CN103984949A - License plate positioning method and system based on high and low cap transformation and connected domain - Google Patents
License plate positioning method and system based on high and low cap transformation and connected domain Download PDFInfo
- Publication number
- CN103984949A CN103984949A CN201410257831.7A CN201410257831A CN103984949A CN 103984949 A CN103984949 A CN 103984949A CN 201410257831 A CN201410257831 A CN 201410257831A CN 103984949 A CN103984949 A CN 103984949A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- car plate
- connected domain
- module
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, and discloses a license plate positioning method and system based on high and low cap transformation and a connected domain. The method comprises the steps of S1, acquiring a gray image of a vehicle, and preprocessing the gray image; S2, extracting a target image from the preprocessed gray image by using high and low cap transformation, and binarizing the target image; S3, calculating the connected domain of the binarized target image to eliminate the connected domain of a non license plate target and realize coarse positioning of a license plate; S4, performing tilt correction and projection on the coarse positioned license plate to realize fine positioning of the license plate and determine the specific position of the license plate. According to the method and the system, the gray image of the vehicle is preprocessed, and the method of the high and low cap transformation and the connected domain is utilized, so that the influence of non-uniform illumination, breakage or rotation on the positioning of the license plate is reduced, and the accuracy of positioning the license plate is further improved.
Description
Technical field
The present invention relates to technical field of image processing, be specifically related to a kind of license plate locating method and system thereof based on the conversion of height cap and connected domain.
Background technology
Along with the fast development of Chinese national economy, the requirement of control of traffic and road is progressively improved, intelligent transportation system is arisen at the historic moment, and as the important component part of intelligent transportation system, Vehicle License Plate Recognition System has obtained fast development, by Chinese scholars broad research.The motor vehicle trade mark is also that car plate location and identification can be widely used in the occasions such as freeway toll station, Entrance, machine-operated gate, to realize automatic monitoring and the management of vehicle, and then has saved man power and material.
Vehicle License Plate Recognition System is divided into Image Acquisition, car plate location, Character segmentation, character recognition four parts, and wherein, car plate location is the committed step of car plate identification.Aspect the extraction of car plate position, need to suppress various interference, such as uneven illumination is even, the rotation of car plate and deformation, fracture or the situation such as stained, extract stable textural characteristics, finally complete car plate location.For car plate location, current research method mainly contains: the colour edging distribution characteristics of utilizing car plate; Utilize car plate and text color characteristics of combination; Utilize car plate frame area, length breadth ratio; Utilize the angle point information of characters on license plate; Utilize the design feature of the cross-correlation vector figure between characters on license plate stroke left and right edges; Utilize car plate textural characteristics.Under illumination and the desirable condition of weather condition, the existing license plate locating method based on the conversion of height cap and connected domain, for positions such as the headstock tailstock, has been obtained good achievement, and locating accuracy can reach more than 98%.
But, along with the application of Vehicle License Plate Recognition System is more and more extensive, because scene becomes increasingly complex, the intensity of variation of weather condition is increasing, cause the not obvious and car plate marginal information of car plate colouring information to be interfered, collect the license plate image with different quality, reduced the precision that car plate is identified.
Summary of the invention
For the above-mentioned defect existing in prior art, technical matters to be solved by this invention is how to have complex scene and to have the accurate location of realizing car plate under multiple disturbed condition.
For solving the problems of the technologies described above, on the one hand, the invention provides a kind of license plate locating method based on the conversion of height cap and connected domain, comprise step:
S1, obtains the gray level image of vehicle, and gray level image is carried out to pre-service;
S2, utilizes the conversion of height cap from extracting target image through pretreated gray level image, and target image is carried out to binary conversion treatment;
S3, calculates the connected domain of target image after binary conversion treatment, eliminates the connected domain of non-car plate target, realizes the coarse positioning to car plate;
S4, carries out slant correction and projection process to the car plate after coarse positioning, realizes the thin location to car plate, determines the particular location of car plate.
Preferably, described step S1 specifically comprises:
Obtain the gray level image of vehicle, gray level image is carried out to histogram equalization processing, obtain the first image;
Adopt smooth noise technology to carry out denoising to described the first image, obtain the second image.
Preferably, described step S2 specifically comprises:
Background figure, is local minimum by any pixel assignment of Background, completes the process of low cap conversion acquisition Background;
Creating low target figure, is local maximum by any pixel assignment of low target figure, completes the process of cap transformation acquisition low target figure;
Creating high target figure, is low target figure and the corresponding difference value of Background by any pixel assignment of high target figure, completes the process that obtains outstanding high target figure; Wherein, the width of Background, low target figure and high target figure and equal highly respectively;
Utilize inter-class variance to maximize Threshold Segmentation Algorithm, high target figure is carried out to binary conversion treatment, obtain the binary image as the high target figure of the 3rd image.
