CN104156704A - Novel license plate identification method and system - Google Patents
Novel license plate identification method and system Download PDFInfo
- Publication number
- CN104156704A CN104156704A CN201410378116.9A CN201410378116A CN104156704A CN 104156704 A CN104156704 A CN 104156704A CN 201410378116 A CN201410378116 A CN 201410378116A CN 104156704 A CN104156704 A CN 104156704A
- Authority
- CN
- China
- Prior art keywords
- image
- character
- license plate
- row
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The invention provides a novel license plate identification method and a system and particularly relates to an integral license plate identification system which utilizes various image processing algorisms to complete license plate image pretreatment and the final license plate character identification. The identification method comprises the following steps: in license plate positioning, the filtering technique based on the math morphology is adopted; in license plate character segmentation and extraction, the methods such as perpendicular projection are adopted; lastly, the license plate characters are indentified through the template matching technique based on multiple-characteristic extraction. Compared with the prior art, the novel license plate identification method and the system can identify the license plate quickly, robustly and accurately.
Description
Technical field
The present invention relates to image recognition technology field, especially a kind of new licence plate recognition method and system.
Background technology
In recent years, in the time of along with the fast development of China Transportation Industry, also brought the problems such as the traffic hazard that takes place frequently, traffic congestion, traffic pollution, setting up efficient intelligent transport system ITS (Intelligent Transportation System) has become current problem demanding prompt solution.Vehicle license is the tag mark of the identity of the external sign automobile of current unique energy, thereby vehicle license recognition system LPRS (License Plate Recognition System) is one of technology of most critical in ITS system.
Vehicle License Plate Recognition System is comprised of car plate location, License Plate Segmentation and three key modules of character recognition, and three large modules are carried out successively.Car plate location is that car plate is extracted from whole vehicle image, License Plate Segmentation is that license plate image is divided into the process of independent character one by one, character recognition is exactly the characteristic of extracting character, adopts the recognition methods of template matches or neural network to identify, and provides recognition result.Any one the module error that forms Vehicle License Plate Recognition System all can affect the performance of system, the development of the technology such as image processing and computer vision, impelled the future development of Vehicle License Plate Recognition System from theory to practical application, but in actual applications, Vehicle License Plate Recognition System is bad at light, in the situations such as car plate poor quality, versatility and robustness are poor, thereby Vehicle License Plate Recognition System is except the good vehicle image of needs shooting quality, also needs further to study and improve the accuracy of modules in Vehicle License Plate Recognition System.
To late 1990s, foreign study mechanism and company start systematically the technology of car plate identification to be researched and developed to application, a lot of recognizers have been proposed, wherein practical, robustness is good, and the Vehicle License Plate Recognition System that performance is good starts the electronic charging for parking lot, the discrepancy vehicle control at crossing, the real-time monitoring of road running vehicle etc.The development and production of a lot of companies the product of car plate identification aspect, as the VPLRS system of the Insignia system of Singapore Zamir company exploitation and the research and development of Optasia company, the See/Car System series that Israel Hi.Tech company produces, these products have very strong specific aim, as VLPRS product, mainly for Singapore's car plate used, identify, car plate recognition effect to other countries is poor, Insignia system is that the car plate feature adopting according to European and the Far East Area is researched and developed specific Vehicle License Plate Recognition System, although the See/Car System of Hi.Tech company research and development has developed corresponding recognition system according to the car plate feature of country variant, but the accuracy rate of See/Car System series identification Chinese character is very low, can not be for China's Vehicle License Plate Recognition System.
The relatively external starting of China is more late, but development rapidly aspect license plate recognition technology research, China has obtained good development at short notice, and domestic many companies also research and develop the product of having produced car plate identification aspect, mainly comprise: Shenzhen Ke Anxin Industrial Co., Ltd., Gaodewei Intelligent Traffic System Co., Ltd., Shanghai, Chinese automation research Han Wang company and the Asia Visual Co.,Ltd of Beijing The Orchid Pavilion San He scientific & technical corporation and Hong-Kong etc.The Vehicle License Plate Recognition System that at present domestic these companies produce tilts at license plate image, have stain, in the situation such as worn and torn, discrimination is not high, poor to environmental suitability.Therefore, still need at present license plate recognition technology to further investigate, exploitation is applicable to practical, robustness is good licence plate recognition method and system.
Summary of the invention
For the existing problem of current Vehicle License Plate Recognition System, the present invention proposes a kind ofly realize fast, licence plate recognition method and the system of robust and accurate car plate identification.
The technical solution adopted for the present invention to solve the technical problems is:
A new licence plate recognition method, comprises the steps:
A) license plate image pre-service, the binary image of acquisition car plate;
B) car plate location: first utilize vertical Sobel edge detection algorithm pretreated binary image to be carried out to the detection of vertical edge, recycling Mathematical Morphology Method edge image carries out the coarse positioning of license plate area, finally utilizes projecting method to realize the accurate location of license plate area;
C) License Plate Character Segmentation: first the frame of license plate area is removed, and then utilized vertical projection method to carry out separating character;
D) Recognition of License Plate Characters: adopt the template matches recognition methods based on many features, first character picture is normalized, then extract contour feature, the projection properties of character picture, mate contrast with the feature of respective symbols in template base, to reach accurate identification.
The present invention also provides a kind of new Vehicle License Plate Recognition System, comprises license plate image pretreatment module, for to existing the image of car plate to carry out image pre-service, obtains the binary image of car plate; Car plate locating module, for realizing the accurate location of license plate area based on mathematical morphology filter technology; License Plate Character Segmentation module, for utilizing vertical projection method that each character of the license plate area image after having good positioning is extracted and becomes single character; Recognition of License Plate Characters module, for adopting the template matches recognition methods based on many features accurately to identify characters on license plate.
The present invention utilizes multiple image processing algorithm to complete whole car plate identification and identifies this whole Vehicle License Plate Recognition System from pre-service car plate to the last, in car plate location, has adopted based on mathematical morphology filter technology; In License Plate Character Segmentation extracts, the methods such as vertical projection have been used; Finally by the template matching technique based on many feature extractions, completed Recognition of License Plate Characters, this licence plate recognition method and system can realize fast, robust and the identification of accurate car plate.
Embodiment
In the embodiment of the present invention, new licence plate recognition method comprises the committed steps such as license plate image pre-service, car plate location, License Plate Character Segmentation extraction and Recognition of License Plate Characters, narration one by one below.
A) license plate image pre-service
In Vehicle License Plate Recognition System, the image obtaining due to video camera shooting, collecting is often subject to many factors to be caused picture quality cannot be directly used in system processing, therefore need to be first to existing the image of car plate to carry out image pre-service.After pre-service, can remove noise in image, strengthen image key message, also can reduce image processing complexity, can directly improve robustness and the accuracy of whole Vehicle License Plate Recognition System.License plate image pre-service work is mainly divided into following three steps.
Step 1, gray processing: generally, by camera collection to vehicle image be coloured image, due to the speed that image is processed in the large and impact of storage space that takies of color document image, therefore, first need coloured image gray processing to process, be generally shown below:
(1)
In formula (1),
be respectively pixel
the intensity level of the red, green, blue look at place,
for the gray-scale value after gray processing.
Step 2, figure image intensifying: in order further to improve the quality of license plate area in image, make its edge more clear, the present invention has adopted the image enchancing method based on Laplce's sharpening mask.Its principle formula is:
(2)
In formula (2),
for the gray level image of input,
for strengthening rear image,
for the Laplace operator of input picture, it is defined as:
(3)
Generally can utilize Laplace operator matrix
carrying out convolution with input picture obtained above-mentioned afterwards
, therefore, can utilize sharpening mask matrix
carry out can obtaining final enhancing image after convolution with input picture
.
Step 3, image binaryzation: the process of image binaryzation is that gradation of image value is set to 0 or 1 process, make whole image present non-black be white effect.Image binaryzation can make successive image process simplification, and the data volume of processing reduces.The key of Binary Sketch of Grey Scale Image is choosing of image threshold, and threshold value is chosen and too highly can be lost marginal information, too lowly can produce false edge.The present invention has invented a kind of simple adapting to image Binarization methods, sees shown in following formula:
(4)
In formula (4),
for optimal threshold,
the maximum gradation value of the enhancing image that expression obtains,
for strengthening the minimum gradation value of image.Final binary image can obtain by following formula:
(5)
Input in above formula (5)
be the image after enhancing,
for final binary image.
B) car plate location
Car plate location is the key link of Vehicle License Plate Recognition System, and accurate car plate location is the basic premise of follow-up License Plate Character Segmentation and identification, therefore its accuracy is related to the accuracy rate of whole Vehicle License Plate Recognition System.Due to the impact of the factors such as environment, license plate image quality declines to some extent, and in image, also has the interference in similar characters on license plate region, and these have all increased the difficulty of locating true license plate area greatly.First the present invention utilizes vertical Sobel edge detection algorithm pretreated binary image to be carried out to the detection of vertical edge, recycling Mathematical Morphology Method edge image carries out the coarse positioning of license plate area, finally utilize projecting method to realize the accurate location of license plate area, be specially following three steps.
Step 1, vertical edge extract: image border is the key character that in image, pixel grey scale has step to change.The character of license plate area and background just have obvious edge, and the number at edge is also a lot, this is one of fundamental difference of other non-license plate areas in license plate area and input picture, and according to the observation, the character of car plate be take vertical edge mostly as main, therefore, the present invention is based on vertical Sobel edge detection algorithm and extract the vertical edge in image, namely utilize the vertical convolution kernel of the Sobel shown in following table to come in the binary image with input each pixel to carry out convolution algorithm and obtained afterwards vertical edge image:
| -1 | 0 | 1 |
| -2 | 0 | 2 |
| -1 | 0 | 1 |
The vertical convolution kernel of table 1 Sobel
Step 2, mathematical morphology filter: after having obtained vertical edge binary image, then adopt mathematical morphology filter method fortune to convert image, give prominence to needed image information.Concrete grammar is exactly to adopt structural elements and the binary image possess required geometric shape to carry out set operation, and first the present invention uses closed operation, re-uses opening operation, finally obtains image after needed morphologic filtering, and concrete steps are as follows:
(1) first to vertical edge binary image
carry out structural elements
closed operation, i.e. structural elements
being defined as is highly 5 pixels, and length is the rectangle of 19 pixels.Closed operation is to use same structural element to first the expand computing of post-etching of image, therefore, and target image
in structural elements
under closed operation be defined as follows:
(6)
In formula (6),
be defined as expansive working, for given target image
and structural elements
, will
middle every bit
expand as
operation; And
be defined as corrosion operation, meet
point
all formation structural elements
with target image
maximal correlation point set.
(2) again to carrying out the image after closed operation
carry out structural elements
opening operation, i.e. structural elements
being defined as is highly 5 pixels, and length is the rectangle of 19 pixels.Opening operation is to use same structural element image first to be corroded to the computing of rear expansion, therefore, and target image
in structural elements
under opening operation be defined as follows:
(7)
In formula (7),
for expansive working, and
for corrosion operation.
Step 3, according to car plate priori, determine license plate area: after having completed mathematical morphology filter, the candidate region of only depositing in target image is rectangle, the candidate region that the present invention utilizes the priori of the car plate of China only to deposit from these, determines real license plate area.At present, the car plate length breadth ratio of China is all 3.14, and size is all 440 * 140 (mm), utilizes this priori to carry out length and width signature analysis to candidate region, thereby finally obtains real license plate area.Specific practice is: first, calculate the length breadth ratio of each candidate's rectangular area, if length breadth ratio is defined as possible license plate area between 2.5 and 4.5, if there are a plurality of possible license plate areas, measure the width in this region and actual width and the length that length judges whether to meet car plate, so just can finally determine unique real license plate area, last according to extracted region apex coordinate, by real license plate area, from original input gray level input picture, cutting is out.
C) License Plate Character Segmentation
The task of License Plate Character Segmentation is that each character in the license plate area image after having good positioning is extracted and becomes single character, and correctly cutting apart characters on license plate is the committed step before character recognition.First the present invention removes the frame of license plate area, and then utilizes vertical projection method to carry out separating character.Concrete steps are as follows:
Step 1, removal license plate area frame: in the license plate area image after having completed car plate location, not only have frame, also have rivet, therefore, before carrying out Character segmentation, need to remove frame and rivet up and down, obtain pure character zone.The present invention removes upper and lower side frame and rivet according to the saltus step rule of character, and specific practice is:
(1) removing upper and lower side frame is mainly every row pixel value transition times in statistics license plate area, establishes saltus step threshold value and is
;
(2) from the middle row of the license plate area image scan statistics that makes progress, when running into every row transition times for the first time, be less than threshold value
time, stop, getting the next line position that the upper bound of character zone is this row;
(3) from the downward scan statistics of middle row of license plate area image, when running into every row transition times for the first time, be less than threshold value
time, stop, getting the lastrow position that the lower bound of character zone is this row;
(4) up-and-down boundary that so far, has removed the character zone of upper and lower side frame and rivet has just been determined;
(5) remove the number that left and right side frame is mainly the every row non-zero pixels value of statistics, and establish threshold value and be
,
width for license plate area image;
(6) from the centre one row scanning left of license plate area image, when running into every row non-zero pixels value number for the first time, be greater than threshold value
time, stop scanning, and the left margin of new character zone is classified on the right one of getting these row as;
(7) from the centre one row scanning to the right of license plate area image, when running into every row non-zero pixels value number for the first time, be greater than threshold value
time, stop scanning, and the right margin of new character zone is classified on the left side one of getting these row as;
Step 2, vertical projection separating character: through the background dot of the pure characters on license plate area image after binaryzation be black picture element 0, character point is white pixel 1, add up the pixel number that every row pixel value is 1, it is exactly the vertical projection distribution plan of white characters pixel, the white pixel at character place is more, and the black picture element that space between character is background, boundary that therefore can be using the trough of vertical projection distribution plan as separating character.Specific practice is as follows:
(1) the vertical projection distribution plan of establishing characters on license plate region bianry image is
, wherein
represented columns, and
be listed as all pixel values and be the number of 1 pixel, the vertical projection from the Far Left of characters on license plate area image to the every row of rightmost scan statistics;
(2), according to vertical projection diagram, set a threshold value
, wherein
for the width of license plate area image, order
, from car plate, scan successively from left to right vertical projection
, satisfy condition one:
and
, condition two:
and
time, be labeled as critical point
, critical point has been recorded and has been cut apart the position that is listed as place, according to these row, just each Character segmentation can be opened.
D) Recognition of License Plate Characters
Character recognition is the final step in car plate identifying, is also that whole system is most crucial, difficulty the best part.In order to reach Recognition of License Plate Characters quickly and accurately, the present invention has adopted the template matches recognition methods based on many features, first character picture is normalized, then extract the various features such as contour feature, projection properties of character picture, mate contrast with the feature of respective symbols in template base, finally reach accurate identification, specifically comprise following three steps:
The normalization of step 1, character picture: inconsistent through the character size that each step is extracted above, so need do normalized to the character picture after cutting apart, character size to be identified is normalized to the size of standard character template, be convenient to the unified feature of extracting character.Suppose that original character picture size is
, the character boundary after normalization conversion is
.Certain pixel in original character
position be
, position after normalization
can be obtained by following formula:
(8)
First add up vertical projection and the horizontal projection of each character, from upper and lower, left and right, four bezel locations are to interscan respectively, and record occurs that projection value is more than or equal to 1 position for the first time, and using this position as new frame.Then utilizing formula (8) and arest neighbors method of interpolation is 48 * 24 by character boundary unification.
The feature extraction of step 2, character picture: feature extraction is exactly by analyzing structure or the statistical framework of character, extract some feature of character, and these features can reflect key and the Global Information of character, reduced the data volume of participating in calculating, improve recognition speed, and reduced to a certain extent the misclassification rate of character recognition.If character picture
size is
.
(1) contour feature extracts: by its procession is scanned, while recording scanning, in image, pixel value has the coordinate position of saltus step, extracts the contour feature of character.Concrete steps are: note from left to right, from right to left, from top to bottom, pixel value is respectively from 0 coordinate that changes to first pixel of 1 during scan image from bottom to up successively
,
with
,
;
(2) projection properties extracts: first utilize following formula (9) to ask horizontal projection and the vertical projection of character
(9)
Obtain again the vertical projection density of character, establish the total pixel value of character
, vertical projection density
, in vertical direction, vertical projection density is to obtain divided by the total pixel value of character according to the projection value of every row, what this density feature reflected is the ratio of the shared whole character pixels value of each row pixel value.Size after character normalizing is 48 * 24, therefore the number of its horizontal projection value is 48, these 48 values is formed a line successively, totally 48 dimensions.The horizontal projection feature of this vector representation character.Vertical projection density amounts to 24, forms a line successively, totally 24 ties up, and represents the vertical projection density of character.
Step 3, the identification of the character match based on many features: the contour feature and the projection properties that extract respectively character Q to be identified and standard form P, and calculate the Euclidean distance value between these features, carry out template matches, apart from reckling, think that these two characters are the most similar, output template character.
In the present embodiment, new Vehicle License Plate Recognition System comprises:
License plate image pretreatment module, for to existing the image of car plate to carry out image pre-service;
Car plate locating module, for realizing the accurate location of license plate area based on mathematical morphology filter technology;
License Plate Character Segmentation module, for utilizing vertical projection method that each character of the license plate area image after having good positioning is extracted and becomes single character;
Recognition of License Plate Characters module, for adopting the template matches recognition methods based on many features accurately to identify characters on license plate.
Wherein, license plate image pretreatment module comprises:
Gray processing unit, for processing coloured image gray processing;
Image enhancing unit, for strengthening image;
Image binaryzation unit, is set to 0 or 1 for gradation of image value, make whole image present non-black be white effect.
Wherein, car plate locating module comprises:
Vertical edge extraction unit, for extracting the vertical edge of image based on vertical Sobel edge detection algorithm;
Mathematical morphology filter unit, for adopting mathematical morphology filter method fortune to convert image, gives prominence to needed image information;
License plate area determining unit, for determining license plate area according to car plate priori.
Wherein, License Plate Character Segmentation module comprises:
Remove license plate area frame unit, for the saltus step rule according to character, remove upper and lower side frame and rivet;
Vertical projection separating character unit, for opening each Character segmentation.
Wherein, Recognition of License Plate Characters module comprises:
Character picture normalization unit, for the character picture after cutting apart is done to normalized, normalizes to character size to be identified the size of standard character template;
Character picture feature extraction unit, for by analyzing structure or the statistical framework of character, extracts contour feature and the projection properties of character;
Character match recognition unit based on many features, for extracting respectively contour feature and the projection properties of character Q to be identified and standard form P, and calculates the Euclidean distance value between these features, carries out template matches, output template character.
The present invention has utilized multiple image processing algorithm to complete whole car plate identification and has identified this whole Vehicle License Plate Recognition System from pre-service car plate to the last.In car plate location, adopted based on mathematical morphology filter technology; In License Plate Character Segmentation extracts, the methods such as vertical projection have been used; Finally by the template matching technique based on many feature extractions, completed Recognition of License Plate Characters.
More than show and described ultimate principle of the present invention, principal character and advantage of the present invention.The technician of the industry should understand; the present invention is not restricted to the described embodiments; that in above-described embodiment and instructions, describes just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.Claimed scope of the present invention is defined by appending claims and equivalent thereof.
Claims (10)
1. a new licence plate recognition method, is characterized in that comprising the steps:
A) license plate image pre-service, the binary image of acquisition car plate;
B) car plate location: first utilize vertical Sobel edge detection algorithm pretreated binary image to be carried out to the detection of vertical edge, recycling Mathematical Morphology Method edge image carries out the coarse positioning of license plate area, finally utilizes projecting method to realize the accurate location of license plate area;
C) License Plate Character Segmentation: first the frame of license plate area is removed, and then utilized vertical projection method to carry out separating character;
D) Recognition of License Plate Characters: adopt the template matches recognition methods based on many features, first character picture is normalized, then extract contour feature, the projection properties of character picture, mate contrast with the feature of respective symbols in template base, to reach accurate identification.
2. new licence plate recognition method according to claim 1, is characterized in that: license plate image pre-service comprises the following steps:
Step 1, gray processing: coloured image gray processing is processed, be shown below:
(1)
In formula (1),
be respectively pixel
the intensity level of the red, green, blue look at place,
for the gray-scale value after gray processing;
Step 2, figure image intensifying: adopt the image enchancing method based on Laplce's sharpening mask, its formula is:
(2)
In formula (2),
for the gray level image of input,
for strengthening rear image,
for the Laplace operator of input picture, it is defined as:
(3)
Utilize Laplace operator matrix
carrying out convolution with input picture obtained above-mentioned afterwards
, utilize sharpening mask matrix
carry out obtaining final enhancing image after convolution with input picture
;
Step 3, image binaryzation: gradation of image value is set to 0 or 1 process, make whole image present non-black be white effect, see shown in following formula:
(4)
In formula (4),
for optimal threshold,
the maximum gradation value of the enhancing image that expression obtains,
for strengthening the minimum gradation value of image, final binary image obtains by following formula:
(5)
Input in above formula (5)
be the image after enhancing,
for final binary image.
3. new licence plate recognition method according to claim 1, is characterized in that, car plate location comprises the following steps:
Step 1, vertical edge extracts: based on vertical Sobel edge detection algorithm, extract the vertical edge in image, namely utilize the vertical convolution kernel of Sobel to come in the binary image with input each pixel to carry out convolution algorithm and obtained afterwards vertical edge image;
Step 2, mathematical morphology filter: adopt mathematical morphology filter method fortune to convert image, give prominence to needed image information, concrete grammar is exactly to adopt the structural elements and the binary image that possess required geometric shape to carry out set operation, first use closed operation, re-use opening operation, finally obtain image after needed morphologic filtering, concrete steps are as follows:
(1) first to vertical edge binary image
carry out structural elements
closed operation, i.e. structural elements
being defined as is highly 5 pixels, and length is the rectangle of 19 pixels, and closed operation is to use same structural element to first the expand computing of post-etching of image, therefore, and target image
in structural elements
under closed operation be defined as follows:
(6)
In formula (6),
be defined as expansive working, for given target image
and structural elements
, will
middle every bit
expand as
operation; And
be defined as corrosion operation, meet
point
all formation structural elements
with target image
maximal correlation point set;
(2) again to carrying out the image after closed operation
carry out structural elements
opening operation, i.e. structural elements
being defined as is highly 5 pixels, and length is the rectangle of 19 pixels, and opening operation is to use same structural element image first to be corroded to the computing of rear expansion, therefore, and target image
in structural elements
under opening operation be defined as follows:
(7)
In formula (7),
for expansive working, and
for corrosion operation;
Step 3, according to car plate priori, determine license plate area: the candidate region that utilizes the priori of the car plate of China only to deposit from these, determine real license plate area, first, calculate the length breadth ratio of each candidate's rectangular area, if length breadth ratio is defined as possible license plate area between 2.5 and 4.5, if there are a plurality of possible license plate areas, measure the width in this region and actual width and the length that length judges whether to meet car plate, so just can finally determine unique real license plate area, last according to extracted region apex coordinate, by real license plate area, from original input gray level input picture, cutting is out.
4. new licence plate recognition method according to claim 1, is characterized in that, License Plate Character Segmentation comprises the following steps:
Step 1, remove license plate area frame: according to the saltus step rule of character, remove upper and lower side frame and rivet, specific practice is:
(1) removing upper and lower side frame is mainly every row pixel value transition times in statistics license plate area, establishes saltus step threshold value and is
;
(2) from the middle row of the license plate area image scan statistics that makes progress, when running into every row transition times for the first time, be less than threshold value
time, stop, getting the next line position that the upper bound of character zone is this row;
(3) from the downward scan statistics of middle row of license plate area image, when running into every row transition times for the first time, be less than threshold value
time, stop, getting the lastrow position that the lower bound of character zone is this row;
(4) up-and-down boundary that so far, has removed the character zone of upper and lower side frame and rivet has just been determined;
(5) remove the number that left and right side frame is mainly the every row non-zero pixels value of statistics, and establish threshold value and be
,
width for license plate area image;
(6) from the centre one row scanning left of license plate area image, when running into every row non-zero pixels value number for the first time, be greater than threshold value
time, stop scanning, and the left margin of new character zone is classified on the right one of getting these row as;
(7) from the centre one row scanning to the right of license plate area image, when running into every row non-zero pixels value number for the first time, be greater than threshold value
time, stop scanning, and the right margin of new character zone is classified on the left side one of getting these row as;
Step 2, vertical projection separating character, specific practice is as follows:
(1) the vertical projection distribution plan of establishing characters on license plate region bianry image is
, wherein
represented columns, and
be listed as all pixel values and be the number of 1 pixel, the vertical projection from the Far Left of characters on license plate area image to the every row of rightmost scan statistics;
(2), according to vertical projection diagram, set a threshold value
, wherein
for the width of license plate area image, order
, from car plate, scan successively from left to right vertical projection
, satisfy condition one:
and
, condition two:
and
time, be labeled as critical point
, critical point has been recorded and has been cut apart the position that is listed as place, according to these row, just each Character segmentation is opened.
5. new licence plate recognition method according to claim 1, is characterized in that, Recognition of License Plate Characters comprises the following steps:
Step 1, the normalization of character picture: suppose that original character picture size is
, the character boundary after normalization conversion is
, certain pixel in original character
position be
, position after normalization
by following formula, obtained:
(8)
First add up vertical projection and the horizontal projection of each character, respectively from upper and lower, left and right four bezel locations to interscan, record occurs that projection value is more than or equal to 1 position for the first time, and using this position as new frame, then utilizing formula (8) and arest neighbors method of interpolation is 48 * 24 by character boundary unification;
Step 2, the feature extraction of character picture: establish character picture
size is
, be the detailed process of feature extraction below:
(1) contour feature extracts: by its procession is scanned, recording pixel value in when scanning image has the coordinate position of saltus step, the contour feature that extracts character, concrete steps are: note from left to right, from right to left, from top to bottom, pixel value is respectively from 0 coordinate that changes to first pixel of 1 during scan image from bottom to up successively
,
with
,
;
(2) projection properties extracts: first utilize following formula (9) to ask horizontal projection and the vertical projection of character
(9)
Obtain again the vertical projection density of character, establish the total pixel value of character
, vertical projection density
in vertical direction, vertical projection density is to obtain divided by the total pixel value of character according to the projection value of every row, what this density feature reflected is the ratio of the shared whole character pixels value of each row pixel value, size after character normalizing is 48 * 24, therefore the number of its horizontal projection value is 48, these 48 values are formed a line successively, totally 48 tie up, the horizontal projection feature of this vector representation character, vertical projection density amounts to 24, forms a line successively, totally 24 tie up, represent the vertical projection density of character;
Step 3, the character match identification based on many features: extract respectively contour feature and the projection properties of character Q to be identified and standard form P, and calculate the Euclidean distance value between these features, carry out template matches, apart from reckling, think that these two characters are the most similar, output template character.
6. a new Vehicle License Plate Recognition System, is characterized in that, comprising:
License plate image pretreatment module, for to existing the image of car plate to carry out image pre-service, obtains the binary image of car plate;
Car plate locating module, for realizing the accurate location of license plate area based on mathematical morphology filter technology;
License Plate Character Segmentation module, for utilizing vertical projection method that each character of the license plate area image after having good positioning is extracted and becomes single character;
Recognition of License Plate Characters module, for adopting the template matches recognition methods based on many features accurately to identify characters on license plate.
7. new Vehicle License Plate Recognition System according to claim 6, is characterized in that, license plate image pretreatment module comprises:
Gray processing unit, for processing coloured image gray processing;
Image enhancing unit, for strengthening image;
Image binaryzation unit, is set to 0 or 1 for gradation of image value, make whole image present non-black be white effect.
8. new Vehicle License Plate Recognition System according to claim 6, is characterized in that, car plate locating module comprises:
Vertical edge extraction unit, for extracting the vertical edge of image based on vertical Sobel edge detection algorithm;
Mathematical morphology filter unit, for adopting mathematical morphology filter method fortune to convert image, gives prominence to needed image information;
License plate area determining unit, for determining license plate area according to car plate priori.
9. new Vehicle License Plate Recognition System according to claim 6, is characterized in that, License Plate Character Segmentation module comprises:
Remove license plate area frame unit, for the saltus step rule according to character, remove upper and lower side frame and rivet;
Vertical projection separating character unit, for opening each Character segmentation.
10. new Vehicle License Plate Recognition System according to claim 6, is characterized in that, Recognition of License Plate Characters module comprises:
Character picture normalization unit, for the character picture after cutting apart is done to normalized, normalizes to character size to be identified the size of standard character template;
Character picture feature extraction unit, for by analyzing structure or the statistical framework of character, extracts contour feature and the projection properties of character;
Character match recognition unit based on many features, for extracting respectively contour feature and the projection properties of character Q to be identified and standard form P, and calculates the Euclidean distance value between these features, carries out template matches, output template character.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410378116.9A CN104156704A (en) | 2014-08-04 | 2014-08-04 | Novel license plate identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410378116.9A CN104156704A (en) | 2014-08-04 | 2014-08-04 | Novel license plate identification method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104156704A true CN104156704A (en) | 2014-11-19 |
Family
ID=51882201
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410378116.9A Pending CN104156704A (en) | 2014-08-04 | 2014-08-04 | Novel license plate identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104156704A (en) |
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104778450A (en) * | 2015-03-30 | 2015-07-15 | 深圳市道通科技股份有限公司 | Preprocessing method and device for VIN (vehicle identification number) image |
| CN104809790A (en) * | 2015-04-29 | 2015-07-29 | 无锡市崇安区科技创业服务中心 | Parking management method based on image analysis |
| CN104992152A (en) * | 2015-06-30 | 2015-10-21 | 深圳訾岽科技有限公司 | Character recognition method and system based on template character library |
| CN105117680A (en) * | 2015-05-04 | 2015-12-02 | 小米科技有限责任公司 | Method and device for recognizing information of information card |
| CN105354573A (en) * | 2015-12-15 | 2016-02-24 | 重庆凯泽科技有限公司 | Container license plate identification method and system |
| CN105448100A (en) * | 2015-12-15 | 2016-03-30 | 重庆凯泽科技有限公司 | Reverse vehicle searching method and reverse vehicle searching system |
| CN105488576A (en) * | 2015-12-03 | 2016-04-13 | 小米科技有限责任公司 | Method and apparatus for determining vehicle maintenance and repair expense |
| CN105528606A (en) * | 2015-10-30 | 2016-04-27 | 小米科技有限责任公司 | Region identification method and device |
| CN105528607A (en) * | 2015-10-30 | 2016-04-27 | 小米科技有限责任公司 | Region extraction method and model training method and device |
| CN105550686A (en) * | 2015-12-09 | 2016-05-04 | 小米科技有限责任公司 | Character segmentation method and device |
| CN105590113A (en) * | 2015-12-16 | 2016-05-18 | 深圳市华德安科技有限公司 | Information-acquiring method based on law enforcement recorder |
| CN105608903A (en) * | 2015-12-15 | 2016-05-25 | 重庆凯泽科技有限公司 | Traffic violation detection method and system |
| CN105809166A (en) * | 2016-03-04 | 2016-07-27 | 深圳市佳信捷技术股份有限公司 | Vehicle license plate recognition method, device and system |
| CN106033316A (en) * | 2015-03-13 | 2016-10-19 | 北京搜狗科技发展有限公司 | Method and device for hand input |
| CN106355205A (en) * | 2016-08-31 | 2017-01-25 | 西安西拓电气股份有限公司 | Recognition method and device for figures in ultraviolet image |
| CN106446898A (en) * | 2016-09-14 | 2017-02-22 | 宇龙计算机通信科技(深圳)有限公司 | Extraction method and extraction device of character information in image |
| CN106446889A (en) * | 2015-08-10 | 2017-02-22 | Tcl集团股份有限公司 | Local identification method and local identification device for station logo |
| CN106485234A (en) * | 2016-10-21 | 2017-03-08 | 合肥哦走信息技术有限公司 | One kind is based on vehicle identification method in intelligent transportation system |
| CN106650741A (en) * | 2016-12-30 | 2017-05-10 | 深圳市捷顺科技实业股份有限公司 | License plate character recognition method and device |
| CN106652466A (en) * | 2016-11-18 | 2017-05-10 | 德阳九鼎智远知识产权运营有限公司 | Vehicle positioning system based on vehicle license plate recognition and vehicle positioning method thereof |
| CN106650732A (en) * | 2016-12-27 | 2017-05-10 | 深圳市捷顺科技实业股份有限公司 | License plate identification method and apparatus |
| CN107220981A (en) * | 2017-06-22 | 2017-09-29 | 深圳怡化电脑股份有限公司 | Character segmentation method, device, equipment and storage medium |
| CN107273889A (en) * | 2017-04-27 | 2017-10-20 | 浙江工业大学 | A kind of licence plate recognition method based on statistics |
| CN107273884A (en) * | 2017-05-05 | 2017-10-20 | 广州恒巨信息科技有限公司 | A kind of License Plate Identification method based on mobile terminal camera |
| CN107358236A (en) * | 2017-08-11 | 2017-11-17 | 南京云计趟信息技术有限公司 | A kind of number-plate number identifying system and method based on camera device |
| CN107358795A (en) * | 2017-07-17 | 2017-11-17 | 陈剑桃 | A kind of effective Vehicle License Plate Recognition System |
| CN107766862A (en) * | 2017-10-19 | 2018-03-06 | 成都博智维讯信息技术股份有限公司 | Item price label digit recognition method |
| CN107798323A (en) * | 2016-08-29 | 2018-03-13 | 北京君正集成电路股份有限公司 | A kind of license plate image localization method and equipment |
| CN107817461A (en) * | 2016-09-12 | 2018-03-20 | 中国电力科学研究院 | Electric instrument automatic Proofreading device |
| CN108073926A (en) * | 2016-11-17 | 2018-05-25 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method and device |
| CN108537223A (en) * | 2018-04-16 | 2018-09-14 | 广东工业大学 | A kind of detection method of license plate, system and equipment and storage medium |
| CN109086767A (en) * | 2018-07-11 | 2018-12-25 | 于洋 | The dividing method of fuzzy license plate in a kind of monitoring scene |
| CN109348179A (en) * | 2018-11-06 | 2019-02-15 | 常州信息职业技术学院 | A kind of road monitoring detection system and method based on artificial intelligence |
| CN109509157A (en) * | 2018-11-09 | 2019-03-22 | 重庆巴奥科技有限公司 | A kind of information of vehicles enhancing recognition methods based on large nuber of images information |
| CN109543613A (en) * | 2018-11-22 | 2019-03-29 | 西安科技大学 | Vehicle Speed and Vehicle License Plate Recognition System and method based on TOF imaging |
| CN109800752A (en) * | 2018-08-09 | 2019-05-24 | 辽宁工业大学 | A kind of automotive license plate Character segmentation identification algorithm based on machine vision |
| CN109829460A (en) * | 2019-04-02 | 2019-05-31 | 安徽省安瑞机电科技有限公司 | A kind of License Plate Character Segmentation recognition methods and its system applied to parking management system |
| CN110135260A (en) * | 2019-04-15 | 2019-08-16 | 青岛小鸟看看科技有限公司 | The determination method, apparatus and electronic equipment on the boundary of area-of-interest in image |
| CN110426401A (en) * | 2019-08-07 | 2019-11-08 | 北京赛育达科教有限责任公司 | A kind of robot vision character machining system and method for real training |
| CN110458183A (en) * | 2019-06-25 | 2019-11-15 | 上海圭目机器人有限公司 | A kind of characteristic matching optimization algorithm of image adaptive |
| CN110543883A (en) * | 2019-08-27 | 2019-12-06 | 河海大学 | license plate recognition method based on deep learning |
| CN111027546A (en) * | 2019-12-05 | 2020-04-17 | 北京嘉楠捷思信息技术有限公司 | Character segmentation method and device and computer readable storage medium |
| CN111046862A (en) * | 2019-12-05 | 2020-04-21 | 北京嘉楠捷思信息技术有限公司 | Character segmentation method and device and computer readable storage medium |
| CN111046876A (en) * | 2019-12-18 | 2020-04-21 | 南京航空航天大学 | License plate character rapid recognition method and system based on texture detection technology |
| CN111696364A (en) * | 2020-06-22 | 2020-09-22 | 沈志强 | Roadside parking management method |
| CN112070081A (en) * | 2020-08-20 | 2020-12-11 | 广州杰赛科技股份有限公司 | Intelligent license plate recognition method based on high-definition video |
| CN112101343A (en) * | 2020-08-17 | 2020-12-18 | 广东工业大学 | License plate character segmentation and recognition method |
| CN112163581A (en) * | 2020-09-09 | 2021-01-01 | 华南理工大学 | License plate letter recognition method, system, device and storage medium |
| CN112381088A (en) * | 2020-11-11 | 2021-02-19 | 上善智城(苏州)信息科技有限公司 | License plate recognition method and system for oil tank truck |
| WO2021109697A1 (en) * | 2019-12-05 | 2021-06-10 | 嘉楠明芯(北京)科技有限公司 | Character segmentation method and apparatus, and computer-readable storage medium |
| CN112991432A (en) * | 2021-04-15 | 2021-06-18 | 重庆大学 | Icing shape identification method based on image processing |
| CN114549811A (en) * | 2021-12-31 | 2022-05-27 | 哈尔滨理工大学 | License plate identification method based on license plate spacer body characteristics |
| CN114821078A (en) * | 2022-05-05 | 2022-07-29 | 北方工业大学 | License plate recognition method and device, electronic equipment and storage medium |
| CN114934467A (en) * | 2022-07-08 | 2022-08-23 | 江苏永达电力金具有限公司 | Parking space barrier gate control method, parking space barrier gate system and medium |
| CN115761713A (en) * | 2022-07-05 | 2023-03-07 | 广西北投信创科技投资集团有限公司 | License plate recognition method and system, electronic equipment and readable storage medium |
| CN116309296A (en) * | 2022-12-31 | 2023-06-23 | 中山市天柏包装制品有限公司 | Intelligent detection method and detection system for defects of packaging box |
| CN117351438A (en) * | 2023-10-24 | 2024-01-05 | 武汉无线飞翔科技有限公司 | Real-time vehicle position tracking method and system based on image recognition |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065138A (en) * | 2012-12-06 | 2013-04-24 | 新疆公众信息产业股份有限公司 | Recognition method of license plate number of motor vehicle |
-
2014
- 2014-08-04 CN CN201410378116.9A patent/CN104156704A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065138A (en) * | 2012-12-06 | 2013-04-24 | 新疆公众信息产业股份有限公司 | Recognition method of license plate number of motor vehicle |
Non-Patent Citations (4)
| Title |
|---|
| 刘攀: "复杂光照条件下的通用车牌定位系统的研究与实现", 《中国优秀硕士学位论文全文数据库》 * |
| 张吉斌: "基于图像处理及支持向量机的车牌识别技术研究", 《中国优秀硕士学位论文全文数据库》 * |
| 张金艳: "基于神经网络的车牌识别算法研究", 《中国优秀硕士学位论文全文数据库》 * |
| 徐安原: "基于FPGA与DSP车牌识别系统的研究与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (78)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106033316A (en) * | 2015-03-13 | 2016-10-19 | 北京搜狗科技发展有限公司 | Method and device for hand input |
| CN104778450B (en) * | 2015-03-30 | 2019-04-02 | 深圳市道通科技股份有限公司 | A kind of preprocess method and device of vehicle identification code image |
| CN104778450A (en) * | 2015-03-30 | 2015-07-15 | 深圳市道通科技股份有限公司 | Preprocessing method and device for VIN (vehicle identification number) image |
| CN104809790A (en) * | 2015-04-29 | 2015-07-29 | 无锡市崇安区科技创业服务中心 | Parking management method based on image analysis |
| CN105117680A (en) * | 2015-05-04 | 2015-12-02 | 小米科技有限责任公司 | Method and device for recognizing information of information card |
| CN105117680B (en) * | 2015-05-04 | 2019-06-14 | 小米科技有限责任公司 | A kind of method and apparatus of the information of ID card |
| CN104992152A (en) * | 2015-06-30 | 2015-10-21 | 深圳訾岽科技有限公司 | Character recognition method and system based on template character library |
| CN106446889B (en) * | 2015-08-10 | 2019-09-17 | Tcl集团股份有限公司 | A kind of local recognition methods of logo and device |
| CN106446889A (en) * | 2015-08-10 | 2017-02-22 | Tcl集团股份有限公司 | Local identification method and local identification device for station logo |
| RU2639668C2 (en) * | 2015-10-30 | 2017-12-21 | Сяоми Инк. | Method and device for region identification |
| CN105528606B (en) * | 2015-10-30 | 2019-08-06 | 小米科技有限责任公司 | Area recognizing method and device |
| CN105528607B (en) * | 2015-10-30 | 2019-02-15 | 小米科技有限责任公司 | Method for extracting region, model training method and device |
| US10157326B2 (en) | 2015-10-30 | 2018-12-18 | Xiaomi Inc. | Method and device for character area identification |
| CN105528606A (en) * | 2015-10-30 | 2016-04-27 | 小米科技有限责任公司 | Region identification method and device |
| WO2017071063A1 (en) * | 2015-10-30 | 2017-05-04 | 小米科技有限责任公司 | Area identification method and device |
| CN105528607A (en) * | 2015-10-30 | 2016-04-27 | 小米科技有限责任公司 | Region extraction method and model training method and device |
| CN105488576A (en) * | 2015-12-03 | 2016-04-13 | 小米科技有限责任公司 | Method and apparatus for determining vehicle maintenance and repair expense |
| CN105550686A (en) * | 2015-12-09 | 2016-05-04 | 小米科技有限责任公司 | Character segmentation method and device |
| CN105448100A (en) * | 2015-12-15 | 2016-03-30 | 重庆凯泽科技有限公司 | Reverse vehicle searching method and reverse vehicle searching system |
| CN105354573B (en) * | 2015-12-15 | 2019-03-22 | 重庆凯泽科技股份有限公司 | A kind of container licence plate recognition method and system |
| CN105608903A (en) * | 2015-12-15 | 2016-05-25 | 重庆凯泽科技有限公司 | Traffic violation detection method and system |
| CN105354573A (en) * | 2015-12-15 | 2016-02-24 | 重庆凯泽科技有限公司 | Container license plate identification method and system |
| CN105590113A (en) * | 2015-12-16 | 2016-05-18 | 深圳市华德安科技有限公司 | Information-acquiring method based on law enforcement recorder |
| CN105809166A (en) * | 2016-03-04 | 2016-07-27 | 深圳市佳信捷技术股份有限公司 | Vehicle license plate recognition method, device and system |
| CN107798323B (en) * | 2016-08-29 | 2020-12-29 | 北京君正集成电路股份有限公司 | License plate image positioning method and device |
| CN107798323A (en) * | 2016-08-29 | 2018-03-13 | 北京君正集成电路股份有限公司 | A kind of license plate image localization method and equipment |
| CN106355205A (en) * | 2016-08-31 | 2017-01-25 | 西安西拓电气股份有限公司 | Recognition method and device for figures in ultraviolet image |
| CN107817461A (en) * | 2016-09-12 | 2018-03-20 | 中国电力科学研究院 | Electric instrument automatic Proofreading device |
| CN106446898A (en) * | 2016-09-14 | 2017-02-22 | 宇龙计算机通信科技(深圳)有限公司 | Extraction method and extraction device of character information in image |
| CN106485234A (en) * | 2016-10-21 | 2017-03-08 | 合肥哦走信息技术有限公司 | One kind is based on vehicle identification method in intelligent transportation system |
| CN108073926A (en) * | 2016-11-17 | 2018-05-25 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method and device |
| CN108073926B (en) * | 2016-11-17 | 2020-04-03 | 杭州海康威视数字技术股份有限公司 | License plate recognition method and device |
| CN106652466A (en) * | 2016-11-18 | 2017-05-10 | 德阳九鼎智远知识产权运营有限公司 | Vehicle positioning system based on vehicle license plate recognition and vehicle positioning method thereof |
| CN106650732B (en) * | 2016-12-27 | 2020-04-28 | 深圳市捷顺科技实业股份有限公司 | License plate recognition method and device |
| CN106650732A (en) * | 2016-12-27 | 2017-05-10 | 深圳市捷顺科技实业股份有限公司 | License plate identification method and apparatus |
| CN106650741A (en) * | 2016-12-30 | 2017-05-10 | 深圳市捷顺科技实业股份有限公司 | License plate character recognition method and device |
| CN107273889A (en) * | 2017-04-27 | 2017-10-20 | 浙江工业大学 | A kind of licence plate recognition method based on statistics |
| CN107273884A (en) * | 2017-05-05 | 2017-10-20 | 广州恒巨信息科技有限公司 | A kind of License Plate Identification method based on mobile terminal camera |
| CN107220981A (en) * | 2017-06-22 | 2017-09-29 | 深圳怡化电脑股份有限公司 | Character segmentation method, device, equipment and storage medium |
| CN107358795A (en) * | 2017-07-17 | 2017-11-17 | 陈剑桃 | A kind of effective Vehicle License Plate Recognition System |
| CN107358236A (en) * | 2017-08-11 | 2017-11-17 | 南京云计趟信息技术有限公司 | A kind of number-plate number identifying system and method based on camera device |
| CN107766862A (en) * | 2017-10-19 | 2018-03-06 | 成都博智维讯信息技术股份有限公司 | Item price label digit recognition method |
| CN108537223B (en) * | 2018-04-16 | 2022-02-15 | 广东工业大学 | License plate detection method, system and equipment and storage medium |
| CN108537223A (en) * | 2018-04-16 | 2018-09-14 | 广东工业大学 | A kind of detection method of license plate, system and equipment and storage medium |
| CN109086767A (en) * | 2018-07-11 | 2018-12-25 | 于洋 | The dividing method of fuzzy license plate in a kind of monitoring scene |
| CN109800752A (en) * | 2018-08-09 | 2019-05-24 | 辽宁工业大学 | A kind of automotive license plate Character segmentation identification algorithm based on machine vision |
| CN109800752B (en) * | 2018-08-09 | 2022-01-07 | 辽宁工业大学 | Automobile license plate character segmentation and recognition algorithm based on machine vision |
| CN109348179A (en) * | 2018-11-06 | 2019-02-15 | 常州信息职业技术学院 | A kind of road monitoring detection system and method based on artificial intelligence |
| CN109348179B (en) * | 2018-11-06 | 2021-07-06 | 常州信息职业技术学院 | Road monitoring and detecting system and method based on artificial intelligence |
| CN109509157A (en) * | 2018-11-09 | 2019-03-22 | 重庆巴奥科技有限公司 | A kind of information of vehicles enhancing recognition methods based on large nuber of images information |
| CN109509157B (en) * | 2018-11-09 | 2021-04-30 | 江苏中天安驰科技有限公司 | Vehicle information enhanced identification method based on massive image information |
| CN109543613A (en) * | 2018-11-22 | 2019-03-29 | 西安科技大学 | Vehicle Speed and Vehicle License Plate Recognition System and method based on TOF imaging |
| CN109829460A (en) * | 2019-04-02 | 2019-05-31 | 安徽省安瑞机电科技有限公司 | A kind of License Plate Character Segmentation recognition methods and its system applied to parking management system |
| CN110135260A (en) * | 2019-04-15 | 2019-08-16 | 青岛小鸟看看科技有限公司 | The determination method, apparatus and electronic equipment on the boundary of area-of-interest in image |
| CN110458183A (en) * | 2019-06-25 | 2019-11-15 | 上海圭目机器人有限公司 | A kind of characteristic matching optimization algorithm of image adaptive |
| CN110426401A (en) * | 2019-08-07 | 2019-11-08 | 北京赛育达科教有限责任公司 | A kind of robot vision character machining system and method for real training |
| CN110543883A (en) * | 2019-08-27 | 2019-12-06 | 河海大学 | license plate recognition method based on deep learning |
| CN111046862B (en) * | 2019-12-05 | 2023-10-27 | 嘉楠明芯(北京)科技有限公司 | Character segmentation method, device and computer readable storage medium |
| CN111027546B (en) * | 2019-12-05 | 2024-03-26 | 嘉楠明芯(北京)科技有限公司 | Character segmentation method, device and computer readable storage medium |
| WO2021109697A1 (en) * | 2019-12-05 | 2021-06-10 | 嘉楠明芯(北京)科技有限公司 | Character segmentation method and apparatus, and computer-readable storage medium |
| CN111046862A (en) * | 2019-12-05 | 2020-04-21 | 北京嘉楠捷思信息技术有限公司 | Character segmentation method and device and computer readable storage medium |
| CN111027546A (en) * | 2019-12-05 | 2020-04-17 | 北京嘉楠捷思信息技术有限公司 | Character segmentation method and device and computer readable storage medium |
| CN111046876B (en) * | 2019-12-18 | 2022-05-31 | 南京航空航天大学 | License plate character rapid recognition method and system based on texture detection technology |
| CN111046876A (en) * | 2019-12-18 | 2020-04-21 | 南京航空航天大学 | License plate character rapid recognition method and system based on texture detection technology |
| CN111696364A (en) * | 2020-06-22 | 2020-09-22 | 沈志强 | Roadside parking management method |
| CN112101343A (en) * | 2020-08-17 | 2020-12-18 | 广东工业大学 | License plate character segmentation and recognition method |
| CN112070081B (en) * | 2020-08-20 | 2024-01-09 | 广州杰赛科技股份有限公司 | Intelligent license plate recognition method based on high-definition video |
| CN112070081A (en) * | 2020-08-20 | 2020-12-11 | 广州杰赛科技股份有限公司 | Intelligent license plate recognition method based on high-definition video |
| CN112163581B (en) * | 2020-09-09 | 2023-08-22 | 华南理工大学 | License plate letter recognition method, system, device and storage medium |
| CN112163581A (en) * | 2020-09-09 | 2021-01-01 | 华南理工大学 | License plate letter recognition method, system, device and storage medium |
| CN112381088A (en) * | 2020-11-11 | 2021-02-19 | 上善智城(苏州)信息科技有限公司 | License plate recognition method and system for oil tank truck |
| CN112991432A (en) * | 2021-04-15 | 2021-06-18 | 重庆大学 | Icing shape identification method based on image processing |
| CN114549811A (en) * | 2021-12-31 | 2022-05-27 | 哈尔滨理工大学 | License plate identification method based on license plate spacer body characteristics |
| CN114821078A (en) * | 2022-05-05 | 2022-07-29 | 北方工业大学 | License plate recognition method and device, electronic equipment and storage medium |
| CN115761713A (en) * | 2022-07-05 | 2023-03-07 | 广西北投信创科技投资集团有限公司 | License plate recognition method and system, electronic equipment and readable storage medium |
| CN114934467A (en) * | 2022-07-08 | 2022-08-23 | 江苏永达电力金具有限公司 | Parking space barrier gate control method, parking space barrier gate system and medium |
| CN116309296A (en) * | 2022-12-31 | 2023-06-23 | 中山市天柏包装制品有限公司 | Intelligent detection method and detection system for defects of packaging box |
| CN117351438A (en) * | 2023-10-24 | 2024-01-05 | 武汉无线飞翔科技有限公司 | Real-time vehicle position tracking method and system based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104156704A (en) | Novel license plate identification method and system | |
| CN101398894B (en) | Automobile license plate automatic recognition method and implementing device thereof | |
| CN101872416B (en) | Vehicle license plate recognition method and system of road image | |
| Prabhakar et al. | Automatic vehicle number plate detection and recognition | |
| CN103136528B (en) | A kind of licence plate recognition method based on dual edge detection | |
| Li et al. | A HOG feature and SVM based method for forward vehicle detection with single camera | |
| CN106650553A (en) | License plate recognition method and system | |
| CN103870803A (en) | Vehicle license plate recognition method and system based on coarse positioning and fine positioning fusion | |
| CN104680161A (en) | Digit recognition method for identification cards | |
| CN106815583B (en) | Method for positioning license plate of vehicle at night based on combination of MSER and SWT | |
| CN103198315A (en) | License plate character segmentation algorithm based on character outline and template matching | |
| CN105205480A (en) | Complex scene human eye locating method and system | |
| Soomro et al. | Vehicle number recognition system for automatic toll tax collection | |
| Azad et al. | New method for optimization of license plate recognition system with use of edge detection and connected component | |
| Garz et al. | A binarization-free clustering approach to segment curved text lines in historical manuscripts | |
| CN109086772A (en) | A kind of recognition methods and system distorting adhesion character picture validation code | |
| Patel | Vehicle license plate recognition using morphology and neural network | |
| Pandya et al. | Morphology based approach to recognize number plates in India | |
| CN104463091A (en) | Face image recognition method based on LGBP feature subvectors of image | |
| Li et al. | Vision-based method for forward vehicle detection and tracking | |
| CN110046618B (en) | License plate recognition method based on machine learning and maximum extremum stable region | |
| Rastegar et al. | An intelligent control system using an efficient License Plate Location and Recognition Approach | |
| Goel et al. | Vehicle registration plate recognition system using template matching | |
| CN112528740A (en) | Pressing plate state identification method | |
| Deb et al. | Automatic vehicle identification by plate recognition for intelligent transportation system applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141119 |