CN105608455A - License plate tilt correction method and apparatus - Google Patents
License plate tilt correction method and apparatus Download PDFInfo
- Publication number
- CN105608455A CN105608455A CN201510968421.8A CN201510968421A CN105608455A CN 105608455 A CN105608455 A CN 105608455A CN 201510968421 A CN201510968421 A CN 201510968421A CN 105608455 A CN105608455 A CN 105608455A
- Authority
- CN
- China
- Prior art keywords
- license plate
- candidate
- contour point
- plate area
- area image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/242—Aligning, centring, orientation detection or correction of the image by image rotation, e.g. by 90 degrees
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a license plate tilt correction method and apparatus. The method comprises the steps of determining a license plate region image in a monitored image of a target vehicle; searching for an upper edge point and a lower edge point of a license plate in the upper part and lower part of the license plate region image respectively; performing linear fitting on the upper edge point and the lower edge point separately to obtain a slope and a variance corresponding to license plate boundaries in the upper and lower parts of the license plate region image; according to the texture concentration degree of the upper and lower parts of the license plate region image as well as the slope and the variance corresponding to the license plate boundaries in the upper and lower parts of the license plate region image, determining a license plate tilt angle; and according to the license plate tilt angle, performing license plate tilt correction. By applying embodiments of the invention, the license plate tilt correction effect can be optimized.
Description
Technical field
The present invention relates to technical field of image processing, relate in particular to a kind of license plate sloped correcting method and device.
Background technology
Along with the rapid growth of China's vehicle population and traffic trip, various ask relevant to traffic administrationTopic also highlights further, and traffic monitoring, vehicle are located in real time, parking lot vehicle management, motor vehicle are disobeyedThe efficient solution of the problems such as chapter investigation is more and more subject to the attention of relevant department, intelligent transportation systemThe appearance of (IntelligentTransportationSystem, be called for short ITS) effectively alleviated traffic loading,Ensure traffic safety, improved conevying efficiency, become the main direction of current traffic administration development.
License plate recognition technology is as one of key technology of intelligent transportation system, in expressway tol lcollection, cityThe fields such as city's road monitoring, parking lot management, vehicle peccancy investigation are all widely used. Car plate identificationTechnology mainly comprises image pretreatment, car plate location, license plate sloped correction, Character segmentation, character recognitionEtc. several parts, wherein license plate sloped correction is the important foundation that successive character is effectively cut apart and identified, and inclinesThe quality of tiltedly proofreading and correct will directly have influence on the overall discrimination of car plate.
Summary of the invention
The invention provides a kind of license plate sloped correcting method and device, to optimize the effect of license plate sloped correction.
According to the first aspect of the embodiment of the present invention, a kind of license plate sloped correcting method is provided, comprising:
Determine the license plate area image in the monitoring image of target vehicle;
In two parts up and down of license plate area image, search for respectively up contour point and the down contour point of car plate;
Respectively described up contour point and down contour point are carried out to linear fit, obtain described license plate area imageTwo parts up and down in corresponding slope and the variance in car plate border;
According to the upper and lower two-part texture concentration degree of described license plate area image and described license plate area imageTwo parts up and down in corresponding slope and the variance in car plate border, determine license plate sloped angle;
Carry out license plate sloped correction according to described license plate sloped angle.
According to the second aspect of the embodiment of the present invention, a kind of license plate sloped means for correcting is provided, comprising:
The first determining unit, for determining the license plate area image of monitoring image of target vehicle;
Search unit, for searching for respectively the top edge of car plate at two parts up and down of license plate area imagePoint and down contour point;
Matching unit, for respectively described up contour point and down contour point being carried out to linear fit, obtains instituteState corresponding slope and the variance in car plate border in two parts up and down of license plate area image;
The second determining unit, for according to the upper and lower two-part texture concentration degree of described license plate area image withAnd corresponding slope and the variance in car plate border in two parts up and down of described license plate area image, determine carBoard angle of inclination;
Correcting unit, for carrying out license plate sloped correction according to described license plate sloped angle.
The application embodiment of the present invention, after the license plate area image in the monitoring image of determining target vehicle,By search for respectively up contour point and the down contour point of car plate in two parts up and down of license plate area image,And then, respectively up contour point and down contour point are carried out to linear fit, obtain the upper and lower of license plate area imageCorresponding slope and the variance in car plate border in two parts, and two-part up and down according to license plate area imageThe slope that car plate border in two parts up and down of texture concentration degree and described license plate area image is corresponding andVariance, determines license plate sloped angle, carries out license plate sloped correction according to this license plate sloped angle, has optimizedLicense plate sloped calibration result.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of a kind of license plate sloped correcting method of providing of the embodiment of the present invention;
Fig. 2 is the structural representation of a kind of license plate sloped means for correcting of providing of the embodiment of the present invention;
Fig. 3 is the structural representation of the license plate sloped means for correcting of another kind that provides of the embodiment of the present invention;
Fig. 4 is the structural representation of the license plate sloped means for correcting of another kind that provides of the embodiment of the present invention.
Detailed description of the invention
In order to make those skilled in the art person understand better the technical scheme in the embodiment of the present invention, and makeAbove-mentioned purpose, the feature and advantage of the embodiment of the present invention can become apparent more, below in conjunction with accompanying drawing pairIn the embodiment of the present invention, technical scheme is described in further detail.
Refer to Fig. 1, the flow process of a kind of license plate sloped correcting method that Fig. 1 provides for the embodiment of the present inventionSchematic diagram, as shown in Figure 1, this license plate sloped correcting method can comprise:
Step 101, determine the license plate area image in the monitoring image of target vehicle.
In the embodiment of the present invention, said method can be applied in intelligent transportation system, for example, is applied toBackground server in intelligent transportation system. For ease of describing, below taking the executive agent of said method asServer is that example is described.
In the embodiment of the present invention, target vehicle does not refer in particular to a certain vehicle, but can refer to arbitrary carrying outThe vehicle of license plate sloped correction, the embodiment of the present invention is follow-up no longer to be repeated.
In the embodiment of the present invention, collection of server, after the monitoring image of target vehicle, can be located prisonLicense plate area in control image, to obtain license plate area image. Wherein, at the monitoring image of target vehicleThe specific implementation in middle positioning licence plate region can be with reference in existing license plate locating method (as coarse positioning algorithm)Related realization, the embodiment of the present invention does not repeat.
Step 102, in two parts up and down of license plate area image, search for respectively car plate up contour point andDown contour point.
In the embodiment of the present invention, server determines after license plate area image, can be at license plate area imageIn upper and lower two parts, search for up contour point and the down contour point of car plate, so that determine the particular location of car plate.
As the optional embodiment of one, in above-mentioned steps 102, upper at license plate area image respectivelyUp contour point and the down contour point of in lower two parts, searching for car plate, can comprise the following steps:
11), license plate area image is divided into upper and lower two parts;
12), from left to right scan by column the first half image and the latter half image, and respectively by the first halfIn the each row of partial image, be more than or equal to continuously the most top white pixel in the white pixel point of predetermined numberPoint, and in the each row of the latter half image, be more than or equal to continuously in the white pixel point of predetermined number and lean on mostUnder white pixel point, be defined as candidate's up contour point and candidate's down contour point;
13), the credible marginal point screening mode of employing based on statistical information is under candidate marginal and candidateMarginal point screens, to determine up contour point and the down contour point of car plate.
In this embodiment, server can be taking 1/2 of picture altitude as boundary, by license plate area imageBe divided into upper and lower two parts, and the candidate marginal of searching for respectively in upper and lower two parts image (may beThe point at car plate edge), its concrete way of search can be as follows:
A), for the first half image: server can scan by column from left to right to the first half image,For each row, individual element point scans from top to bottom, to detect white pixel point, ifIn a certain row, find and be more than or equal to continuously predetermined number (this predetermined number can be according to concreteScape set, be below described as N) white pixel point, this is more than or equal to continuously N white look likeThe most top white pixel point in vegetarian refreshments (first white pixel point) is from top to bottom defined as candidateUp contour point, wherein, the transverse and longitudinal coordinate of i candidate's up contour point can be designated as respectively LocationUpX[i]And LocationUpY[i].
B, for the latter half image: server can scan by column from left to right to the latter half image,For each row, individual element point scans from the bottom up, to detect white pixel point, ifIn a certain row, find and be more than or equal to continuously N white pixel point, this is greater than continuously or etc.White pixel point most on the lower in N white pixel point (first white pixel point) from the bottom upBe defined as candidate's down contour point, wherein, the transverse and longitudinal coordinate of i candidate's down contour point can be designated as respectivelyLocationDnX[i] and LocationDnY[i].
By above-mentioned processing, server can search on some candidates respectively from upper and lower two parts imageMarginal point and candidate's down contour point.
In this embodiment, after server search arrives candidate's up contour point and candidate's down contour point, canAccording to positional information, texture information etc., candidate marginal is screened, as adopted based on statistical informationCredible marginal point screening mode candidate's up contour point and candidate's down contour point are screened, it is concreteRealization can comprise the following steps:
21), calculate respectively the white pixel point number of upper candidate marginal and lower candidate marginal columnMean value, and the number of the white pixel point of column is less than to candidate's up contour point and the candidate of mean valueDown contour point is deleted;
22) be, 0 or car by ordinate value in the candidate's up contour point after primary screening and candidate's down contour pointCandidate's up contour point of board area image height and candidate's down contour point are deleted;
23), poor by abscissa value in the candidate's up contour point after postsearch screening and candidate's down contour point respectivelyThe absolute value of value is less than candidate's up contour point and candidate's down contour point of predeterminable range threshold value and deletes;
24), respectively ordinate value in the candidate's up contour point after three screenings and candidate's down contour point is not inconsistentClosing candidate's up contour point and candidate's down contour point of monotonicity deletes.
Taking server, candidate's up contour point is screened that (screening of candidate's down contour point can be in like manner as exampleObtain it), in this embodiment, server search, after candidate's up contour point, can calculate on each candidateThe mean value AverUp of the white pixel point number of marginal point column:
Wherein, m is the sum of candidate's up contour point, HistUp[i] be i candidate's up contour point placeThe number of white pixel point in row.
Server calculates after AverUp, the number of the white pixel point of column can be less than to thisCandidate's top edge pixel of AverUp is deleted, and remaining candidate's up contour point can be called after primary screeningCandidate's up contour point.
In this embodiment, consider that car plate location may not cause the vertical of car plate candidate marginal entirelyCoordinate be 0 or license plate area picture altitude (taking the summit of license plate area image as initial point, license plate area figurePicture institute planar horizontal direction is that axis of abscissas, vertical direction are axis of ordinates), these points incline to car plateThe definite of rake angle can produce larger interference, thereby server can be by the candidate top after primary screeningIn edge point ordinate be 0 or candidate's up contour point of license plate area picture altitude delete, on remaining candidateMarginal point can be called the candidate's up contour point after postsearch screening.
For the candidate's up contour point after postsearch screening, server can further calculate each candidate's top edgePoint distance in the horizontal direction, for example, i+1 candidate's up contour point and i candidate's up contour pointDistance be in the horizontal direction: LocationUpX[i+1]-LocationUpX[i]. Server can be byIn candidate's up contour point after postsearch screening, the distance of horizontal direction is less than the candidate top of predeterminable range threshold valueEdge point deletion, remaining candidate's up contour point can be called the candidate's up contour point after screening three times.
In this embodiment, consider that when license plate sloped, the ordinate of car plate candidate marginal should toolHave certain monotonicity (as monotonic increase or monotone decreasing), thereby server can be by after three screeningsCandidate's up contour point in do not meet monotonicity candidate's up contour point delete, ensure residue candidate top edgeThe monotonicity of some ordinate.
For instance, suppose that three candidate's up contour points after screening comprise 10 candidate's up contour points altogether(ordinate is respectively y1～y10), by order from left to right, these 10 candidate's up contour points from a left side toThe right side, supposes that y1 increases progressively successively to y6, and y8 also increases progressively successively to y10, but y6 is greater than y7, and y6Be less than y8, server can determine that the 7th candidate's up contour point do not meet monotonicity, can be deletedRemove.
By above-mentioned processing, server can be realized the screening to candidate's up contour point and candidate's down contour point,Obtaining credible down contour point (can referred to as up contour point) and credible down contour point (can be referred to as lower limbPoint).
As the optional embodiment of one, in embodiments of the present invention, above-mentioned respectively at license plate area figureBefore searching for the up contour point and down contour point of car plate in two parts up and down of picture, can also comprise:
By Binarization methods, license plate area image is gone to disturb and process.
In this embodiment, consider and in monitoring image, locate the car obtaining by car plate location technologyIn board area image, can there is a lot of definite information in car plate angle of inclination of disturbing, thereby, in order to improve carThe accuracy that board angle of inclination is definite, need to go to disturb to license plate area image and process, to reduce carThe definite interference of interfere information to license plate sloped angle in board area image.
As the optional embodiment of one, by Binarization methods, license plate area image is gone to disturbProcess, can comprise the following steps:
31), obtain the gray value of each pixel in license plate area image, and add up the ash of license plate area imageDegree histogram;
32), determine the binaryzation threshold value of license plate area image according to this grey level histogram;
33), license plate area image is mapped as to hsv color space from RGB color space, and obtainThe colourity of each pixel, saturation degree and brightness in license plate area image;
34), according to the colourity of each pixel in license plate area image, saturation degree, brightness, gray value, twoValue threshold value and default Chroma threshold, saturation degree threshold value, luminance threshold, determine license plate areaForeground pixel in image and background pixel;
35), the pixel value of the foreground pixel in license plate area image is set to 255, the pixel of background pixelValue is set to 0, to obtain the binary image that license plate area image is corresponding;
36), binary image is carried out to expansion process.
In this embodiment, server gets after license plate area image, on the one hand, and can be to this carBoard area image carries out gray processing processing, obtains the gray value of each pixel in this license plate area image, andAdd up the grey level histogram of this license plate area image, and then, determine license plate area according to this grey level histogramThe binaryzation threshold value (Thres) of image, for example, server can be counted according to large Tianjin method (Ostu)Calculate the two-value threshold value of license plate area.
Alternatively, in this embodiment, server is being determined the binaryzation thresholding threshold of license plate area imageAfter value, can also be according to the gray value of each pixel in license plate area image, and this binaryzation thresholdingThreshold value, for each pixel arranges iForeGround flag bit, this iForeGround flag bit is for markPixel is the possibility of foreground pixel, for example, the mark as the iForeGround of pixel can be setThe value of position is 1 o'clock, shows that this pixel may be foreground pixel point; As the iForeGround of pixelThe value of flag bit is 0 o'clock, shows that this pixel can not be foreground pixel point. And then server canBy gray value and the binaryzation threshold value of more each pixel, gray value is greater than to this binaryzation thresholdingThe iForeGround mark position 1 of the pixel of threshold value, on the contrary set to 0.
On the other hand, server can by license plate area image from RGB (Red, Green, Blue, redTurquoise) color space is mapped as HSV (Hue, Saturation, Value, colourity, saturation degree, brightness)Color space, and obtain colourity, saturation degree and the monochrome information of each pixel in license plate area image.
Based on above-mentioned information, server can adopt the binarization method based on many information fusion to enter imageRow binary conversion treatment, specifically can comprise:
A), collect some (can set according to concrete scene, as several thousand, several ten thousand) different colours,The license plate image of different scenes, adds up respectively blue car plate (being called for short blue board), yellow car plate (abbreviation yellow card)And colourity, saturation degree and the monochrome information of white car plate (being called for short person who is not a member of any political party), and true according to statisticsFixed blue board, yellow card, person who is not a member of any political party's Chroma threshold, saturation degree threshold value and luminance threshold;
B), when car plate color (RM the present invention of car plate color does not limit) when blue, willIn license plate area image colourity, saturation degree and brightness respectively at the Chroma threshold of blue board, saturation degree threshold value andWithin the scope of luminance threshold, and the pixel that the value of iForeGround flag bit is 1 is defined as foreground pixelPoint, and the pixel value of this pixel is set to 255; Otherwise, determine that this pixel is background pixel point,And the pixel value of this pixel is set to 0;
C), when car plate color is while being yellow, colourity, saturation degree and brightness in license plate area image are dividedOther within the scope of Chroma threshold, saturation degree threshold value and the luminance threshold of yellow card, and iForeGround markThe pixel that the value of position is 1 is defined as foreground pixel point, and the pixel value of this pixel is set to 255;Otherwise, determine that this pixel is background pixel point, and the pixel value of this pixel be set to 0;
B), when car plate color is when white, colourity, saturation degree and brightness in license plate area image are dividedOther within the scope of person who is not a member of any political party's Chroma threshold, saturation degree threshold value and luminance threshold, and iForeGround markThe pixel that the value of position is 1 is defined as foreground pixel point, and the pixel value of this pixel is set to 255;Otherwise, determine that this pixel is background pixel point, and the pixel value of this pixel be set to 0;
Through above-mentioned processing, can obtain the binary image that license plate area image is corresponding.
In this embodiment, server obtains after binary image that license plate area image is corresponding, canThis binary image is carried out to expansion process, to obtain the binary image after expansion.
Should be realized that, the above-mentioned specific implementation that license plate area image is carried out to binary conversion treatment is only thisA kind of example of the technical scheme that inventive embodiments provides, and be not limiting the scope of the present invention,Based on the embodiment of the present invention, those skilled in the art are not in the situation that paying creative work, to above-mentionedBinary conversion treatment realizes the modification of carrying out all should belong to protection scope of the present invention.
Step 103, respectively up contour point and down contour point are carried out to linear fit, obtain license plate area figureCorresponding slope and the variance in car plate border in two parts up and down of picture.
In the embodiment of the present invention, server search after the up contour point and down contour point of car plate, Ke YifenOther up contour point and down contour point are carried out to linear fit, obtain in two parts up and down of license plate area imageCorresponding slope and the variance in car plate border; Wherein, the car plate limit in the first half of license plate area imageThe slope that the slope that bound pair is answered and variance are the straight line of up contour point institute matching (is that place, car plate coboundary is straightThe slope of line, hereinafter referred to as the first slope) and variance (hereinafter referred to as first variance), license plate area imageThe latter half in slope corresponding to car plate border and the variance slope that is the straight line of down contour point institute matching(being the slope of car plate lower boundary place straight line, hereinafter referred to as the second slope) and variance are (hereinafter referred to asTwo variances). Wherein, the first slope, the second slope, first variance, second variance can be designated as respectivelyAngleUp、AngleDn、VarianceUp、VarianceDn。
Step 104, according to the upper and lower two-part texture concentration degree of license plate area image and described car plate districtCorresponding slope and the variance in car plate border in two parts up and down of area image, determines license plate sloped angle.
In the embodiment of the present invention, server, license plate area image is divided into after upper and lower two parts, also canTo calculate respectively license plate area image the first half and the latter half texture concentration degree, UpTexture andDownTexture, and calculate the departure degree TextureScale of upper and lower two-part texture concentration degree:
TextureScale＝DownTexture/UpTexture；
Wherein, when TextureScale be greater than the first departure degree threshold value (can set according to concrete scene,Below be described as α) time, think that the texture of the latter half is abundanter; When TextureScale is less than secondWhen departure degree threshold value (can set according to concrete scene, below be described as β), think the first halfTexture is abundanter; When TextureScale is in the time that [β, α] is interval, think that upper and lower two parts texture is suitable.Wherein, α is greater than β.
What deserves to be explained is, in embodiments of the present invention, server calculates upper and lower two of license plate area imagePoint operation and the server of departure degree of texture concentration degree in license plate area, search for up contour point with underBetween the operation of marginal point, do not have inevitable sequential relationship, server can first be carried out calculating car plateThe operation of the departure degree of the upper and lower two-part texture concentration degree of area image, rear execution is in license plate areaBetween search up contour point and the operation of down contour point, on the contrary also can, can also the two walk abreast and carry out,The embodiment of the present invention does not limit this.
In the embodiment of the present invention, server obtains license plate area image upper and lower two-part texture concentration degreeAfter the corresponding slope and variance in car plate border in two parts up and down of departure degree, license plate area image,Can be according in two parts up and down of the departure degree of this texture concentration degree and described license plate area imageThe slope that car plate border is corresponding and variance are determined license plate sloped angle, can comprise:
If exist texture abundanter in the upper and lower two parts of license plate area image, and variance corresponding to car plate borderLarger part, determines license plate sloped angle according to slope corresponding to car plate border in another part;
If exist texture abundanter in the upper and lower two parts of license plate area image, but variance corresponding to car plate borderLittle part, the variance size corresponding according to the car plate border in another part, and this another partIn the corresponding angular dimension of slope corresponding to car plate border, determine license plate sloped angle;
If the upper and lower two parts texture of license plate area image is suitable, divide according to the car plate border in these two partsThe corresponding corresponding angular dimension of slope is not determined license plate sloped angle.
As the optional embodiment of one, above-mentioned implementation can comprise following several situation:
1, when departure degree is greater than the first departure degree threshold value, and second variance is while being greater than first variance, willThe angle that the first slope is corresponding is defined as license plate sloped angle;
2, when departure degree is less than the second departure degree threshold value, and first variance is while being greater than second variance, willThe angle that the second slope is corresponding is defined as license plate sloped angle; Wherein, the first departure degree threshold value is greater thanTwo departure degree threshold values;
3, when departure degree is greater than the first departure degree threshold value, first variance is greater than second variance, and firstWhen variance is less than default variance threshold values, if angle corresponding to the first slope is less than the first angle threshold, willThe angle that the first slope is corresponding is defined as license plate sloped angle; If the angle that the first slope is corresponding is greater than or etc.In the first angle threshold, angle corresponding the second slope is defined as to license plate sloped angle;
4, when departure degree is less than the second departure degree threshold value, second variance is greater than first variance, and secondWhen variance is less than default variance threshold values, if angle corresponding to the second slope is less than the first angle threshold, willThe angle that the second slope is corresponding is defined as license plate sloped angle; If the angle that the second slope is corresponding is greater than or etc.In the first angle threshold, angle corresponding the first slope is defined as to license plate sloped angle;
5, when departure degree is greater than the first departure degree threshold value, first variance is greater than second variance, and firstWhen variance is more than or equal to default variance threshold values, angle corresponding the second slope is defined as to license plate sloped angleDegree;
6, when departure degree is less than the second departure degree threshold value, second variance is greater than first variance, and secondWhen variance is more than or equal to default variance threshold values, angle corresponding the first slope is defined as to license plate sloped angleDegree;
7, when departure degree is more than or equal to the second departure degree threshold value, be less than or equal to the first departure degreeThreshold value, and the absolute value of the difference of angle corresponding to the first slope angle corresponding with the second slope is less thanWhen two angle thresholds, the mean value of angle corresponding with the second slope angle corresponding the first slope is determinedFor license plate sloped angle;
8, when departure degree is more than or equal to the second departure degree threshold value, be less than or equal to the first departure degreeThreshold value, and the absolute value of the difference of angle corresponding to the first slope angle corresponding with the second slope is greater than orWhile equaling the second angle threshold, if first variance is less than second variance, by angle corresponding the first slopeBe defined as license plate sloped angle; If first variance is greater than second variance, by inclination corresponding the second slopeAngle is defined as license plate sloped angle.
In this embodiment, when a part (the first half or the latter half) texture in upper and lower two partsConcentration degree little (being that another part texture concentration degree is abundanter), and the marginal point of this part institute fitting a straight lineVariance hour, is defined as license plate sloped by the angle corresponding to slope of the straight line of the marginal point matching of this partAngle. For example,, as TextureScale > α and VarianceDn > when VarianceUp, get firstAngle corresponding to slope is as the angle of slant correction; When TextureScale < β and VarianceDn <When VarianceUp, get angle that the second slope the is corresponding angle as slant correction.
When certain part texture concentration degree is little, and the variance of the marginal point of this part institute fitting a straight line is when large,TextureScale>α and VarianceDn<VarianceUp or TextureScale<β andVarianceDn > VarianceUp, server can further judge that the marginal point institute matching of this part is straightWhether the variance of line is less than default variance threshold values (can set according to concrete scene, below be described as T);If be less than T, further judge angle that the slope of marginal point institute fitting a straight line of this part is corresponding whetherBe less than the first angle threshold (can set according to concrete scene, as 30 degree, 35 degree etc., below with 30Degree is example), if be less than 30 degree, server can be by the slope of the marginal point institute fitting a straight line of this partCorresponding angle is as license plate sloped angle; As be more than or equal to 30 degree, server can be by another portionPoint angle corresponding to the slope of marginal point institute fitting a straight line as angle of inclination; If the marginal point of this partThe variance of institute's fitting a straight line is greater than T, what server can be by the marginal point institute fitting a straight line of another partAngle corresponding to slope is as license plate sloped angle.
When upper and lower two parts texture concentration degree is suitable, TextureScale, in the time that [β, α] is interval, servesDevice can calculate the absolute value of the difference of angle that the first slope is corresponding and angle corresponding to the second slope, andWhether the absolute value that judges this difference is less than the second angle threshold (can set according to concrete scene), if littleIn, using the average of angle corresponding the first slope and angle corresponding to the second slope as license plate sloped angleDegree; Otherwise, by the angle corresponding to slope of the straight line of marginal point institute the matching little variance of the straight line of institute's matchingDegree is as license plate sloped angle.
Step 105, carry out license plate sloped correction according to this license plate sloped angle.
In the embodiment of the present invention, server is being determined after license plate sloped angle, can inclined according to this car plateRake angle carries out license plate sloped correction to the car plate of target vehicle, and its specific implementation the present invention do not limitFixed.
Can find out by above description, in the technical scheme providing in the embodiment of the present invention, at definite orderAfter license plate area image in the monitoring image of mark vehicle, by respectively at up and down two of license plate area imageIn part, search for up contour point and the down contour point of car plate, and then, respectively to up contour point and down contour pointCarry out linear fit, obtain slope corresponding to car plate border in two parts up and down of license plate area image andVariance, and according to the upper and lower two-part texture concentration degree of license plate area image and described license plate area imageTwo parts up and down in corresponding slope and the variance in car plate border, determine license plate sloped angle, according to thisLicense plate sloped angle is carried out license plate sloped correction, in the implementation procedure of license plate sloped correction, can eliminateThe car plate interference of other textures to license plate sloped angle calculation around, and be not subject to local overexposure, the ring such as owe to expose to the sunBorder affects, and has strengthened the robustness of license plate sloped angle calculation, has optimized license plate sloped calibration result.
Refer to Fig. 2, the structural representation of a kind of license plate sloped means for correcting providing for the embodiment of the present inventionFigure, wherein, this license plate sloped means for correcting can be applied to the intelligent transportation system in said method embodimentSystem, for example, is applied in the background server of intelligent transportation system, and as shown in Figure 2, this is license plate slopedMeans for correcting can comprise:
The first determining unit 210, for determining the license plate area image of monitoring image of target vehicle;
Search unit 220, searches for car plate for two parts up and down at license plate area image respectivelyMarginal point and down contour point;
Matching unit 230, for respectively described up contour point and down contour point being carried out to linear fit,To corresponding slope and the variance in car plate border in two parts up and down of described license plate area image;
The second determining unit 240, for concentrating according to the upper and lower two-part texture of described license plate area imageCorresponding slope and the variance in car plate border in two parts up and down of degree and described license plate area image, reallyFixed license plate sloped angle;
Correcting unit 250, for carrying out license plate sloped correction according to described license plate sloped angle.
Please also refer to Fig. 3, the knot of the license plate sloped means for correcting of another kind providing for the embodiment of the present inventionStructure schematic diagram, this embodiment is on aforementioned basis embodiment illustrated in fig. 2, and described device can also comprise:
Binary conversion treatment unit 260, for going described license plate area image by Binarization methodsDisturb and process;
In optional embodiment, in the license plate sloped means for correcting shown in Fig. 3, binary conversion treatment unit 260,Can comprise:
Statistics subelement 261, for obtaining the gray value of the each pixel of described license plate area image, andAdd up the grey level histogram of described license plate area image;
Determine subelement 262, for determining two of described license plate area image according to described grey level histogramValue threshold value;
Obtain subelement 263, for described license plate area image is shone upon from RGB RGB color spaceFor colourity saturation degree brightness hsv color space, and obtain the look of each pixel in described license plate area imageDegree, saturation degree and brightness;
Described definite subelement 262, also for according to the colourity of the each pixel of described license plate area image,Saturation degree, brightness, gray value, described binaryzation threshold value and default Chroma threshold, saturation degreeThreshold value, luminance threshold, determine foreground pixel and background pixel in described license plate area image;
Subelement 264 is set, for the pixel value of the foreground pixel of described license plate area image is set to255, the pixel value of background pixel is set to 0, to obtain binary image corresponding to described license plate area image;
Process subelement 265, for described binary image is carried out to expansion process.
Please also refer to Fig. 4, the knot of the license plate sloped means for correcting of another kind providing for the embodiment of the present inventionStructure schematic diagram, this embodiment on aforementioned basis embodiment illustrated in fig. 2, search unit in described device220, can comprise:
Divide subelement 221, for license plate area image is divided into upper and lower two parts;
Scanning subelement 222, for from left to right scanning by column the first half image and the latter half image,And will in each described the first half image row, be more than or equal to continuously in the white pixel point of predetermined number respectivelyThe most top white pixel point, and be more than or equal to continuously present count in the each row of described the latter half imageWhite pixel point most on the lower in the white pixel point of amount, is defined as candidate's up contour point and candidate followingEdge point;
Screening subelement 223, for adopting credible marginal point screening mode based on statistical information to describedCandidate marginal and candidate's down contour point screen, to determine up contour point and the down contour point of car plate.
In optional embodiment, described screening subelement 223, can be specifically for:
The white pixel point number of the upper candidate marginal of calculating and lower candidate marginal column is average respectivelyValue, and it is following that the number of the white pixel point of column is less than to candidate's up contour point and the candidate of mean valueEdge point deletion;
Be 0 or car plate district by ordinate value in the candidate's up contour point after primary screening and candidate's down contour pointCandidate's up contour point of area image height and candidate's down contour point are deleted;
Respectively by the difference of abscissa value in the candidate's up contour point after postsearch screening and candidate's down contour pointAbsolute value is less than candidate's up contour point and candidate's down contour point of predeterminable range threshold value and deletes;
Respectively ordinate value in the candidate's up contour point after three screenings and candidate's down contour point is not met to listCandidate's up contour point of tonality and candidate's down contour point are deleted.
In optional embodiment, described the second determining unit 240, specifically for:
If exist texture abundanter in the upper and lower two parts of license plate area image, and variance corresponding to car plate borderLarger part, determines license plate sloped angle according to slope corresponding to car plate border in another part;
If exist texture abundanter in the upper and lower two parts of license plate area image, but variance corresponding to car plate borderLittle part, according to the corresponding variance size on the car plate border in another part, and this another portionDivide the corresponding corresponding angular dimension of slope, determine license plate sloped angle;
If the upper and lower two parts texture of license plate area image is suitable, divide according to the car plate border in these two partsThe corresponding corresponding angle of slope is not determined license plate sloped angle.
In said apparatus, the implementation procedure of the function and efficacy of unit specifically refers in said method correspondingThe implementation procedure of step, does not repeat them here.
For device embodiment, because it corresponds essentially to embodiment of the method, so relevant part ginsengSee the part explanation of embodiment of the method. Device embodiment described above is only schematically,The wherein said unit as separating component explanation can or can not be also physically to separate, and doesThe parts that show for unit can be or can not be also physical locations, can be positioned at a place,Or also can be distributed on multiple NEs. Can select according to the actual needs part wherein orThe whole modules of person realize the present invention program's object. Those of ordinary skill in the art are not paying creativenessIn the situation of work, be appreciated that and implement.
As seen from the above-described embodiment, after the license plate area image in the monitoring image of determining target vehicle,By search for respectively up contour point and the down contour point of car plate in two parts up and down of license plate area image,And then, respectively up contour point and down contour point are carried out to linear fit, obtain the upper and lower of license plate area imageCorresponding slope and the variance in car plate border in two parts, and two-part up and down according to license plate area imageThe slope that car plate border in two parts up and down of texture concentration degree and described license plate area image is corresponding andVariance, determines license plate sloped angle, carries out license plate sloped correction according to this license plate sloped angle, has optimizedLicense plate sloped calibration result.
Those skilled in the art are considering description and are putting into practice after invention disclosed herein, will easily expect thisOther embodiment of invention. The application is intended to contain any modification of the present invention, purposes or adaptabilityChange, these modification, purposes or adaptations are followed general principle of the present invention and are comprised thisBright undocumented common practise or conventional techniques means in the art. Description and embodiment only byBe considered as exemplary, true scope of the present invention and spirit pointed out by claim below.
Should be understood that, the present invention is not limited to described above and illustrated in the accompanying drawings accurateStructure, and can carry out various amendments and change not departing from its scope. Scope of the present invention is only by instituteAttached claim limits.
Claims (10)
1. a license plate sloped correcting method, is characterized in that, comprising:
Determine the license plate area image in the monitoring image of target vehicle;
In two parts up and down of license plate area image, search for respectively up contour point and the down contour point of car plate;
Respectively described up contour point and down contour point are carried out to linear fit, obtain described license plate area imageTwo parts up and down in corresponding slope and the variance in car plate border;
Upper according to the upper and lower two-part texture concentration degree of license plate area image and described license plate area imageCorresponding slope and the variance in car plate border in lower two parts, determines license plate sloped angle;
Carry out license plate sloped correction according to described license plate sloped angle.
2. method according to claim 1, is characterized in that, described respectively at license plate area imageTwo parts up and down in search for the up contour point and down contour point of car plate before, also comprise:
By Binarization methods, described license plate area image is gone to disturb and process;
Wherein, described by Binarization methods to license plate area go disturb process, comprising:
Obtain the gray value of each pixel in described license plate area image, and add up described license plate area imageGrey level histogram;
Determine the binaryzation threshold value of described license plate area image according to described grey level histogram;
Described license plate area image is mapped as to the brightness of colourity saturation degree from RGB RGB color spaceHsv color space, and colourity, saturation degree and the brightness of obtaining each pixel in described license plate area image;
According to colourity, saturation degree, brightness, gray value, the institute of each pixel in described license plate area imageState binaryzation threshold value and default Chroma threshold, saturation degree threshold value, luminance threshold, described in determiningForeground pixel in license plate area image and background pixel;
The pixel value of the foreground pixel in described license plate area image is set to 255, the pixel of background pixelValue is set to 0, to obtain binary image corresponding to described license plate area image;
Described binary image is carried out to expansion process.
3. method according to claim 2, is characterized in that, described respectively at license plate area imageTwo parts up and down in search for up contour point and the down contour point of car plate, comprising:
License plate area image is divided into upper and lower two parts;
From left to right scan by column the first half image and the latter half image, and respectively by the described first halfIn the each row of partial image, be more than or equal to continuously the most top white pixel in the white pixel point of predetermined numberPoint, and be more than or equal to continuously in the white pixel point of predetermined number in the each row of described the latter half imageWhite pixel point most on the lower, is defined as candidate's up contour point and candidate's down contour point;
The credible marginal point screening mode of employing based on statistical information is under described candidate marginal and candidateMarginal point screens, to determine up contour point and the down contour point of car plate.
4. method according to claim 3, is characterized in that, described employing is based on statistical informationCredible marginal point screening mode screens described candidate's up contour point and candidate's down contour point, comprising:
The white pixel point number of the upper candidate marginal of calculating and lower candidate marginal column is average respectivelyValue, and it is following that the number of the white pixel point of column is less than to candidate's up contour point and the candidate of mean valueEdge point deletion;
Be 0 or car plate district by ordinate value in the candidate's up contour point after primary screening and candidate's down contour pointCandidate's up contour point of area image height and candidate's down contour point are deleted;
Respectively by the difference of abscissa value in the candidate's up contour point after postsearch screening and candidate's down contour pointAbsolute value is less than candidate's up contour point and candidate's down contour point of predeterminable range threshold value and deletes;
Respectively ordinate value in the candidate's up contour point after three screenings and candidate's down contour point is not met to listCandidate's up contour point of tonality and candidate's down contour point are deleted.
5. method according to claim 1, is characterized in that, described according on license plate area imageCar plate border pair in two parts up and down of lower two-part texture concentration degree and described license plate area imageThe slope of answering and variance, determine license plate sloped angle, comprising:
If exist texture abundanter in the upper and lower two parts of license plate area image, and variance corresponding to car plate borderLarger part, determines license plate sloped angle according to slope corresponding to car plate border in another part;
If exist texture abundanter in the upper and lower two parts of license plate area image, but variance corresponding to car plate borderLittle part, the variance size corresponding according to the car plate border in another part, and this another partIn the corresponding angular dimension of slope corresponding to car plate border, determine license plate sloped angle;
If the upper and lower two parts texture of license plate area image is suitable, divide according to the car plate border in these two partsThe corresponding corresponding angular dimension of slope is not determined license plate sloped angle.
6. a license plate sloped means for correcting, is characterized in that, comprising:
The first determining unit, for determining the license plate area image of monitoring image of target vehicle;
Search unit, for searching for respectively the top edge of car plate at two parts up and down of license plate area imagePoint and down contour point;
Matching unit, for respectively described up contour point and down contour point being carried out to linear fit, obtains instituteState corresponding slope and the variance in car plate border in two parts up and down of license plate area image;
The second determining unit, for according to the upper and lower two-part texture concentration degree of license plate area image and instituteState corresponding slope and the variance in car plate border in two parts up and down of license plate area image, determine that car plate inclinesRake angle;
Correcting unit, for carrying out license plate sloped correction according to described license plate sloped angle.
7. device according to claim 6, is characterized in that, described device also comprises:
Binary conversion treatment unit, for going to disturb to described license plate area image by Binarization methodsProcess;
Described binary conversion treatment unit, comprising:
Statistics subelement, for obtaining the gray value of the each pixel of described license plate area image, and statisticsThe grey level histogram of described license plate area image;
Determine subelement, for determine the binaryzation of described license plate area image according to described grey level histogramThreshold value;
Obtain subelement, for described license plate area image is mapped as from RGB RGB color spaceColourity saturation degree brightness hsv color space, and obtain the look of each pixel in described license plate area imageDegree, saturation degree and brightness;
Described definite subelement, also for according to the colourity of the each pixel of described license plate area image, fullWith degree, brightness, gray value, described binaryzation threshold value and default Chroma threshold, saturation degree thresholdValue, luminance threshold, determine foreground pixel and background pixel in described license plate area image;
Subelement is set, for the pixel value of the foreground pixel of described license plate area image is set to 255,The pixel value of background pixel is set to 0, to obtain binary image corresponding to described license plate area image;
Process subelement, for described binary image is carried out to expansion process.
8. device according to claim 7, is characterized in that, described search unit, comprising:
Divide subelement, for license plate area image is divided into upper and lower two parts;
Scanning subelement, for from left to right scanning by column the first half image and the latter half image, andTo in each described the first half image row, be more than or equal to continuously in the white pixel point of predetermined number the most respectivelyTop white pixel point, and be more than or equal to continuously predetermined number in the each row of described the latter half imageWhite pixel point in white pixel point most on the lower, be defined as candidate's up contour point and candidate's lower limbPoint;
Screening subelement, for adopting credible marginal point screening mode based on statistical information to described candidateMarginal point and candidate's down contour point screen, to determine up contour point and the down contour point of car plate.
9. device according to claim 8, is characterized in that, described screening subelement, specifically usesIn:
The white pixel point number of the upper candidate marginal of calculating and lower candidate marginal column is average respectivelyValue, and it is following that the number of the white pixel point of column is less than to candidate's up contour point and the candidate of mean valueEdge point deletion;
Be 0 or car plate district by ordinate value in the candidate's up contour point after primary screening and candidate's down contour pointCandidate's up contour point of area image height and candidate's down contour point are deleted;
Respectively by the difference of abscissa value in the candidate's up contour point after postsearch screening and candidate's down contour pointAbsolute value is less than candidate's up contour point and candidate's down contour point of predeterminable range threshold value and deletes;
Respectively ordinate value in the candidate's up contour point after three screenings and candidate's down contour point is not met to listCandidate's up contour point of tonality and candidate's down contour point are deleted.
10. device according to claim 6, is characterized in that, described the second determining unit, toolBody is used for:
If exist texture abundanter in the upper and lower two parts of license plate area image, and variance corresponding to car plate borderLarger part, determines license plate sloped angle according to slope corresponding to car plate border in another part;
If exist texture abundanter in the upper and lower two parts of license plate area image, but variance corresponding to car plate borderLittle part, the variance size corresponding according to the car plate border in another part, and this another partThe corresponding corresponding angular dimension of slope, determines license plate sloped angle;
If the upper and lower two parts texture of license plate area image is suitable, divide according to the car plate border in these two partsThe corresponding corresponding angle of slope is not determined license plate sloped angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510968421.8A CN105608455B (en) | 2015-12-18 | 2015-12-18 | A kind of license plate sloped correcting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510968421.8A CN105608455B (en) | 2015-12-18 | 2015-12-18 | A kind of license plate sloped correcting method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105608455A true CN105608455A (en) | 2016-05-25 |
| CN105608455B CN105608455B (en) | 2019-06-11 |
Family
ID=55988381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510968421.8A Active CN105608455B (en) | 2015-12-18 | 2015-12-18 | A kind of license plate sloped correcting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105608455B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203433A (en) * | 2016-07-13 | 2016-12-07 | 西安电子科技大学 | In a kind of vehicle monitoring image, car plate position automatically extracts and the method for perspective correction |
| CN106485248A (en) * | 2016-10-20 | 2017-03-08 | 江苏商贸职业学院 | A kind of slag-soil truck license plate sloped correcting method |
| CN106504407A (en) * | 2016-11-01 | 2017-03-15 | 深圳怡化电脑股份有限公司 | A kind of method and device for processing banknote image |
| CN107506765A (en) * | 2017-10-13 | 2017-12-22 | 厦门大学 | A kind of method of the license plate sloped correction based on neutral net |
| CN108241859A (en) * | 2016-12-26 | 2018-07-03 | 浙江宇视科技有限公司 | The bearing calibration of car plate and device |
| CN108647678A (en) * | 2018-04-24 | 2018-10-12 | 山东华软金盾软件股份有限公司 | A kind of quick horizontal tilt car plate antidote in monitor video |
| CN108805122A (en) * | 2017-05-03 | 2018-11-13 | 迅驰（北京）视讯科技有限公司 | Car plate horizontal tilt antidote, system and car license recognition equipment |
| WO2019019250A1 (en) * | 2017-07-28 | 2019-01-31 | 平安科技（深圳）有限公司 | Method and apparatus for acquiring inclination value of inclined image, and terminal and storage medium |
| CN109948521A (en) * | 2019-03-18 | 2019-06-28 | 北京华宇信息技术有限公司 | Image correcting error method and device, equipment and storage medium |
| CN110213629A (en) * | 2019-06-27 | 2019-09-06 | 腾讯科技（深圳）有限公司 | A kind of Information Embedding method, apparatus, server and storage medium |
| CN113139409A (en) * | 2020-01-20 | 2021-07-20 | 浙江宇视科技有限公司 | Vehicle identification method, device, equipment and storage medium |
| CN113228033A (en) * | 2018-12-29 | 2021-08-06 | 浙江大华技术股份有限公司 | License plate recognition system and method |
| CN113259598A (en) * | 2021-07-16 | 2021-08-13 | 深圳市赛菲姆科技有限公司 | Camera horizontal adjustment control method, system, terminal and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496019A (en) * | 2011-12-08 | 2012-06-13 | 银江股份有限公司 | License plate character segmenting method |
| CN103500327A (en) * | 2013-10-16 | 2014-01-08 | 武汉理工大学 | Vehicle type identification method of vehicles of same brand based on space position information |
| CN104318233A (en) * | 2014-10-19 | 2015-01-28 | 温州大学 | Method for horizontal tilt correction of number plate image |
| CN104809715A (en) * | 2014-01-23 | 2015-07-29 | 广州南沙资讯科技园有限公司博士后科研工作站 | Banknote image inclination correction and region extraction method |
-
2015
- 2015-12-18 CN CN201510968421.8A patent/CN105608455B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102496019A (en) * | 2011-12-08 | 2012-06-13 | 银江股份有限公司 | License plate character segmenting method |
| CN103500327A (en) * | 2013-10-16 | 2014-01-08 | 武汉理工大学 | Vehicle type identification method of vehicles of same brand based on space position information |
| CN104809715A (en) * | 2014-01-23 | 2015-07-29 | 广州南沙资讯科技园有限公司博士后科研工作站 | Banknote image inclination correction and region extraction method |
| CN104318233A (en) * | 2014-10-19 | 2015-01-28 | 温州大学 | Method for horizontal tilt correction of number plate image |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203433A (en) * | 2016-07-13 | 2016-12-07 | 西安电子科技大学 | In a kind of vehicle monitoring image, car plate position automatically extracts and the method for perspective correction |
| CN106485248A (en) * | 2016-10-20 | 2017-03-08 | 江苏商贸职业学院 | A kind of slag-soil truck license plate sloped correcting method |
| CN106504407A (en) * | 2016-11-01 | 2017-03-15 | 深圳怡化电脑股份有限公司 | A kind of method and device for processing banknote image |
| CN106504407B (en) * | 2016-11-01 | 2019-06-07 | 深圳怡化电脑股份有限公司 | A kind of method and device handling banknote image |
| CN108241859A (en) * | 2016-12-26 | 2018-07-03 | 浙江宇视科技有限公司 | The bearing calibration of car plate and device |
| CN108805122A (en) * | 2017-05-03 | 2018-11-13 | 迅驰（北京）视讯科技有限公司 | Car plate horizontal tilt antidote, system and car license recognition equipment |
| US11074443B2 (en) | 2017-07-28 | 2021-07-27 | Ping An Technology (Shenzhen) Co., Ltd. | Method and device for acquiring slant value of slant image, terminal and storage medium |
| WO2019019250A1 (en) * | 2017-07-28 | 2019-01-31 | 平安科技（深圳）有限公司 | Method and apparatus for acquiring inclination value of inclined image, and terminal and storage medium |
| CN107506765A (en) * | 2017-10-13 | 2017-12-22 | 厦门大学 | A kind of method of the license plate sloped correction based on neutral net |
| CN107506765B (en) * | 2017-10-13 | 2020-09-01 | 厦门大学 | License plate inclination correction method based on neural network |
| CN108647678A (en) * | 2018-04-24 | 2018-10-12 | 山东华软金盾软件股份有限公司 | A kind of quick horizontal tilt car plate antidote in monitor video |
| CN113228033A (en) * | 2018-12-29 | 2021-08-06 | 浙江大华技术股份有限公司 | License plate recognition system and method |
| US11842535B2 (en) | 2018-12-29 | 2023-12-12 | Zhejiang Dahua Technology Co., Ltd. | Systems and methods for license plate recognition |
| CN109948521A (en) * | 2019-03-18 | 2019-06-28 | 北京华宇信息技术有限公司 | Image correcting error method and device, equipment and storage medium |
| CN109948521B (en) * | 2019-03-18 | 2021-05-11 | 北京华宇信息技术有限公司 | Image deviation rectifying method and device, equipment and storage medium |
| CN110213629A (en) * | 2019-06-27 | 2019-09-06 | 腾讯科技（深圳）有限公司 | A kind of Information Embedding method, apparatus, server and storage medium |
| CN113139409A (en) * | 2020-01-20 | 2021-07-20 | 浙江宇视科技有限公司 | Vehicle identification method, device, equipment and storage medium |
| CN113259598A (en) * | 2021-07-16 | 2021-08-13 | 深圳市赛菲姆科技有限公司 | Camera horizontal adjustment control method, system, terminal and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105608455B (en) | 2019-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105608455A (en) | License plate tilt correction method and apparatus | |
| JP4928310B2 (en) | License plate recognition device, control method thereof, computer program | |
| CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
| CN109871763B (en) | Specific target tracking method based on YOLO | |
| CN101443791B (en) | Improved foreground/background separation in digitl images | |
| US20150294163A1 (en) | Image processing device | |
| CN105373794A (en) | Vehicle license plate recognition method | |
| CN106971185B (en) | License plate positioning method and device based on full convolution network | |
| US6377711B1 (en) | Methods and systems for detecting the edges of objects in raster images using diagonal edge detection | |
| CN101154271A (en) | License plate character segmentation method based on fast area labeling algorithm and license plate large-spacing locating method | |
| CN105308944A (en) | Classifying objects in images using mobile devices | |
| CN111967313B (en) | Unmanned aerial vehicle image annotation method assisted by deep learning target detection algorithm | |
| EP0843275A2 (en) | Pattern extraction apparatus and method for extracting patterns | |
| Yang et al. | A vehicle license plate recognition system based on fixed color collocation | |
| US11164031B2 (en) | System and method for analyzing an image of a vehicle | |
| CN102915544A (en) | Video image motion target extracting method based on pattern detection and color segmentation | |
| CN102542244A (en) | Face detection method and system and computer program product | |
| CN106951898A (en) | Recommend method and system, electronic equipment in a kind of vehicle candidate region | |
| CN104463134A (en) | License plate detection method and system | |
| CN109977941A (en) | Licence plate recognition method and device | |
| CN112528994B (en) | Free angle license plate detection method, license plate recognition method and recognition system | |
| CN113628202A (en) | Determination method, cleaning robot and computer storage medium | |
| CN111178359A (en) | License plate number recognition method, device and equipment and computer storage medium | |
| CN103295238A (en) | ROI (region of interest) motion detection based real-time video positioning method for Android platform | |
| CN106815580A (en) | A kind of license plate locating method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |