognition method and device and electronic equipment Download PDFInfo
- Publication number
- CN110956169A CN110956169A CN201811133867.9A CN201811133867A CN110956169A CN 110956169 A CN110956169 A CN 110956169A CN 201811133867 A CN201811133867 A CN 201811133867A CN 110956169 A CN110956169 A CN 110956169A
- Authority
- CN
- China
- Prior art keywords
- license plate
- classification information
- recognition result
- character recognition
- fine classification
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000013135 deep learning Methods 0.000 abstract description 13
- 238000003384 imaging method Methods 0.000 abstract description 11
- 238000012545 processing Methods 0.000 abstract description 6
- 239000010410 layer Substances 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000012937 correction Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 239000002356 single layer Substances 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000012549 training Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000003702 image correction Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/146—Aligning or centring of the image pick-up or image-field
- G06V30/1475—Inclination or skew detection or correction of characters or of image to be recognised
- G06V30/1478—Inclination or skew detection or correction of characters or of image to be recognised of characters or characters lines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- 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)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Character Input (AREA)
- Character Discrimination (AREA)
- Image Analysis (AREA)
Abstract
The application provides a license plate recognition method, a license plate recognition device and electronic equipment, wherein the method comprises the following steps: obtaining coordinate information and rough classification information of a license plate in a target image based on a YOLO model, then obtaining a license plate region image of the license plate in the target image by using the coordinate information, correcting the license plate region image based on an STN model, and then obtaining a character recognition result in the corrected license plate region image by using an attention model; further, fine classification information of the license plate is determined based on the character recognition result and the coarse classification information. According to the method, the license plate region image is corrected through a deep learning method, the influence of the imaging quality of the license plate on license plate recognition is reduced, more accurate rough classification information and character recognition results are obtained, fine classification information is obtained through further processing on the basis, and the recognition rate of license plate recognition is improved.
Description
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to a license plate recognition method and apparatus, and an electronic device.
Background
The license plate is the 'ID card' of the vehicle and can uniquely indicate the vehicle. In the fields of public security monitoring, traffic management and the like, the license plate recognition technology is widely applied to obtain the license plate information of vehicles in scenes such as public security gates, entrances and exits, parking lots and the like. Most license plates in various regions in China are composed of numbers, letters and provinces, which are abbreviated as numbers and license plate purposes, and can be divided into a plurality of license plate types.
In the related technology, features of an image are generally extracted by a deep learning method, and then the features of the image are identified, so that license plate information in the image is obtained.
However, although the types of license plates are relatively fixed, the imaging quality of the license plates in different scenes is very different, and the imaging quality can greatly affect the recognition effect of the deep learning-based method. Therefore, in practical applications, the effect of license plate recognition is not ideal.
Disclosure of Invention
In view of this, the present application provides a license plate recognition method, a license plate recognition device, and an electronic device, which reduce the influence of imaging quality on recognition, and further process the license plate on the basis of deep learning to realize more accurate license plate recognition.
Specifically, the method is realized through the following technical scheme:
a license plate recognition method includes:
inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image;
acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, and inputting the license plate region image into a preset STN model so as to correct the license plate region image by the STN model;
inputting the corrected license plate region image into a trained attention model, and outputting a character recognition result of the license plate by the attention model based on the license plate region image;
and determining fine classification information of the license plate based on the character recognition result and the coarse classification information, and recognizing the license plate according to the fine classification information and the character recognition result.
In the license plate recognition method, before recognizing the license plate according to the fine classification information and the character recognition result, the method further includes:
correcting the character recognition result of the license plate based on a preset license plate inherent rule;
the recognizing the license plate according to the fine classification information and the character recognition result comprises the following steps:
and identifying the license plate according to the fine classification information and the corrected character identification result.
In the license plate recognition method, the recognizing the license plate according to the fine classification information and the character recognition result includes:
correcting color information of the license plate in the coarse classification information based on the character recognition result and the fine classification information;
and identifying the license plate according to the fine classification information, the character identification result and the corrected color information.
In the license plate recognition method, the determining fine classification information of the license plate based on the character recognition result and the coarse classification information includes:
calculating the confidence corresponding to the character recognition result of the license plate;
determining whether the confidence coefficient reaches a preset confidence coefficient threshold value;
and if so, determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate.
In the license plate recognition method, the coarse classification information is multi-classification information divided according to color information and the number of layers of license plate characters, and the fine classification information is a license plate classification actually divided based on relevant regulations;
the determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate comprises the following steps:
checking a preset fine classification information matching rule based on the character recognition result and the coarse classification information of the license plate; wherein, the fine classification information matching rule is a series of inherent rules for the fine classification of the license plate;
and determining the fine classification information corresponding to the matched fine classification information matching rule as the fine classification information of the license plate.
A license plate recognition device comprising:
the positioning classification unit is used for inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image;
the correcting unit is used for acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, inputting the license plate region image into a preset STN model, and correcting the license plate region image through the STN model;
the first recognition unit is used for inputting the corrected license plate region image into a trained attention model so as to output a character recognition result of the license plate based on the license plate region image through the attention model;
and the second identification unit is used for determining the fine classification information of the license plate based on the character identification result and the rough classification information and identifying the license plate according to the fine classification information and the character identification result.
In the license plate recognition device, the second recognition unit is further configured to:
correcting the character recognition result of the license plate based on a preset license plate inherent rule;
when the license plate is identified according to the fine classification information and the character identification result, the second identification unit is further configured to:
and identifying the license plate according to the fine classification information and the corrected character identification result.
In the license plate recognition device, the second recognition unit is further configured to:
correcting color information of the license plate in the coarse classification information based on the character recognition result and the fine classification information;
and identifying the license plate according to the fine classification information, the character identification result and the corrected color information.
