CN103729619B - The automatic identifying method of car plate frame in preventing road monitoring system - Google Patents
The automatic identifying method of car plate frame in preventing road monitoring system Download PDFInfo
- Publication number
- CN103729619B CN103729619B CN201310671424.6A CN201310671424A CN103729619B CN 103729619 B CN103729619 B CN 103729619B CN 201310671424 A CN201310671424 A CN 201310671424A CN 103729619 B CN103729619 B CN 103729619B
- Authority
- CN
- China
- Prior art keywords
- line segment
- isosceles triangle
- drift angle
- cog region
- line
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
- Character Discrimination (AREA)
Abstract
Car license recognition applied technical field during the automatic identifying method of car plate frame belongs to urban road monitor system in preventing road monitoring system, in this preventing road monitoring system, the automatic identifying method of car plate frame is by defining the scanning recognition district of an isosceles triangle, and make this cog region move along given identification path respectively, realize the screening to the topography including the doubtful frame of car plate in entire image, and by acquired multiple straight line set being identified and judgement process, finally realize filtering out the profile of car plate frame from view picture photo.This automatic identifying method also has that operand is little, speed fast, the most reliable and be easy to advantages such as popularizing.
Description
Technical field
The invention belongs to Car license recognition applied technical field in urban road monitor system, be specifically related to a kind of road
The automatic identifying method of car plate frame in the monitoring system of road.
Background technology
In the last few years, urban highway traffic quickly grows, for more preferable management road traffic, quickly
Accurately detecting vehicles peccancy, Vehicular intelligent supervisory systems is set up in each city successively.In the case of unmanned participation,
Monitoring road can be carried out by existing urban road monitor system by vehicle speed measuring device and shooting camera system
Automatic camera, obtains the photo including target vehicle and car plate thereof.
But automatically identify that car plate violating the regulations is then a more complicated image recognition processing process, existing car
Board recognition methods many employings edge extracting image processing algorithm, is identified view picture photograph image processing, this
Kind of method not only operand is huge, processes the time long, is also easily subject in photo other beyond car plate simultaneously
The interference of image information, causes the error of recognition result even to directly result in mistake.
Therefore, research and develop a kind of taking the lead in from the view picture photograph image including car plate and identify and filter out car plate wheel
Wide boundary line, is accurately positioned car plate present position in photo, in order to get rid of the image district of non-car plate in photo
Territory, the most again by subsequent treatment, completes the quickly identification to the number-plate number image within car plate border
Method, becomes problem demanding prompt solution.
Summary of the invention
In order to solve existing licence plate recognition method many employings edge extracting image processing algorithm, to view picture photo
Image is identified processing, and not only operand is huge, processes the time long, is also easily subject to car in photo simultaneously
The interference of other image information beyond board, causes the error of recognition result even to directly result in mistake, therefore
Need to research and develop a kind of can taking the lead in from the view picture photograph image including car plate identify and filter out license plate outline limit
The technical problem of the licence plate recognition method in boundary line, the present invention provides car plate frame in a kind of preventing road monitoring system
Automatic identifying method.
It is as follows that the present invention solves the technical scheme that technical problem taked:
In a kind of preventing road monitoring system, the automatic identifying method of car plate frame comprises the steps:
The first step, the original image that the resolution taken by preventing road monitoring system is 768 × 576 pixels is entered
Row edge extracting, it is thus achieved that bianry image E, in bianry image E gray value be 1 point represent marginal point, ash
Angle value be 0 point represent non-edge point, with the upper left corner of bianry image E as initial point, set up x-axis level to
Right, y-axis pixel planes rectangular coordinate system straight down;
Second step, acquisition comprise the straight line set U of doubtful horizontal sides wire section1, it specifically includes following sub-step
Rapid:
Step 2.1: two scanning patterns on the cog region of definition isosceles triangle and horizontal direction:
Defining the cog region of horizontal isosceles triangle, the drift angle of this isosceles triangle is 30 degree and towards x-axis just
Direction, the pixel (729,0) that the straight line at the place, base of isosceles triangle is perpendicular in x-axis;Isosceles three
The center line of dihedral is parallel with x-axis;58 oblique line sections, described 58 oblique lines are also comprised in this isosceles triangle
Duan Jun is with the drift angle of isosceles triangle as starting point, and base is terminal, and is positioned at coordinate system range described in step one
Within;In this isosceles triangle in 61 line segments altogether including two stringcourses and a center line, arbitrarily
The adjacent angle between two is 0.5 degree;
Choose with pixel a1(40,40) are starting point with pixel a728(729,40) are the line of terminal
Section a, and choose with pixel (40,537) and be starting point and be the line of terminal with pixel (729,537)
B is as the scanning pattern in two horizontal directions for section;The length of line segment a and line segment b is 690 pixels
Point, and both is parallel with x-axis;
Step 2.2: the cog region drift angle of isosceles triangle is along the scanning recognition process of horizontal line section a, and it is concrete
Including following sub-step:
Step 2.2.1: upper a certain to fixation by the drift angle of the cog region of horizontal isosceles triangle and line segment a
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of horizontal isosceles triangle described in step 2.1 and a certain given pixel on line segment a
Point overlaps, and the sequence number of the drift angle position of the cog region of the most horizontal isosceles triangle is defined as an, its
Middle n=x-39, in formula, x is the abscissa value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of horizontal isosceles triangle is anTime, define this isosceles three
Serial number Fa of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61);By oblique line section Fan-01
On the coordinate figure of each pixel be all stored in line segment array data [an-01In], by same method, by it
Its each oblique line section an-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fan-kCorresponding
Line segment array data [an-k] in (k sequentially takes 01,02,03 ... 61);
Step 2.2.2: when completing the cog region drift angle current given coordinate points on horizontal line section a, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 2.2.2.1: the drift angle of the cog region of horizontal isosceles triangle is in starting point a of line segment a1(40,40)
Time, select oblique line section Fa in current cog region1-01;
Step 2.2.2.2: judge acquired by step 2.2.1 and oblique line section Fan-kCorresponding line segment array
data[an-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
2.2.2.3;If no, then entering step 2.2.2.4;
Step 2.2.2.3: linea oblique anterior section Fa will be worked asn-kLinear equation La corresponding under pixel coordinate systemn-kIt is stored in
Array U1, complete when linea oblique anterior section Fan-kThe identification screening of upper doubtful frame, and enter step 2.2.2.4;
Step 2.2.2.4: judge current line segment array data [an-k] (k sequentially takes 01,02,03 ... 61)
In, whether k is 61, if k ≠ 61, then enters step 2.2.2.5;If k=61, then enter step 2.2.3;
Step 2.2.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 2.2.2.2 institute
State oblique line section Fan-kAnd line segment array data [a corresponding thereton-k] sequence number, re-execute step 2.2.2.2
To step 2.2.2.4;
Step 2.2.3: make the drift angle of cog region of horizontal isosceles triangle from starting point a of line segment a1(40,40)
A pixel unit is sequentially moved horizontally, after each drift angle moves a new position, all to x-axis positive direction
Update the drift angle a of the cog region of horizontal isosceles triangle described in step 2.2.2.1n, and repeated execution of steps 2.2.2,
Until the drift angle of the cog region of horizontal isosceles triangle moves the terminal a to line segment a729Time (729,40),
By anIt is updated to a690, subsequently into step 2.3;
Step 2.3: by the cog region drift angle of isosceles triangle by the terminal a of the line segment a described in step 2.1729
(729,40) change the starting point (40,537) to line segment b, and according to complete with the method as step 2.2
Becoming the cog region drift angle scanning recognition process along horizontal line section b of isosceles triangle, it specifically includes following son
Step:
Step 2.3.1: upper a certain to fixation by the drift angle of the cog region of horizontal isosceles triangle and line segment b
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of horizontal isosceles triangle described in step 2.1 and a certain given pixel on line segment b
Point overlaps, and the sequence number of the drift angle position of the cog region of the most horizontal isosceles triangle is defined as bn, its
Middle n=x-39, in formula, x is the abscissa value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of horizontal isosceles triangle is bnTime, define this isosceles three
Serial number Fb of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61);By oblique line section
Fbn-01On the coordinate figure of each pixel be all stored in line segment array data [bn-01In], by same method,
By other each oblique line section bn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fbn-k
Corresponding line segment array data [bn-k] in (k sequentially takes 01,02,03 ... 61);
Step 2.3.2: when completing the cog region drift angle current given coordinate points on horizontal line section b, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 2.3.2.1: the drift angle of the cog region of horizontal isosceles triangle is in starting point b of line segment b1(40,40)
Time, select oblique line section Fb in current cog region1-01;
Step 2.3.2.2: judge acquired by step 2.3.1 and oblique line section Fbn-kCorresponding line segment array
data[bn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
2.3.2.3;If no, then entering step 2.3.2.4;
Step 2.3.2.3: linea oblique anterior section Fb will be worked asn-kLinear equation Lb corresponding under pixel coordinate systemn-kIt is stored in
Array U1, complete when linea oblique anterior section Fbn-kThe identification screening of upper doubtful frame, and enter step 2.3.2.4;
Step 2.3.2.4: judge present segment array data [bn-k] in (k sequentially takes 01,02,03 ... 61),
Whether k is 61, if k ≠ 61, then enters step 2.3.2.5;If k=61, then enter step 2.3.3;
Step 2.3.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 2.3.2.2 institute
State oblique line section Fbn-kAnd line segment array data [b corresponding thereton-k] sequence number, re-execute step 2.3.2.2
To step 2.3.2.4;
Step 2.3.3: make the drift angle of cog region of horizontal isosceles triangle from starting point b of line segment b1(40,537)
A pixel unit is sequentially moved horizontally, after each drift angle moves a new position, all to x-axis positive direction
Update the drift angle b of the cog region of horizontal isosceles triangle described in step 2.3.2.1n, and repeated execution of steps 2.3.2,
Until the drift angle of the cog region of horizontal isosceles triangle moves the terminal b to line segment b729Time (729,40),
By bnIt is updated to b690, complete the cog region drift angle scanning recognition process along horizontal line section b of isosceles triangle,
Subsequently into step 3;
Step 3, acquisition comprise the straight line set U of doubtful vertical edge wire section2, it specifically includes following sub-step
Rapid: step 3.1: two scanning patterns on the cog region of definition isosceles triangle and vertical direction:
Defining the cog region of vertical isosceles triangle, the drift angle of this isosceles triangle is 30 degree and towards y-axis just
Direction, the pixel (0,537) that the straight line at the place, base of isosceles triangle is perpendicular in y-axis;Isosceles three
The center line of dihedral is parallel with y-axis;58 oblique line sections, described 58 oblique lines are also comprised in this isosceles triangle
Duan Jun is with the drift angle of isosceles triangle as starting point, and base is terminal, and is positioned at coordinate system range described in step one
Within;In this isosceles triangle in 61 line segments altogether including two stringcourses and a center line, arbitrarily
The adjacent angle between two is 0.5 degree;
Choosing with pixel (40,40) is starting point with the line segment c that pixel (40,537) is terminal,
And choose with pixel (729,40) and be starting point and make with line segment d that pixel (729,537) is terminal
It it is the scanning pattern on two vertical directions;The length of line segment c and line segment d is 498 pixels, and
Both is parallel with y-axis;
Step 3.2: the cog region drift angle of isosceles triangle is along the scanning recognition process of vertical line segment c, and it is concrete
Including following sub-step:
Step 3.2.1: upper a certain to fixation by the drift angle of the cog region of vertical isosceles triangle and line segment c
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of vertical isosceles triangle described in step 3.1 and a certain given pixel on line segment c
Point overlaps, and the sequence number of the now drift angle position of the cog region of vertical isosceles triangle is defined as cn, its
Middle n=y-39, in formula, y is the ordinate value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of vertical isosceles triangle is cnTime, define this isosceles three
Serial number Fc of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61);By oblique line section Fcn-01
On the coordinate figure of each pixel be all stored in line segment array data [cn-01In], by same method, by it
Its each oblique line section cn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fcn-kCorresponding
Line segment array data [cn-k] in (k sequentially takes 01,02,03 ... 61);
Step 3.2.2: when completing the cog region drift angle current given coordinate points on vertical line segment c, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 3.2.2.1: vertically the drift angle of the cog region of isosceles triangle is in starting point c of line segment c1(40,40)
Time, select oblique line section Fc in current cog region1-01;
Step 3.2.2.2: judge acquired by step 3.2.1 and oblique line section Fcn-kCorresponding line segment array
data[cn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
3.2.2.3;If no, then entering step 3.2.2.4;
Step 3.2.2.3: linea oblique anterior section Fc will be worked asn-kLinear equation Lc corresponding under pixel coordinate systemn-kIt is stored in
Array U2, complete when linea oblique anterior section Fcn-kThe identification screening of upper doubtful frame, and enter step 3.2.2.4;
Step 3.2.2.4: judge current line segment array data [cn-k] (k sequentially takes 01,02,03 ... 61)
In, whether k is 61, if k ≠ 61, then enters step 3.2.2.5;If k=61, then enter step 3.2.3;
Step 3.2.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 3.2.2.2 institute
State oblique line section Fcn-kAnd line segment array data [c corresponding thereton-k] sequence number, re-execute step 3.2.2.2
To step 3.2.2.4;
Step 3.2.3: make the drift angle of cog region of vertical isosceles triangle from starting point c of line segment c1(40,40)
A pixel unit is the most vertically moved, after each drift angle moves a new position, all to y-axis positive direction
Update the drift angle c of the vertically cog region of isosceles triangle described in step 3.2.2.1n, and repeated execution of steps 3.2.2,
Until vertically the drift angle of the cog region of isosceles triangle moves the terminal c to line segment c498Time (40,537),
By cnIt is updated to c498, subsequently into step 3.3;
Step 3.3: by the cog region drift angle of isosceles triangle by the terminal c of the line segment c described in step 3.1498
(40,537) change the starting point (729,40) to line segment d, and according to complete with the method as step 3.2
Becoming the cog region drift angle scanning recognition process along vertical line segment d of isosceles triangle, it specifically includes following son
Step:
Step 3.3.1: upper a certain to fixation by the drift angle of the cog region of vertical isosceles triangle and line segment d
Vegetarian refreshments overlaps, and corresponding with oblique line section by the ordinate value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of vertical isosceles triangle described in step 3.1 and a certain given pixel on line segment d
Point overlaps, and the sequence number of the now drift angle position of the cog region of vertical isosceles triangle is defined as dn, its
Middle n=y-39, in formula, y is the ordinate value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of vertical isosceles triangle is dnTime, define this isosceles three
Serial number Fd of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61);By oblique line section
Fdn-01On the coordinate figure of each pixel be all stored in line segment array data [dn-01In], by same method,
By other each oblique line section dn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fdn-k
Corresponding line segment array data [dn-k] in (k sequentially takes 01,02,03 ... 61);
Step 3.3.2: when completing the cog region drift angle current given coordinate points on vertical line segment d, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 3.3.2.1: vertically the drift angle of the cog region of isosceles triangle is in starting point d of line segment d1(729,40)
Time, select oblique line section Fd in current cog region1-01;
Step 3.3.2.2: judge acquired by step 3.3.1 and oblique line section Fdn-kCorresponding line segment array
data[dn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
3.3.2.3;If no, then entering step 3.3.2.4;
Step 3.3.2.3: linea oblique anterior section Fd will be worked asn-kLinear equation Ld corresponding under pixel coordinate systemn-kIt is stored in
Array U2, complete when linea oblique anterior section Fdn-kThe identification screening of upper doubtful frame, and enter step 3.3.2.4;
Step 3.3.2.4: judge present segment array data [dn-k] in (k sequentially takes 01,02,03 ... 61),
Whether k is 61, if k ≠ 61, then enters step 3.3.2.5;If k=61, then enter step 3.3.3;
Step 3.3.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 3.3.2.2 institute
State oblique line section Fdn-kAnd line segment array data [d corresponding thereton-k] sequence number, re-execute step 3.3.2.2
To step 3.3.2.4;
Step 3.3.3: make the drift angle of cog region of vertical isosceles triangle from starting point d of line segment d1(729,40)
A pixel unit is the most vertically moved, after each drift angle moves a new position, all to y-axis positive direction
Update the drift angle d of the vertically cog region of isosceles triangle described in step 3.3.2.1n, and repeated execution of steps 3.3.2,
Until vertically the drift angle of the cog region of isosceles triangle moves the terminal d to line segment d498Time (729,537),
By dnIt is updated to d498, complete the cog region drift angle scanning recognition process along vertical line segment d of isosceles triangle,
Subsequently into step 4;
Step 4: straight line set U described in read step two respectively1With straight line set U described in step 32, and sentence
Disconnected logical formula U1&U2Result, if its result is true, then enter step 5, if its result is false, the most directly
Connect the identification process terminating this method;
Step 5: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, from U1Middle screening retains energy and U2In the most vertical straight line of two straight lines, by qualified
Straight line is stored in set U3, subsequently into step 6;
Step 6: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, from U2Middle screening retains energy and U3In the most vertical straight line of two straight lines, by qualified
Straight line is stored in set U4, subsequently into step 7;
Step 7: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, it is judged that set U4In whether have two straight lines can simultaneously and U3In two straight lines vertical, then
Enter step 8;
Step 8: obtained the linear equation of rectangle four edges line by step 7 respectively, the four edges line of this rectangle
It is the car plate doubtful boundary line in photo.
The slope result of product according to two intersecting straight lines described in step 5 or step 6 or step 7, sentences
Disconnected whether meet condition perpendicular to one another both it criterion be:
Any two linear equations are denoted as respectively: straight line 1:y1=k1x+b1;Straight line 2:y2=k2x+b2;Then
Judgement formula: E=k can be obtained1×k2;
If the result value of E is in interval [-1.05 ,-0.95], then it is assumed that two straight lines are perpendicular to one another.
The invention has the beneficial effects as follows: in this preventing road monitoring system, the automatic identifying method of car plate frame is by fixed
The scanning recognition district of one isosceles triangle of justice, and make this cog region move along given identification path respectively,
Realize the screening to the topography including the doubtful frame of car plate in entire image, and by acquired
Multiple straight line set are identified and judgement processes, and finally realize filtering out car plate frame from view picture photo
Profile.This automatic identifying method also has that operand is little, speed fast, the most reliable and be easy to promote general
And etc. advantage.
Accompanying drawing explanation
Fig. 1 is the block schematic illustration of the cog region of isosceles triangle of the present invention;
Fig. 2 is cog region and the schematic diagram of scanning pattern thereof of isosceles triangle of the present invention;
Fig. 3 is the general flow chart of the automatic identifying method of car plate frame in preventing road monitoring system of the present invention;
Fig. 4 is the sub-process figure of step 2 of the present invention;
Fig. 5 is the sub-process figure of step 3 of the present invention;
Fig. 6 is the sub-process figure of step 2.2 of the present invention;
Fig. 7 is the sub-process figure of step 2.2.2 of the present invention;
Fig. 8 is the sub-process figure of step 2.3 of the present invention;
Fig. 9 is the sub-process figure of step 2.3.2 of the present invention;
Figure 10 is the sub-process figure of step 3.2 of the present invention;
Figure 11 is the sub-process figure of step 3.2.2 of the present invention;
Figure 12 is the sub-process figure of step 3.3 of the present invention;
Figure 13 is the sub-process figure of step 3.3.2 of the present invention;
Detailed description of the invention
Below in conjunction with the accompanying drawings the present invention is described in further details.
As shown in figure 1 to figure 13, the automatic identifying method bag of car plate frame in the preventing road monitoring system of the present invention
Include following steps:
The first step, the original image that the resolution taken by preventing road monitoring system is 768 × 576 pixels is entered
Row edge extracting, it is thus achieved that bianry image E, in bianry image E gray value be 1 point represent marginal point, ash
Angle value be 0 point represent non-edge point, with the upper left corner of bianry image E as initial point, set up x-axis level to
Right, y-axis pixel planes rectangular coordinate system straight down.
Second step, acquisition comprise the straight line set U of doubtful horizontal sides wire section1, it specifically includes following sub-step
Rapid:
Step 2.1: two scanning patterns on the cog region of definition isosceles triangle and horizontal direction:
Defining the cog region of horizontal isosceles triangle, the drift angle of this isosceles triangle is 30 degree and towards x-axis just
Direction, the pixel (729,0) that the straight line at the place, base of isosceles triangle is perpendicular in x-axis.Isosceles three
The center line of dihedral is parallel with x-axis.58 oblique line sections, described 58 oblique lines are also comprised in this isosceles triangle
Duan Jun is with the drift angle of isosceles triangle as starting point, and base is terminal, and is positioned at coordinate system range described in step one
Within.In this isosceles triangle in 61 line segments altogether including two stringcourses and a center line, arbitrarily
The adjacent angle between two is 0.5 degree.
Choose with pixel a1(40,40) are starting point with pixel a728(729,40) are the line of terminal
Section a, and choose with pixel (40,537) and be starting point and be the line of terminal with pixel (729,537)
B is as the scanning pattern in two horizontal directions for section.The length of line segment a and line segment b is 690 pixels
Point, and both is parallel with x-axis.
Step 2.2: the cog region drift angle of isosceles triangle is along the scanning recognition process of horizontal line section a, and it is concrete
Including following sub-step:
Step 2.2.1: upper a certain to fixation by the drift angle of the cog region of horizontal isosceles triangle and line segment a
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation.Its specific practice is:
Make the drift angle of the cog region of horizontal isosceles triangle described in step 2.1 and a certain given pixel on line segment a
Point overlaps, and the sequence number of the drift angle position of the cog region of the most horizontal isosceles triangle is defined as an, its
Middle n=x-39, in formula, x is the abscissa value that current corner place gives pixel.That is, horizontal isoceles triangle
The drift angle of the cog region of shape is in starting point a of line segment a1Time (40,40), the sequence number of drift angle position is also
a1.When the drift angle of cog region of horizontal isosceles triangle is at the terminal a of line segment a690Time (729,40), drift angle
The sequence number of position is a690。
When the sequence number of the drift angle position of the cog region of horizontal isosceles triangle is anTime, define this isosceles three
Serial number Fa of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61).Then Article 1 waist
Serial number Fa of linen-01, serial number Fa of center linen-31, serial number Fa of another stringcoursen-61.By oblique line section
Fan-01On the coordinate figure of each pixel be all stored in line segment array data [an-01In], by same method,
By other each oblique line section an-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fan-k
Corresponding line segment array data [an-k] in (k sequentially takes 01,02,03 ... 61).
Such as, the drift angle of the cog region of horizontal isosceles triangle is in starting point a of line segment a1Time (40,40),
Serial number a of each bar oblique line section in isosceles triangle1-k(k sequentially takes 01,02,03 ... 61).Then
Article one, serial number Fa of stringcourse1-01, serial number Fa of center line1-31, serial number Fa of another stringcourse1-61.Will
Oblique line section Fa1-01On the coordinate figure of each pixel be all stored in line segment array data [a1-01In], with same
Method, by other each oblique line section a1-kOn the coordinate figure of each pixel be all stored in and this line segment sequence
Number a1-kCorresponding line segment array data [a1-k] in (k sequentially takes 01,02,03 ... 61).
Step 2.2.2: when completing the cog region drift angle current given coordinate points on horizontal line section a, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment.It specifically wraps
Include following sub-step:
Step 2.2.2.1: the drift angle of the cog region of horizontal isosceles triangle is in starting point a of line segment a1(40,40)
Time, select oblique line section Fa in current cog region1-01。
Step 2.2.2.2: judge acquired by step 2.2.1 and oblique line section Fan-kCorresponding line segment array
data[an-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
2.2.2.3.If no, then entering step 2.2.2.4.
Step 2.2.2.3: linea oblique anterior section Fa will be worked asn-kLinear equation La corresponding under pixel coordinate systemn-kIt is stored in
Array U1, complete when linea oblique anterior section Fan-kThe identification screening of upper doubtful frame, and enter step 2.2.2.4.
Step 2.2.2.4: judge current line segment array data [an-k] (k sequentially takes 01,02,03 ... 61)
In, whether k is 61, if k ≠ 61, then enters step 2.2.2.5.If k=61, then illustrate to have completed working as n
During for some explicit value, to oblique line section Fan-01To Fan-61The identification screening of upper doubtful frame, and own
With screened go out oblique line section Fa including car plate doubtful frame line segmentn-Linear equation La corresponding for kn-k(k
Sequentially take 01,02,03 ... 61) the most it is stored into array U1In.Enter step 2.2.3.
Step 2.2.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 2.2.2.2 institute
State oblique line section Fan-kAnd line segment array data [a corresponding thereton-k] sequence number, re-execute step 2.2.2.2
To step 2.2.2.4.
Step 2.2.3: make the drift angle of cog region of horizontal isosceles triangle from starting point a of line segment a1(40,40)
A pixel unit is sequentially moved horizontally, after each drift angle moves a new position, all to x-axis positive direction
Update the drift angle a of the cog region of horizontal isosceles triangle described in step 2.2.2.1n, and repeated execution of steps 2.2.2,
Until the drift angle of the cog region of horizontal isosceles triangle moves the terminal a to line segment a729Time (729,40),
By anIt is updated to a690, subsequently into step 2.3.
Step 2.3: by the cog region drift angle of isosceles triangle by the terminal a of the line segment a described in step 2.1729
(729,40) change the starting point (40,537) to line segment b, and according to complete with the method as step 2.2
Becoming the cog region drift angle scanning recognition process along horizontal line section b of isosceles triangle, it specifically includes following son
Step:
Step 2.3.1: upper a certain to fixation by the drift angle of the cog region of horizontal isosceles triangle and line segment b
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation.Its specific practice is:
Make the drift angle of the cog region of horizontal isosceles triangle described in step 2.1 and a certain given pixel on line segment b
Point overlaps, and the sequence number of the drift angle position of the cog region of the most horizontal isosceles triangle is defined as bn, its
Middle n=x-39, in formula, x is the abscissa value that current corner place gives pixel.That is, horizontal isoceles triangle
The drift angle of the cog region of shape is in starting point b of line segment b1Time (40,537), the sequence number of drift angle position is also
b1.When the drift angle of cog region of horizontal isosceles triangle is at the terminal b of line segment b690Time (729,537), top
The sequence number of position, angle is b690。
When the sequence number of the drift angle position of the cog region of horizontal isosceles triangle is bnTime, define this isosceles three
Serial number Fb of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61).Then Article 1 waist
Serial number Fb of linen-01, serial number Fb of center linen-31, serial number Fb of another stringcoursen-61.By oblique line section
Fbn-01On the coordinate figure of each pixel be all stored in line segment array data [bn-01In], by same method,
By other each oblique line section bn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fbn-k
Corresponding line segment array data [bn-k] in (k sequentially takes 01,02,03 ... 61).
Such as, the drift angle of the cog region of horizontal isosceles triangle is in starting point b of line segment b1Time (40,40),
Serial number b of each bar oblique line section in isosceles triangle1-k(k sequentially takes 01,02,03 ... 61).Then
Article one, serial number Fb of stringcourse1-01, serial number Fb of center line1-31, serial number Fb of another stringcourse1-61.Will
Oblique line section Fb1-01On the coordinate figure of each pixel be all stored in line segment array data [b1-01In], with same
Method, by other each oblique line section b1-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number
b1-kCorresponding line segment array data [b1-k] in (k sequentially takes 01,02,03 ... 61).
Step 2.3.2: when completing the cog region drift angle current given coordinate points on horizontal line section b, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment.It specifically wraps
Include following sub-step:
Step 2.3.2.1: the drift angle of the cog region of horizontal isosceles triangle is in starting point b of line segment b1(40,40)
Time, select oblique line section Fb in current cog region1-01。
Step 2.3.2.2: judge acquired by step 2.3.1 and oblique line section Fbn-kCorresponding line segment array
data[bn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
2.3.2.3.If no, then entering step 2.3.2.4.
Step 2.3.2.3: linea oblique anterior section Fb will be worked asn-kLinear equation Lb corresponding under pixel coordinate systemn-kIt is stored in
Array U1, complete when linea oblique anterior section Fbn-kThe identification screening of upper doubtful frame, and enter step 2.3.2.4.
Step 2.3.2.4: judge present segment array data [bn-k] in (k sequentially takes 01,02,03 ... 61),
Whether k is 61, if k ≠ 61, then enters step 2.3.2.5.If k=61, then illustrate to have completed to as n be
During some explicit value, to oblique line section Fbn-01To Fbn-61The identification screening of upper doubtful frame, and all with
Screened go out oblique line section Fb including car plate doubtful frame line segmentn-Linear equation Lb corresponding for kn-k(k is suitable
Secondary take 01,02,03 ... 61) the most it is stored into array U1In.Enter step 2.3.3.
Step 2.3.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 2.3.2.2 institute
State oblique line section Fbn-kAnd line segment array data [b corresponding thereton-k] sequence number, re-execute step 2.3.2.2
To step 2.3.2.4.
Step 2.3.3: make the drift angle of cog region of horizontal isosceles triangle from starting point b of line segment b1(40,537)
A pixel unit is sequentially moved horizontally, after each drift angle moves a new position, all to x-axis positive direction
Update the drift angle b of the cog region of horizontal isosceles triangle described in step 2.3.2.1n, and repeated execution of steps 2.3.2,
Until the drift angle of the cog region of horizontal isosceles triangle moves the terminal b to line segment b729Time (729,40),
By bnIt is updated to b690, complete the cog region drift angle scanning recognition process along horizontal line section b of isosceles triangle,
Subsequently into step 3.
Step 3, acquisition comprise the straight line set U of doubtful vertical edge wire section2, it specifically includes following sub-step
Rapid: step 3.1: two scanning patterns on the cog region of definition isosceles triangle and vertical direction:
Defining the cog region of vertical isosceles triangle, the drift angle of this isosceles triangle is 30 degree and towards y-axis just
Direction, the pixel (0,537) that the straight line at the place, base of isosceles triangle is perpendicular in y-axis.Isosceles three
The center line of dihedral is parallel with y-axis.58 oblique line sections, described 58 oblique lines are also comprised in this isosceles triangle
Duan Jun is with the drift angle of isosceles triangle as starting point, and base is terminal, and is positioned at coordinate system range described in step one
Within.In this isosceles triangle in 61 line segments altogether including two stringcourses and a center line, arbitrarily
The adjacent angle between two is 0.5 degree.
Choosing with pixel (40,40) is starting point with the line segment c that pixel (40,537) is terminal,
And choose with pixel (729,40) and be starting point and make with line segment d that pixel (729,537) is terminal
It it is the scanning pattern on two vertical directions.The length of line segment c and line segment d is 498 pixels, and
Both is parallel with y-axis.
Step 3.2: the cog region drift angle of isosceles triangle is along the scanning recognition process of vertical line segment c, and it is concrete
Including following sub-step:
Step 3.2.1: upper a certain to fixation by the drift angle of the cog region of vertical isosceles triangle and line segment c
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation.Its specific practice is:
Make the drift angle of the cog region of vertical isosceles triangle described in step 3.1 and a certain given pixel on line segment c
Point overlaps, and the sequence number of the now drift angle position of the cog region of vertical isosceles triangle is defined as cn, its
Middle n=y-39, in formula, y is the ordinate value that current corner place gives pixel.That is, vertical isoceles triangle
The drift angle of the cog region of shape is in starting point c of line segment c1Time (40,40), the sequence number of drift angle position is also
c1.When the drift angle of cog region of vertical isosceles triangle is at the terminal c of line segment c498Time (40,537), drift angle
The sequence number of position is c498。
When the sequence number of the drift angle position of the cog region of vertical isosceles triangle is cnTime, define this isosceles three
Serial number Fc of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61).Then Article 1 waist
Serial number Fc of linen-01, serial number Fc of center linen-31, serial number Fc of another stringcoursen-61.By oblique line section
Fcn-01On the coordinate figure of each pixel be all stored in line segment array data [cn-01In], by same method,
By other each oblique line section cn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fcn-k
Corresponding line segment array data [cn-k] in (k sequentially takes 01,02,03 ... 61).
Such as, the drift angle of the cog region of vertical isosceles triangle is in starting point c of line segment c1Time (40,40),
Serial number c of each bar oblique line section in isosceles triangle1-k(k sequentially takes 01,02,03 ... 61).Then
Article one, serial number Fc of stringcourse1-01, serial number Fc of center line1-31, serial number Fc of another stringcourse1-61.Will
Oblique line section Fc1-01On the coordinate figure of each pixel be all stored in line segment array data [c1-01In], with same
Method, by other each oblique line section c1-kOn the coordinate figure of each pixel be all stored in and this line segment sequence
Number c1-kCorresponding line segment array data [c1-k] in (k sequentially takes 01,02,03 ... 61).
Step 3.2.2: when completing the cog region drift angle current given coordinate points on vertical line segment c, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment.It specifically wraps
Include following sub-step:
Step 3.2.2.1: vertically the drift angle of the cog region of isosceles triangle is in starting point c of line segment c1(40,40)
Time, select oblique line section Fc in current cog region1-01。
Step 3.2.2.2: judge acquired by step 3.2.1 and oblique line section Fcn-kCorresponding line segment array
data[cn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
3.2.2.3.If no, then entering step 3.2.2.4.
Step 3.2.2.3: linea oblique anterior section Fc will be worked asn-kLinear equation Lc corresponding under pixel coordinate systemn-kIt is stored in
Array U2, complete when linea oblique anterior section Fcn-kThe identification screening of upper doubtful frame, and enter step 3.2.2.4.
Step 3.2.2.4: judge current line segment array data [cn-k] (k sequentially takes 01,02,03 ... 61)
In, whether k is 61, if k ≠ 61, then enters step 3.2.2.5.If k=61, then illustrate to have completed working as n
During for some explicit value, to oblique line section Fcn-01To Fcn-61The identification screening of upper doubtful frame, and own
With screened go out oblique line section Fc including car plate doubtful frame line segmentn-Linear equation Lc corresponding for kn-k(k
Sequentially take 01,02,03 ... 61) the most it is stored into array U2In.Enter step 3.2.3.
Step 3.2.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 3.2.2.2 institute
State oblique line section Fcn-kAnd line segment array data [c corresponding thereton-k] sequence number, re-execute step 3.2.2.2
To step 3.2.2.4.
Step 3.2.3: make the drift angle of cog region of vertical isosceles triangle from starting point c of line segment c1(40,40)
A pixel unit is the most vertically moved, after each drift angle moves a new position, all to y-axis positive direction
Update the drift angle c of the vertically cog region of isosceles triangle described in step 3.2.2.1n, and repeated execution of steps 3.2.2,
Until vertically the drift angle of the cog region of isosceles triangle moves the terminal c to line segment c498Time (40,537),
By cnIt is updated to c498, subsequently into step 3.3.
Step 3.3: by the cog region drift angle of isosceles triangle by the terminal c of the line segment c described in step 3.1498
(40,537) change the starting point (729,40) to line segment d, and according to complete with the method as step 3.2
Becoming the cog region drift angle scanning recognition process along vertical line segment d of isosceles triangle, it specifically includes following son
Step:
Step 3.3.1: upper a certain to fixation by the drift angle of the cog region of vertical isosceles triangle and line segment d
Vegetarian refreshments overlaps, and corresponding with oblique line section by the ordinate value renewal array of current given pixel, oblique line section
Linear equation.Its specific practice is:
Make the drift angle of the cog region of vertical isosceles triangle described in step 3.1 and a certain given pixel on line segment d
Point overlaps, and the sequence number of the now drift angle position of the cog region of vertical isosceles triangle is defined as dn, its
Middle n=y-39, in formula, y is the ordinate value that current corner place gives pixel.That is, vertical isoceles triangle
The drift angle of the cog region of shape is in starting point d of line segment d1Time (729,40), the sequence number of drift angle position is also
d1.When the drift angle of cog region of vertical isosceles triangle is at the terminal d of line segment d690Time (729,537), top
The sequence number of position, angle is d498。
When the sequence number of the drift angle position of the cog region of vertical isosceles triangle is dnTime, define this isosceles three
Serial number Fd of each bar oblique line section in dihedraln-k(k sequentially takes 01,02,03 ... 61).Then Article 1 waist
Serial number Fd of linen-01, serial number Fd of center linen-31, serial number Fd of another stringcoursen-61.By oblique line section
Fdn-01On the coordinate figure of each pixel be all stored in line segment array data [dn-01In], by same method,
By other each oblique line section dn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fdn-k
Corresponding line segment array data [dn-k] in (k sequentially takes 01,02,03 ... 61).
Such as, the drift angle of the cog region of vertical isosceles triangle is in starting point d of line segment d1Time (40,40),
Serial number d of each bar oblique line section in isosceles triangle1-k(k sequentially takes 01,02,03 ... 61).Then
Article one, serial number Fd of stringcourse1-01, serial number Fd of center line1-31, serial number Fd of another stringcourse1-61.Will
Oblique line section Fd1-01On the coordinate figure of each pixel be all stored in line segment array data [d1-01In], with same
Method, by other each oblique line section d1-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number
d1-kCorresponding line segment array data [d1-k] in (k sequentially takes 01,02,03 ... 61).
Step 3.3.2: when completing the cog region drift angle current given coordinate points on vertical line segment d, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment.It specifically wraps
Include following sub-step:
Step 3.3.2.1: vertically the drift angle of the cog region of isosceles triangle is in starting point d of line segment d1(729,40)
Time, select oblique line section Fd in current cog region1-01。
Step 3.3.2.2: judge acquired by step 3.3.1 and oblique line section Fdn-kCorresponding line segment array
data[dn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
3.3.2.3.If no, then entering step 3.3.2.4.
Step 3.3.2.3: linea oblique anterior section Fd will be worked asn-kLinear equation Ld corresponding under pixel coordinate systemn-kIt is stored in
Array U2, complete when linea oblique anterior section Fdn-kThe identification screening of upper doubtful frame, and enter step 3.3.2.4.
Step 3.3.2.4: judge present segment array data [dn-k] in (k sequentially takes 01,02,03 ... 61),
Whether k is 61, if k ≠ 61, then enters step 3.3.2.5.If k=61, then illustrate to have completed to as n be
During some explicit value, to oblique line section Fdn-01To Fdn-61The identification screening of upper doubtful frame, and all with
Screened go out oblique line section Fd including car plate doubtful frame line segmentn-Linear equation Ld corresponding for kn-k(k is suitable
Secondary take 01,02,03 ... 61) the most it is stored into array U2In.Enter step 3.3.3.
Step 3.3.2.5: make k=k+1(k sequentially take 01,02,03 ... 61), update step 3.3.2.2 institute
State oblique line section Fdn-kAnd line segment array data [d corresponding thereton-k] sequence number, re-execute step 3.3.2.2
To step 3.3.2.4.
Step 3.3.3: make the drift angle of cog region of vertical isosceles triangle from starting point d of line segment d1(729,40)
A pixel unit is the most vertically moved, after each drift angle moves a new position, all to y-axis positive direction
Update the drift angle d of the vertically cog region of isosceles triangle described in step 3.3.2.1n, and repeated execution of steps 3.3.2,
Until vertically the drift angle of the cog region of isosceles triangle moves the terminal d to line segment d498Time (729,537),
By dnIt is updated to d498, complete the cog region drift angle scanning recognition process along vertical line segment d of isosceles triangle,
Subsequently into step 4.
Step 4: straight line set U described in read step two respectively1With straight line set U described in step 32, and sentence
Disconnected logical formula U1&U2Result, if its result is true, then enter step 5, if its result is false, the most directly
Connect the identification process terminating this method.
Step 5: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, from U1Middle screening retains energy and U2In the most vertical straight line of two straight lines, by qualified
Straight line is stored in set U3, subsequently into step 6.
Step 6: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, from U2Middle screening retains energy and U3In the most vertical straight line of two straight lines, by qualified
Straight line is stored in set U4, subsequently into step 7.
Step 7: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, it is judged that set U4In whether have two straight lines can simultaneously and U3In two straight lines vertical, then
Enter step 8.
Step 8: obtained the linear equation of rectangle four edges line by step 7 respectively, the four edges line of this rectangle
It is the car plate doubtful boundary line in photo.
The slope result of product according to two intersecting straight lines described in step 5 or step 6 or step 7, sentences
Disconnected whether meet condition perpendicular to one another both it criterion be:
Any two linear equations are denoted as respectively: straight line 1:y1=k1x+b1.Straight line 2:y2=k2x+b2.Then
Judgement formula: E=k can be obtained1×k2。
If the result value of E is in interval [-1.05 ,-0.95], then it is assumed that two straight lines are perpendicular to one another.
Claims (2)
1. the automatic identifying method of car plate frame in preventing road monitoring system, it is characterised in that the method include as
Lower step:
The first step, the original image that the resolution taken by preventing road monitoring system is 768 × 576 pixels is entered
Row edge extracting, it is thus achieved that bianry image E, in bianry image E gray value be 1 point represent marginal point, ash
Angle value be 0 point represent non-edge point, with the upper left corner of bianry image E as initial point, set up x-axis level to
Right, y-axis pixel planes rectangular coordinate system straight down;
Second step, acquisition comprise the straight line set U of doubtful horizontal sides wire section1, it specifically includes following sub-step
Rapid:
Step 2.1: two scanning patterns on the cog region of definition isosceles triangle and horizontal direction:
Defining the cog region of horizontal isosceles triangle, the drift angle of this isosceles triangle is 30 degree and towards x-axis just
Direction, the pixel (729,0) that the straight line at the place, base of isosceles triangle is perpendicular in x-axis;Isosceles three
The center line of dihedral is parallel with x-axis;58 oblique line sections, described 58 oblique lines are also comprised in this isosceles triangle
Duan Jun is with the drift angle of isosceles triangle as starting point, and base is terminal, and is positioned at coordinate system range described in step one
Within;In this isosceles triangle in 61 line segments altogether including two stringcourses and a center line, arbitrarily
The adjacent angle between two is 0.5 degree;
Choose with pixel a1(40,40) are starting point with pixel a728(729,40) are the line of terminal
Section a, and choose with pixel (40,537) and be starting point and be the line of terminal with pixel (729,537)
B is as the scanning pattern in two horizontal directions for section;The length of line segment a and line segment b is 690 pixels
Point, and both is parallel with x-axis;
Step 2.2: the cog region drift angle of isosceles triangle is along the scanning recognition process of horizontal line section a, and it is concrete
Including following sub-step:
Step 2.2.1: upper a certain to fixation by the drift angle of the cog region of horizontal isosceles triangle and line segment a
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of horizontal isosceles triangle described in step 2.1 and a certain given pixel on line segment a
Point overlaps, and the sequence number of the drift angle position of the cog region of the most horizontal isosceles triangle is defined as an, its
Middle n=x-39, in formula, x is the abscissa value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of horizontal isosceles triangle is anTime, define this isosceles three
Serial number Fa of each bar oblique line section in dihedraln-k, k sequentially takes 01,02,03 ... 61;By oblique line section Fan-01
On the coordinate figure of each pixel be all stored in line segment array data [an-01In], by same method, by it
Its each oblique line section an-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fan-kCorresponding
Line segment array data [an-kIn], k sequentially takes 01,02,03 ... 61;
Step 2.2.2: when completing the cog region drift angle current given coordinate points on horizontal line section a, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 2.2.2.1: the drift angle of the cog region of horizontal isosceles triangle is in starting point a of line segment a1(40,40)
Time, select oblique line section Fa in current cog region1-01;
Step 2.2.2.2: judge acquired by step 2.2.1 and oblique line section Fan-kCorresponding line segment array
data[an-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
2.2.2.3;If no, then entering step 2.2.2.4;
Step 2.2.2.3: linea oblique anterior section Fa will be worked asn-kLinear equation La corresponding under pixel coordinate systemn-kIt is stored in
Array U1, complete when linea oblique anterior section Fan-kThe identification screening of upper doubtful frame, and enter step 2.2.2.4;
Step 2.2.2.4: judge current line segment array data [an-kIn], k sequentially takes 01,02,03 ... 61,
Whether k is 61, if k ≠ 61, then enters step 2.2.2.5;If k=61, then enter step 2.2.3;
Step 2.2.2.5: make k=k+1, k sequentially take 01,02,03 ... 61, updates step 2.2.2.2 institute
State oblique line section Fan-kAnd line segment array data [a corresponding thereton-k] sequence number, re-execute step 2.2.2.2
To step 2.2.2.4;
Step 2.2.3: make the drift angle of cog region of horizontal isosceles triangle from starting point a of line segment a1(40,40)
A pixel unit is sequentially moved horizontally, after each drift angle moves a new position, all to x-axis positive direction
Update the drift angle a of the cog region of horizontal isosceles triangle described in step 2.2.2.1n, and repeated execution of steps 2.2.2,
Until the drift angle of the cog region of horizontal isosceles triangle moves the terminal a to line segment a729Time (729,40),
By anIt is updated to a690, subsequently into step 2.3;
Step 2.3: by the cog region drift angle of isosceles triangle by the terminal a of the line segment a described in step 2.1729
(729,40) change the starting point (40,537) to line segment b, and according to complete with the method as step 2.2
Becoming the cog region drift angle scanning recognition process along horizontal line section b of isosceles triangle, it specifically includes following son
Step:
Step 2.3.1: upper a certain to fixation by the drift angle of the cog region of horizontal isosceles triangle and line segment b
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of horizontal isosceles triangle described in step 2.1 and a certain given pixel on line segment b
Point overlaps, and the sequence number of the drift angle position of the cog region of the most horizontal isosceles triangle is defined as bn, its
Middle n=x-39, in formula, x is the abscissa value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of horizontal isosceles triangle is bnTime, define this isosceles three
Serial number Fb of each bar oblique line section in dihedraln-k, k sequentially takes 01,02,03 ... 61;By oblique line section Fbn-01
On the coordinate figure of each pixel be all stored in line segment array data [bn-01In], by same method, by it
Its each oblique line section bn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fbn-kCorresponding
Line segment array data [bn-kIn], k sequentially takes 01,02,03 ... 61;
Step 2.3.2: when completing the cog region drift angle current given coordinate points on horizontal line section b, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 2.3.2.1: the drift angle of the cog region of horizontal isosceles triangle is in starting point b of line segment b1(40,40)
Time, select oblique line section Fb in current cog region1-01;
Step 2.3.2.2: judge acquired by step 2.3.1 and oblique line section Fbn-kCorresponding line segment array
data[bn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
2.3.2.3;If no, then entering step 2.3.2.4;
Step 2.3.2.3: linea oblique anterior section Fb will be worked asn-kLinear equation Lb corresponding under pixel coordinate systemn-kIt is stored in
Array U1, complete when linea oblique anterior section Fbn-kThe identification screening of upper doubtful frame, and enter step 2.3.2.4;
Step 2.3.2.4: judge present segment array data [bn-k] in (k sequentially takes 01,02,03 ... 61),
Whether k is 61, if k ≠ 61, then enters step 2.3.2.5;If k=61, then enter step 2.3.3;
Step 2.3.2.5: make k=k+1, k sequentially take 01,02,03 ... 61, updates step 2.3.2.2 institute
State oblique line section Fbn-kAnd line segment array data [b corresponding thereton-k] sequence number, re-execute step 2.3.2.2
To step 2.3.2.4;
Step 2.3.3: make the drift angle of cog region of horizontal isosceles triangle from starting point b of line segment b1(40,537)
A pixel unit is sequentially moved horizontally, after each drift angle moves a new position, all to x-axis positive direction
Update the drift angle b of the cog region of horizontal isosceles triangle described in step 2.3.2.1n, and repeated execution of steps 2.3.2,
Until the drift angle of the cog region of horizontal isosceles triangle moves the terminal b to line segment b729Time (729,40),
By bnIt is updated to b690, complete the cog region drift angle scanning recognition process along horizontal line section b of isosceles triangle,
Subsequently into step 3;
Step 3, acquisition comprise the straight line set U of doubtful vertical edge wire section2, it specifically includes following sub-step
Rapid: step 3.1: two scanning patterns on the cog region of definition isosceles triangle and vertical direction:
Defining the cog region of vertical isosceles triangle, the drift angle of this isosceles triangle is 30 degree and towards y-axis just
Direction, the pixel (0,537) that the straight line at the place, base of isosceles triangle is perpendicular in y-axis;Isosceles three
The center line of dihedral is parallel with y-axis;58 oblique line sections, described 58 oblique lines are also comprised in this isosceles triangle
Duan Jun is with the drift angle of isosceles triangle as starting point, and base is terminal, and is positioned at coordinate system range described in step one
Within;In this isosceles triangle in 61 line segments altogether including two stringcourses and a center line, arbitrarily
The adjacent angle between two is 0.5 degree;
Choosing with pixel (40,40) is starting point with the line segment c that pixel (40,537) is terminal,
And choose with pixel (729,40) and be starting point and make with line segment d that pixel (729,537) is terminal
It it is the scanning pattern on two vertical directions;The length of line segment c and line segment d is 498 pixels, and
Both is parallel with y-axis;
Step 3.2: the cog region drift angle of isosceles triangle is along the scanning recognition process of vertical line segment c, and it is concrete
Including following sub-step:
Step 3.2.1: upper a certain to fixation by the drift angle of the cog region of vertical isosceles triangle and line segment c
Vegetarian refreshments overlaps, and corresponding with oblique line section by the abscissa value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of vertical isosceles triangle described in step 3.1 and a certain given pixel on line segment c
Point overlaps, and the sequence number of the now drift angle position of the cog region of vertical isosceles triangle is defined as cn, its
Middle n=y-39, in formula, y is the ordinate value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of vertical isosceles triangle is cnTime, define this isosceles three
Serial number Fc of each bar oblique line section in dihedraln-k, k sequentially takes 01,02,03 ... 61;By oblique line section Fcn-01
On the coordinate figure of each pixel be all stored in line segment array data [cn-01In], by same method, by it
Its each oblique line section cn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fcn-kCorresponding
Line segment array data [cn-kIn], k sequentially takes 01,02,03 ... 61;
Step 3.2.2: when completing the cog region drift angle current given coordinate points on vertical line segment c, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 3.2.2.1: vertically the drift angle of the cog region of isosceles triangle is in starting point c of line segment c1(40,40)
Time, select oblique line section Fc in current cog region1-01;
Step 3.2.2.2: judge acquired by step 3.2.1 and oblique line section Fcn-kCorresponding line segment array
data[cn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
3.2.2.3;If no, then entering step 3.2.2.4;
Step 3.2.2.3: linea oblique anterior section Fc will be worked asn-kLinear equation Lc corresponding under pixel coordinate systemn-kIt is stored in
Array U2, complete when linea oblique anterior section Fcn-kThe identification screening of upper doubtful frame, and enter step 3.2.2.4;
Step 3.2.2.4: judge current line segment array data [cn-kIn], k sequentially takes 01,02,03 ... 61,
Whether k is 61, if k ≠ 61, then enters step 3.2.2.5;If k=61, then enter step 3.2.3;
Step 3.2.2.5: make k=k+1, k sequentially take 01,02,03 ... 61, updates step 3.2.2.2 institute
State oblique line section Fcn-kAnd line segment array data [c corresponding thereton-k] sequence number, re-execute step 3.2.2.2
To step 3.2.2.4;
Step 3.2.3: make the drift angle of cog region of vertical isosceles triangle from starting point c of line segment c1(40,40)
A pixel unit is the most vertically moved, after each drift angle moves a new position, all to y-axis positive direction
Update the drift angle c of the vertically cog region of isosceles triangle described in step 3.2.2.1n, and repeated execution of steps 3.2.2,
Until vertically the drift angle of the cog region of isosceles triangle moves the terminal c to line segment c498Time (40,537),
By cnIt is updated to c498, subsequently into step 3.3;
Step 3.3: by the cog region drift angle of isosceles triangle by the terminal c of the line segment c described in step 3.1498
(40,537) change the starting point (729,40) to line segment d, and according to complete with the method as step 3.2
Becoming the cog region drift angle scanning recognition process along vertical line segment d of isosceles triangle, it specifically includes following son
Step:
Step 3.3.1: upper a certain to fixation by the drift angle of the cog region of vertical isosceles triangle and line segment d
Vegetarian refreshments overlaps, and corresponding with oblique line section by the ordinate value renewal array of current given pixel, oblique line section
Linear equation;Its specific practice is:
Make the drift angle of the cog region of vertical isosceles triangle described in step 3.1 and a certain given pixel on line segment d
Point overlaps, and the sequence number of the now drift angle position of the cog region of vertical isosceles triangle is defined as dn, its
Middle n=y-39, in formula, y is the ordinate value that current corner place gives pixel;
When the sequence number of the drift angle position of the cog region of vertical isosceles triangle is dnTime, define this isosceles three
Serial number Fd of each bar oblique line section in dihedraln-k, k sequentially takes 01,02,03 ... 61;By oblique line section Fdn-01
On the coordinate figure of each pixel be all stored in line segment array data [dn-01In], by same method, by it
Its each oblique line section dn-kOn the coordinate figure of each pixel be all stored in and this line segment sequence number Fdn-kCorresponding
Line segment array data [dn-kIn], k sequentially takes 01,02,03 ... 61;
Step 3.3.2: when completing the cog region drift angle current given coordinate points on vertical line segment d, to identification
Whether each oblique line section in district has judgement identification and the screening of doubtful car plate frame line segment;It specifically wraps
Include following sub-step:
Step 3.3.2.1: vertically the drift angle of the cog region of isosceles triangle is in starting point d of line segment d1(729,40)
Time, select oblique line section Fd in current cog region1-01;
Step 3.3.2.2: judge acquired by step 3.3.1 and oblique line section Fdn-kCorresponding line segment array
data[dn-kIn], on a bit of line segment being made up of every 40 continuous image vegetarian refreshments, if at least 36 steps
Marginal point described in rapid one, if having, then containing doubtful car plate frame line segment in this oblique line section, enters step
3.3.2.3;If no, then entering step 3.3.2.4;
Step 3.3.2.3: linea oblique anterior section Fd will be worked asn-kLinear equation Ld corresponding under pixel coordinate systemn-kIt is stored in
Array U2, complete when linea oblique anterior section Fdn-kThe identification screening of upper doubtful frame, and enter step 3.3.2.4;
Step 3.3.2.4: judge present segment array data [dn-kIn], k sequentially takes 01,02,03 ... 61,
Whether k is 61, if k ≠ 61, then enters step 3.3.2.5;If k=61, then enter step 3.3.3;
Step 3.3.2.5: make k=k+1, k sequentially take 01,02,03 ... 61, updates step 3.3.2.2 institute
State oblique line section Fdn-kAnd line segment array data [d corresponding thereton-k] sequence number, re-execute step 3.3.2.2
To step 3.3.2.4;
Step 3.3.3: make the drift angle of cog region of vertical isosceles triangle from starting point d of line segment d1(729,40)
A pixel unit is the most vertically moved, after each drift angle moves a new position, all to y-axis positive direction
Update the drift angle d of the vertically cog region of isosceles triangle described in step 3.3.2.1n, and repeated execution of steps 3.3.2,
Until vertically the drift angle of the cog region of isosceles triangle moves the terminal d to line segment d498Time (729,537),
By dnIt is updated to d498, complete the cog region drift angle scanning recognition process along vertical line segment d of isosceles triangle,
Subsequently into step 4;
Step 4: straight line set U described in read step two respectively1With straight line set U described in step 32, and sentence
Disconnected logical formula U1&U2Result, if its result is true, then enter step 5, if its result is false, the most directly
Connect the identification process terminating this method;
Step 5: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, from U1Middle screening retains energy and U2In the most vertical straight line of two straight lines, by qualified
Straight line is stored in set U3, subsequently into step 6;
Step 6: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, from U2Middle screening retains energy and U3In the most vertical straight line of two straight lines, by qualified
Straight line is stored in set U4, subsequently into step 7;
Step 7: according to the slope result of product of two intersecting straight lines, it is judged that whether meet both it and hang down each other
Straight condition, it is judged that set U4In whether have two straight lines can simultaneously and U3In two straight lines vertical, then
Enter step 8;
Step 8: obtained the linear equation of rectangle four edges line by step 7 respectively, the four edges line of this rectangle
It is the car plate doubtful boundary line in photo.
The automatic identifying method of car plate frame in preventing road monitoring system the most according to claim 1, it is special
Levy and be, the slope result of product according to two intersecting straight lines described in step 5 or step 6 or step 7,
Judge that the criterion whether meeting condition perpendicular to one another both it is:
Any two linear equations are denoted as respectively: straight line 1:y1=k1x+b1;Straight line 2:y2=k2x+b2;Then
Judgement formula: E=k can be obtained1×k2;
If the result value of E is in interval [-1.05 ,-0.95], then it is assumed that two straight lines are perpendicular to one another.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310671424.6A CN103729619B (en) | 2013-12-11 | 2013-12-11 | The automatic identifying method of car plate frame in preventing road monitoring system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310671424.6A CN103729619B (en) | 2013-12-11 | 2013-12-11 | The automatic identifying method of car plate frame in preventing road monitoring system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103729619A CN103729619A (en) | 2014-04-16 |
CN103729619B true CN103729619B (en) | 2016-08-17 |
Family
ID=50453685
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310671424.6A Expired - Fee Related CN103729619B (en) | 2013-12-11 | 2013-12-11 | The automatic identifying method of car plate frame in preventing road monitoring system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103729619B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106898083A (en) * | 2017-03-01 | 2017-06-27 | 深圳怡化电脑股份有限公司 | A kind of image processing method and device for bank note |
CN110059702B (en) * | 2019-03-29 | 2021-10-22 | 北京奇艺世纪科技有限公司 | Object contour recognition method and device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0347090A2 (en) * | 1988-06-15 | 1989-12-20 | Eev Limited | Vehicle monitoring system |
CN202077142U (en) * | 2011-05-17 | 2011-12-14 | 成都凯智科技有限公司 | Vehicle-mounted intelligent video detecting and analyzing system |
-
2013
- 2013-12-11 CN CN201310671424.6A patent/CN103729619B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0347090A2 (en) * | 1988-06-15 | 1989-12-20 | Eev Limited | Vehicle monitoring system |
CN202077142U (en) * | 2011-05-17 | 2011-12-14 | 成都凯智科技有限公司 | Vehicle-mounted intelligent video detecting and analyzing system |
Non-Patent Citations (1)
Title |
---|
自适应窗口选取的Mean-Shift目标跟踪;王明佳 等;《光子学报》;20120131;第41卷(第1期);67-71 * |
Also Published As
Publication number | Publication date |
---|---|
CN103729619A (en) | 2014-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105206109B (en) | A kind of vehicle greasy weather identification early warning system and method based on infrared CCD | |
CN107341453B (en) | Lane line extraction method and device | |
CN101633356B (en) | System and method for detecting pedestrians | |
CN104933409B (en) | A kind of parking stall recognition methods based on panoramic picture dotted line feature | |
US8041079B2 (en) | Apparatus and method for detecting obstacle through stereovision | |
CN104700072B (en) | Recognition methods based on lane line historical frames | |
CN107633220A (en) | A kind of vehicle front target identification method based on convolutional neural networks | |
CN105868687A (en) | Traveling road surface detection apparatus and traveling road surface detection method | |
CN107066953A (en) | It is a kind of towards the vehicle cab recognition of monitor video, tracking and antidote and device | |
CN109521019A (en) | A kind of bridge bottom crack detection method based on unmanned plane vision | |
CN103927526A (en) | Vehicle detecting method based on Gauss difference multi-scale edge fusion | |
CN103903005A (en) | License plate image identification system and method | |
CN109815906A (en) | Method for traffic sign detection and system based on substep deep learning | |
CN104318206A (en) | Barrier detection method and apparatus | |
CN103679205A (en) | Preceding car detection method based on shadow hypothesis and layered HOG (histogram of oriented gradient) symmetric characteristic verification | |
Kim et al. | Road and lane detection using stereo camera | |
CN109886175A (en) | A kind of method for detecting lane lines that straight line is combined with circular arc | |
CN109887273A (en) | A kind of bridge mobile load Optimum Identification Method based on multi-source redundancy | |
CN109886088A (en) | The determination method and Related product in congestion lane | |
CN103729619B (en) | The automatic identifying method of car plate frame in preventing road monitoring system | |
CN103679121A (en) | Method and system for detecting roadside using visual difference image | |
CN110443142B (en) | Deep learning vehicle counting method based on road surface extraction and segmentation | |
CN105654038A (en) | Vehicle lamp identification method and device | |
CN110826364B (en) | Library position identification method and device | |
Zhang et al. | Vehicle detection in UAV aerial images based on improved YOLOv3 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160817 Termination date: 20201211 |
|
CF01 | Termination of patent right due to non-payment of annual fee |