CN103914675B - A kind of 2 D code of quick response matrix recognition methods of clothing - Google Patents
A kind of 2 D code of quick response matrix recognition methods of clothing Download PDFInfo
- Publication number
- CN103914675B CN103914675B CN201410097740.1A CN201410097740A CN103914675B CN 103914675 B CN103914675 B CN 103914675B CN 201410097740 A CN201410097740 A CN 201410097740A CN 103914675 B CN103914675 B CN 103914675B
- Authority
- CN
- China
- Prior art keywords
- code
- image
- quick response
- response matrix
- point
- 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)
Abstract
The invention discloses the 2 D code of quick response matrix recognition methods of a kind of clothing, comprise the steps: that a. gathers the image in 2 D code of patch on clothing raw material；B. image is carried out gray processing, Fast Median Filtering, binary conversion treatment；D. binary image is carried out edge extracting；E. two pieces of side areas of edge extracting QR code are isolated；F. hough transform detection side dotted line is used to obtain QR code deviation angle；G. bilinear interpolation rotates binary image, preliminary corrections；H. QR code is positioned, determine the order of three view finding figures, by the method for image array transposition upset, QR code is adjusted to correct orientation；I. it is decoded according to country's Quick Response Code standard.The present invention uses the projection method that is combined with hough transform of cutting can quickly, accurately obtain QR code deviation angle, while the noise in image of collection can be overcome to pollute, the problem of uneven illumination, be converted into image in 2 D code easy to identify.
Description
Technical field
The invention discloses the 2 D code of quick response matrix recognition methods of a kind of clothing, particularly relate to a kind of clothing
The two-dimensional code identification method of 2 D code of quick response matrix offset correction.
Background technology
Modern wears the clothes and stresses fashion, individual character, and clothing industry particularly fashionable dress producer also to walk on a large scale, the product of small lot
Product production line, the discrepancy cargo type class in warehouse and shipment frequency also can continue to increase, the work if clothing enter, put in storage and make an inventory
Still follow traditional manually make out an invoice, mode of manually checking over goods, its efficiency and accuracy all can not be guaranteed, and Finance Department may
Receiving the document of mistake, administration section can not timely and accurately obtain required inventory data.
Barcode technology is one of current most convenient, practical data input means, has wide development in China and answers
Use prospect.But, traditional bar code can only identify " article ", and can not describe " article " so that bar code is in every profession and trade
Use be greatly limited.And the appearance of two-dimensional bar code well solves the problem that one-dimension code cann't be solved.In state
Outward, it just has the title of " move database " (Portable data file).It has that information density is high, information capacity big,
The advantage that capacity of resisting disturbance is strong, error correcting capability is strong.And these advantages can solve due to one-dimension code finite capacity well, and not
The problem that can describe product.
QR Code(2 D code of quick response matrix, is called for short QR code) it is a kind of common Quick Response Code, it can only pass through image
Know and distinguish otherwise, and image recognition mode is easily affected by outside factor, as light is uneven, the most secretly, CCD can be caused to take the photograph
As the image of head shooting has noise.Additionally, due to when obtaining QR Code view data, may because of placement location not
Equally, obtain QR Code image to offset, it is impossible to location QR Code view finding graph position smoothly, directly set up sampling grid.
So it is unavoidable problem in many application that solution sound pollution, uneven illumination, QR Code offset.
Summary of the invention
In view of the above problems, the technical problem to be solved is to provide the quick response matrix two dimension of a kind of clothing
The two-dimensional code identification method of code offset correction.
For solving above-mentioned technical problem, the present invention is achieved by the following technical solutions:
The 2 D code of quick response matrix recognition methods of a kind of clothing of the present invention, comprises the following steps:
A) CCD camera gathers the image in 2 D code of quick response matrix of patch on clothing raw material；
B) coloured image of the 2 D code of quick response matrix to be identified gathered is converted to 256 color shade images, and
The background noise point of above-mentioned gray level image is filtered off with Fast median filtering algorithm；
C) using maximum variance between clusters to find optimal segmentation threshold, the principle of this algorithm is, when separate by threshold value
When the weighted sum of two pixel group variances of foreground and background minimizes, this threshold value is optimal segmentation threshold；
D) according to optimal segmentation threshold, by the gray value of each pixel of above-mentioned gray level image compared with threshold value,
If more than threshold value, then assignment 1, if less than threshold value, then assignment 0, obtaining binary image；
E) use Sobel edge edge detective operators that binary image is done edge extracting, two phases with different gray value
Always there is gray-scale edges between neighbouring region, gray-scale edges is the discontinuous result of gray value, and this discontinuous generally may utilize is asked
Derivative is conveniently detected；
E1) Sobel edge edge detective operators is a kind of gradient operator, and it uses two templates, and one is level, another
Being vertical, be both 3*3 masterplate, each template approaches a partial derivative, respectively with horizontal shuttering and vertical formwork pair
Image carries out convolution, obtains two matrixes, and they represent the image two partial derivatives at same position respectively；
E2) using the maximum of two numbers of the two matrix correspondence position as the output giving point, image in 2 D code is obtained
Edge；
F) the adaptive two pieces of sides isolating 2 D code of quick response matrix based on histogram projection cutting-out method are used
Region；
F1) image after edge extracting is carried out the horizontal direction i.e. projection of X-direction, gray value on statistics X=i straight line
It is the number of the pixel of 255, is stored in array xProjection [i].Start to scan from left to right from i=0, if condition
((xProjection [i+1] > 0) && (xProjection [i+2] > 1) is true to xProjection [i]==0) &&, then it is assumed that look for
To the abscissa on a summit of 2 D code of quick response matrix, defining this point is A point, obtains xA=i+1.The most from top to bottom
Scanned straight lines X=xA, finding first gray value is the pixel of 255, the vertical coordinate of this pointFor the vertical coordinate of A, i.e.The coordinate obtaining A point isIf condition (xProjection [i] > 1) && (xProjection [i+1] >
0) && (xProjection [i+2]==0) it is true, then it is assumed that find the horizontal seat on another summit of 2 D code of quick response matrix
Mark, defining this point is C point, obtains xC=i+1.Scanned straight lines X=x the most from the bottom upC, finding first gray value is the picture of 255
Vegetarian refreshments, the vertical coordinate of this pointFor the vertical coordinate of C, i.e.Obtaining C point coordinates is
F2) image after edge extracting is carried out the vertical square of projection to i.e. Y direction, gray scale on statistics Y=j straight line
Value is the number of the pixel of 255, is stored in array yProjection [j], then starts to scan from left to right from j=0, if bar
((yProjection [j+1] > 0) && (yProjection [j+2] > 1) is true to yProjection [j]==0) && to part, then think
Finding the vertical coordinate on the 3rd summit of 2 D code of quick response matrix, defining this point is D point, obtains yD=j+1, then from the right side
Turn left scanned straight lines Y=yD, finding first gray value is the pixel of 255, the abscissa of this pointFor the abscissa of D, i.e.The coordinate obtaining D point isIf condition (yProjection [j] > 1) && (yProjection [j+1] >
0) && (yProjection [j+2]==0) it is true, then it is assumed that find the vertical seat on the 4th summit of 2 D code of quick response matrix
Mark, defining this point is B point, obtains yB=j+1, the most from left to right scanned straight lines Y=yB, finding first gray value is the picture of 255
Vegetarian refreshments, the abscissa of this pointFor the abscissa of B, i.e.Obtaining B point coordinates is
F3) analyze D point and B point coordinates position, set an offset direction decision criteria: work as xD≥xBTime judge quickly to ring
Answering matrix two-dimensional code image is to left avertence, works as xD<xBTime judge QR code image be to right avertence；
F4) according to the 2 D code of quick response matrix offset direction of above-mentioned decision criteria gained, cutting method is automatically selected:
When judging 2 D code of quick response matrix to left avertence, it is embodied in the abscissa x of D pointDAbscissa x more than or equal to B pointB,
Straight line X=xBRight side area and straight line Y=yAThe intersection area in region of lower section crop；
When judging 2 D code of quick response matrix to right avertence, it is embodied in the abscissa x of D pointAAbscissa than B point
xBLittle, straight line X=xBLeft field and Y=yAThe intersection area of upper area crop, isolate quick response matrix two
Two pieces of side areas of dimension code；
The regulation of the cutting method of above-mentioned offset direction decision criteria and corresponding offset direction is to farthest cut out
Cut 2 D code of quick response matrix interior zone, obtain long and narrow two piece side area, reduce hough transform detection side line
Time amount of calculation, it also avoid the interference to detection side line of the Quick Response Code internal line segment simultaneously；
G) hough transform algorithm detection side dotted line is used to obtain 2 D code of quick response matrix deviation angle；
G1) initializing a polar coordinate ρ, the array in θ space, the number that quantifies on ρ direction is image diagonal direction
Pixel countWherein N1And N2Being respectively width and the height of image, the quantization number on θ direction is 1800, and angle is from 0 °
～180 °, every lattice 0.1 °；
G2) setting up an accumulator A (ρ, θ), juxtaposition finite element is 0；
G3) pixel (x that each gray value is 255 on sequential search imagei,yi) set up equation ρ=xicos(θ)+
yiSin (θ), calculates ρ value to each centrifugal pump of θ parameter, obtains this pixel (xi,yi) sine on polar coordinate
Curve, if curve is in too small lattice (ρ, θ), this accumulator A (ρ, θ) adds 1；
G4) finding out accumulator A (ρ, θ) local maximum above, this value is the conllinear parameter of image collinear points, it
Polar theta is the deviation angle of side dotted line；
H) bilinear interpolation is used to rotate image, preliminary corrections image in 2 D code of quick response matrix；
H1) find image after above-mentioned binaryzation central point (a, b), determine postrotational centre coordinate for (c, d), according to
Reverse rotation transformation matrix expression formula: Image is entered
Row rotates, point (x0,y0) rotate θ angle become point (x1,y1), wherein (x0,y0) it is not rotate any point in front image,
(x1,y1) it is by point (x0,y0) so that (a is b) that center of rotation rotates the point that obtains of θ angle；
H2) image being carried out bilinear interpolation, image is by (x after rotation transformation1,y1(x corresponding to)0,y0) coordinate
It is not necessarily integer, finds (x0,y0) four, vicinity pixel (x, y), (x+1, y), (x, y+1), (x+1, y+1), according to double
Linear interpolation formula, in both the horizontal and vertical directions to its interpolation, formula is:
I) 2 D code of quick response matrix is positioned, determine the order of three view finding figures, propose one image array
The method of transposition upset is adjusted to correct orientation 2 D code of quick response matrix, it is achieved the school of 2 D code of quick response matrix
Just；
I1) first horizontal sweep image in 2 D code of quick response matrix the most from left to right, records and meets (1:1:3:1:1) view finding
The line segment of graphics proportion feature, permits the deviation having 0.5, it is most important that record this line segment to the width of each pixel
Starting point Pr and terminal Pl, the abscissa of the centre coordinate of view finding figure isThen along straight lineFrom vertical scan direction, record the such line segment meeting 1:1:3:1:1, obtain in view finding figure
The vertical coordinate of heart coordinate；Record according to this mode that image is carried out horizontal and vertical directions scanning and meet 1:1:3:
Such line segment of 1:1, just can obtain the center point coordinate of three view finding figures successively；
I2) central point of three view finding figures is coupled together, constitute an isosceles right triangle, it is stipulated that seeking of left side
That view finding figure that image pattern is on isosceles right triangle right angle；A left side is determined by calculating the length of side and triangular angles
Side view finding figure position in the picture；
I3) 2 D code of quick response matrix is adjusted to correct orientation, with suitable by the method for image array transposition upset
Clockwise is as the criterion, if view finding figure in left side is in the image upper left corner, it is determined that 2 D code of quick response matrix is correct orientation, nothing
Need correction；If view finding figure in left side is in the image upper right corner, it is determined that 2 D code of quick response matrix offset by 90 degree, passes through image moment
Battle array transposition (i, j) → (j, i) exchange with up-downgoing (i, j) → (n+1-i, j), correction 2 D code of quick response matrix is to correct side
Position；If view finding figure in left side is in the image lower right corner, it is determined that 2 D code of quick response matrix offset by 180 degree, by (i, j) → (m
+ 1-j, n+1-i) matrix element transposed transform, correction 2 D code of quick response matrix is to correct orientation；If left side view finding is at image
The lower left corner, it is determined that 2 D code of quick response matrix offset by 270 degree, by image array transposition (i, j) → (j, i) and left and right arrange
Exchange (i, j) → (i, m+1-j), correction 2 D code of quick response matrix to correct orientation；
J) 2 D code of quick response matrix to correction to correct orientation, is carried out according to standard GB/T/T18284-2000
Decoding, it is achieved the 2 D code of quick response matrix identification of clothing.
The 2 D code of quick response matrix recognition methods of a kind of clothing as above, comprises clothes in described image in 2 D code
The dress title of raw material, the place of production, coroner, specification, model and unit price information.
The 2 D code of quick response matrix recognition methods of a kind of clothing as above, senses in described CCD camera
Device pixel is 300,000, and its coloured image highest resolution gathered is 480*640；
The 2 D code of quick response matrix recognition methods of a kind of clothing as above, described coloured image is converted to
The formula of 256 color shade images is:
Gray scale=0.3* red component+0.59* green component+0.11* blue component.
The 2 D code of quick response matrix recognition methods of a kind of clothing as above, described Fast Median Filtering filters off
The noise spot of image background refers to, selects a current pixel as central point, chooses a 3*3 size of point centered by it
Matrix sampling window, first carries out ascending sort, more often row is carried out ascending sort, take cornerwise three afterwards each column in window
Number, is ranked up, and takes intermediate value, and this value moves up and down in the picture as current output valve, window, to view picture figure
As filtering.
Present invention also offers the 2 D code of quick response matrix identification device of a kind of clothing, described identification device includes
Equipped with PC and the CCD camera of video frequency collection card, CCD camera connects PC by USB data line, and CCD takes the photograph
After the image in 2 D code of quick response matrix of head acquisition clothing, pass to PC by USB data line, PC takes
2 D code information is decoded by the 2 D code of quick response matrix identification software of dress, and information decoding obtained is stored in data base
In；The 2 D code of quick response matrix identification software of above-mentioned clothing include image in 2 D code acquisition module, image correction module,
Quick Response Code decoder module and raw material information import DBM；
Image in 2 D code acquisition module is used for obtaining CCD camera by video flowing in the incoming PC of USB data line
The image in 2 D code of quick response matrix of frame clothing；
Image correction module is for being converted to 256 by the Quick Response Code coloured image that above-mentioned image in 2 D code acquisition module obtains
Color shade image, and with Fast median filtering algorithm picture noise point, use maximum variance between clusters to find optimal segmenting threshold,
Image binaryzation, uses Sobel edge edge detective operators that the image after binaryzation is done edge extracting, utilizes adaptive based on directly
Side's figure projection cutting-out method isolates two pieces of side areas of 2 D code of quick response matrix, uses hough transform detection side dotted line
Obtain 2 D code of quick response matrix deviation angle, use bilinear interpolation to rotate image, preliminary corrections quick response matrix
Image in 2 D code, positions 2 D code of quick response matrix afterwards, determines the order of three view finding figures, use image array transposition
The method of upset is adjusted to correct orientation 2 D code of quick response matrix, it is achieved the correction of 2 D code of quick response matrix；
Quick Response Code decoder module for according to standard GB/T/T18284-2000 to quickly responding square after above-mentioned correction
Battle array image in 2 D code is decoded, and obtains title, the place of production, coroner, specification, model and list in the Quick Response Code pasted on clothing raw material
The information such as valency；
Raw material information imports DBM for by the 2 D code information automated import of data storehouse on above-mentioned clothing raw material
In, it is achieved information-based and paperless management.
Beneficial effect
The present invention is adaptive based on histogram projection cutting-out method, isolate QR Code exactly two pieces by using
Side area, then obtains deviation angle, preliminary corrections QR Code image with hough transform, positions QR Code afterwards, really
The order of fixed three view finding figures, is adjusted to correct orientation by the method for image array transposition upset QR Code, it is achieved QR
The correction of Code, the purpose being correctly decoded.Use Quick Response Code automatic identification technology, it is achieved clothing raw material information is led automatically simultaneously
Enter in stock's Management System Data storehouse.Most important or Finance Department, administration section, can obtain required quickly and accurately
Inventory data.Finance Department can error correction in time, corresponding adjustment can be made according to data by administration section, it is achieved automatization manages
Reason.
Accompanying drawing explanation
Fig. 1 is present system structure chart；
Binaryzation schematic diagram when Fig. 2 is QR Code left shift of the present invention less than 90 °；
Fig. 3 is the QR Code of the present invention binaryzation schematic diagram when offseting to the right more than 90 °；
Fig. 4 is the QR Code of the present invention binaryzation schematic diagram when offseting to the right 180 °；
Fig. 5 is vertical-horizontal perspective view after QR Code edge extracting in Fig. 2；
Fig. 6 is vertical-horizontal perspective view after QR Code edge extracting in Fig. 3；
Fig. 7 is vertical-horizontal perspective view after QR Code edge extracting in Fig. 4；
Two pieces of side areas when Fig. 8 is QR Code left shift of the present invention less than 90 ° obtain schematic diagrams；
Fig. 9 is that the QR Code of the present invention two pieces of side areas when offseting to the right more than 90 ° obtain schematic diagrams；
Figure 10 is that the QR Code of the present invention two pieces of side areas when offseting to the right 180 ° obtain schematic diagrams；
Figure 11 is the schematic diagram after the QR Code preliminary corrections offset in Fig. 2；
Figure 12 is the schematic diagram after the QR Code preliminary corrections offset in Fig. 3；
Figure 13 is the schematic diagram after the QR Code preliminary corrections offset in Fig. 4；
Figure 14 is the QR Code image that offsets in Fig. 2 schematic diagram after cumulative correction after matrix transpose overturns；
Figure 15 is the QR Code image that offsets in Fig. 3 schematic diagram after cumulative correction after matrix transpose overturns；
Figure 16 is the QR Code image that offsets in Fig. 4 schematic diagram after cumulative correction after matrix transpose overturns.
Detailed description of the invention
Below in conjunction with detailed description of the invention, the present invention is expanded on further.Should be understood that these embodiments are merely to illustrate this
Bright rather than limit the scope of the present invention.In addition, it is to be understood that after having read the content that the present invention lectures, art technology
The present invention can be made various changes or modifications by personnel, and these equivalent form of values fall within the application appended claims equally and limited
Fixed scope.
Refer to system shown in Figure 1 structure chart, including: image in 2 D code acquisition module S101, image correction module S102,
Quick Response Code decoder module S103, decoded information import DBM S104；Image in 2 D code acquisition module S101 is taken the photograph by CCD
As the image in 2 D code of patch on head shooting clothing raw material, it is transferred to PC by USB data line, obtains by the C# program write
The image in 2 D code of quick response matrix of one frame clothing of video flowing；Image correction module S102 is to image in 2 D code acquisition module
The image in 2 D code that S101 is collected carries out gray processing, and Fast Median Filtering removes picture noise point, uses maximum between-cluster variance
Find optimal segmentation threshold values, it is achieved image binaryzation, use Sobel edge edge detective operators that the image after binaryzation is done edge and carry
Take, and with the adaptive two pieces of side areas isolating QR Code based on histogram projection cutting-out method, use hough transform inspection
Survey side dotted line and obtain QR Code deviation angle, use bilinear interpolation to rotate image, preliminary corrections QR Code image, it
Afterwards QR Code is positioned, determine the order of three view finding figures, by the method for image array transposition upset, QR Code is adjusted
To correct orientation, it is achieved the correction of QR Code；Quick Response Code decoder module S103 is for obtaining image correction module S102
QR Code image, according to standard GB/T/T18284-2000, decoding obtains the wherein title of clothing raw material, the place of production, inspection
The information such as member, specification, model, unit price；Raw material information imports DBM S104 for the information in S103 being automatically imported
In clothing inventory management system data base, it is achieved automated management.
Embodiment 1
Introduce one embodiment of the present of invention below in conjunction with accompanying drawing 2,5,8,11,14, collect on clothing raw material to
Left avertence is moved the image in 2 D code of quick response matrix less than 90 ° and is corrected to correct identification direction.
The coloured image gathered is turned according to gray scale=0.3* red component+0.59* green component+0.11* blue component
It is changed to 256 color shade images, and finds optimal partition threshold T=82 with maximum between-cluster variance, each by above-mentioned gray level image
The gray value of individual pixel is compared with threshold value, if more than threshold value, then assignment 255, if less than threshold value, then assignment 0, changing into two-value
Change image, as shown in Figure 2.
Again binary image Sobel edge edge detective operators is done edge extracting, hang down with 3*3 horizontal shuttering and 3*3 respectively
Straight template carries out convolution to image, obtains two matrixes, and they represent the image two partial derivatives at same position respectively,
The maximum of two numbers of the two matrix correspondence position is as the output giving point, it is achieved the edge extracting of binary image, as
Shown in Fig. 5.
Image after edge extracting carries out the horizontal direction i.e. projection of X-direction again, and on statistics X=i straight line, gray value is
The number of the pixel of 255, is stored in array xProjection [i].Start to scan from left to right from i=0, if condition
((xProjection [i+1] > 0) && (xProjection [i+2] > 1) is true to xProjection [i]==0) &&, then it is assumed that look for
To the abscissa on a summit of 2 D code of quick response matrix, defining this point is A point, obtains xA=i+1.The most from top to bottom
Scanned straight lines X=xA, finding first gray value is the pixel of 255, the vertical coordinate of this pointFor the vertical coordinate of A, i.e.The coordinate obtaining A point is.If condition (xProjection [i] > 1) && (xProjection [i+1] >
0) && (xProjection [i+2]==0) it is true, then it is assumed that find the horizontal seat on another summit of 2 D code of quick response matrix
Mark, defining this point is C point, obtains xC=i+1.Scanned straight lines X=x the most from the bottom upC, finding first gray value is the picture of 255
Vegetarian refreshments, the vertical coordinate of this pointFor the vertical coordinate of C, i.e.Obtaining C point coordinates isThen, edge is carried
Image after taking carries out the vertical square of projection to i.e. Y direction, and on statistics Y=j straight line, gray value is the individual of the pixel of 255
Number, is stored in array yProjection [j], then starts to scan from left to right from j=0, if condition (yProjection [j]==
(yProjection [j+1] > 0) && (yProjection [j+2] > 1) is true to 0) &&, then think and find quick response matrix two dimension
The vertical coordinate on the 3rd summit of code, defining this point is D point, obtains yD=j+1, then turn left scanned straight lines Y=y from the right sideD, find
First gray value is the pixel of 255, the abscissa of this pointFor the abscissa of D, i.e.Obtain the coordinate of D point
ForIf condition (yProjection [j] > 1) && (yProjection [j+1] > 0) && (yProjection [j+2]=
=0) being true, then it is assumed that find the vertical coordinate on the 4th summit of 2 D code of quick response matrix, defining this point is B point, obtains yB
=j+1, the most from left to right scanned straight lines Y=yB, finding first gray value is the pixel of 255, the abscissa of this pointFor B
Abscissa, i.e.Obtaining B point coordinates is.Thus determine 2 D code of quick response matrix four top
Point A, B, C, D coordinate position (x in the pictureA,yA), (xB,yB), (xC,yC), (xD,yD), and available xD>xB, such as Fig. 5 institute
Show.
Further according to the offset direction decision criteria of regulation: x in technical schemeD≥xBTime image in 2 D code of quick response matrix
For left avertence, xD<xBSpeed per hour response matrix image in 2 D code is right avertence.Judge that this image in 2 D code, as left avertence, and selects left avertence cutting
Mode:
Straight line X=xBRight side area and straight line Y=yAThe intersection area in region of lower section crop, as shown in Figure 8.
The regulation of the cutting method of above-mentioned offset direction decision criteria and corresponding offset direction is to farthest crop quickly
Response matrix Quick Response Code interior zone, obtains long and narrow two piece side area, reduces calculating during hough transform detection side line
Amount it also avoid the interference to detection side line of the Quick Response Code internal line segment simultaneously.
Use hough transform algorithm to detect the side dotted line in two pieces of long and narrow side areas again, first initialize a pole
The array in coordinate ρ, θ space, the pixel count that quantization number is image diagonal direction on ρ directionWherein N1With
N2Being respectively the width of image and height, the quantization number on θ direction is 1800, angle from 0 °～180 °, every lattice 0.1 °.Then, build
A vertical accumulator A (ρ, θ), juxtaposition finite element is 0.The pixel that each gray value is 255 on sequential search image
(xi,yi) set up equation ρ=xicos(θ)+yiSin (θ), calculates ρ value to each centrifugal pump of θ parameter, obtains this pixel
(xi,yi) sine curve on polar coordinate, if curve is in too small lattice (ρ, θ), this accumulator A (ρ, θ) adds 1.?
After, find out the maximum on accumulator A (ρ, θ), this value is the conllinear parameter of image side dotted line, obtains θ=7.5°I.e. two dimension
The deviation angle of code is 7.5 °.
Then, finding the center point P of above-mentioned binary image, the coordinate of P isAs shown in Figure 2.With P point it is
Center of rotation, according to the reverse rotation transformation matrix expression formula in technical scheme and bilinear interpolation formula, uses bilinear interpolation
Turn clockwise image, preliminary corrections image in 2 D code of quick response matrix, as shown in figure 11.
Then, horizontal sweep preliminary corrections image in 2 D code of quick response matrix from left to right, record and meet (1:1:3:
1:1) the line segment of view finding graphics proportion feature, permits the deviation having 0.5, it is most important that record to the width of each pixel
Starting point P of this line segmentrWith terminal Pl, the abscissa of the centre coordinate of view finding figure is, further along straight lineFrom vertical scan direction, record the such line segment meeting 1:1:3:1:1, obtain view finding figure
The vertical coordinate of centre coordinate；Record according to this mode that image is carried out horizontal and vertical directions scanning and meet 1:1:
Such line segment of 3:1:1, obtains the central point of three view finding figures successively, such as the P1 in Figure 11, P2, P3.Three view findings
The center point P 1, P2 of figure, P3 couples together, and constitutes an isosceles right triangle, it is stipulated that the view finding figure in left side is in
That view finding figure on isosceles right triangle right angle.By calculating length of side L12, the ratio between L23, L13 two determines
View finding figure position in the picture, left side, that point that ratio is approximately folded by two limits of 1 is left side view finding figure
Central point, P1 is view finding centre of figure position in the picture, left side, as shown in figure 11.
Finally, by the method for image array transposition upset, 2 D code of quick response matrix is adjusted to correct orientation, with
Being as the criterion clockwise, view finding centre of figure point P1 in left side is in the image upper left corner, it is determined that the quick response matrix after preliminary corrections
Quick Response Code has been correct orientation, it is not necessary to again correct, as shown in figure 14.Subsequently can be by standard GB/T/T18284-2000
2 D code of quick response matrix is decoded thus realizes the 2 D code of quick response matrix identification of patch on clothing raw material.
Embodiment 2
Introduce one embodiment of the present of invention below in conjunction with accompanying drawing 3,6,9,12,15, collect on clothing raw material to
Right avertence is moved the image in 2 D code of quick response matrix more than 90 ° and is corrected to correct identification direction, the method that this embodiment uses
It is identical with embodiment 1.
According to the method identical with embodiment 1, the coloured image gathered is changed into binary image, as shown in Figure 3.
Use Sobel edge edge detective operators that binary image is done edge extracting, the image after edge extracting is carried out horizontal direction i.e. X
Axial projection and the vertical square of projection to i.e. Y direction, scan from four direction up and down the most respectively, determines fast
Speed response matrix Quick Response Code four summits A, B, C, D coordinate position (x in the pictureA,yA), (xB,yB), (xC,yC), (xD,
yD), and available xD<xB, as shown in Figure 6.According to the offset direction decision criteria of regulation: x in technical schemeD≥xBTime quickly ring
Answering matrix two-dimensional code image is left avertence, xD<xBSpeed per hour response matrix image in 2 D code is right avertence.Judge that this image in 2 D code is as the right side
Partially, and select right avertence cutting method: straight line X=xBLeft field and straight line Y=yAThe intersection area cutting in region of top
Fall, as shown in Figure 9.Employing hough transform algorithm detection side dotted line obtains 2 D code of quick response matrix deviation angle and is
79.7°.Finding the center point P of above-mentioned binary image, the coordinate of P isAs shown in Figure 3.In with P point for rotation
The heart, turns clockwise image with bilinear interpolation, preliminary corrections image in 2 D code of quick response matrix, as shown in figure 12.Root
According to Quick Response Code GB decoding algorithm, utilize horizontal and vertical scanning, 2 D code of quick response matrix is positioned, determines three view findings
The central point of figure, such as the P1 in Figure 12, P2, P3.The center point P 1, P2 of three view finding figures, P3 couples together, and constitutes one
Individual isosceles right triangle, it is stipulated that view finding figure that the view finding figure in left side is on isosceles right triangle right angle.
By calculating length of side L12, the ratio between L23, L13 two determines view finding figure position in the picture, left side, and ratio approximates
Be 1 two limits folded by that point be left side view finding figure central point, P1 be left side view finding centre of figure in the picture
Position, as shown in figure 12.Finally, by the method for image array transposition upset, 2 D code of quick response matrix is adjusted to correctly
Orientation, be as the criterion in a clockwise direction, view finding centre of figure point P1 in left side is in the image lower right corner, it is determined that quick after preliminary corrections
Response matrix Quick Response Code offset by 180 degree, by (i, j) → (m+1-j, n+1-i) matrix element transposed transform, correction quickly rings
Answer matrix two-dimensional code to correct orientation, as shown in figure 15.Subsequently can be by standard GB/T/T18284-2000 to quickly response
Matrix two-dimensional code is decoded thus realizes the 2 D code of quick response matrix identification of patch on clothing raw material.
Embodiment 3
Introduce one embodiment of the present of invention below in conjunction with accompanying drawing 4,7,10,13,16, collect on clothing raw material to
Right avertence is moved the image in 2 D code of quick response matrix of 180 ° and is corrected to correct identification direction, the method for this embodiment use and reality
It is identical for executing example 1.
According to the method identical with embodiment 1, the coloured image gathered is changed into binary image, as shown in Figure 4,
Use Sobel edge edge detective operators that binary image is done edge extracting.Image after edge extracting is carried out horizontal direction i.e. X
Axial projection and the vertical square of projection to i.e. Y direction, scan from four direction up and down the most respectively, determines fast
Speed response matrix Quick Response Code four summits A, B, C, D coordinate position (x in the pictureA,yA), (xB,yB), (xC,yC), (xD,
yD), and available xD>xB, as shown in Figure 7.According to the offset direction decision criteria of regulation: x in technical schemeD≥xBTime quickly ring
Answering matrix two-dimensional code image is left avertence, xD<xBSpeed per hour response matrix image in 2 D code is right avertence.Judge that this image in 2 D code is as a left side
Partially, and select left avertence cutting method: straight line X=xBRight side area and straight line Y=yAThe intersection area cutting in region of lower section
Fall, as shown in Figure 10.Then, hough transform algorithm detection side dotted line is used to obtain 2 D code of quick response matrix deviation angle
It it is 90.1 °.Finding the center point P of above-mentioned binary image, the coordinate of P isAs shown in Figure 4.In with P point for rotation
The heart, turns clockwise image with bilinear interpolation, preliminary corrections image in 2 D code of quick response matrix, as shown in figure 13.Root
According to Quick Response Code GB decoding algorithm, utilize horizontal and vertical scanning, 2 D code of quick response matrix is positioned, determines three view findings
The central point of figure, such as the P1 in Figure 13, P2, P3.The center point P 1, P2 of three view finding figures, P3 couples together, and constitutes one
Individual isosceles right triangle, it is stipulated that view finding figure that the view finding figure in left side is on isosceles right triangle right angle.
By calculating length of side L12, the ratio between L23, L13 two determines view finding figure position in the picture, left side, and ratio approximates
Be 1 two limits folded by that point be left side view finding figure central point, P1 be left side view finding centre of figure in the picture
Position, as shown in figure 13.Finally, by the method for image array transposition upset, 2 D code of quick response matrix is adjusted to correctly
Orientation, be as the criterion in a clockwise direction, view finding centre of figure point P1 in left side is in the image lower left corner, it is determined that quick after preliminary corrections
Response matrix Quick Response Code offset by 270 degree, by image array transposition (i, j) → (j, i) and left and right row exchange (i, j) → (i, m
+ 1-j), correction 2 D code of quick response matrix is to correct orientation, as shown in figure 16.Subsequently can by standard GB/T/
2 D code of quick response matrix is decoded thus realizes the quick response matrix two dimension of patch on clothing raw material by T18284-2000
Code identifies.
Claims (6)
1. a 2 D code of quick response matrix recognition methods for clothing, is characterized in that comprising the following steps:
A) CCD camera gathers the image in 2 D code of quick response matrix of patch on clothing raw material；
B) coloured image of the 2 D code of quick response matrix to be identified gathered is converted to 256 color shade images, and with fast
Speed median filtering method filters off the background noise point of above-mentioned gray level image；
C) using maximum variance between clusters to find optimal segmentation threshold, principle is, when by the separate foreground and background of threshold value two
When the weighted sum of individual pixel group variance minimizes, this threshold value is optimal segmentation threshold；
D) according to optimal segmentation threshold, by the gray value of each pixel of above-mentioned gray level image compared with threshold value, if greatly
In threshold value, then assignment 1, if less than threshold value, then assignment 0, obtaining binary image；
E) use Sobel edge edge detective operators that binary image is done edge extracting, two adjacent regions with different gray value
Always there is gray-scale edges between territory, gray-scale edges is the discontinuous result of gray value, this discontinuous generally utilization side of differentiating
Just detect；
E1) Sobel edge edge detective operators is a kind of gradient operator, and it uses two templates, and one is level, and another is vertical
Straight, it is both 3*3 masterplate, each template approaches a partial derivative, uses horizontal shuttering and vertical formwork to image respectively
Carrying out convolution, obtain two matrixes, they represent the image two partial derivatives at same position respectively；
E2) using the maximum of two numbers of the two matrix correspondence position as output, the edge of image in 2 D code is obtained；
F) the adaptive two pieces of side areas isolating 2 D code of quick response matrix based on histogram projection cutting-out method are used；
F1) image after edge extracting carrying out the horizontal direction i.e. projection of X-direction, on statistics X=i straight line, gray value is
The number of the pixel of 255, is stored in array xProjection [i]；Start to scan from left to right from i=0, if condition
(xProjection [i]==0) && (xProjection [i+1] > 0) && (xProjection [i+2] > 1) is true, then it is assumed that
Finding the abscissa on a summit of 2 D code of quick response matrix, defining this point is A point, obtains xA=i+1；Then from upper past
Lower scanned straight lines X=xA, finding first gray value is the pixel of 255, the vertical coordinate of this pointFor the vertical coordinate of A, i.e.The coordinate obtaining A point isIf condition (xProjection [i] > 1) && (xProjection [i+1] >
0) && (xProjection [i+2]==0) it is true, then it is assumed that find the horizontal stroke on another summit of 2 D code of quick response matrix
Coordinate, defining this point is C point, obtains xC=i+1；Scanned straight lines X=x the most from the bottom upC, finding first gray value is 255
Pixel, the vertical coordinate of this pointFor the vertical coordinate of C, i.e.Obtaining C point coordinates is
F2) image after edge extracting carrying out the vertical direction i.e. projection of Y direction, on statistics Y=j straight line, gray value is
The number of the pixel of 255, is stored in array yProjection [j], then starts to scan from left to right from j=0, if condition
((yProjection [j+1] > 0) && (yProjection [j+2] > 1) is true to yProjection [j]==0) &&, then think
Finding the vertical coordinate on the 3rd summit of 2 D code of quick response matrix, defining this point is D point, obtains yD=j+1, then from the right side
Turn left scanned straight lines Y=yD, finding first gray value is the pixel of 255, the abscissa of this pointFor the abscissa of D, i.e.The coordinate obtaining D point isIf condition (yProjection [j] > 1) && (yProjection [j+1] >
0) && (yProjection [j+2]==0) is true, then it is assumed that find the 4th summit vertical of 2 D code of quick response matrix
Coordinate, defining this point is B point, obtains yB=j+1, the most from left to right scanned straight lines Y=yB, finding first gray value is 255
Pixel, the abscissa of this pointFor the abscissa of B, i.e.Obtaining B point coordinates is
F3) analyze D point and B point coordinates position, set an offset direction decision criteria: work as xD≥xBTime judge quickly respond square
Battle array image in 2 D code is to left avertence, works as xD＜ xBTime judge QR code image be to right avertence；
F4) according to the 2 D code of quick response matrix offset direction of above-mentioned decision criteria gained, cutting method is automatically selected:
When judging 2 D code of quick response matrix to left avertence, it is embodied in the abscissa x of D pointDHorizontal seat more than or equal to B point
Mark xB, straight line X=xBRight side area and straight line Y=yAThe intersection area in region of lower section crop；
When judging 2 D code of quick response matrix to right avertence, it is embodied in the abscissa x of D pointAAbscissa x than B pointBIt is little,
Straight line X=xBLeft field and Y=yAThe intersection area of upper area crop, isolate quick response matrix two dimension
Two pieces of side areas of code；
The regulation of the cutting method of above-mentioned offset direction decision criteria and corresponding offset direction is to farthest crop
2 D code of quick response matrix interior zone, obtains long and narrow two piece side area, during minimizing hough transform detection side line
Amount of calculation, it also avoid the interference to detection side line of the Quick Response Code internal line segment simultaneously；
G) hough transform algorithm detection side dotted line is used to obtain 2 D code of quick response matrix deviation angle；
G1) a polar coordinate ρ is initialized, the array in θ space, the pixel that quantization number is image diagonal direction on ρ direction
NumberWherein N1And N2Being respectively the width of image and height, the quantization number on θ direction is 1800, angle from 0 °～
180 °, every lattice 0.1 °；
G2) setting up an accumulator A (ρ, θ), juxtaposition finite element is 0；
G3) pixel (x that each gray value is 255 on sequential search imagei,yi) set up equation ρ=xi cos(θ)+yi
Sin (θ), calculates ρ value to each centrifugal pump of θ parameter, obtains this pixel (xi,yi) on polar coordinate one sinusoidal bent
Line, if curve is in too small lattice (ρ, θ), this accumulator A (ρ, θ) adds 1；
G4) finding out accumulator A (ρ, θ) local maximum above, this value is the conllinear parameter of image collinear points, and its pole is sat
Mark θ is the deviation angle of side dotted line；
H) bilinear interpolation is used to rotate image, preliminary corrections image in 2 D code of quick response matrix；
H1) (a b), determines that postrotational centre coordinate is for (c, d), according to reverse rotation to find the central point of above-mentioned binary image
Transformation matrix expression formula:Image is rotated,
Point (x0,y0) rotate θ angle become point (x1,y1), wherein (x0,y0) it is not rotate any point in front image, (x1,y1)
It is by point (x0,y0) so that (a is b) that center of rotation rotates the point that obtains of θ angle；
H2) image being carried out bilinear interpolation, image is by (x after rotation transformation1,y1(x corresponding to)0,y0) coordinate differ
Surely it is integer, finds (x0,y0) four, vicinity pixel (x, y), (x+1, y), (x, y+1), (x+1, y+1), according to bilinearity
Formula for interpolation, in both the horizontal and vertical directions to its interpolation, formula is:
Wherein α=x0-x, β=y0-
Y, obtains point (x0,y0) gray value f (x0,y0), this value is exactly point (x1,y1) gray value；
I) 2 D code of quick response matrix is positioned, determine the order of three view finding figures, propose one and use image array transposition
The method of upset is adjusted to correct orientation 2 D code of quick response matrix, it is achieved the correction of 2 D code of quick response matrix；
I1) first horizontal sweep image in 2 D code of quick response matrix the most from left to right, records and meets (1:1:3:1:1) view finding figure ratio
The line segment of example feature, permits the deviation having 0.5, it is most important that record starting point P of this line segment to the width of each pixelr
With terminal Pl, the abscissa of the centre coordinate of view finding figure isThen along straight line
From vertical scan direction, record the such line segment meeting 1:1:3:1:1, obtain the vertical seat of the centre coordinate of view finding figure
Mark；Record according to this mode that image is carried out horizontal and vertical directions scanning and meet the such of 1:1:3:1:1
Line segment, just can obtain the center point coordinate of three view finding figures successively；
I2) central point of three view finding figures is coupled together, constitute an isosceles right triangle, it is stipulated that the view finding figure in left side
That view finding figure that shape is on isosceles right triangle right angle；Determine that left side is sought by calculating the length of side and triangular angles
Image pattern position in the picture；
I3) 2 D code of quick response matrix is adjusted to correct orientation, with clockwise by the method for image array transposition upset
Direction is as the criterion, if view finding figure in left side is in the image upper left corner, it is determined that 2 D code of quick response matrix is correct orientation, it is not necessary to school
Just；If view finding figure in left side is in the image upper right corner, it is determined that 2 D code of quick response matrix offset by 90 degree, is turned by image array
Put (i, j) → (j, i) exchange with up-downgoing (i, j) → (n+1-i, j), correction 2 D code of quick response matrix is to correct orientation；
If view finding figure in left side is in the image lower right corner, it is determined that 2 D code of quick response matrix offset by 180 degree, by (i, j) → (m+1-
J, n+1-i) matrix element transposed transform, correction 2 D code of quick response matrix is to correct orientation；If left side view finding is left at image
Inferior horn, it is determined that 2 D code of quick response matrix offset by 270 degree, by image array transposition (i, j) → (j, i) and left and right row right
(i, j) → (i, m+1-j), correction 2 D code of quick response matrix is to correct orientation for tune；
J) 2 D code of quick response matrix to correction to correct orientation, solves according to standard GB/T/T 18284-2000
Code, it is achieved the 2 D code of quick response matrix identification of clothing.
The 2 D code of quick response matrix recognition methods of a kind of clothing the most according to claim 1, it is characterised in that described
Image in 2 D code comprises the title of clothing raw material, the place of production, coroner, specification, model and unit price information.
The 2 D code of quick response matrix recognition methods of a kind of clothing the most according to claim 1, it is characterised in that described
CCD camera in sensor pixel be 300,000, its gather coloured image highest resolution be 480*640.
The 2 D code of quick response matrix recognition methods of a kind of clothing the most according to claim 1, it is characterised in that described
Coloured image be converted to the formula of 256 color shade images and be:
Gray scale=0.3* red component+0.59* green component+0.11* blue component.
The 2 D code of quick response matrix recognition methods of a kind of clothing the most according to claim 1, it is characterised in that described
Fast Median Filtering filter off the noise spot of image background and refer to, select a current pixel as central point, choose during it is
The matrix sampling window of one 3*3 size of heart point, first carries out ascending sort, more often row is carried out ascending order row each column in window
Sequence, takes cornerwise three numbers afterwards, is ranked up, takes intermediate value, and this value is entered in the picture as current output valve, window
Row moves up and down, and filters entire image.
6. the 2 D code of quick response matrix identification of a kind of clothing of the recognition methods as according to any one of Claims 1 to 5
Device, is characterized in that: device includes the PC equipped with video frequency collection card and CCD camera, CCD camera in described identifying
Connect PC by USB data line, after the image in 2 D code of quick response matrix of CCD camera acquisition clothing, pass through USB
Data wire passes to PC, and in PC, 2 D code information is carried out by the 2 D code of quick response matrix identification software of clothing
Decoding, information decoding obtained is stored in data base；The 2 D code of quick response matrix identification software of above-mentioned clothing includes
Image in 2 D code acquisition module, image correction module, Quick Response Code decoder module and raw material information import DBM；
Image in 2 D code acquisition module is for obtaining CCD camera by the incoming PC of USB data line the one of video flowing
The image in 2 D code of quick response matrix of frame clothing；
Image correction module for being converted to 256 color ashes by the Quick Response Code coloured image that above-mentioned image in 2 D code acquisition module obtains
Degree image, and with Fast median filtering algorithm picture noise point, use maximum variance between clusters to find optimal segmenting threshold, image
Binaryzation, uses Sobel edge edge detective operators that the image after binaryzation is done edge extracting, utilizes adaptive based on rectangular histogram
Projection cutting-out method isolates two pieces of side areas of 2 D code of quick response matrix, uses hough transform detection side dotted line to obtain
2 D code of quick response matrix deviation angle, uses bilinear interpolation to rotate image, preliminary corrections quick response matrix two dimension
Code image, positions 2 D code of quick response matrix afterwards, determines the order of three view finding figures, overturns with image array transposition
Method 2 D code of quick response matrix is adjusted to correct orientation, it is achieved the correction of 2 D code of quick response matrix；
Quick Response Code decoder module is used for according to standard GB/T/T 18284-2000 the 2 D code of quick response matrix after correction
Image is decoded, and obtains title, the place of production, coroner, specification, model and unit price information in the Quick Response Code pasted on clothing raw material；
Raw material information imports DBM for by the 2 D code information automated import of data storehouse on clothing raw material.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410097740.1A CN103914675B (en) | 2014-03-17 | 2014-03-17 | A kind of 2 D code of quick response matrix recognition methods of clothing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410097740.1A CN103914675B (en) | 2014-03-17 | 2014-03-17 | A kind of 2 D code of quick response matrix recognition methods of clothing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103914675A CN103914675A (en) | 2014-07-09 |
| CN103914675B true CN103914675B (en) | 2016-11-16 |
Family
ID=51040345
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410097740.1A Expired - Fee Related CN103914675B (en) | 2014-03-17 | 2014-03-17 | A kind of 2 D code of quick response matrix recognition methods of clothing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103914675B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292212A (en) * | 2017-04-26 | 2017-10-24 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of Quick Response Code localization method under low signal-to-noise ratio environment |
Families Citing this family (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104112153B (en) * | 2014-07-17 | 2017-10-10 | 上海透云物联网科技有限公司 | Bar dimension code recognition methods and its system based on mobile terminal |
| CN104200187B (en) * | 2014-08-25 | 2017-06-16 | 北京慧眼智行科技有限公司 | A kind of quick exclusion QR code positions detect the method and system of figure error detection |
| CN104268498B (en) * | 2014-09-29 | 2017-09-19 | 杭州华为数字技术有限公司 | A kind of recognition methods of Quick Response Code and terminal |
| CN104363400A (en) * | 2014-10-29 | 2015-02-18 | 广东欧珀移动通信有限公司 | Method and device for scanning local two-dimensional codes |
| CN104484659B (en) * | 2014-12-30 | 2018-08-07 | 南京巨鲨显示科技有限公司 | A method of to Color medical and gray scale image automatic identification and calibration |
| CN104573606A (en) * | 2014-12-30 | 2015-04-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | QR two-dimensional code processing method and system |
| CN106156676B (en) * | 2015-04-15 | 2019-04-19 | 广州诚迈伟讯物联网技术有限公司 | A kind of image-type CCD bar code method for quickly identifying and device |
| CN105095822B (en) * | 2015-09-07 | 2018-07-06 | 福建联迪商用设备有限公司 | A kind of Chinese letter co pattern image detection method and system |
| CN105138945B (en) * | 2015-09-25 | 2017-12-08 | 三明学院 | A kind of rebar surface two-dimensional code identification method |
| CN105260694B (en) * | 2015-10-22 | 2017-12-01 | 佛山科学技术学院 | A kind of two-dimension code area localization method based on multistage key extraction with analysis |
| CN105405204B (en) * | 2015-11-04 | 2018-02-02 | 东方通信股份有限公司 | The paper money number recognition methods of cash inspecting machine |
| CN105389883B (en) * | 2015-11-04 | 2018-01-12 | 东方通信股份有限公司 | A kind of paper money number recognition methods of cash inspecting machine |
| CN105335744B (en) * | 2015-11-10 | 2018-09-21 | 佛山科学技术学院 | A kind of one-dimension code zone location extracting band distribution characteristics based on image backbone |
| CN105654019B (en) * | 2016-02-23 | 2018-09-04 | 海信集团有限公司 | Quick Response Code fast decoding method and device |
| CN107404721A (en) * | 2016-05-20 | 2017-11-28 | 阿里巴巴集团控股有限公司 | Internet of things equipment matches somebody with somebody network method, image-pickup method and equipment |
| CN106339653A (en) * | 2016-08-24 | 2017-01-18 | 乐视控股（北京）有限公司 | QR code scanning processing method and QR code scanning processing device |
| CN106447604B (en) * | 2016-09-30 | 2021-07-13 | 北京奇虎科技有限公司 | Method and device for transforming face picture in video |
| CN107096720B (en) * | 2017-06-22 | 2022-11-29 | 西安科技大学 | Express item bar code automatic identification system and method based on image processing |
| CN107545259A (en) * | 2017-08-31 | 2018-01-05 | 华南理工大学 | A kind of Quick Response Code reconstructing method based on Da-Jin algorithm |
| CN108055116A (en) * | 2017-10-25 | 2018-05-18 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Quick Response Code duplex communication method |
| CN108280384B (en) * | 2017-12-30 | 2019-03-15 | 广州市钛码电子科技有限公司 | A kind of two-dimensional code identification method |
| CN108345816A (en) * | 2018-01-29 | 2018-07-31 | 广州中大微电子有限公司 | A kind of Quick Response Code extracting method and system in the case where uneven illumination is even |
| CN108388825B (en) * | 2018-04-04 | 2021-06-08 | 厦门码灵半导体技术有限公司 | Fast response code searching method and device |
| CN108734164A (en) * | 2018-05-04 | 2018-11-02 | 北京物灵智能科技有限公司 | Card, identification card method, paint this reading machine people and storage device |
| CN108629789A (en) * | 2018-05-14 | 2018-10-09 | 华南理工大学 | A kind of well-marked target detection method based on VggNet |
| CN109359999B (en) * | 2018-08-21 | 2021-10-29 | 同济大学 | Encoding and decoding method for two-dimensional line segment graph of commodity outer package |
| CN111435531B (en) * | 2018-12-25 | 2023-08-08 | Tcl科技集团股份有限公司 | Image deflection angle detection method, intelligent device and storage medium |
| CN109711223A (en) * | 2018-12-28 | 2019-05-03 | 福州符号信息科技有限公司 | A kind of promotion QR code decoding rate method and apparatus |
| CN109934320A (en) * | 2019-02-27 | 2019-06-25 | 上海交通大学 | A kind of antifalsification label and its method for anti-counterfeit based on microsize particle |
| CN110032907A (en) * | 2019-04-15 | 2019-07-19 | 苏州国芯科技股份有限公司 | A kind of two-dimensional code identification method, system and electronic equipment and storage medium |
| CN110348264B (en) * | 2019-07-04 | 2023-12-26 | 北京电子工程总体研究所 | QR two-dimensional code image correction method and system |
| CN110415177A (en) * | 2019-07-26 | 2019-11-05 | 四川长虹电器股份有限公司 | A kind of image rotation sanction drawing method based on Java |
| CN111046692B (en) * | 2019-11-26 | 2023-04-14 | 北京奇艺世纪科技有限公司 | Method and device for determining two-dimensional code area |
| CN111553341B (en) * | 2019-12-24 | 2023-08-29 | 西安元智系统技术有限责任公司 | Cultural relic disassembly relative position verification method based on two-dimensional code |
| CN113033743A (en) * | 2019-12-25 | 2021-06-25 | 北京极智嘉科技股份有限公司 | Positioning identifier, robot for identifying positioning identifier and positioning method |
| CN111523341B (en) * | 2020-04-03 | 2023-07-11 | 青岛进化者小胖机器人科技有限公司 | Binarization method and device for two-dimensional code image |
| CN112036205A (en) * | 2020-08-03 | 2020-12-04 | 浙江大华技术股份有限公司 | Two-dimensional code identification method, electronic device and storage medium |
| CN112055010B (en) * | 2020-08-31 | 2022-07-05 | 中国平安财产保险股份有限公司 | Two-dimensional code picture intercepting method and device, electronic equipment and storage medium |
| CN112528700B (en) * | 2020-12-11 | 2024-04-16 | 马鞍山市博浪热能科技有限公司 | Two-dimensional code recognition analysis system based on big data |
| CN112818425A (en) * | 2021-02-04 | 2021-05-18 | 李小辉 | Intelligent integrated garment design system |
| CN112966533A (en) * | 2021-02-26 | 2021-06-15 | 陕西科技大学 | Two-dimensional code password verification system and method based on multiple data verification |
| CN114548136A (en) * | 2021-11-18 | 2022-05-27 | 福建星网天合智能科技有限公司 | Method, device, equipment and medium for analyzing light-reflecting two-dimensional code picture |
| CN114819025B (en) * | 2022-06-28 | 2022-09-13 | 深圳市国人光速科技有限公司 | Ink jet image generating method and device containing multiple variable information and computer |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101710407A (en) * | 2009-12-29 | 2010-05-19 | 江西科技师范学院 | Payment method capable of consumption payment by mobile phone on basis of two-dimension code and payment system thereof |
| CN101930532A (en) * | 2010-10-29 | 2010-12-29 | 福州中路网络技术开发有限公司 | Mobile phone-based camera quick response matrix code recognizing and reading method |
-
2014
- 2014-03-17 CN CN201410097740.1A patent/CN103914675B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101710407A (en) * | 2009-12-29 | 2010-05-19 | 江西科技师范学院 | Payment method capable of consumption payment by mobile phone on basis of two-dimension code and payment system thereof |
| CN101930532A (en) * | 2010-10-29 | 2010-12-29 | 福州中路网络技术开发有限公司 | Mobile phone-based camera quick response matrix code recognizing and reading method |
Non-Patent Citations (1)
| Title |
|---|
| 《QR码抗倾斜识别方法的研究与应用》;付裕等;《计算机应用与软件》;20120118;100-102 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292212A (en) * | 2017-04-26 | 2017-10-24 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of Quick Response Code localization method under low signal-to-noise ratio environment |
| CN107292212B (en) * | 2017-04-26 | 2020-10-23 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Two-dimensional code positioning method under low signal-to-noise ratio environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103914675A (en) | 2014-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103914675B (en) | A kind of 2 D code of quick response matrix recognition methods of clothing | |
| CN101443791B (en) | Improved foreground/background separation in digitl images | |
| CN106570436B (en) | A kind of substation equipment kilowatt meter reading-out system and its meter register method based on two dimensional code | |
| CN100456314C (en) | QR two-dimensional bar code recognition method based on pickup head for chatting | |
| CN105931295B (en) | A kind of geologic map Extracting Thematic Information method | |
| CN111753577B (en) | Apple identification and positioning method in automatic picking robot | |
| CN107689061A (en) | Rule schema shape code and localization method for indoor mobile robot positioning | |
| CN105719306B (en) | A kind of building rapid extracting method in high-resolution remote sensing image | |
| CN102646193B (en) | Segmentation method of character images distributed in ring shape | |
| CN103646249B (en) | A kind of greenhouse intelligent mobile robot vision navigation path identification method | |
| CN107122737A (en) | A kind of road signs automatic detection recognition methods | |
| CN105426844B (en) | A kind of answer card identification method | |
| CN103426164B (en) | Based on scallop size computing method and the scallop sorting system of Opencv graphical analysis | |
| CN103258201A (en) | Form line extraction method integrating global information and local information | |
| CN103294980A (en) | Micro-QR (Quick Response) code recognition method based on picture processing | |
| CN108509928A (en) | For Cold region apple jujube garden field pipe operation vision guided navigation path extraction method | |
| CN109191520B (en) | Plant leaf area measuring method and system based on color calibration | |
| CN102072882A (en) | Method for quantitatively analyzing plant nutrient content index based on leaf image characteristics | |
| CN103871072B (en) | Orthography based on project digital elevation model inlays line extraction method | |
| CN103632137A (en) | Human iris image segmentation method | |
| CN107220577A (en) | A kind of Quick Response Code localization method and system based on machine learning | |
| CN107729853A (en) | A kind of automatic identifying method suitable for the narrow tuning drive gear formula instrument of transformer station | |
| CN110414308A (en) | A kind of target identification method for dynamic foreign matter on transmission line of electricity | |
| CN109977944A (en) | A kind of recognition methods of digital water meter reading | |
| CN103942555A (en) | Method for detecting nitrogen content of plant through images |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20161116 Termination date: 20190317 |