In the license plate recognition device, the second recognition unit is further configured to:
calculating the confidence corresponding to the character recognition result of the license plate;
determining whether the confidence coefficient reaches a preset confidence coefficient threshold value;
and if so, determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate.
In the license plate recognition device, the coarse classification information is multi-classification information divided according to color information and the number of layers of license plate characters, and the fine classification information is a license plate classification actually divided based on relevant regulations;
the second identification unit is further configured to:
checking a preset fine classification information matching rule based on the character recognition result and the coarse classification information of the license plate; wherein, the fine classification information matching rule is a series of inherent rules for the fine classification of the license plate;
and determining the fine classification information corresponding to the matched fine classification information matching rule as the fine classification information of the license plate.
An electronic device comprising a memory, a processor, and machine-executable instructions stored on the memory and executable on the processor, wherein the processor when executing the machine-executable instructions implements a method comprising:
inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image;
acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, and inputting the license plate region image into a preset STN model so as to correct the license plate region image by the STN model;
inputting the corrected license plate region image into a trained attention model, and outputting a character recognition result of the license plate by the attention model based on the license plate region image;
and determining fine classification information of the license plate based on the character recognition result and the coarse classification information, and recognizing the license plate according to the fine classification information and the character recognition result.
In the embodiment of the application, a deep learning method is adopted in a plurality of steps, the STN model is used for correcting the license plate region image, the influence of the imaging quality of the license plate on license plate recognition is reduced, accurate rough classification information and character recognition results can be obtained, in addition, the rough classification information and the character recognition results are further processed to obtain fine classification information, and the inaccurate result caused by the overfitting of a network model when the fine classification information is directly obtained through the deep learning method is avoided.
Drawings
FIG. 1 is a schematic diagram of a license plate region image in a natural scene shown in the present application;
FIG. 2 is a flow chart of a license plate recognition method shown in the present application;
FIG. 3 is a schematic diagram of a license plate sub-category information shown in the present application;
FIG. 4 is a block diagram of an embodiment of a license plate recognition device shown in the present application;
fig. 5 is a hardware configuration diagram of an electronic device shown in the present application.
Detailed Description
In order to make the technical solutions in the embodiments of the present invention better understood and make the above objects, features and advantages of the embodiments of the present invention more comprehensible, the following description of the prior art and the technical solutions in the embodiments of the present invention with reference to the accompanying drawings is provided.
Most license plates in various regions in China can be divided into a plurality of fixed types, but the imaging quality of the license plates is different under different natural scenes. Fig. 1 is a schematic diagram of a license plate region image in a natural scene shown in the present application. As shown in fig. 1, there are many differences in the definition, size, and identifiability of the license plate in the license plate region image due to the influence of objective factors such as shooting angle, illumination, distance, and the like.
In the related art, the license plate recognition method based on deep learning mainly comprises the following parts: license plate positioning/vehicle positioning-license plate positioning, license plate region image correction, character segmentation, feature extraction and character recognition. Generally, the license plate recognition result can be obtained by processing the above links by using a deep learning method.
However, in the related art, features of an image are extracted by using a deep learning method, and then the features of the image are recognized, so as to obtain license plate information in the image. However, because the imaging quality difference of the license plate in a natural scene is very large, in the training process of the network model for license plate recognition, even if a huge number of image samples are used, the network model can be over-fitted, so that an accurate recognition result cannot be obtained.
In view of this, in the technical scheme of the application, the license plate region image is corrected in the process of identifying the target image based on the depth learning method, so that the influence of the imaging effect on the identification is reduced; in addition, after the coarse classification information and the character recognition result of the license plate in the target image are obtained based on the deep learning method, the coarse classification information and the character recognition result are further processed, and therefore the accuracy of license plate recognition is improved. The target image generally refers to any image used for license plate recognition.
Firstly, a network model for license plate recognition in the technical scheme of the application needs to be built.
In one embodiment, the network model applied in the present application may include a yolo (young Only lookone) model, an STN (Spatial Transformer Networks) model, and an AM (attention model) model.
The YOLO model is used for detecting a license plate in a target image and determining coordinate information and rough classification information of the license plate in the image. The above-mentioned rough classification information is a variety of kinds of information divided according to color information (including blue, yellow, white, green and black) and the number of layers of license plate characters (including single layer and double layer). In the present application, the rough classification information may include nine license plate categories, such as blue single layer, yellow double layer, white single layer, white double layer, green single layer, green double layer, black single layer, and others.
In the application, a neural network-based YOLO model can be constructed, and then a sample image is obtained. The sample image is marked with rough classification information and coordinate information of the license plate in the image, wherein the coordinate information comprises a center coordinate of a bounding box (bounding box) of the license plate and the width and height of the bounding box. And outputting the coarse classification information and the coordinate information of the license plate in the sample image by using the YOLO model, and then training the network parameters of the neural network according to the difference between the coarse classification information output by the YOLO model and the marked coarse classification information and the difference between the coordinate information output by the YOLO model and the marked coordinate information. And training the neural network through a certain number of sample images to obtain a YOLO model capable of realizing license plate positioning and rough classification.
The STN model is used for correcting the license plate region image extracted from the target image and aligning the license plate region image on the space, thereby reducing the influence of the license plate outline on license plate recognition due to the geometric transformation such as rotation, translation, distortion and the like on the space and solving the problems of license plate inclination or overlarge shooting angle and the like in the target image.
In the application, an STN model based on a neural network can be constructed, and then a sample image is obtained. The sample image is marked with the license plate in the image for affine transformation of six angle values. And training the Network parameters of the neural Network by utilizing the difference between six angle values obtained by calculating the sample image by using a positioning Network (localization Network) of the STN model and the six angle values of the marks. And training the neural network through a certain number of sample images to obtain the STN model capable of realizing license plate area image correction.
