CN106355180A - Method for positioning license plates on basis of combination of color and edge features - Google Patents
Method for positioning license plates on basis of combination of color and edge features Download PDFInfo
- Publication number
- CN106355180A CN106355180A CN201610807379.6A CN201610807379A CN106355180A CN 106355180 A CN106355180 A CN 106355180A CN 201610807379 A CN201610807379 A CN 201610807379A CN 106355180 A CN106355180 A CN 106355180A
- Authority
- CN
- China
- Prior art keywords
- license plate
- car plate
- plate
- car
- svm
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for positioning license plates on the basis of combination of color and edge features, and belongs to the technical field of intelligent traffic. The method includes positioning the license plates for input images by the aid of the color features of the license plates, judging the license plates by the aid of SVM (support vector machine) models and acquiring the license plates; positioning the license plates for the input images by the aid of the edge features of the license plates if the license plates are not positioned, judging the license plates by the aid of the SVM models and acquiring the license plates. The method has the advantage that the license plates can be quickly and effectively positioned for the license plate images with complicated backgrounds.
Description
Technical field
The present invention relates to a kind of license plate locating method combining with edge feature based on color, belong to intelligent transport technology
Field.
Background technology
With the fast development of Modern Traffic, the modern management to vehicle increasingly tends to automatization, the automatic inspection to vehicle
Survey, the demand of identifying system increasingly increases, for example high speed crossing charging aperture, cell automatic parking lot management etc..Existing in order to adapt to
For the development of traffic, intelligentized traffic control system is arisen at the historic moment, and one of important step is namely based on the identification of car plate
System.License plate recognition technology is a very important research topic in modern intelligent transportation field, and license plate recognition technology is main
Including three parts such as License Plate, License Plate Character Segmentation and Recognition of License Plate Characters, and during License Plate is Car license recognition
One link of most critical, directly affects the performance of Vehicle License Plate Recognition System.
For the license plate image that background is complicated, how correctly, quickly and efficiently carry out License Plate, be that current car plate is known
One of most important task in other system.At present, license plate locating method is broadly divided into two classes, and that is, the car plate based on gray level image is fixed
Method for position and the license plate locating method based on coloured image.License plate locating method based on gray level image mainly adopts rim detection
Operator carries out rim detection to gray level image, thus obtaining candidate license plate.Common edge detection operator has roberts edge to examine
Measuring and calculating, sobel edge detection operator and prewitt boundary operator etc..Experiment shows, sobel edge detection operator can not only
Preferably project car plate edge feature, and speed.Based on the localization method of sobel edge detection operator, idiographic flow
As indicated with 1, the method carries out Gaussian Blur to the coloured image of input and filters interference noise figure first, and the image of denoising is entered
Row gray processing;Then adopt sobel edge detection operator to extract the vertical edge of gray level image, two are carried out to edge image simultaneously
Value is processed, and extracts all profiles of image using closed operation;Finally according to the area of car plate, length-width ratio etc. to profile
Little boundary rectangle is screened, thus extracting candidate license plate.But, be have ignored based on the license plate locating method of sobel operator
The colouring information of coloured image is although many car plates can fast and effeciently be oriented, but interlocks in license plate image vertical edge
In the case of, the method cannot orient car plate.Meanwhile, the method can only cursorily orient car plate.
Compared with coloured image, gray level image have lost colouring information, so coloured image is more conducive to License Plate, and
Car plate can be accurately positioned out when complex background, different illumination.In Digital Image Processing, conventional color model has
Rgb model and hsv model, 3 components r, g, b of rgb model change with the change of brightness, and hsv model colourity
H, saturation s, 3 representation in components of brightness v, wherein h, s contains the colour information of image, is not affected by brightness, and v component
Only contain monochrome information, 3 components are separate.Therefore, hsv model is mainly adopted using the license plate locating method of color.
License plate locating method based on hsv space is as shown in Fig. 2 the rgb image of input is converted into the figure in hsv space by the method first
Picture;Then h, s, v threshold range being respectively adopted setting carries out binaryzation to the image in hsv space, and using closed operation and carries
Take all profiles;Finally according to the angle of inclination of car plate, length-width ratio and edge segmentation error, the minimum enclosed rectangle of profile is entered
Row screening, thus obtain car plate.Localization method profit compared with the license plate locating method based on sobel operator, based on hsv space
With colouring information, improve the success rate of License Plate.But for car plate color is identical with body color, illumination is not enough or
Situations such as exposure, the method cannot efficiently locate out car plate.Meanwhile, this localization method mainly according to artificial experience to candidate's car
Board is screened, and robustness is poor.
Content of the invention
The technical problem to be solved is the defect overcoming prior art, provides a kind of special with edge based on color
Levy the license plate locating method combining, can fast and effeciently orient car plate in the complicated license plate image of background.
For solving above-mentioned technical problem, the present invention provides a kind of License Plate side combining based on color with edge feature
Method, comprises the following steps:
1) License Plate is carried out to input picture using the color characteristic of car plate, then entered using svm car plate discrimination model
Driving board judges, obtains car plate;
2) if the car plate quantity oriented by color characteristic is more than maxplate, do not need to carry out sobel twice
Positioning, is otherwise carried out License Plate using the edge feature of car plate to input picture, is then carried out using svm car plate discrimination model
Car plate judges, obtains car plate;Described maxplate represents the number of car plate in a width license plate image.
Aforesaid stating carries out License Plate using color characteristic to input picture, specifically comprises the following steps that
1-1) by input license plate image color space from rgb spatial transformation be hsv space, carry out histogram equalization simultaneously
Change is processed;
1-2) it is respectively adopted blue matching template, with yellow matching template, binaryzation is carried out to pretreated license plate image
Process;
1-3) closed operation is carried out to the license plate image of binaryzation, then extract all profiles of this image;
1-4) minimum enclosed rectangle is taken to each profile, then the angle of inclination according to car plate and the preliminary filtration one of length-width ratio
Underproof candidate license plate a bit, is finally standardized to the size of car plate processing;
1-5) using the svm car plate discrimination model training to through step 1-4) filter after remaining candidate license plate carry out
Judge, obtain car plate.
The aforesaid edge feature using car plate carries out License Plate to input picture, specifically comprises the following steps that
2-1) license plate image of input is carried out with Gaussian mode gelatinizing and process filtration interference noise, then ash is carried out to this image
Degreeization;
2-2) level and vertical edge are extracted respectively by sobel operator to the image of gray processing, obtain the edge of car plate
Image;
2-3) to step 2-2) the car plate edge image that obtains carries out binaryzation successively, closed operation is processed, and then extracts institute
There is profile, finally repairing treatment carried out to the chain rupture part in profile, edge image is obtained again using sobel operator simultaneously,
Thus extracting all profiles;
2-4) minimum enclosed rectangle is taken to each profile, simultaneously the angle of inclination according to car plate and the preliminary filtration one of length-width ratio
Underproof candidate license plate a bit, is then standardized to the size of remaining candidate license plate;
2-5) using the svm car plate discrimination model training to through step 2-4) filter after remaining candidate license plate carry out
Judge, obtain car plate.
The aforesaid angle of inclination according to car plate and length-width ratio tentatively filter some underproof candidate license plate and refer to, if
The absolute value of license plate sloped angle is less than 30 degree, and the length-width ratio of car plate between 2-4, then retains this candidate license plate, otherwise gives up
Abandon this candidate license plate.
The acquisition of aforesaid svm car plate discrimination model, specifically comprises the following steps that
3-1) produce substantial amounts of candidate license plate, be respectively adopted color characteristic and with edge feature, input license plate image carried out determining
Position, obtains substantial amounts of candidate license plate;
3-2) label for candidate license plate;
3-3) build car plate training set, candidate license plate includes the real car plate of two classes and non-car plate, respectively from real car plate figure
A number of picture composition car plate training set is extracted, remaining picture constitutes car plate test in piece collection and non-car plate pictures
Collection;
3-4) train svm car plate discrimination model, by car plate training set, svm model is trained, obtain svm car plate and sentence
Other model, adopts car plate test set to verify svm car plate discrimination model simultaneously.
Aforesaid step 3-2) in, the method using successive iteration automatic labeling label is labelled for each candidate license plate, tool
Body step is as follows:
3-2-1) choose the candidate license plate of a number of unlabelled, artificial labeling is carried out to these candidate license plate simultaneously
Sign;
3-2-2) using svm model, the candidate license plate labeled is trained, obtains a svm car plate and differentiate
Model;
3-2-3) choose the candidate license plate of a number of unlabelled again, using the svm car plate discrimination model training
Candidate license plate automatic labeling label to these unlabelleds, simultaneously manual confirmation label mistake candidate license plate;
3-2-4) candidate license plate correctly labeled is combined, then carries out step 3-2-2), until
All candidate license plate all correctly post label.
Aforesaid step 3-3) in, the picture extracting 70% from real car plate pictures with non-car plate pictures constitutes car
Board training set.
Aforesaid maxplate value takes 1.
The aforesaid size to car plate is standardized referring to, the size of pick-up board is 136*36.
The beneficial effect that the present invention is reached:
The present invention is directed to the complicated license plate image of background, can fast and effeciently orient car plate.
Brief description
Fig. 1 is the license plate locating method flow chart based on sobel boundary operator;
Fig. 2 is the license plate locating method flow chart based on hsv space;
Fig. 3 is the training of svm car plate discrimination model and the process judging car plate;
Fig. 4 is the flow chart of the license plate locating method of the present invention.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings.Following examples are only used for clearly illustrating the present invention
Technical scheme, and can not be limited the scope of the invention with this.
The license plate locating method being combined with edge feature based on color of the present invention, as shown in figure 4, include following walking
Rapid:
1st, License Plate is carried out to input picture using the color characteristic of car plate, then entered using svm car plate discrimination model
Driving board judges, obtains car plate.License Plate is carried out to input picture using color characteristic, specifically comprises the following steps that
1-1) by input license plate image color space from rgb spatial transformation be hsv space, carry out histogram equalization simultaneously
Change is processed;
1-2) it is respectively adopted blue matching template, with yellow matching template, binaryzation is carried out to pretreated license plate image
Process;For example: for each pixel in image, if the value of h falls between 95-140 and s value respectively falls in 70- with v value
Between 255 and 60-255, then this pixel is set to white pixel, otherwise for black picture element.Table 1 is respectively blue and yellow
Join h, s, v span of template.
Table 1 yellow and blue matching template
h_min | h_max | s_min | s_max | v_min | v_max | |
Blue matching template | 95 | 140 | 70 | 255 | 60 | 255 |
Yellow matching template | 12 | 40 | 70 | 255 | 60 | 255 |
1-3) closed operation is carried out to the license plate image of binaryzation, then extract all profiles of this image;
1-4) minimum enclosed rectangle is taken to each profile, then the angle of inclination according to car plate and the preliminary filtration of length-width ratio are big
Partly underproof candidate license plate, is finally standardized to the size of car plate processing;The size of typically Chinese car plate is
440mm*140mm, length-width ratio is that the angle of inclination of car plate in 3.14, and practical application also will not be very big, typically spends -30
Between 30 degree.If the absolute value of license plate sloped angle is less than 30 degree, and the length-width ratio of car plate between 2-4, then retains this time
Select car plate, otherwise give up this candidate license plate.Therefore, this process can filter 80% underproof candidate license plate.
1-5) using the svm car plate discrimination model training to through step 1-4) after remaining candidate license plate judge,
Obtain car plate.
Svm car plate discrimination model requires the size of every car plate identical, otherwise car plate cannot be learnt and differentiate.Root
The mean breadth of a few thousand sheets car plates and average height according to statistics, the size of pick-up board of the present invention is 136*36.
If the car plate quantity that 2 are oriented by color characteristic is more than maxplate, do not need to carry out sobel twice
Positioning, is otherwise carried out License Plate using the edge feature of car plate to input picture, is then carried out using svm car plate discrimination model
Car plate judges, obtains car plate.Maxplate represents the number of car plate in a width license plate image, and in the present invention, this value is 1.
Edge feature using car plate carries out License Plate to input picture, specifically comprises the following steps that
2-1) license plate image of input is carried out with Gaussian mode gelatinizing and process filtration interference noise, then ash is carried out to this image
Degreeization;
2-2) level and vertical edge are extracted respectively by sobel operator to the image of gray processing, obtain the edge of car plate
Image;
2-3) to step 2-2) the car plate edge image that obtains carries out binaryzation successively, closed operation is processed, and then extracts institute
There is profile, finally repairing treatment is carried out to the chain rupture part in profile, adopt sobel operator to obtain more accurately edge simultaneously
Image, thus extract all profiles;
2-4) minimum enclosed rectangle is taken to each profile, the length-width ratio according to car plate is preliminary with angle of inclination simultaneously filters one
Underproof candidate license plate a bit, is then standardized to the size of remaining candidate license plate;Filter principle and step 1-4) identical;
2-5) using the svm car plate discrimination model training to through step 2-4) after remaining candidate license plate sentence
Disconnected, obtain car plate.
In this step, the license plate locating method based on sobel boundary operator can be realized using hardware.
Step 1-5) and step 2-5) in svm car plate discrimination model acquisition, as shown in figure 3, specifically comprising the following steps that
3-1) produce substantial amounts of candidate license plate.It is respectively adopted color characteristic with edge feature, input license plate image to be carried out determining
Position, obtains substantial amounts of candidate license plate.Candidate license plate comprises real car plate and non-car plate two class.
3-2) label for candidate license plate.Artificial labelling for each candidate license plate is a very big workload, this
Invention is labelled for each candidate license plate using a kind of method of successive iteration automatic labeling label.The method is first to having pasted
The candidate license plate collection training of label obtains a svm car plate discrimination model, then adopts candidate's car to unlabelled for this model
Board collection carries out automatic labeling label.Constantly repeat this process, until all candidate license plate are all correctly labeled, comprise the following steps that
Shown:
These candidate license plate are manually labelled by the 3-2-1) candidate license plate of selected part unlabelled simultaneously;
3-2-2) using svm model, the candidate license plate labeled is trained, obtains a svm car plate and differentiate
Model;
These are waited by the 3-2-3) candidate license plate of selected part unlabelled using the svm car plate discrimination model training
Select car plate automatic labeling label, manual confirmation labels the candidate license plate of mistake simultaneously;
3-2-4) candidate license plate correctly labeled is combined, then carries out step 3-2-2), until
All candidate license plate all correctly post label.
3-3) build car plate training set.Candidate license plate includes the real car plate of two classes and non-car plate, respectively from real car plate figure
Extract a number of picture in piece collection and non-car plate pictures and constitute car plate training set, that is, respectively from real car plate picture with non-
The picture obtaining 70% in car plate picture constitutes car plate training set, and remaining picture constitutes car plate test set, test training svm car
Board discrimination model.
3-4) train svm car plate discrimination model.By car plate training set, svm model is trained, obtains svm car plate and sentence
Other model, adopts car plate test set to verify svm car plate discrimination model simultaneously.
The present invention is differentiated to candidate license plate using svm model, and svm model is machine learning algorithm.Therefore, also may be used
To be differentiated to candidate license plate using the others machine learning algorithm such as neutral net (bp).
The above is only the preferred embodiment of the present invention it is noted that ordinary skill people for the art
For member, on the premise of without departing from the technology of the present invention principle, some improvement can also be made and deform, these improve and deform
Also should be regarded as protection scope of the present invention.
Claims (9)
1. a kind of license plate locating method being combined with edge feature based on color is it is characterised in that comprise the following steps:
1) License Plate is carried out to input picture using the color characteristic of car plate, then driving is entered using svm car plate discrimination model
Board judges, obtains car plate;
2) if the car plate quantity oriented by color characteristic is more than maxplate, do not need to carry out sobel positioning twice,
Otherwise License Plate is carried out to input picture using the edge feature of car plate, then car plate is carried out using svm car plate discrimination model
Judge, obtain car plate;Described maxplate represents the number of car plate in a width license plate image.
2. a kind of license plate locating method being combined with edge feature based on color according to claim 1, its feature exists
In, described License Plate is carried out to input picture using color characteristic, specifically comprise the following steps that
1-1) by input license plate image color space from rgb spatial transformation be hsv space, carry out at histogram equalization simultaneously
Reason;
1-2) it is respectively adopted blue matching template, with yellow matching template, binary conversion treatment is carried out to pretreated license plate image;
1-3) closed operation is carried out to the license plate image of binaryzation, then extract all profiles of this image;
1-4) minimum enclosed rectangle is taken to each profile, then the angle of inclination according to car plate and length-width ratio tentatively filter some not
Qualified candidate license plate, is finally standardized to the size of car plate processing;
1-5) using the svm car plate discrimination model training to through step 1-4) filter after remaining candidate license plate sentence
Disconnected, obtain car plate.
3. a kind of license plate locating method being combined with edge feature based on color according to claim 1, its feature exists
In the described edge feature using car plate carries out License Plate to input picture, specifically comprises the following steps that
2-1) license plate image of input is carried out with Gaussian mode gelatinizing and process filtration interference noise, then gray scale is carried out to this image
Change;
2-2) level and vertical edge are extracted respectively by sobel operator to the image of gray processing, obtain the edge image of car plate;
2-3) to step 2-2) the car plate edge image that obtains carries out binaryzation successively, closed operation is processed, and then extracts all wheels
Exterior feature, finally carries out repairing treatment to the chain rupture part in profile, obtains edge image again using sobel operator simultaneously, thus
Extract all profiles;
2-4) minimum enclosed rectangle is taken to each profile, the angle of inclination according to car plate and length-width ratio tentatively filter some not simultaneously
Qualified candidate license plate, is then standardized to the size of remaining candidate license plate;
2-5) using the svm car plate discrimination model training to through step 2-4) filter after remaining candidate license plate sentence
Disconnected, obtain car plate.
4. a kind of license plate locating method being combined with edge feature based on color according to Claims 2 or 3, its feature
It is, the described angle of inclination according to car plate and length-width ratio tentatively filter some underproof candidate license plate and refer to, if car plate
The absolute value at angle of inclination is less than 30 degree, and the length-width ratio of car plate between 2-4, then retains this candidate license plate, otherwise gives up this
Candidate license plate.
5. a kind of license plate locating method being combined with edge feature based on color according to Claims 2 or 3, its feature
It is that the acquisition of described svm car plate discrimination model specifically comprises the following steps that
3-1) produce substantial amounts of candidate license plate, be respectively adopted color characteristic and with edge feature, input license plate image positioned,
Obtain substantial amounts of candidate license plate;
3-2) label for candidate license plate;
3-3) build car plate training set, candidate license plate includes the real car plate of two classes and non-car plate, respectively from real car plate pictures
Constitute car plate training set with extracting a number of picture in non-car plate pictures, remaining picture constitutes car plate test set;
3-4) train svm car plate discrimination model, by car plate training set, svm model is trained, obtain svm car plate and differentiate mould
Type, adopts car plate test set to verify svm car plate discrimination model simultaneously.
6. a kind of license plate locating method being combined with edge feature based on color according to claim 5, its feature exists
In described step 3-2) in, the method using successive iteration automatic labeling label is labelled for each candidate license plate, and concrete steps are such as
Shown in lower:
3-2-1) choose the candidate license plate of a number of unlabelled, these candidate license plate are manually labelled simultaneously;
3-2-2) using svm model, the candidate license plate labeled is trained, obtains a svm car plate and differentiate mould
Type;
3-2-3) choose the candidate license plate of a number of unlabelled again, using the svm car plate discrimination model training to this
The candidate license plate automatic labeling label of a little unlabelleds, simultaneously manual confirmation label wrong candidate license plate;
3-2-4) candidate license plate correctly labeled is combined, then carries out step 3-2-2), until all
Candidate license plate all correctly posts label.
7. a kind of license plate locating method being combined with edge feature based on color according to claim 5, its feature exists
In described step 3-3) in, the picture extracting 70% from real car plate pictures with non-car plate pictures constitutes car plate training
Collection.
8. a kind of license plate locating method being combined with edge feature based on color according to claim 1, its feature exists
In described maxplate value takes 1.
9. a kind of license plate locating method being combined with edge feature based on color according to Claims 2 or 3, its feature
It is, the described size to car plate is standardized referring to, the size of pick-up board is 136*36.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610807379.6A CN106355180B (en) | 2016-09-07 | 2016-09-07 | A kind of license plate locating method combined based on color with edge feature |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610807379.6A CN106355180B (en) | 2016-09-07 | 2016-09-07 | A kind of license plate locating method combined based on color with edge feature |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106355180A true CN106355180A (en) | 2017-01-25 |
CN106355180B CN106355180B (en) | 2019-07-02 |
Family
ID=57859712
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610807379.6A Active CN106355180B (en) | 2016-09-07 | 2016-09-07 | A kind of license plate locating method combined based on color with edge feature |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106355180B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109239900A (en) * | 2018-11-07 | 2019-01-18 | 华东师范大学 | A kind of full-automatic quick focusing method for the big visual field acquisition of microscopic digital image |
CN109359651A (en) * | 2018-11-08 | 2019-02-19 | 东风商用车有限公司 | A kind of License Plate processor and its location processing method |
CN109871849A (en) * | 2019-01-11 | 2019-06-11 | 西安艾润物联网技术服务有限责任公司 | A kind of method and system of vehicle identification |
CN110533027A (en) * | 2019-07-22 | 2019-12-03 | 浙江省北大信息技术高等研究院 | A kind of mobile device-based text detection and recognition methods and system |
CN111079744A (en) * | 2019-12-06 | 2020-04-28 | 鲁东大学 | Intelligent vehicle license plate identification method and device suitable for complex illumination environment |
CN111652222A (en) * | 2020-07-13 | 2020-09-11 | 深圳市智搜信息技术有限公司 | License plate positioning method and device, computer equipment and storage medium |
CN112488031A (en) * | 2020-12-11 | 2021-03-12 | 华能华家岭风力发电有限公司 | Safety helmet detection method based on color segmentation |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012145909A1 (en) * | 2011-04-28 | 2012-11-01 | 中国科学院自动化研究所 | Method for detecting tampering with color digital image based on chroma of image |
CN103324958A (en) * | 2013-06-28 | 2013-09-25 | 浙江大学苏州工业技术研究院 | License plate locating method based on projection method and SVM under complex background |
CN105488797A (en) * | 2015-11-25 | 2016-04-13 | 安徽创世科技有限公司 | License plate location method for HSV space |
-
2016
- 2016-09-07 CN CN201610807379.6A patent/CN106355180B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012145909A1 (en) * | 2011-04-28 | 2012-11-01 | 中国科学院自动化研究所 | Method for detecting tampering with color digital image based on chroma of image |
CN103324958A (en) * | 2013-06-28 | 2013-09-25 | 浙江大学苏州工业技术研究院 | License plate locating method based on projection method and SVM under complex background |
CN105488797A (en) * | 2015-11-25 | 2016-04-13 | 安徽创世科技有限公司 | License plate location method for HSV space |
Non-Patent Citations (4)
Title |
---|
HUILI HAN 等: "Method of license plate location based on edge detection and color information", 《PROCEEDINGS 2011 INTERNATIONAL CONFERENCE ON TRANSPORTATION, MECHANICAL, AND ELECTRICAL ENGINEERING (TMEE)》 * |
苗姣姣 等: "HSV空间和形态学处理相结合的车牌定位方法", 《电视技术》 * |
郭延祥 等: "基于边缘检测和颜色纹理直方图的车牌定位方法", 《计算机科学与探索》 * |
郭捷 等: "基于颜色和纹理分析的车牌定位方法", 《中国图象图形学报》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109239900A (en) * | 2018-11-07 | 2019-01-18 | 华东师范大学 | A kind of full-automatic quick focusing method for the big visual field acquisition of microscopic digital image |
CN109239900B (en) * | 2018-11-07 | 2020-10-16 | 华东师范大学 | Full-automatic rapid focusing method for large-field acquisition of microscopic digital image |
CN109359651A (en) * | 2018-11-08 | 2019-02-19 | 东风商用车有限公司 | A kind of License Plate processor and its location processing method |
CN109871849A (en) * | 2019-01-11 | 2019-06-11 | 西安艾润物联网技术服务有限责任公司 | A kind of method and system of vehicle identification |
CN110533027A (en) * | 2019-07-22 | 2019-12-03 | 浙江省北大信息技术高等研究院 | A kind of mobile device-based text detection and recognition methods and system |
CN110533027B (en) * | 2019-07-22 | 2022-09-02 | 杭州未名信科科技有限公司 | Text detection and identification method and system based on mobile equipment |
CN111079744A (en) * | 2019-12-06 | 2020-04-28 | 鲁东大学 | Intelligent vehicle license plate identification method and device suitable for complex illumination environment |
CN111079744B (en) * | 2019-12-06 | 2020-09-01 | 鲁东大学 | Intelligent vehicle license plate identification method and device suitable for complex illumination environment |
CN111652222A (en) * | 2020-07-13 | 2020-09-11 | 深圳市智搜信息技术有限公司 | License plate positioning method and device, computer equipment and storage medium |
CN112488031A (en) * | 2020-12-11 | 2021-03-12 | 华能华家岭风力发电有限公司 | Safety helmet detection method based on color segmentation |
Also Published As
Publication number | Publication date |
---|---|
CN106355180B (en) | 2019-07-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106355180B (en) | A kind of license plate locating method combined based on color with edge feature | |
CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
CN110197589B (en) | Deep learning-based red light violation detection method | |
CN104392205B (en) | A kind of recognition methods and system of improper license plate | |
CN107545239A (en) | A kind of deck detection method matched based on Car license recognition with vehicle characteristics | |
CN103198315B (en) | Based on the Character Segmentation of License Plate of character outline and template matches | |
CN102968646B (en) | A kind of detection method of license plate based on machine learning | |
CN102509098B (en) | Fisheye image vehicle identification method | |
CN106022232A (en) | License plate detection method based on deep learning | |
CN107273896A (en) | A kind of car plate detection recognition methods based on image recognition | |
CN102999753B (en) | License plate locating method | |
CN105373794A (en) | Vehicle license plate recognition method | |
CN102663760B (en) | Location and segmentation method for windshield area of vehicle in images | |
CN104732227A (en) | Rapid license-plate positioning method based on definition and luminance evaluation | |
CN105913041A (en) | Pre-marked signal lights based identification method | |
CN104715239A (en) | Vehicle color identification method based on defogging processing and weight blocking | |
CN102880863B (en) | Method for positioning license number and face of driver on basis of deformable part model | |
CN106650731A (en) | Robust license plate and logo recognition method | |
CN106886987A (en) | A kind of train license plate binary image interfusion method | |
CN103824081A (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
CN104598885A (en) | Method for detecting and locating text sign in street view image | |
CN105741324A (en) | Moving object detection identification and tracking method on moving platform | |
CN104766344B (en) | Vehicle checking method based on movement edge extractor | |
CN106203237A (en) | The recognition methods of container-trailer numbering and device | |
CN111191611A (en) | Deep learning-based traffic sign label identification method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP03 | Change of name, title or address |
Address after: Room 410, Building A3, Wuhan University Science Park, No. 3, Wudayuan 4th Road, East Lake New Technology Development Zone, Wuhan City, Hubei Province 430000 Patentee after: Wuhan Lantao Technology Co.,Ltd. Address before: Room 410, Building A3, Wuhan University Science Park, No. 3, Wudayuan 4th Road, Donghu New Technology Development Zone, Wuhan City, Hubei Province, 430223 Patentee before: WUHAN ANKE WEISHI TECHNOLOGY CO.,LTD. |
|
CP03 | Change of name, title or address |