Preferably, described step S3 specifically comprises:
To each connected domain in the 3rd image, calculate width and height and the foreground pixel number of minimum boundary rectangle in this connected domain, eliminate the connected domain of non-car plate target;
Eliminate after the connected domain of non-car plate target, obtain the minimum boundary rectangle of remaining connected domain, realize the coarse positioning to car plate.
Preferably, in described step S3, eliminating non-car plate target connection specifically comprises:
By setting the width, height of minimum boundary rectangle and the threshold value of size of connected domain, width, height and the size threshold value of the minimum boundary rectangle of width, height and big or small and the setting of the minimum boundary rectangle calculating described in relatively;
If judge the width of this minimum boundary rectangle and be highly not less than setting width and height threshold value, this connected domain of cancellation; If or judge minimum boundary rectangle in connected domain be not less than or be not more than the big or small threshold value of setting minimum boundary rectangle, this connected domain of cancellation.
Preferably, described step S4 specifically comprises:
The 3rd image in minimum boundary rectangle is carried out to horizontal tilt and proofread and correct processing, obtain the 4th image;
The 4th image is carried out to vertical skew correction processing, obtain image the 5th image;
The 5th image is carried out to horizontal projection processing, obtain the upper and lower marginal position of car plate;
The 5th image is carried out to vertical projection processing, obtain car plate left and right edges position;
According to the marginal position up and down of car plate, realize the thin location to car plate, determine the particular location of car plate.
On the other hand, the present invention also provides a kind of Position System of automobile license plate location based on the conversion of height cap and connected domain simultaneously, and this system comprises: pretreatment module, for obtaining the gray level image of vehicle, and gray level image is carried out to pre-service;
Binarization block, for utilizing the conversion of height cap from extracting target image through pretreated gray level image, and carries out binary conversion treatment to target image;
Coarse positioning module, for calculating the connected domain of the target image after binary conversion treatment, eliminates the connected domain of non-car plate target, realizes the coarse positioning to car plate;
Thin locating module, for the car plate after coarse positioning is done to slant correction and projection, realizes the thin location to car plate, determines the particular location of car plate.
Preferably, described binarization block specifically comprises:
Low cap processing module, for background figure, is local minimum by any pixel assignment of Background, completes the process of low cap conversion acquisition Background;
High cap processing module, for creating low target figure, is local maximum by any pixel assignment of low target figure, completes the process of cap transformation acquisition low target figure;
The first acquisition module, for creating high target figure, is low target figure and the corresponding difference value of Background by any pixel assignment of high target figure, completes the process that obtains outstanding high target figure; Wherein, the width of Background, low target figure and high target figure and equal highly respectively;
The second acquisition module, for utilizing inter-class variance to maximize Threshold Segmentation Algorithm, carries out binary conversion treatment to high target figure, obtains the binary image as the high target figure of the 3rd image.
Preferably, described coarse positioning module specifically comprises:
Computing module, for each connected domain to the 3rd image, calculates width and height and the foreground pixel number of minimum boundary rectangle in this connected domain, eliminates the connected domain of non-car plate target;
Cancellation module, for eliminating after the connected domain of non-car plate target, obtains the minimum boundary rectangle of remaining connected domain, realizes the coarse positioning to car plate.
Preferably, described thin locating module specifically comprises:
Level correction module, proofreaies and correct processing for the 3rd image of minimum boundary rectangle being carried out to horizontal tilt, obtains the 4th image;
Vertical correction module, for the 4th image is carried out to vertical skew correction processing, obtains image the 5th image;
Horizontal projection module, for the 5th image is carried out to horizontal projection processing, obtains the upper and lower marginal position of car plate;
Vertical projection module, carries out vertical projection processing to the 5th image, obtains car plate left and right edges position;
Determination module, for according to the marginal position up and down of car plate, realizes the thin location to car plate, determines the particular location of car plate.