The AM model is used for coding and decoding the license plate region image so as to obtain a character recognition result of the license plate in the license plate region image.
In the application, an AM model based on a neural network can be established, and then a sample image is obtained, wherein the sample image is a license plate region image marked with character information in the image, and the character information marked by the sample image can comprise 7 to 10 characters because the number of characters of an actual license plate is not fixed. And training the network parameters of the neural network according to the character recognition result output by the AM model and the difference between the character recognition result and the marked character information, and training the AM model through a certain number of sample images so as to obtain the AM model capable of recognizing the characters in the license plate region image.
After the network model is built and trained, license plate recognition can be carried out based on the network model.
Referring to fig. 2, a flowchart of a license plate recognition method according to the present application is shown, and as shown in fig. 2, the method includes the following steps:
step 201: inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image.
The license plate recognition method can be applied to electronic equipment related to license plate recognition, such as monitoring equipment of a traffic gate or a background server of a traffic monitoring system.
In an embodiment, the target image may be input to a trained YOLO model, and the YOLO model may divide the target image into a plurality of meshes, and predict each mesh, so as to obtain coordinate information and rough classification information of the license plate in the target image.
The above YOLO model predicts B bounding boxes and the confidence of each bounding box for each mesh, and thus each bounding box includes 5 elements (x, y, w, h, c), where (x, y) represents the coordinates of the bounding box in the target image, w and h represent the width and height of the bounding box, and c represents the confidence of the bounding box. The boundary frame is the coordinate information of the license plate.
In addition, the probability values corresponding to each type of rough classification information are given to the B bounding boxes, and in the present application, 9 probability values are given to each bounding box, and correspond to one type of rough classification information respectively.
When the confidence of any bounding box reaches a preset confidence threshold and the probability value corresponding to one type of rough classification information reaches a preset probability value threshold, determining that the license plate exists in the bounding box, and determining that the rough classification information is the rough classification information of the license plate. The rough classification information is a plurality of types of information divided according to color information (including blue, yellow, white, green and black) and the number of layers of license plate characters (including single-layer and double-layer).
Step 202: acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, and inputting the license plate region image into a preset STN model so as to correct the license plate region image by the STN model.
In order to reduce the influence of the imaging quality of the license plate on the license plate recognition, the license plate region image can be input into the STN model, and the STN model corrects the license plate region image.
As an embodiment, in order to avoid the detection error of the YOLO model and ensure that a complete license plate region image is extracted, the license plate region image to be extracted may be determined after extending a specified distance to the four sides of the bounding box based on the bounding box where the license plate is located.
Such as: if the bounding box of the license plate is (x, y, w, h), the position of the license plate region image can be (x, y, 1.4 xw, 1.4 xh), which means that the image is one fifth extended to the four sides on the basis of the bounding box.
The STN model described above includes a positioning network, a Grid Generator (Grid Generator), and a Sampler (Sampler).
The positioning network can perform a series of convolution calculation and full connection processing on the license plate region image to obtain six angle values for affine transformation, and the six angle values form a 2 x 3 matrix.
And the grid generator calculates the coordinate position in the corrected license plate region image by using the matrix to obtain the coordinate position in the license plate region image before correction corresponding to each coordinate position in the corrected license plate region image.
Such as: the (0,1) and (0,2) … … in the corrected license plate region image correspond to (6,2) and (6,3) … … in the license plate region image before correction, respectively
Further, the sampler can obtain the values of the pixel points at each coordinate position from the license plate region image before correction by using the corresponding relationship, and then fill the values into the corresponding positions of the license plate region image after correction.
Such as: if (0,1) in the corrected license plate region image corresponds to (6,2) in the license plate region image before correction, the sampler may obtain a value of a pixel point at the (6,2) position of the license plate region image before correction, and then fill the pixel point at the (0,1) position of the license plate region image after correction.
Through the above processing, the STN model corrects the license plate region image.
Referring to fig. 1, after the license plate region image of the 1 st row and the 2 nd column is corrected, the imaging effect is similar to that of the license plate region image of the 2 nd row and the 3 rd column. Through correction, the problem that the license plate in the license plate region image is inclined or the shooting angle is too large can be solved, so that a subsequent AM model can obtain a better character recognition effect.
Step 203: inputting the corrected license plate region image into a trained attention model, and outputting a character recognition result of the license plate by the attention model based on the license plate region image.
In practical applications, before the corrected license plate region image is input to the attention model, the license plate region image may be compressed into an image feature sequence through convolution calculation, and then the image feature sequence is input to the attention model. The attention model can encode and decode the image feature sequence to obtain target information and then classify the target information to obtain the character recognition result of the license plate. Specifically, the encoding and decoding processes can refer to the related art, and are not described herein.
Step 204: and determining fine classification information of the license plate based on the character recognition result and the coarse classification information, and recognizing the license plate according to the fine classification information and the character recognition result.
The fine classification information is a license plate classification actually classified based on relevant regulations. Referring to fig. 3, a schematic diagram of the license plate fine classification information shown in the present application is provided, specifically, the fine classification information may include 20 kinds in total, such as a blue civil license plate, a yellow civil license plate, a black civil license plate, a yellow civil tail license plate, a 02-type license plate, a police license plate, a single-line military license plate, a double-line military license plate, a single-line armed police license plate, a double-line armed police license plate, an armed police headquarters license plate, a port and australian license plate, a coach license plate, an embassy license plate, a civil aviation license plate, an agricultural 1225 license plate, an agricultural 1325 license plate, a motorcycle front license plate, a motorcycle tail license plate, a new energy license.
After the character recognition result and the rough classification information of the license plate in the target image are obtained based on the deep learning method, the fine classification information of the license plate can be determined based on the character recognition result and the rough classification information. Or, the fine classification information of the license plate is determined directly based on the character recognition result.
In practical application, the network model may recognize other things in the natural scene as a license plate, and output corresponding rough classification information and character recognition results.
As an embodiment, in order to reduce the workload of license plate recognition, before determining the fine classification information of the license plate, a part of invalid license plates may be filtered according to the confidence degree corresponding to the character recognition result.
The character recognition result output by the attention model is a plurality of groups of characters and the confidence degree corresponding to each character in each group of characters. Wherein any one group of characters represents a possible recognition result of one character in the license plate.
Such as: if any license plate comprises 7 characters, the attention model outputs 7 groups of characters, and each character of each group of characters has a corresponding confidence coefficient. If the confidence of the first character in the first group of characters is 0.3, the probability that the character is actually the first character of the license plate is 0.3.
First, the confidence corresponding to the character recognition result of the license plate can be calculated. As an example, the maximum confidence in each set of characters may be accumulated and then averaged.
Next, it may be determined whether the confidence reaches a preset confidence threshold. The confidence threshold may be configured based on the actual application effect, and may be 0.65, for example.
On one hand, if the confidence coefficient does not reach the confidence coefficient threshold value, the license plate can be considered as a result of recognition error, and the license plate is stopped being recognized.
On the other hand, if the confidence reaches the confidence threshold, the license plate can be considered to exist in the target image, and the license plate can be further identified. In the application, the fine classification information of the license plate is determined.
In an embodiment shown, a preset fine classification information matching rule may be checked based on the character recognition result of the license plate and the rough classification information, and the fine classification information corresponding to the matched fine classification information matching rule may be determined as the fine classification information of the license plate.
The fine classification information matching rules are a series of inherent rules for fine classification of the license plate.
Such as: if the coarse classification information is 'white single-layer' and the last character of the character recognition result is 'alarm', the corresponding fine classification information is 'police license plate'.
And if the second character to the sixth character of the character recognition result are numbers, the corresponding fine classification information is the 'license plate of the embassy hall'.
And if the first character of the character recognition result is 'civil' and the third character to the sixth character are numbers or letters, the corresponding fine classification information is 'civil aviation license plate'.
If the character recognition result has eight bits and the rough classification information is 'green single layer', the corresponding fine classification information is 'new energy license plate'.
Of course, there are other fine classification information matching rules, which are not described herein.
After the fine classification information of the license plate is obtained, the license plate can be identified according to the fine classification information and the character identification result.
In addition, in order to obtain a more accurate recognition result, as an embodiment, the character recognition result of the license plate may be corrected based on a preset license plate intrinsic rule.
The license plate inherent rules are a series of inherent rules for correcting character recognition results.
Such as: if the character "ji" exists in the character recognition result, the first letter of the character recognition result cannot be "I". If the first letter of the character recognition result is "I", it may be corrected to "J".
If the character "ji" exists in the character recognition result, the first letter of the character recognition result cannot be "K". If the first letter of the character recognition result is "K", it may be corrected to "R".
If the character "ji" exists in the character recognition result, the first letter of the character recognition result cannot be "M" or "N". If the first letter of the character recognition result is "M" or "N", it may be corrected to "H".
If the character "promotion" is present in the character recognition result, the first letter of the character recognition result may not be "I". If the first letter of the character recognition result is "I", it may be corrected to "J".
If the character "promotion" is present in the character recognition result, the first letter of the character recognition result may not be "N". If the first letter of the character recognition result is "N", it may be corrected to "H".
If the character "Mongolia" exists in the character recognition result, the first letter of the character recognition result may not be "Q". If the first letter of the character recognition result is "Q", it may be corrected to D.
Of course, there are other rules inherent in the license plate and they are not described here.
And after the fine classification information of the license plate and the corrected character recognition result are obtained, the license plate can be recognized.
In one embodiment, the color information of the license plate may be corrected based on the character recognition result and the fine classification information of the license plate in order to enrich the recognition result.
Referring to fig. 3, the color information may be corrected according to the correspondence between the fine classification information and the color information in fig. 3.
Such as: if the fine classification information is 'blue civil license plate', the color information is 'blue'.
If the fine classification information is 'yellow civil license plate', the color information is 'yellow'.
If the fine classification information is 'police license plate', the color information is 'white'.
If the fine classification information is 'Hongkong and Australia license plate', the color information is 'black'.
And if the fine classification information is 'civil aviation license plate', the color information is 'green'.
And after the corrected color information is obtained, recognizing the license plate based on the fine classification information, the corrected character recognition result and the corrected color information.
And finally, outputting the license plate recognition result of the license plate. The license plate recognition result comprises a character recognition result and fine classification information. Optionally, the license plate recognition result may further include color information.
In summary, in the technical solution of the present application, coordinate information and rough classification information of a license plate are determined from a target image through a trained YOLO model, then a license plate region image of the license plate is obtained from the target image based on the coordinate information, and the license plate region image is corrected by using an STN model, so that characters can be recognized from the corrected license plate region image in the following process, and the influence of imaging quality on license plate recognition is reduced;
further, inputting the license plate region image into a trained attention model, so that the attention model encodes and decodes the license plate region image and outputs a character recognition result; after the coarse classification information and the character recognition result of the license plate in the target image are obtained through a deep learning method, fine classification information of the license plate can be determined based on the coarse classification information and the character recognition result, and in addition, the character recognition result can be corrected based on the inherent rule of the license plate, so that a more accurate character recognition result is obtained;
in addition, the coarse classification information and the character recognition result are further processed to obtain fine classification information, so that the condition that the result is inaccurate due to overfitting of a network model when the fine classification information is directly obtained by the deep learning method is avoided, and the recognition rate of license plate recognition can be further improved by correcting the character recognition result;
in addition, the license plate recognition result including the color information can be output, and the license plate recognition result is richer and is applicable to subsequent more flexible processing.
Corresponding to the embodiment of the license plate recognition method, the application also provides an embodiment of the license plate recognition device.
Referring to fig. 4, a block diagram of an embodiment of a license plate recognition device shown in the present application is shown:
as shown in fig. 4, the license plate recognition device 40 includes:
the positioning and classifying unit 410 is configured to input a target image into a trained YOLO model, so that the YOLO model detects a license plate in the target image, and outputs coordinate information and rough classification information of the license plate in the target image.
A correcting unit 420, configured to obtain a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, and input the license plate region image to a preset STN model, so that the license plate region image is corrected by the STN model.
The first recognition unit 430 is configured to input the corrected license plate region image to a trained attention model, so that the attention model outputs a character recognition result of the license plate based on the license plate region image.
A second identifying unit 440, configured to determine fine classification information of the license plate based on the character recognition result and the coarse classification information, and identify the license plate according to the fine classification information and the character recognition result.
In this example, the second identifying unit 440 is further configured to:
correcting the character recognition result of the license plate based on a preset license plate inherent rule;
when the license plate is identified according to the fine classification information and the character identification result, the second identification unit 440 is further configured to:
and identifying the license plate according to the fine classification information and the corrected character identification result.
In this example, the second identifying unit 440 is further configured to:
correcting color information of the license plate in the coarse classification information based on the character recognition result and the fine classification information;
and identifying the license plate according to the fine classification information, the character identification result and the corrected color information.
In this example, the second identifying unit 440 is further configured to:
calculating the confidence corresponding to the character recognition result of the license plate;
determining whether the confidence coefficient reaches a preset confidence coefficient threshold value;
and if so, determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate.
In this example, the rough classification information is various classification information divided according to color information and the number of layers of license plate characters, and the fine classification information is a license plate classification actually divided based on relevant regulations;
the second identifying unit 440 is further configured to:
checking a preset fine classification information matching rule based on the character recognition result and the coarse classification information of the license plate; wherein, the fine classification information matching rule is a series of inherent rules for the fine classification of the license plate;
and determining the fine classification information corresponding to the matched fine classification information matching rule as the fine classification information of the license plate.
The embodiment of the license plate recognition device can be applied to electronic equipment. The device embodiments may be implemented by software, or by hardware, or by a combination of hardware and software. In the case of software implementation, as a logical device, a processor of the electronic device reads corresponding machine executable instructions in the machine readable storage medium into the memory for execution.
From a hardware level, as shown in fig. 5, a hardware structure diagram of an electronic device where the license plate recognition apparatus of the present application is located may include a processor 501 and a machine-readable storage medium 502 storing machine-executable instructions. The processor 501 and the machine-readable storage medium 502 may communicate via a system bus 503. The processor 501 is capable of implementing the license plate recognition described above by loading and executing machine executable instructions stored by the machine-readable storage medium 502.
The machine-readable storage medium 502 referred to herein may be any electronic, magnetic, optical, or other physical storage device that can contain or store information such as executable instructions, data, and the like. For example, the machine-readable storage medium may be: a RAM (random Access Memory), a volatile Memory, a non-volatile Memory, a flash Memory, a storage drive (e.g., a hard drive), a solid state drive, any type of storage disk (e.g., an optical disk, a dvd, etc.), or similar storage medium, or a combination thereof.
The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the application. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only exemplary of the present application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the scope of protection of the present application.
Claims (11)
1. A license plate recognition method is characterized by comprising the following steps:
inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image;
acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, and inputting the license plate region image into a preset STN model so as to correct the license plate region image by the STN model;
inputting the corrected license plate region image into a trained attention model, and outputting a character recognition result of the license plate by the attention model based on the license plate region image;
and determining fine classification information of the license plate based on the character recognition result and the coarse classification information, and recognizing the license plate according to the fine classification information and the character recognition result.
2. The method of claim 1, wherein before identifying the license plate according to the fine classification information and the character recognition result, the method further comprises:
correcting the character recognition result of the license plate based on a preset license plate inherent rule;
the recognizing the license plate according to the fine classification information and the character recognition result comprises the following steps:
and identifying the license plate according to the fine classification information and the corrected character identification result.
3. The method of claim 1, wherein the recognizing the license plate according to the fine classification information and the character recognition result comprises:
correcting color information of the license plate in the coarse classification information based on the character recognition result and the fine classification information;
and identifying the license plate according to the fine classification information, the character identification result and the corrected color information.
4. The method of claim 1, wherein the determining fine classification information of the license plate based on the character recognition result and the coarse classification information comprises:
calculating the confidence corresponding to the character recognition result of the license plate;
determining whether the confidence coefficient reaches a preset confidence coefficient threshold value;
and if so, determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate.
5. The method according to claim 1 or 4, wherein the rough classification information is multi-classification information divided according to color information and the number of layers of license plate characters, and the fine classification information is a license plate classification actually divided based on relevant regulations;
the determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate comprises the following steps:
checking a preset fine classification information matching rule based on the character recognition result and the coarse classification information of the license plate; wherein, the fine classification information matching rule is a series of inherent rules for the fine classification of the license plate;
and determining the fine classification information corresponding to the matched fine classification information matching rule as the fine classification information of the license plate.
6. A license plate recognition device, comprising:
the positioning classification unit is used for inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image;
the correcting unit is used for acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, inputting the license plate region image into a preset STN model, and correcting the license plate region image through the STN model;
the first recognition unit is used for inputting the corrected license plate region image into a trained attention model so as to output a character recognition result of the license plate based on the license plate region image through the attention model;
and the second identification unit is used for determining the fine classification information of the license plate based on the character identification result and the rough classification information and identifying the license plate according to the fine classification information and the character identification result.
7. The apparatus of claim 6, wherein the second identification unit is further configured to:
correcting the character recognition result of the license plate based on a preset license plate inherent rule;
when the license plate is identified according to the fine classification information and the character identification result, the second identification unit is further configured to:
and identifying the license plate according to the fine classification information and the corrected character identification result.
8. The apparatus of claim 6, wherein the second identification unit is further configured to:
correcting color information of the license plate in the coarse classification information based on the character recognition result and the fine classification information;
and identifying the license plate according to the fine classification information, the character identification result and the corrected color information.
9. The apparatus of claim 6, wherein the second identification unit is further configured to:
calculating the confidence corresponding to the character recognition result of the license plate;
determining whether the confidence coefficient reaches a preset confidence coefficient threshold value;
and if so, determining the fine classification information of the license plate based on the character recognition result and the coarse classification information of the license plate.
10. The apparatus according to claim 6 or 9, wherein the rough classification information is multi-classification information divided according to color information and the number of layers of license plate characters, and the fine classification information is a license plate classification actually divided based on relevant regulations;
the second identification unit is further configured to:
checking a preset fine classification information matching rule based on the character recognition result and the coarse classification information of the license plate; wherein, the fine classification information matching rule is a series of inherent rules for the fine classification of the license plate;
and determining the fine classification information corresponding to the matched fine classification information matching rule as the fine classification information of the license plate.
11. An electronic device comprising a memory, a processor, and machine-executable instructions stored on the memory and executable on the processor, wherein the processor when executing the machine-executable instructions implements a method comprising:
inputting a target image into a trained YOLO model, detecting a license plate in the target image by the YOLO model, and outputting coordinate information and rough classification information of the license plate in the target image;
acquiring a license plate region image of the license plate from the target image based on the coordinate information of the license plate in the target image, and inputting the license plate region image into a preset STN model so as to correct the license plate region image by the STN model;
inputting the corrected license plate region image into a trained attention model, and outputting a character recognition result of the license plate by the attention model based on the license plate region image;
and determining fine classification information of the license plate based on the character recognition result and the coarse classification information, and recognizing the license plate according to the fine classification information and the character recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811133867.9A CN110956169B (en) | 2018-09-27 | 2018-09-27 | License plate recognition method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811133867.9A CN110956169B (en) | 2018-09-27 | 2018-09-27 | License plate recognition method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110956169A true CN110956169A (en) | 2020-04-03 |
| CN110956169B CN110956169B (en) | 2024-09-06 |
Family
ID=69975155
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811133867.9A Active CN110956169B (en) | 2018-09-27 | 2018-09-27 | License plate recognition method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110956169B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539337A (en) * | 2020-04-26 | 2020-08-14 | 上海眼控科技股份有限公司 | Vehicle posture correction method, device and equipment |
| CN111860539A (en) * | 2020-07-20 | 2020-10-30 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
| CN111881741A (en) * | 2020-06-22 | 2020-11-03 | 浙江大华技术股份有限公司 | License plate recognition method and device, computer equipment and computer-readable storage medium |
| CN112052845A (en) * | 2020-10-14 | 2020-12-08 | 腾讯科技（深圳）有限公司 | Image recognition method, device, equipment and storage medium |
| CN112132139A (en) * | 2020-09-22 | 2020-12-25 | 深兰科技（上海）有限公司 | Character recognition method and device |
| CN112381129A (en) * | 2020-11-10 | 2021-02-19 | 浙江大华技术股份有限公司 | License plate classification method and device, storage medium and electronic equipment |
| CN112508018A (en) * | 2020-12-14 | 2021-03-16 | 北京澎思科技有限公司 | License plate recognition method and device and storage medium |
| CN112560586A (en) * | 2020-11-27 | 2021-03-26 | 国家电网有限公司大数据中心 | Method and device for obtaining structured data of pole and tower signboard and electronic equipment |
| CN113038001A (en) * | 2021-02-26 | 2021-06-25 | 维沃移动通信有限公司 | Display method and device and electronic equipment |
| CN113128507A (en) * | 2021-05-12 | 2021-07-16 | 锐捷网络股份有限公司 | License plate recognition method and device, electronic equipment and storage medium |
| CN113505790A (en) * | 2021-07-15 | 2021-10-15 | 普联技术有限公司 | License plate processing method, device and equipment and computer readable storage medium |
| CN113591835A (en) * | 2020-04-30 | 2021-11-02 | 成都鼎桥通信技术有限公司 | Method and device for detecting key information of license plate |
| CN113674329A (en) * | 2021-08-13 | 2021-11-19 | 上海同温层智能科技有限公司 | Vehicle driving behavior detection method and system |
| CN114973230A (en) * | 2022-06-27 | 2022-08-30 | 成都佳华物链云科技有限公司 | License plate detection method, electronic equipment and computer readable storage medium |
| CN116453071A (en) * | 2023-04-17 | 2023-07-18 | 北京睿芯通量科技发展有限公司 | Identification method and device of vehicle attribute information, electronic equipment and storage medium |
| CN117173686A (en) * | 2023-07-07 | 2023-12-05 | 深圳市美通视讯科技有限公司 | Novel motor vehicle license plate detection and identification method, system and readable storage medium |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006163446A (en) * | 2004-12-02 | 2006-06-22 | Canon Software Inc | Information processing apparatus, data retrieval method, program, and recording medium |
| CN102693431A (en) * | 2012-05-31 | 2012-09-26 | 信帧电子技术（北京）有限公司 | Method and device for identifying type of white number plate |
| CN102722733A (en) * | 2012-05-31 | 2012-10-10 | 信帧电子技术（北京）有限公司 | Identification method and device of license plate types |
| CN104050450A (en) * | 2014-06-16 | 2014-09-17 | 西安通瑞新材料开发有限公司 | Vehicle license plate recognition method based on video |
| CN104464302A (en) * | 2014-12-22 | 2015-03-25 | 南京中兴力维软件有限公司 | License plate recognition and intelligent error correction method and system |
| CN105118300A (en) * | 2015-08-28 | 2015-12-02 | 浙江工业大学 | Method of correcting license plate number automatically recognized in traffic snapshot |
| CN105224962A (en) * | 2014-07-03 | 2016-01-06 | 浙江宇视科技有限公司 | A kind of similar vehicle license plate extraction method and device |
| CN105528139A (en) * | 2015-12-04 | 2016-04-27 | 上海博康智能信息技术有限公司 | License plate number identification result correction method based on full mouse operation |
| CN106203418A (en) * | 2016-07-14 | 2016-12-07 | 北京精英智通科技股份有限公司 | A kind of method and device of car plate detection |
| CN106650553A (en) * | 2015-10-30 | 2017-05-10 | 比亚迪股份有限公司 | License plate recognition method and system |
| US20170249524A1 (en) * | 2016-02-25 | 2017-08-31 | Xerox Corporation | Method and system for detection-based segmentation-free license plate recognition |
| CN107679531A (en) * | 2017-06-23 | 2018-02-09 | 平安科技（深圳）有限公司 | Licence plate recognition method, device, equipment and storage medium based on deep learning |
| CN108009543A (en) * | 2017-11-29 | 2018-05-08 | 深圳市华尊科技股份有限公司 | A kind of licence plate recognition method and device |
| CN108108729A (en) * | 2016-11-25 | 2018-06-01 | 杭州海康威视数字技术股份有限公司 | A kind of recognition methods of car plate type and device |
| CN108257602A (en) * | 2018-01-30 | 2018-07-06 | 海信集团有限公司 | License plate number character string antidote, device, server and terminal |
| US20180211117A1 (en) * | 2016-12-20 | 2018-07-26 | Jayant Ratti | On-demand artificial intelligence and roadway stewardship system |
| JP2018120445A (en) * | 2017-01-26 | 2018-08-02 | 株式会社 日立産業制御ソリューションズ | Car number recognition apparatus |
-
2018
- 2018-09-27 CN CN201811133867.9A patent/CN110956169B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006163446A (en) * | 2004-12-02 | 2006-06-22 | Canon Software Inc | Information processing apparatus, data retrieval method, program, and recording medium |
| CN102693431A (en) * | 2012-05-31 | 2012-09-26 | 信帧电子技术（北京）有限公司 | Method and device for identifying type of white number plate |
| CN102722733A (en) * | 2012-05-31 | 2012-10-10 | 信帧电子技术（北京）有限公司 | Identification method and device of license plate types |
| CN104050450A (en) * | 2014-06-16 | 2014-09-17 | 西安通瑞新材料开发有限公司 | Vehicle license plate recognition method based on video |
| CN105224962A (en) * | 2014-07-03 | 2016-01-06 | 浙江宇视科技有限公司 | A kind of similar vehicle license plate extraction method and device |
| CN104464302A (en) * | 2014-12-22 | 2015-03-25 | 南京中兴力维软件有限公司 | License plate recognition and intelligent error correction method and system |
| CN105118300A (en) * | 2015-08-28 | 2015-12-02 | 浙江工业大学 | Method of correcting license plate number automatically recognized in traffic snapshot |
| CN106650553A (en) * | 2015-10-30 | 2017-05-10 | 比亚迪股份有限公司 | License plate recognition method and system |
| CN105528139A (en) * | 2015-12-04 | 2016-04-27 | 上海博康智能信息技术有限公司 | License plate number identification result correction method based on full mouse operation |
| US20170249524A1 (en) * | 2016-02-25 | 2017-08-31 | Xerox Corporation | Method and system for detection-based segmentation-free license plate recognition |
| CN106203418A (en) * | 2016-07-14 | 2016-12-07 | 北京精英智通科技股份有限公司 | A kind of method and device of car plate detection |
| CN108108729A (en) * | 2016-11-25 | 2018-06-01 | 杭州海康威视数字技术股份有限公司 | A kind of recognition methods of car plate type and device |
| US20180211117A1 (en) * | 2016-12-20 | 2018-07-26 | Jayant Ratti | On-demand artificial intelligence and roadway stewardship system |
| JP2018120445A (en) * | 2017-01-26 | 2018-08-02 | 株式会社 日立産業制御ソリューションズ | Car number recognition apparatus |
| CN107679531A (en) * | 2017-06-23 | 2018-02-09 | 平安科技（深圳）有限公司 | Licence plate recognition method, device, equipment and storage medium based on deep learning |
| CN108009543A (en) * | 2017-11-29 | 2018-05-08 | 深圳市华尊科技股份有限公司 | A kind of licence plate recognition method and device |
| CN108257602A (en) * | 2018-01-30 | 2018-07-06 | 海信集团有限公司 | License plate number character string antidote, device, server and terminal |
Non-Patent Citations (7)
| Title |
|---|
| JIANLIN WANG 等: "Sequence recognition of Chinese license plates", NEUROCOMPUTING, 20 August 2018 (2018-08-20), pages 149 - 158, XP085470992, DOI: 10.1016/j.neucom.2018.08.023 * |
| SÉRGIO MONTAZZOLLI 等: "Real-Time Brazilian License Plate Detection and Recognition Using Deep Convolutional Neural Networks", 《2017 30TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI)》 * |
| SÉRGIO MONTAZZOLLI 等: "Real-Time Brazilian License Plate Detection and Recognition Using Deep Convolutional Neural Networks", 《2017 30TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI)》, 7 November 2017 (2017-11-07), pages 55 - 62 * |
| 张世乾: "基于改进SVM的矿用地磅车牌识别技术的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 张世乾: "基于改进SVM的矿用地磅车牌识别技术的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, 15 April 2014 (2014-04-15), pages 138 - 841 * |
| 李珺 等: "基于Hausdorff距离的车牌字符识别方法研究", 《甘肃科学学报》 * |
| 李珺 等: "基于Hausdorff距离的车牌字符识别方法研究", 《甘肃科学学报》, 30 June 2009 (2009-06-30), pages 144 - 147 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539337A (en) * | 2020-04-26 | 2020-08-14 | 上海眼控科技股份有限公司 | Vehicle posture correction method, device and equipment |
| CN113591835A (en) * | 2020-04-30 | 2021-11-02 | 成都鼎桥通信技术有限公司 | Method and device for detecting key information of license plate |
| CN111881741A (en) * | 2020-06-22 | 2020-11-03 | 浙江大华技术股份有限公司 | License plate recognition method and device, computer equipment and computer-readable storage medium |
| CN111860539A (en) * | 2020-07-20 | 2020-10-30 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
| CN111860539B (en) * | 2020-07-20 | 2024-05-10 | 济南博观智能科技有限公司 | License plate color recognition method, device and medium |
| CN112132139A (en) * | 2020-09-22 | 2020-12-25 | 深兰科技（上海）有限公司 | Character recognition method and device |
| CN112052845A (en) * | 2020-10-14 | 2020-12-08 | 腾讯科技（深圳）有限公司 | Image recognition method, device, equipment and storage medium |
| CN112381129A (en) * | 2020-11-10 | 2021-02-19 | 浙江大华技术股份有限公司 | License plate classification method and device, storage medium and electronic equipment |
| CN112560586A (en) * | 2020-11-27 | 2021-03-26 | 国家电网有限公司大数据中心 | Method and device for obtaining structured data of pole and tower signboard and electronic equipment |
| CN112560586B (en) * | 2020-11-27 | 2024-05-10 | 国家电网有限公司大数据中心 | Method and device for obtaining structural data of pole and tower signboard and electronic equipment |
| CN112508018A (en) * | 2020-12-14 | 2021-03-16 | 北京澎思科技有限公司 | License plate recognition method and device and storage medium |
| CN113038001A (en) * | 2021-02-26 | 2021-06-25 | 维沃移动通信有限公司 | Display method and device and electronic equipment |
| CN113128507B (en) * | 2021-05-12 | 2023-03-21 | 锐捷网络股份有限公司 | License plate recognition method and device, electronic equipment and storage medium |
| CN113128507A (en) * | 2021-05-12 | 2021-07-16 | 锐捷网络股份有限公司 | License plate recognition method and device, electronic equipment and storage medium |
| CN113505790A (en) * | 2021-07-15 | 2021-10-15 | 普联技术有限公司 | License plate processing method, device and equipment and computer readable storage medium |
| CN113674329A (en) * | 2021-08-13 | 2021-11-19 | 上海同温层智能科技有限公司 | Vehicle driving behavior detection method and system |
| CN114973230A (en) * | 2022-06-27 | 2022-08-30 | 成都佳华物链云科技有限公司 | License plate detection method, electronic equipment and computer readable storage medium |
| CN116453071A (en) * | 2023-04-17 | 2023-07-18 | 北京睿芯通量科技发展有限公司 | Identification method and device of vehicle attribute information, electronic equipment and storage medium |
| CN117173686A (en) * | 2023-07-07 | 2023-12-05 | 深圳市美通视讯科技有限公司 | Novel motor vehicle license plate detection and identification method, system and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110956169B (en) | 2024-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110956169A (en) | License plate recognition method and device and electronic equipment | |
| CN108596166B (en) | Container number identification method based on convolutional neural network classification | |
| CN105512660B (en) | License plate number identification method and device | |
| CN112686812B (en) | Bank card inclination correction detection method and device, readable storage medium and terminal | |
| CN113591967B (en) | Image processing method, device, equipment and computer storage medium | |
| CN108268867B (en) | License plate positioning method and device | |
| CN111191611B (en) | Traffic sign label identification method based on deep learning | |
| CN105787466B (en) | A kind of fine recognition methods and system of type of vehicle | |
| CN107818322A (en) | A kind of vehicle VIN code tampering detection system and methods for vehicle annual test | |
| CN113947766B (en) | Real-time license plate detection method based on convolutional neural network | |
| CN113128507B (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN112052845A (en) | Image recognition method, device, equipment and storage medium | |
| CN106897681A (en) | A kind of remote sensing images comparative analysis method and system | |
| CN108052929A (en) | Parking space state detection method, system, readable storage medium storing program for executing and computer equipment | |
| CN113658192A (en) | Multi-target pedestrian track acquisition method, system, device and medium | |
| US11574482B2 (en) | Bidirectional statistical consolidation of multi-modal grouped observations | |
| CN115880683B (en) | Urban waterlogging ponding intelligent water level detection method based on deep learning | |
| CN113642577A (en) | Low-contrast license plate recognition method, system, equipment and storage medium | |
| CN109523570B (en) | Motion parameter calculation method and device | |
| CN112329886A (en) | Double-license plate recognition method, model training method, device, equipment and storage medium | |
| CN112699711B (en) | Lane line detection method and device, storage medium and electronic equipment | |
| CN112784494B (en) | Training method of false positive recognition model, target recognition method and device | |
| CN111402185A (en) | Image detection method and device | |
| CN114724119B (en) | Lane line extraction method, lane line detection device, and storage medium | |
| CN116563786A (en) | TEDS jumper fault identification detection method, storage medium and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |