CN106250892A - Method of Automatic Recognition for Character of Lcecse Plate - Google Patents
Method of Automatic Recognition for Character of Lcecse Plate Download PDFInfo
- Publication number
- CN106250892A CN106250892A CN201610654054.9A CN201610654054A CN106250892A CN 106250892 A CN106250892 A CN 106250892A CN 201610654054 A CN201610654054 A CN 201610654054A CN 106250892 A CN106250892 A CN 106250892A
- Authority
- CN
- China
- Prior art keywords
- image
- license plate
- light intensity
- character
- vehicle
- 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.)
- Withdrawn
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
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/141—Control of illumination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
-
- 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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/247—Aligning, centring, orientation detection or correction of the image by affine transforms, e.g. correction due to perspective effects; Quadrilaterals, e.g. trapezoids
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Input (AREA)
Abstract
The invention provides a kind of Method of Automatic Recognition for Character of Lcecse Plate, comprising: input color vehicle image;Obtain current light intensity;If current light intensity intensity is higher than presetting light intensity, uses, based on color dot, search and the morphologic algorithm of locating license plate of vehicle of mathematics are carried out License Plate to colored vehicle image, obtain license plate image;If current light intensity intensity is less than presetting light intensity, uses algorithm of locating license plate of vehicle based on gray level image that colored vehicle image is carried out License Plate, obtain license plate image;License plate image is carried out horizontal tilt rectification and vertical tilt is corrected;In conjunction with vertical projection and Drop fall algorithm, license plate image is carried out Character segmentation, obtain character block image;By character block image normalization, extract the original thick meshed feature of character block image;Use trained BP neutral net that each character block image is identified, generate the number-plate number identified.By the way, the present invention can accurately identify characters on license plate.
Description
Technical field
The present invention relates to field of license plate recognition, particularly relate to a kind of Method of Automatic Recognition for Character of Lcecse Plate.
Background technology
Along with the expansion of city size, the vehicles number inside city increases sharply, and traditional labor management traffic is
Through not adapting to this change.Therefore, intelligent transportation system by large-scale application in urban traffic control and scheduling in.Wherein,
Car license recognition is the core in intelligent transportation system.
Car license recognition is the important component part in modern intelligent transportation system, applies quite varied.It identifies in mode,
Based on the technology such as computer vision, Digital Image Processing.By a series of process of car plate data can be realized traffic flow
Amount Con trolling index measurement, vehicle location, the supervision of high way super speed automatization, break in traffic rules and regulations candid photograph, toll station and parking
The functions such as field toll administration.Automatically identifying for safeguarding traffic safety and urban public security of license plate, prevents traffic jam, real
Existing traffic automation management has great significance.Accordingly, it would be desirable to a kind of Method of Automatic Recognition for Character of Lcecse Plate.
Summary of the invention
The technical problem that present invention mainly solves is to provide a kind of Method of Automatic Recognition for Character of Lcecse Plate, it is possible to accurately identify car
Board character.
For solving above-mentioned technical problem, the technical scheme that the present invention uses is: provide a kind of characters on license plate automatically to know
Other method, including: input color vehicle image;Obtain current light intensity, and by described current light intensity intensity and default light
Strong intensity contrasts;If described current light intensity intensity is higher than described default light intensity, use based on color dot search
Algorithm of locating license plate of vehicle morphologic with mathematics carries out License Plate to described colored vehicle image, obtains license plate image;If institute
State current light intensity intensity and be less than described default light intensity, use algorithm of locating license plate of vehicle based on gray level image to described colored car
Image carries out License Plate, obtains license plate image;Described license plate image is carried out horizontal tilt rectification and vertical tilt is corrected;
In conjunction with vertical projection and Drop fall algorithm, described license plate image is carried out Character segmentation, obtain character block image;By character block image
Normalization, extracts the original thick meshed feature of character block image;Use trained BP neutral net to each character block figure
As being identified, generate the number-plate number identified.
It is different from the situation of prior art, the invention has the beneficial effects as follows: 1, can be by day, it is also possible at night to car
Board character is identified;2, combining the character segmentation method of vertical projection and Drop fall algorithm, robustness is good, it is possible to preferably solve
The problem such as Characters Stuck, fracture, segmentation effect is good;3, the practical situation of China's car plate is combined, for the particularity of Chinese character, for matter
Measure low Chinese character image and set up multiple training sample, improve robustness and the recognition correct rate of neural network recognization Chinese character;
4, BP neutral net design science is reasonable, and overall flow achieves the balance of real-time and accuracy.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of embodiment of the present invention Method of Automatic Recognition for Character of Lcecse Plate.
Detailed description of the invention
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 carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of protection of the invention.
See Fig. 1, be the schematic flow sheet of embodiment of the present invention Method of Automatic Recognition for Character of Lcecse Plate.The car plate of the present embodiment
Character automatic identifying method comprises the following steps:
S11: input color vehicle image.
S12: obtain current light intensity, and current light intensity intensity is contrasted with default light intensity.
Light intensity can use Fibre Optical Sensor to obtain.
S13: if current light intensity intensity is higher than presetting light intensity, use based on color dot search and mathematical morphology
Algorithm of locating license plate of vehicle colored vehicle image is carried out License Plate, obtain license plate image;If current light intensity intensity is less than pre-
If light intensity, use algorithm of locating license plate of vehicle based on gray level image that colored vehicle image is carried out License Plate, obtain car plate
Image.
License Plate is always the difficult point in Vehicle License Recognition System, and in license plate image, substantial amounts of interference and light irradiate
The factors such as power, car plate are reflective all can affect the accuracy of License Plate.The present invention proposes two kinds of algorithm of locating license plate of vehicle and ties mutually
The algorithm of locating license plate of vehicle closed: in the case of light is reasonable by day, uses based on color dot morphologic to search and mathematics
Algorithm of locating license plate of vehicle;Light is the best by day or uses algorithm of locating license plate of vehicle based on gray level image in the case of night.This
Bright default light intensity is preferably 8000lx, when current light intensity intensity is more than 8000lx, uses based on color dot search sum
Learn morphologic algorithm of locating license plate of vehicle and the colored vehicle image obtained is carried out License Plate;Current light intensity intensity is less than 8000lx
Time, use algorithm of locating license plate of vehicle based on gray level image that the colored vehicle image obtained is carried out License Plate.
S14: license plate image is carried out horizontal tilt rectification and vertical tilt is corrected.
Fix due to the position of video camera and result in the difference of shooting angle, the license plate image that obtains after location or many or
Few situation that there is inclination.Inclination can affect follow-up License Plate Character Segmentation, causes segmentation errors, thus right before Character segmentation
License plate image carries out slant correction, accurately splits ready for character below.The license plate sloped Three models that is divided into: level side
To tilt, superposition inclined vertically, horizontal inclined vertically, the most license plate sloped correction can be from horizontal and vertical two
Direction is carried out, and the most first carries out horizontal tilt correction, determines the up-and-down boundary of characters on license plate, finally carries out vertical skew correction.
The present invention uses the car plate horizontal tilt correcting algorithm minimum based on vertical edge spot projection variance and based on horizontal edge
The car plate vertical skew correction algorithm that some variance is minimum.
S15: combine vertical projection and Drop fall algorithm and license plate image is carried out Character segmentation, obtain character block image.
The local minimum foundation as coarse segmentation of white pixel projection can be obtained, if slightly at characters on license plate gap area
The character block picture traverse that segmentation obtains exceedes width threshold value, then use Drop fall algorithm to carry out secondary Accurate Segmentation, find character
Closure edge profile, carry out cutting along gap, carry out splitting on the basis of the marginal point of the upper left corner and last cell and
Squaring expansion, and the target character image split is screened, remove the non-words such as vertical edge frame, rivet, separator
Symbol image.
S16: by character block image normalization, extract the original thick meshed feature of character block image.
Character block image normalization is that 32 × 16 pixels are big by the distribution according to horizontal and vertical directions character pixels
Little, character zone background is black, and character prospect is white, using each pixel of the character pattern after normalization as one
Grid, extracts the original thick meshed feature of character block image.
S17: use trained BP neutral net to be identified each character block image, generates the car plate identified
Number.
According to the actual features of China's car plate, design BP neutral net: comprise Chinese character network, letter network, alphanumeric
Network and digital network, classify to dissimilar character, and sets up two grade network for the character that similarity is high, determines simultaneously
The parameter such as neuron number, activation primitive, learning rate in input and output neuron number, hidden layer number, hidden layer.
The foregoing is only embodiments of the invention, not thereby limit the scope of the claims of the present invention, every utilize this
Equivalent structure or equivalence flow process that bright description and accompanying drawing content are made convert, or are directly or indirectly used in other relevant skills
Art field, is the most in like manner included in the scope of patent protection of the present invention.