The invention provides a kind of license plate locating method and system thereof based on the conversion of height cap and connected domain, by vehicle gray level image is carried out to pre-service, reduce uneven illumination, fracture or stainedly car plate located to the impact causing, utilize the conversion of height cap to extract target image, and target image is done to binary conversion treatment, utilize connected domain to obtain the coarse positioning of car plate.Car plate in coarse positioning is done to slant correction, and the projection of recycling horizontal and vertical direction accurately navigates to the position of car plate.This license plate locating method based on the conversion of height cap and connected domain based on the conversion of height cap and connected domain, has solved car plate orientation problem under various interference preferably, has improved the precision of car plate location.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the license plate locating method based on the conversion of height cap and connected domain in one embodiment of the present of invention;
Fig. 2 is the structural representation of the Position System of automobile license plate location based on the conversion of height cap and connected domain in an alternative embodiment of the invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described.Obviously, described embodiment is for implementing preferred embodiments of the present invention, and described description is to illustrate that rule of the present invention is object, not in order to limit scope of the present invention.Protection scope of the present invention should with claim the person of being defined be as the criterion, based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under the prerequisite of not making creative work, belongs to the scope of protection of the invention.
In license plate locating method based on height cap conversion and connected domain in prior art, because uneven illumination is even, the interference of the rotation of car plate and distortion, fracture or the factor such as stained, can reduce the degree of accuracy that car plate is located.The present invention is by carrying out pre-service to vehicle gray level image, reduce even, the fracture of uneven illumination or stainedly car plate located to the impact causing, utilize height cap change detection to go out target image, and target image is carried out to binary conversion treatment, utilize connected domain to obtain the coarse positioning of car plate, and further the car plate in coarse positioning is done to slant correction, the projection of recycling horizontal and vertical direction accurately navigates to the position of car plate, thereby can effectively overcome these disturbing factors, improve the accuracy of car plate location.
Fig. 1 is the schematic flow sheet of the license plate locating method based on the conversion of height cap and connected domain in one embodiment of the present of invention, and as shown in Figure 1, the method comprising the steps of:
Step S1, obtains the gray level image of vehicle, and gray level image is carried out to pre-service.
Preferably, step S1 specifically comprises:
Obtain the gray level image of vehicle, gray level image is carried out to histogram equalization processing, obtain the first image; Adopt smooth noise technology to carry out denoising to the first image, obtain the second image.
Step S2, utilizes the conversion of height cap from extracting target image through pretreated gray level image, and target image is carried out to binary conversion treatment.
Preferably, step S2 specifically comprises:
Background figure, is local minimum by any pixel assignment of Background, completes the process of low cap conversion acquisition Background;
Creating low target figure, is local maximum by any pixel assignment of low target figure, completes the process of cap transformation acquisition low target figure;
Creating high target figure, is low target figure and the corresponding difference value of Background by any pixel assignment of high target figure, completes the process that obtains outstanding high target figure; Wherein, the width of Background, low target figure and high target figure and equal highly respectively;
Utilize inter-class variance to maximize Threshold Segmentation Algorithm, high target figure is carried out to binary conversion treatment, obtain the binary image as the high target figure of the 3rd image.
Step S3, calculates the connected domain of target image after binary conversion treatment, eliminates the connected domain of non-car plate target, realizes the coarse positioning to car plate.
Preferably, step S3 specifically comprises:
To each connected domain in the 3rd image, calculate width and height and the foreground pixel number of minimum boundary rectangle in this connected domain, eliminate the connected domain of non-car plate target;
Eliminate after the connected domain of non-car plate target, obtain the minimum boundary rectangle of remaining connected domain, realize the coarse positioning to car plate.
Preferably, in step S3, eliminating non-car plate target connection specifically comprises:
By setting the width, height of minimum boundary rectangle and the threshold value of size of connected domain, width, height and the size threshold value of the minimum boundary rectangle of width, height and big or small and the setting of the minimum boundary rectangle calculating described in relatively;
If judge the width of this minimum boundary rectangle and be highly not less than setting width and height threshold value, this connected domain of cancellation; If or judge minimum boundary rectangle in connected domain be not less than or be not more than the big or small threshold value of setting minimum boundary rectangle, this connected domain of cancellation.
Step S4, carries out slant correction and projection process to the car plate after coarse positioning, realizes the thin location to car plate, determines the particular location of car plate.
Preferably, step S4 specifically comprises:
The 3rd image in minimum boundary rectangle is carried out to horizontal tilt and proofread and correct processing, obtain the 4th image; The 4th image is carried out to vertical skew correction processing, obtain image the 5th image; The 5th image is carried out to horizontal projection processing, obtain the upper and lower marginal position of car plate; The 5th image is carried out to vertical projection processing, obtain car plate left and right edges position; According to the marginal position up and down of car plate, realize the thin location to car plate, determine the particular location of car plate.
Particularly, obtaining after license plate image, be transformed into the gray-scale map I of license plate image
gray, to gray-scale map I
graycarry out following operation, finally can obtain car plate at the position of image Rect
i(i=1,2 ..., n; N is car plate number), complete the location to car plate,
The gray level image that the first step operated, obtained vehicle, carries out pre-service: obtain the image of vehicle, vehicle image is converted to gray-scale map I
gray, to gray-scale map I
graydo histogram equalization, obtain the first image I
clahe; To I
claheadopt smooth noise technology to carry out denoising, obtain the second image I
denoising.
Second step operation, utilize the conversion of height cap from extracting target image through pretreated gray level image, and target image is carried out to binary conversion treatment comprise:
(1) low cap conversion obtains Background I
backgroud: create image I
backgroud, and to image I
backgroudany pixel Pixel
(i, j)(i=0,1 ..., height-1; J=0,1 ... width-1) assignment is MIN (Pixel'
(i-cLen/2, j), Pixel'
(i-cLen/2+1, j)..., Pixel'
(i+cLen/2, j)), wherein Pixel' is I
denoisingpixel value, height refers to I
denoisingheight, width refers to I
denoisingwidth, cLen gets the proportional numers of character duration, in the time of practical application, character duration can be predicted conventionally, I
backgroudthe width of image and highly equal I
denoisingwidth and height.
(2) cap transformation obtains low target figure I
objective_L: create image I
objective_L, and to image I
objective_Lany pixel I
objective_Lassignment is MAX (Pixel'
(i-cLen/2, j), Pixel'
(i-cLen/2+1, j)..., Pixel'
(i+cLen/2, j)), wherein Pixel' is I
denoisingpixel value, cLen gets the proportional numers of character duration, in the time of practical application, character duration can be predicted conventionally, wherein, I
objective_Limage wide high in I
denoisingwide height.
(3) obtain outstanding high target figure I
objective_H: create image I
objective_H, and to image I
objective_Hany pixel Pixel
(i, j)(i=0,1 ..., height-1; J=0,1 ... width-1) assignment is (Pixel'
(i, j)-Pixel "
(i, j)), wherein, Pixel' is I
objective_Lpixel value, Pixel " is I
backgroudpixel value, I
objective_Hthe width of image and highly equal I
objective_Lwidth and height, height refers to I
objective_Lheight, width refers to I
objective_Lwidth.
(4) obtain high target figure I
objective_Hbinary image I
bin_obj: utilize inter-class variance to maximize Threshold Segmentation Algorithm (Otsu), binary image I
objective_H, obtain image I
bin_obj.
The 3rd step operation, realize car plate coarse positioning and comprise:
(1) calculate connected domain C
i, to image I
bin_objin each connected domain C
i(i=1,2 ..., n; N is the number of connected domain), calculate its minimum boundary rectangle R
iwide (the h of height
i, w
i) and foreground pixel count F
i, eliminate the connected domain C of non-car plate target
i, can eliminate by two kinds of methods the connected domain C of non-car plate target
i:
Method 1, setting threshold T
h maxand T
w maxif, h
i>=T
h maxor w
i>=T
w max, cancellation connected domain C
i;
Method 2, setting threshold T
r maxand T
r minif, R
i>=T
r maxor R
i≤ T
r min, cancellation connected domain C
i;
(2) eliminate the connected domain C after non-car plate target
iminimum boundary rectangle R
iit is exactly the result of car plate coarse positioning.
FOUR EASY STEPS, realize the thin location of car plate and comprise:
1) horizontal tilt is proofreaied and correct: to R
iin image I
bin_objdo horizontal tilt correction and obtain image I
bin_obj_H_i;
2) vertical skew correction: to image I
bin_obj_H_ido vertical skew correction and obtain image I
bin_obj_HV_i;
3) to I
bin_obj_HV_ido horizontal projection, find lower limb Top on car plate
i, Bottom
i;
4) to I
bin_obj_HV_ido vertical projection, find car plate left and right edges Left
i, Right
i.
Now, found the particular location Rect of car plate
i(Top
i, Left
i, Bottom
i, Right
i).
In license plate locating method based on the conversion of height cap and connected domain provided by the invention, by vehicle gray level image is carried out to pre-service, reduce uneven illumination, fracture or stainedly car plate located to the impact causing, utilize the conversion of height cap to extract target image, and target image is done to binary conversion treatment, utilize connected domain to obtain the coarse positioning of car plate.Car plate in coarse positioning is done to slant correction, and the projection of recycling horizontal and vertical direction accurately navigates to the position of car plate.This license plate locating method based on the conversion of height cap and connected domain based on the conversion of height cap and connected domain, has solved car plate orientation problem under various interference preferably, has improved the precision of car plate location.
Fig. 2 is the structural representation of the Position System of automobile license plate location based on the conversion of height cap and connected domain in an alternative embodiment of the invention, and as shown in Figure 2, this system comprises: pretreatment module 201, binarization block 202, coarse positioning module 203 and thin locating module 204.Wherein, pretreatment module 201 is for obtaining the gray level image of vehicle, and gray level image is carried out to pre-service; Binarization block 202 extracts target image for utilizing height cap to convert from the pretreated gray level image of process, and target image is carried out to binary conversion treatment; Coarse positioning module 203, for the target image to after binary conversion treatment, is calculated connected domain and is realized the coarse positioning to license plate image; Thin locating module 204, for the license plate image after coarse positioning being done to slant correction and the license plate image after slant correction being done to the projection of horizontal and vertical direction, is realized the thin location to car plate.
Preferably, pretreatment unit 201 specifically comprises balance module 2011 and denoising module 2012, and wherein, balance module 2011, for obtaining the gray level image of vehicle, carries out histogram equalization processing to gray level image, obtains the first image; Denoising module 2012, for adopting smooth noise technology to carry out as denoising to the first image, obtains the second image.
Preferably, binarization block 202 specifically comprises low cap processing module 2021, high cap processing module 2022, the first acquisition module 2023 and the second acquisition module 2024, wherein, low cap processing module 2021 is for background figure, be local minimum by any pixel assignment of Background, complete the process of low cap conversion acquisition Background; High cap processing module 2022, for creating low target figure, is local maximum by any pixel assignment of low target figure, completes the process of cap transformation acquisition low target figure; The first acquisition module 2023, for creating high target figure, is low target figure and the corresponding difference value of Background by any pixel assignment of high target figure, completes the process that obtains outstanding high target figure; Wherein, the width of Background, low target figure and high target figure and equal highly respectively; The second acquisition module 2024, for utilizing inter-class variance to maximize Threshold Segmentation Algorithm, carries out binary conversion treatment to high target figure, obtains the binary image as the high target figure of the 3rd image.
Preferably, coarse positioning module 203 specifically comprises computing module 2031 and cancellation module 2032, wherein, computing module 2031 is for each connected domain to the 3rd image, calculate width and height and the foreground pixel number of minimum boundary rectangle in this connected domain, eliminate the connected domain of non-car plate target; Cancellation module 2032, for eliminating after the connected domain of non-car plate target, obtains the minimum boundary rectangle of remaining connected domain, realizes the coarse positioning to car plate.
Preferably, computing module 2031 specifically comprises Threshold module 20311 and cancellation module 20312, wherein, Threshold module 20311 is for the threshold value of width, height and the size of the minimum boundary rectangle by setting connected domain, and the relation of width, height and the size threshold value of width, height and the size of the minimum boundary rectangle of gained and the minimum boundary rectangle of setting is calculated in judgement; If cancellation module 20312 for judge the width of this minimum boundary rectangle and be highly not less than setting width and height threshold value, this connected domain of cancellation; If or judge minimum boundary rectangle in connected domain be not less than or be not more than the big or small threshold value of setting minimum boundary rectangle, this connected domain of cancellation.
Preferably, thin locating module 204 specifically comprises level correction module 2041, vertical correction module 2042, horizontal projection module 2043, vertical projection module 2044 and determination module 2045, wherein, level correction module 2041, proofread and correct processing for the 3rd image of minimum boundary rectangle being carried out to horizontal tilt, obtain the 4th image; Vertical correction module 2042, for the 4th image is carried out to vertical skew correction processing, obtains image the 5th image; Horizontal projection module 2043, for the 5th image is carried out to horizontal projection processing, obtains the upper and lower marginal position of car plate; Vertical projection module 2044, carries out vertical projection processing to the 5th image, obtains car plate left and right edges position; Determination module 2045, for according to the marginal position up and down of car plate, realizes the thin location to car plate, determines the particular location of car plate.
In Position System of automobile license plate location based on the conversion of height cap and connected domain provided by the invention, by vehicle gray level image is carried out to pre-service, reduce uneven illumination, fracture or stainedly car plate located to the impact causing, utilize the conversion of height cap to extract target image, and target image is done to binary conversion treatment, utilize connected domain to obtain the coarse positioning of car plate.Car plate in coarse positioning is done to slant correction, and the projection of the gentle vertical direction of recycling water S accurately navigates to the position of car plate.This license plate locating method based on the conversion of height cap and connected domain based on the conversion of height cap and connected domain, has solved car plate orientation problem under various interference preferably, has improved the precision of car plate location.
Below be only the preferred embodiment of the present invention, it should be pointed out that above-mentioned preferred implementation should not be considered as limitation of the present invention, protection scope of the present invention should be as the criterion with claim limited range.For those skilled in the art, without departing from the spirit and scope of the present invention, can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (10)
1. the license plate locating method based on the conversion of height cap and connected domain, is characterized in that, described method comprises step:
S1, obtains the gray level image of vehicle, and gray level image is carried out to pre-service;
S2, utilizes the conversion of height cap from extracting target image through pretreated gray level image, and target image is carried out to binary conversion treatment;
S3, calculates the connected domain of target image after binary conversion treatment, eliminates the connected domain of non-car plate target, realizes the coarse positioning to car plate;
S4, carries out slant correction and projection process to the car plate after coarse positioning, realizes the thin location to car plate, determines the particular location of car plate.
2. the method for claim 1, is characterized in that, described step S1 specifically comprises:
Obtain the gray level image of vehicle, gray level image is carried out to histogram equalization processing, obtain the first image;
Adopt smooth noise technology to carry out denoising to described the first image, obtain the second image.
3. method as claimed in claim 2, is characterized in that, described step S2 specifically comprises:
Background figure, is local minimum by any pixel assignment of Background, completes the process of low cap conversion acquisition Background;
Creating low target figure, is local maximum by any pixel assignment of low target figure, completes the process of cap transformation acquisition low target figure;
Creating high target figure, is low target figure and the corresponding difference value of Background by any pixel assignment of high target figure, completes the process that obtains outstanding high target figure; Wherein, the width of Background, low target figure and high target figure and equal highly respectively;
Utilize inter-class variance to maximize Threshold Segmentation Algorithm, high target figure is carried out to binary conversion treatment, obtain the binary image as the high target figure of the 3rd image.
4. method as claimed in claim 3, is characterized in that, described step S3 specifically comprises:
To each connected domain in the 3rd image, calculate width and height and the foreground pixel number of minimum boundary rectangle in this connected domain, eliminate the connected domain of non-car plate target;
Eliminate after the connected domain of non-car plate target, obtain the minimum boundary rectangle of remaining connected domain, realize the coarse positioning to car plate.
5. method as claimed in claim 4, is characterized in that, eliminates non-car plate target connection and specifically comprise in described step S3:
By setting the width, height of minimum boundary rectangle and the threshold value of size of connected domain, width, height and the size threshold value of the minimum boundary rectangle of width, height and big or small and the setting of the minimum boundary rectangle calculating described in relatively;
If judge the width of this minimum boundary rectangle and be highly not less than setting width and height threshold value, this connected domain of cancellation; If or judge minimum boundary rectangle in connected domain be not less than or be not more than the big or small threshold value of setting minimum boundary rectangle, this connected domain of cancellation.
6. method as claimed in claim 5, is characterized in that, described step S4 specifically comprises:
The 3rd image in minimum boundary rectangle is carried out to horizontal tilt and proofread and correct processing, obtain the 4th image;
The 4th image is carried out to vertical skew correction processing, obtain image the 5th image;
The 5th image is carried out to horizontal projection processing, obtain the upper and lower marginal position of car plate;
The 5th image is carried out to vertical projection processing, obtain car plate left and right edges position;
According to the marginal position up and down of car plate, realize the thin location to car plate, determine the particular location of car plate.
7. the Position System of automobile license plate location based on the conversion of height cap and connected domain, is characterized in that, comprising:
Pretreatment module, for obtaining the gray level image of vehicle, and carries out pre-service to gray level image;
Binarization block, for utilizing the conversion of height cap from extracting target image through pretreated gray level image, and carries out binary conversion treatment to target image;
Coarse positioning module, for calculating the connected domain of the target image after binary conversion treatment, eliminates the connected domain of non-car plate target, realizes the coarse positioning to car plate;
Thin locating module, for the car plate after coarse positioning is done to slant correction and projection, realizes the thin location to car plate, determines the particular location of car plate.
8. system as claimed in claim 7, is characterized in that, described binarization block specifically comprises:
Low cap processing module, for background figure, is local minimum by any pixel assignment of Background, completes the process of low cap conversion acquisition Background;
High cap processing module, for creating low target figure, is local maximum by any pixel assignment of low target figure, completes the process of cap transformation acquisition low target figure;
The first acquisition module, for creating high target figure, any pixel assignment of high target figure is low target figure and the corresponding difference value of Background, completes the process that obtains outstanding high target figure; Wherein, the width of Background, low target figure and high target figure and equal highly respectively;
The second acquisition module, for utilizing inter-class variance to maximize Threshold Segmentation Algorithm, carries out binary conversion treatment to high target figure, obtains the binary image as the high target figure of the 3rd image.
9. system as claimed in claim 7, is characterized in that, described coarse positioning module specifically comprises:
Computing module, for each connected domain to the 3rd image, calculates width and height and the foreground pixel number of minimum boundary rectangle in this connected domain, eliminates the connected domain of non-car plate target;
Cancellation module, for eliminating after the connected domain of non-car plate target, obtains the minimum boundary rectangle of remaining connected domain, realizes the coarse positioning to car plate.
10. system as claimed in claim 7, is characterized in that, described thin locating module specifically comprises:
Level correction module, proofreaies and correct processing for the 3rd image of minimum boundary rectangle being carried out to horizontal tilt, obtains the 4th image;
Vertical correction module, for the 4th image is carried out to vertical skew correction processing, obtains image the 5th image;
Horizontal projection module, for the 5th image is carried out to horizontal projection processing, obtains the upper and lower marginal position of car plate;
Vertical projection module, carries out vertical projection processing to the 5th image, obtains car plate left and right edges position;
Determination module, for according to the marginal position up and down of car plate, realizes the thin location to car plate, determines the particular location of car plate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410257831.7A CN103984949B (en) | 2014-06-11 | 2014-06-11 | License plate locating method and its system based on high bot-hat transformation and connected domain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410257831.7A CN103984949B (en) | 2014-06-11 | 2014-06-11 | License plate locating method and its system based on high bot-hat transformation and connected domain |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103984949A true CN103984949A (en) | 2014-08-13 |
| CN103984949B CN103984949B (en) | 2017-07-21 |
Family
ID=51276911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410257831.7A Active CN103984949B (en) | 2014-06-11 | 2014-06-11 | License plate locating method and its system based on high bot-hat transformation and connected domain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103984949B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104361333A (en) * | 2014-12-10 | 2015-02-18 | 东方网力科技股份有限公司 | Traffic speed limit sign recognition method and device |
| CN104836986A (en) * | 2015-04-15 | 2015-08-12 | 华东师范大学 | Video information intelligent display method and apparatus |
| CN105117726A (en) * | 2015-08-07 | 2015-12-02 | 南京富士通南大软件技术有限公司 | License plate positioning method based on multi-feature area accumulation |
| CN106600620A (en) * | 2016-12-28 | 2017-04-26 | 天津普达软件技术有限公司 | Alarm method of unsuccessful landing of infusion bag in pallet on production line on conveyor belt |
| CN107016389A (en) * | 2017-04-07 | 2017-08-04 | 广东工业大学 | The method and device of a kind of License Plate |
| CN109389566A (en) * | 2018-10-19 | 2019-02-26 | 辽宁奇辉电子系统工程有限公司 | Subway height adjusting valve fastening nut defective mode detection method based on boundary characteristic |
| CN109584266A (en) * | 2018-11-15 | 2019-04-05 | 腾讯科技（深圳）有限公司 | A kind of object detection method and device |
| CN111611995A (en) * | 2020-04-02 | 2020-09-01 | 陕西土豆数据科技有限公司 | Method for positioning license plate of road video image |
| CN113095320A (en) * | 2021-04-01 | 2021-07-09 | 湖南大学 | License plate recognition method and system and computing device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246551A (en) * | 2008-03-07 | 2008-08-20 | 北京航空航天大学 | Fast license plate locating method |
| CN102999753A (en) * | 2012-05-07 | 2013-03-27 | 腾讯科技（深圳）有限公司 | License plate locating method |
| 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
- 2014-06-11 CN CN201410257831.7A patent/CN103984949B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246551A (en) * | 2008-03-07 | 2008-08-20 | 北京航空航天大学 | Fast license plate locating method |
| CN102999753A (en) * | 2012-05-07 | 2013-03-27 | 腾讯科技（深圳）有限公司 | License plate locating method |
| 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 (1)
| Title |
|---|
| 杨萱: "车牌识别系统中定位算法的研究", 《中国优秀硕士学位论文全文数据库（电子期刊）》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104361333A (en) * | 2014-12-10 | 2015-02-18 | 东方网力科技股份有限公司 | Traffic speed limit sign recognition method and device |
| CN104836986A (en) * | 2015-04-15 | 2015-08-12 | 华东师范大学 | Video information intelligent display method and apparatus |
| CN105117726B (en) * | 2015-08-07 | 2018-10-09 | 南京富士通南大软件技术有限公司 | License plate locating method based on multiple features zone-accumulation |
| CN105117726A (en) * | 2015-08-07 | 2015-12-02 | 南京富士通南大软件技术有限公司 | License plate positioning method based on multi-feature area accumulation |
| CN106600620A (en) * | 2016-12-28 | 2017-04-26 | 天津普达软件技术有限公司 | Alarm method of unsuccessful landing of infusion bag in pallet on production line on conveyor belt |
| CN107016389B (en) * | 2017-04-07 | 2020-09-11 | 广东工业大学 | License plate positioning method and device |
| CN107016389A (en) * | 2017-04-07 | 2017-08-04 | 广东工业大学 | The method and device of a kind of License Plate |
| CN109389566A (en) * | 2018-10-19 | 2019-02-26 | 辽宁奇辉电子系统工程有限公司 | Subway height adjusting valve fastening nut defective mode detection method based on boundary characteristic |
| CN109389566B (en) * | 2018-10-19 | 2022-01-11 | 辽宁奇辉电子系统工程有限公司 | Method for detecting bad state of fastening nut of subway height adjusting valve based on boundary characteristics |
| CN109584266A (en) * | 2018-11-15 | 2019-04-05 | 腾讯科技（深圳）有限公司 | A kind of object detection method and device |
| CN109584266B (en) * | 2018-11-15 | 2023-06-09 | 腾讯科技（深圳）有限公司 | Target detection method and device |
| CN111611995A (en) * | 2020-04-02 | 2020-09-01 | 陕西土豆数据科技有限公司 | Method for positioning license plate of road video image |
| CN111611995B (en) * | 2020-04-02 | 2023-05-23 | 土豆数据科技集团有限公司 | Method applied to positioning of highway video image license plate |
| CN113095320A (en) * | 2021-04-01 | 2021-07-09 | 湖南大学 | License plate recognition method and system and computing device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103984949B (en) | 2017-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103984949A (en) | License plate positioning method and system based on high and low cap transformation and connected domain | |
| CN105160309B (en) | Three lanes detection method based on morphological image segmentation and region growing | |
| CN102043950B (en) | Vehicle outline recognition method based on canny operator and marginal point statistic | |
| Li et al. | Nighttime lane markings recognition based on Canny detection and Hough transform | |
| CN103065138B (en) | Recognition method of license plate number of motor vehicle | |
| CN111563412B (en) | Rapid lane line detection method based on parameter space voting and Bessel fitting | |
| CN102419820A (en) | Method for rapidly detecting car logo in videos and images | |
| CN107273896A (en) | A kind of car plate detection recognition methods based on image recognition | |
| CN104700072B (en) | Recognition methods based on lane line historical frames | |
| CN107862290A (en) | Method for detecting lane lines and system | |
| CN105893949A (en) | Lane line detection method under complex road condition scene | |
| CN102693423A (en) | Method for precise positioning of license plate in strong light conditions | |
| CN103136528B (en) | A kind of licence plate recognition method based on dual edge detection | |
| CN103593981B (en) | A kind of model recognizing method based on video | |
| CN107392103A (en) | The detection method and device of road surface lane line, electronic equipment | |
| CN107895151A (en) | Method for detecting lane lines based on machine vision under a kind of high light conditions | |
| CN103116751A (en) | Automatic license plate character recognition method | |
| CN102364496A (en) | Method and system for identifying automobile license plates automatically based on image analysis | |
| CN103927526A (en) | Vehicle detecting method based on Gauss difference multi-scale edge fusion | |
| CN104537348B (en) | Special vehicle recognition methods and its identifying system | |
| CN105528609A (en) | Vehicle license plate location method based on character position | |
| CN104036241A (en) | License plate recognition method | |
| CN107909047A (en) | A kind of automobile and its lane detection method and system of application | |
| CN110334692A (en) | A kind of blind way recognition methods based on image procossing | |
| CN110674812B (en) | Civil license plate positioning and character segmentation method facing complex background |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210721 Address after: No. 6, Jiuhua Road, khuchuang Park, Mianyang, Sichuan Patentee after: Sichuan Jiuzhou Investment Holding Group Co.,Ltd. Address before: No. 6, Jiuhua Road, khuchuang Park, Mianyang, Sichuan Patentee before: SICHUAN JIUZHOU ELECTRIC GROUP Co.,Ltd. |
|
| TR01 | Transfer of patent right |