Claims (1)
1. a Method of Automatic Recognition for Character of Lcecse Plate, it is characterised in that including:
Input color vehicle image;
Obtain current light intensity, and described current light intensity intensity is contrasted with default light intensity;
If described current light intensity intensity is higher than described default light intensity, use based on color dot search and mathematical morphology
Algorithm of locating license plate of vehicle described colored vehicle image is carried out License Plate, obtain license plate image;If described current light intensity is strong
Degree, less than described default light intensity, uses algorithm of locating license plate of vehicle based on gray level image that described colored vehicle image is carried out car
Board positions, and obtains license plate image;
Described license plate image is carried out horizontal tilt rectification and vertical tilt is corrected;
In conjunction with vertical projection and Drop fall algorithm, described license plate image is carried out Character segmentation, obtain character block image;
By character block image normalization, extract the original thick meshed feature of character block image;
Use trained BP neutral net that each character block image is identified, generate the number-plate number identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610654054.9A CN106250892A (en) | 2016-08-11 | 2016-08-11 | Method of Automatic Recognition for Character of Lcecse Plate |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610654054.9A CN106250892A (en) | 2016-08-11 | 2016-08-11 | Method of Automatic Recognition for Character of Lcecse Plate |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106250892A true CN106250892A (en) | 2016-12-21 |
Family
ID=58078249
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610654054.9A Withdrawn CN106250892A (en) | 2016-08-11 | 2016-08-11 | Method of Automatic Recognition for Character of Lcecse Plate |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106250892A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109033921A (en) * | 2017-06-08 | 2018-12-18 | 北京君正集成电路股份有限公司 | A kind of training method and device of identification model |
CN109086772A (en) * | 2018-08-16 | 2018-12-25 | 成都市映潮科技股份有限公司 | A kind of recognition methods and system distorting adhesion character picture validation code |
US10296794B2 (en) | 2016-12-20 | 2019-05-21 | Jayant Rtti | On-demand artificial intelligence and roadway stewardship system |
WO2019175686A1 (en) | 2018-03-12 | 2019-09-19 | Ratti Jayant | On-demand artificial intelligence and roadway stewardship system |
-
2016
- 2016-08-11 CN CN201610654054.9A patent/CN106250892A/en not_active Withdrawn
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10296794B2 (en) | 2016-12-20 | 2019-05-21 | Jayant Rtti | On-demand artificial intelligence and roadway stewardship system |
CN109033921A (en) * | 2017-06-08 | 2018-12-18 | 北京君正集成电路股份有限公司 | A kind of training method and device of identification model |
WO2019175686A1 (en) | 2018-03-12 | 2019-09-19 | Ratti Jayant | On-demand artificial intelligence and roadway stewardship system |
CN109086772A (en) * | 2018-08-16 | 2018-12-25 | 成都市映潮科技股份有限公司 | A kind of recognition methods and system distorting adhesion character picture validation code |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110197589B (en) | Deep learning-based red light violation detection method | |
CN103824081B (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
CN104156731B (en) | Vehicle License Plate Recognition System and method based on artificial neural network | |
CN102637257B (en) | Video-based detection and recognition system and method of vehicles | |
CN104021378B (en) | Traffic lights real-time identification method based on space time correlation Yu priori | |
CN101877050B (en) | Automatic extracting method for characters on license plate | |
CN103324930B (en) | A kind of registration number character dividing method based on grey level histogram binaryzation | |
CN101673338B (en) | Fuzzy license plate identification method based on multi-angle projection | |
CN109255350B (en) | New energy license plate detection method based on video monitoring | |
CN106250892A (en) | Method of Automatic Recognition for Character of Lcecse Plate | |
CN102364496A (en) | Method and system for identifying automobile license plates automatically based on image analysis | |
CN202563526U (en) | Transportation vehicle detection and recognition system based on video | |
CN104050450A (en) | Vehicle license plate recognition method based on video | |
CN105718870A (en) | Road marking line extracting method based on forward camera head in automatic driving | |
CN103116985A (en) | Detection method and device of parking against rules | |
CN103679168A (en) | Detection method and detection device for character region | |
CN104298976A (en) | License plate detection method based on convolutional neural network | |
CN103824452A (en) | Lightweight peccancy parking detection device based on full view vision | |
CN102289686A (en) | Method for identifying classes of moving targets based on transfer learning | |
CN103268489A (en) | Motor vehicle plate identification method based on sliding window searching | |
CN106295643A (en) | Automobile license plate automatic recognition method | |
CN102938057B (en) | A kind of method for eliminating vehicle shadow and device | |
CN104851288B (en) | Traffic light positioning method | |
CN106529461A (en) | Vehicle model identifying algorithm based on integral characteristic channel and SVM training device | |
CN102436578B (en) | Formation method for dog face characteristic detector as well as dog face detection method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20161221 |
|
WW01 | Invention patent application withdrawn after publication |