CN106485183B - A kind of Quick Response Code localization method and system - Google Patents
A kind of Quick Response Code localization method and system Download PDFInfo
- Publication number
- CN106485183B CN106485183B CN201610857210.1A CN201610857210A CN106485183B CN 106485183 B CN106485183 B CN 106485183B CN 201610857210 A CN201610857210 A CN 201610857210A CN 106485183 B CN106485183 B CN 106485183B
- Authority
- CN
- China
- Prior art keywords
- quick response
- response code
- region
- outer boundary
- data
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1417—2D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1443—Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Electromagnetism (AREA)
- General Health & Medical Sciences (AREA)
- Toxicology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
A kind of Quick Response Code localization method and system, prescreening is carried out to view data first with the marginal information in two-dimensional code data area to the image of input, determine to include the image-region of Quick Response Code as candidate region, to each candidate region, the approximate location of Quick Response Code outer boundary is determined using the positioning figure of Quick Response Code to be positioned, so as to obtain the coarse localization of Quick Response Code, the exact boundary of Quick Response Code is finally determined according to the approximate location of Quick Response Code outer boundary, so as to obtain being accurately positioned for Quick Response Code.The present invention automatically can position the Quick Response Code in image so that need not manually interfere when being positioned to Quick Response Code can in rejection image complex background interference, and by prescreening and coarse positioning, reduce the data processing amount during automatic positioning.
Description
Technical field
Image procossing of the present invention, and in particular to the method and system positioned according to image to Quick Response Code.
Background technology
Compared with bar code, it is a variety of excellent that Quick Response Code has that information capacity is big, reliability is high, confidentiality and security are good etc.
Point, therefore planar bar code technology obtains rapid development and extensive use in industrial circle.The key of planar bar code technology application is
Quick Response Code identifies, i.e., obtains bar code information by some scanning devices, then decoded.At present, Quick Response Code identification mostly according to
Rely in specific identification equipment, and need manual-alignment Quick Response Code to be identified, it is necessary to manual intervention.Recently as digitized map
As the development for the treatment of technology, there is the two-dimensional code recognition mode based on camera shooting, it is right after the image of Quick Response Code is got
The automatic identification of Quick Response Code generally includes：(1) the automatic detection positioning of bar code；(2) bar code image is handled；(3) Quick Response Code
Decoding.
The automatic of Quick Response Code detects the first step that positioning is bar code identification, and a complicated process.Due to two
Tie up in code shooting process, Quick Response Code is often mixed with text or other patterns, and Quick Response Code is only shooting image
A part, in position fixing process, it is necessary to filter out text and other patterns, orient Quick Response Code.What is more, once shoots
Multiple image in 2 D code may be photographed, in follow-up position fixing process, it is necessary to identify this multiple Quick Response Code from background
Come and positioned one by one.Therefore, how to be automatically positioned in complicated image and identify that Quick Response Code is a problem to be solved.
The content of the invention
The application provides a kind of Quick Response Code localization method and system, and Quick Response Code can be automatically positioned out in complicated image.
According in a first aspect, provide a kind of Quick Response Code localization method in a kind of embodiment, including：
Input image data；
Prescreening is carried out to view data using the marginal information in two-dimensional code data area, determines that the figure of Quick Response Code may be included
As candidate region, the marginal information refers to that gray value has the region of mutation in picture region；
To each candidate region, the approximate location of Quick Response Code outer boundary is determined using the positioning figure of Quick Response Code to be positioned,
So as to obtain the coarse localization of Quick Response Code；
The exact boundary of Quick Response Code is determined according to the approximate location of Quick Response Code outer boundary, so as to obtain the accurate of Quick Response Code
Positioning.
According to second aspect, a kind of Quick Response Code alignment system is provided in a kind of embodiment, including：
Figure acquiring unit, for input image data；
Prescreening unit, carries out prescreening, determining can for the marginal information using two-dimensional code data area to view data
The image-region of Quick Response Code can be included as candidate region, the marginal information refers to that gray value has the region of mutation；
Coarse localization unit, for each candidate region, Quick Response Code to be determined using the positioning figure of Quick Response Code to be positioned
The approximate location of outer boundary, so as to obtain the coarse localization of Quick Response Code；
Unit is accurately positioned, for determining the exact boundary of Quick Response Code according to the approximate location of Quick Response Code outer boundary, from
And obtain being accurately positioned for Quick Response Code.
In above-described embodiment, Quick Response Code positioning is carried out based on view data, position fixing process includes prescreening, coarse positioning and essence
Determine position, by the way that the marginal information of image and texture information are combined, carry out prescreening, obtain possible two-dimension code area,
Coarse positioning is carried out then in conjunction with the positioning figure of Quick Response Code, the accurate side of Quick Response Code is finally determined on the basis of coarse positioning
Boundary, so as to obtain being accurately positioned for Quick Response Code.This positioning method automatically can position the Quick Response Code in image so that
Need not manually interfere when being positioned to Quick Response Code can in rejection image complex background interference, and by prescreening and
Coarse positioning, reduces the data processing amount during automatic positioning.
Brief description of the drawings
Fig. 1 is the schematic diagram of Data Matrix Quick Response Codes；
Fig. 2 is a kind of structure diagram of the Quick Response Code identification equipment of embodiment；
Fig. 3 is Quick Response Code positioning flow figure in a kind of embodiment；
Fig. 4 is a kind of flow chart of embodiment of prescreening；
Fig. 5 is the schematic diagram that whole image data are divided into some data blocks；
Fig. 6 is the schematic diagram of the positioning figure of Data Matrix Quick Response Codes；
Fig. 7 is a kind of flow chart of embodiment of coarse positioning；
Fig. 8 is a kind of schematic diagram of the empty side rectangular area obtained in specific embodiment；
Fig. 9 is a kind of pinpoint flow chart of embodiment；
Figure 10 a are the schematic diagram of the scanning area of two solid line outer boundaries；
Figure 10 b are the schematic diagram of the scanning area of two dotted line outer boundaries；
Figure 11 is a kind of structure diagram of the alignment system of embodiment.
Embodiment
Quick Response Code can be produced by various coding modes, and for convenience of the positioning of Quick Response Code, every kind of Quick Response Code has respective
Location feature, referred to herein as positions figure, and positioning figure is used for bar code symbol and positions and determine bar code structural information, it is also possible to
It is clear area in solving symbol location and problem of dtmf distortion DTMF, the surrounding for positioning figure, for by bar code and background separation.Such as Data
Matrix Quick Response Codes, as shown in Figure 1, complete Data Matrix Quick Response Codes 10 are rendered as the two-dimension code pattern of a rectangle, have
Four edges, two orthogonal solid line outer boundaries 11 and two orthogonal dotted line outer boundaries 12, centre is chequered with black and white
Data cell form data field 13.In embodiments of the present invention, when the Quick Response Code in image positions, first will
The marginal information and texture information of image are combined, and are carried out prescreening, possible two-dimension code area are obtained, then in conjunction with Quick Response Code
Positioning figure (Pattern Finder) carry out coarse positioning, the accurate side of Quick Response Code is finally determined on the basis of coarse positioning
Boundary, so as to obtain being accurately positioned for Quick Response Code.
The present invention is made by embodiment combination attached drawing by taking Data Matrix Quick Response Codes as an example below further detailed
Describe in detail bright.
Embodiment one：
Please refer to Fig.2, Quick Response Code identification equipment includes image capture module 20 and image processing module 40, Image Acquisition mould
Block 20 can be that industrial camera, cell-phone camera are first-class, send the view data of collection to image procossing by network interface or USB
Module 40, in a further embodiment, Quick Response Code identification equipment can not also include image capture module 20, and view data also may be used
To be directly loaded into image processing module 40 from local picture, or figure downloaded to from other servers by wired or wireless network
As processing module 40.
Image processing module 40 includes image quality measure system 41, pretreatment system 42, alignment system 43, image rectification
And strengthening system 44 conciliates code system 45.
Image quality measure system 41 gives a mark the contrast, illumination nonlinear change, fog-level of input picture,
The value range of fraction can be set to 0~1, and marking result is passed to pretreatment system 42.Contrast evaluation uses region pair
Than degree form, divide the image into as the difference of several parts, the in each area interior maxima and minima for calculating gray scale, then
Divided by maximum, i.e.,
Wherein f (xi,yi) it is coordinate (x in the j of regioni,yi) place gray value.After the contrast for obtaining regional, to institute
The contrast for having pixel region is averaged, and the contrast as whole image is given a mark.The contrast estimation of image can also be selected
Select based on histogram most secretly between value (minimum reflectance) and its most bright value (maximum reflectivity) midpoint it is a kind of extremely simple
Method gives a mark image.
It is two parts by picture breakdown when giving a mark to the nonlinear change of illumination：Luminance picture i (x, y) and reflection
Image r (x, y), gives a mark the nonlinear change of image according to reflected image.
The fog-level evaluation of image mainly analyzes the diffusion breadth at edge, and usual image is fuzzyyer, and edge expands
Scattered width is bigger, and image is more clear, and image border diffusion breadth is narrower, is given a mark according to the diffusion breadth of image border.
The handling result of image quality measure system 41 passes to pretreatment system 42.Pretreatment system is according to picture quality
The score of assessment is judged, if score is less than given threshold, carries out histogram equalization and frequency filtering processing, enhancing figure
The contrast of picture, reduces the nonlinear change of illumination.Pretreated view data will be transmitted to alignment system 43.If score is more than
Or equal to given threshold, then view data is directly transmitted to alignment system 43 by pretreatment system 42.
Alignment system 43 is used to search two-dimension code area in view data, when finding out Quick Response Code area in view data
Quick Response Code is positioned behind domain, behind alignment system 43 will be described in detail.
The Quick Response Code that image rectification and strengthening system 44 are oriented according to alignment system 43 carries out image rectification and enhancing, figure
It is mainly used for eliminating trapezoidal distortion as correcting, carries out perspective correction, reduces to decode caused by image in 2 D code transparent effect and miss
Difference.Image enhancement is mainly used for the contrast of image-region where strengthening Quick Response Code, prepares for binaryzation, improves Quick Response Code solution
The contrast in code region, so as to improve decoded accuracy rate.
Solution code system 45 is used to the data field of Quick Response Code be identified, and Quick Response Code is decoded according to coding rule,
In a kind of specific embodiment, decode and carried out according to international standard ISO/IEC 16022.
It will be understood by those skilled in the art that in the embodiment having, image processing module 40 can also not have image matter
Measure assessment system 41, pretreatment system 42 and one or more of image rectification and strengthening system 44.
In the present embodiment, alignment system 43 method of Quick Response Code positioning is carried out in view data as shown in figure 3, including with
Lower step：
Step 100, input image data.Two-dimension code pattern may be included in the view data that alignment system 43 obtains,
Two-dimension code pattern may not included, it is also possible to include multiple two-dimension code patterns.
Step 110, prescreening.In this step, pre-sifted is carried out to view data using the marginal information in two-dimensional code data area
Choosing, determines that it is prominent to refer to that grey scale pixel value has as candidate region, wherein marginal information for the image-region that may include Quick Response Code
The region of change.A kind of embodiment of prescreening is as shown in figure 4, comprise the following steps：
Step 111, data block is divided.Whole image data are divided into M × N number of data block, M and N can be equal, also may be used
To differ, as shown in figure 5, including c × d pixel in each data block, c and d can be equal, can not also wait.Below to every
A data block is analyzed.
Step 112, each data marginal information in the block is detected.In this step mainly using Quick Response Code data field feature into
Row edge detection.Compared to other backgrounds of image, data field is made of some chequered with black and white data cells, therefore can be formed
Substantial amounts of marginal information.Marginal information can utilize single order information or second order information to be detected.If according to single order information into
Row edge detection, can be detected using such as Sobel, Canny scheduling algorithm, be screened according to threshold value, so as to detect side
Edge information；If carrying out edge detection according to second order information, the null point of second order information may be selected and be used as marginal point.This implementation
In example, to reduce calculation amount, marginal information detection can use following algorithm：
The shade of gray of each pixel is calculated first.For a two dimensional image, each pixel may be produced in X-dimension
Raw gray scale mutation, it is also possible to produce gray scale mutation in Y dimensions, the gradient magnitude of each pixel is calculated by the following formula.
I (x, y) is image in the gray value at pixel (x, y) place, g in formulax(x,y),gy(x, y) is shade of gray in x, y
The component in direction.The gradient magnitude of pixel is：
The edge angle of pixel is：
It will be appreciated by those skilled in the art that in a further embodiment other algorithms can also be used to calculate pixel
Gradient magnitude, such as in a predetermined direction the gray value of adjacent two pixel is subtracted each other to obtain gradient magnitude.
After the gradient magnitude for calculating each pixel, the gradient magnitude of pixel and the second given threshold are compared,
If gradient magnitude is more than the second given threshold, then it is assumed that the pixel is a marginal point, also referred herein as an edge
Information.
Step 113, each data edge parameters in the block are calculated according to marginal information.Edge parameters include marginal frequency and side
Edge differential seat angle, marginal frequency refer to the number of the marginal information detected in data block, i.e., gradient magnitude is more than the in data block
The number of the marginal point of two given thresholds, uses fiRepresent the marginal frequency of i-th of data block.Second given threshold can be with when setting
According to following principles：If contrast is stronger, the threshold value value is larger, if contrast is poor, the threshold value is smaller.Due to two dimension
The data field of code is mixed by monochrome data unit according to coding rule to be formed, therefore data field can form more marginal information.
Edge angle difference is calculated using following methods：
The histogram of the edge angle of statistics block, can be filtered histogram processing, avoid the influence of noise, so
The differential seat angle between obvious wave crest, the differential seat angle in histogram are calculated afterwards is denoted as edge angle difference αi。
For two-dimension code area, the histogram of its edge angle has two obvious wave crests, and the angle of two wave crests
Nearly 90 ° of differential.Therefore, the edge angle difference α of the data block of statisticsiCloser to 90 °, the probability containing Quick Response Code is got in data block
Greatly.
It will be appreciated by those skilled in the art that in a further embodiment, edge parameters can also be defined and include more
One in marginal frequency and edge angle difference is included with the relevant parameter of marginal information, or definition edge parameters.
Step 114, according to the edge parameters of each data block judge the data block whether be with the relevant data block of Quick Response Code,
Judge in data block whether a part that may be containing Quick Response Code or Quick Response Code.Judge number using following steps in the present embodiment
According to whether related to Quick Response Code in block：
Pixel in data block is subjected to binary conversion treatment, the ratio of monochrome pixels, obtains the first ratio in statistics block
Example, is denoted as ri；
The ratio of the marginal frequency of statistics block and all data numbers of blocks of division, obtains calculating percentage shared by edge
Than being denoted as the second ratio, i.e.,：
Edge angle difference and 90 degree of the ratio of statistics block, obtain the 3rd ratio, i.e.,：
The weighted sum of the first ratio, the second ratio and the 3rd ratio is calculated, obtains the Relevance scores of the data block:
Score=w1eratio+w2αratio+w3ri (7)
Wherein, w1,w2,w3The weight of respectively each ratio.
By Relevance scores score compared with the 3rd given threshold, which is marked according to comparative result, if
The Relevance scores score of the data block is more than the 3rd given threshold, then the attribute of the data block is labeled as True, and representing should
A part for Quick Response Code or Quick Response Code may be included in data block, it is related to Quick Response Code, otherwise the attribute of the data block is marked
For False, represent that the data block is unrelated with Quick Response Code, belong to background parts.
It will be appreciated by those skilled in the art that in other specific embodiments, number can also be judged using other methods
Whether be related to Quick Response Code according to block, for example, only according to marginal frequency judge data block whether be it is related to Quick Response Code, if one
A data marginal frequency in the block is less than certain threshold value, illustrates that data field may not be contained in the data block, i.e., in the data block
May there is no Quick Response Code.
Step 115, carry out connecting processing with the relevant data block of Quick Response Code by all, obtain connected region.
In the present embodiment, when data block is carried out regional connectivity analysis, in order to reduce the time of the connectivity of region analysis,
A width figure is regenerated, each data block is represented with a gray value, if data block's attribute is True, gray value is arranged to
255, if data block's attribute is False, gray value is arranged to 0.When being connected only to gray value be 255 data block into
Row connection is handled, and the connected region of acquisition is denoted as Ri。
Step 116, judge whether connected region size meets the first given threshold, if it is, the connected region is can
The candidate region of Quick Response Code can be included, otherwise the connected region is not processed.The connected region obtained in step 115 may have
One, it is also possible to have multiple, presumable big, what is had is small, it may be possible to the region comprising Quick Response Code, it is also possible to the mixed and disorderly back of the body
Scape, to reduce follow-up treating capacity, it is necessary to first judge connected region.In the present embodiment, due to Data Matrix two dimensions
Code is rectangle, therefore judges whether connected region meets the side size range of rectangle.
For example, i-th of connected region R to acquisitioni, first to connected region RiMinimum enclosed rectangle is done, it is minimum to obtain it
The length and width of boundary rectangle is respectively leniAnd widi。
Then by leniAnd widiCompared with given threshold, if l1≤leni≤l2,w1≤widi≤w2, then connected region
Domain RiAs the candidate region of localization region, otherwise ignore the region, wherein l1,l2For the lower threshold and the upper limit of zone length
Threshold value, w1,w2For the lower threshold and upper limit threshold of peak width.
Referring again to Fig. 3 behind acquisition candidate region, step 120 is performed to candidate region, carries out coarse positioning processing.
In the step 120, to each candidate region, Quick Response Code outer boundary is determined using the positioning figure of Quick Response Code to be positioned
Approximate location, so as to obtain the coarse localization of Quick Response Code.
The positioning figure for the Data Matrix Quick Response Codes that the present embodiment defines is the width of a module, is data field
Border, it includes two solid line to intersect vertically outer boundary information and two dotted line outer boundary information to intersect vertically, such as Fig. 6 institutes
Show, wherein the adjacent side being all made of dark module constitutes " L " shape solid line outer boundary 14, " L " edge-to-edge is by alternate
Two adjacent sides of depth module composition constitute dotted line outer boundary 15 again, what solid line outer boundary 14 and dotted line outer boundary 15 enclosed
The surrounding of rectangle is clear area, for by Quick Response Code and background separation.In a kind of specific embodiment, Data Matrix are utilized
The positioning figure of Quick Response Code determines the process of the approximate location of Quick Response Code outer boundary as shown in fig. 7, comprises following steps：
Step 121, line segment is extracted from candidate region.
Line segments extraction can have two methods it can be selected that one kind is Hough transform method, and one kind is region growth method.Nothing
By being Hough transform method or region growth method, processing is all candidate region RiThe information of corresponding image part.Region increases
Regular way is mainly increased using the direction of edge angle, first in candidate region RiIt is middle that seed point is found out according to marginal information,
The point of gradient magnitude maximum is specifically found out in candidate region as seed point, is then searched in 8 neighborhoods of seed point full
The marginal point of sufficient angle conditions is added to growth region, which is：
|θregion-θi|≤τ (8)
θ in formularegionFor the angle of growth region, θiFor the edge angle of pixel in 8 neighborhoods.Initial growth region
An only seed point, therefore, the initial angle of growth region are the edge angle of seed point.
Often increase a point into growth region, then the angle for updating growth region is：
The point of new growth is continued to increase as seed point, until there is no pixel satisfaction in 8 neighborhoods of new seed point
Terminate to increase during angle conditions, fitting a straight line is carried out to the point in growth region, such as using least square fitting, so as to carry
Take out line segment.
In candidate region RiThe pixel not increased in find out the point of gradient magnitude maximum again as seed point,
The region for starting next line segment increases, and next line segment is extracted using same method.So extract candidate region RiIn
All line segments.
Following information can be obtained after line segments extraction：Center of gravity (the c of each line segmentxi,cyi), the length l of line segmenti, line segment
Unit direction vector (nxi,nyi), the endpoint (p of line segment can be calculated according to these informationx1,py1) and (px2,py2)。
All line segments extracted are added into list of line segments, then carry out line segment screening, line segment screening includes two aspects, the
First, the screening to line segment length, second is the screening to line segment quantity.
Step 122, judge whether the length of line segment meets the requirements.
According to positioning figure, solid line outer boundary should have a length range, i.e. length limit lupperWith length lower limit
llowerIf the length of line segment is more than length limit lupperOr less than length lower limit llower, then by the line segment from list of line segments
Middle deletion, so as to filter out the line segment that length meets solid line outer boundary length range.
Step 123, judge whether the quantity of list of line segments middle conductor is less than two, if it is, the candidate region is abandoned,
Otherwise the candidate region is carried out performing step 124.
Step 124, " L " shape line match." L " shape line match includes two aspects, and first is the matching of angle threshold, i.e.,
The two lines section that intersecting and angle meets angle threshold is filtered out in the line segment of candidate region, its judgement formula is：α1<α<α2, its
Middle α1,α2For angle threshold, α is the angle of two line segments, if it is, two line segments meet angle threshold.Second is relative length
Matching, i.e., be denoted as l respectively by the length of two lines section1,l2,l1＞ l2, judge whether to meet l1/l2<τ1, wherein τ1Set for user
The threshold value put.When above-mentioned two aspects all meet, then the two lines section currently judged is probably " L " shape line segment, then performs step
Rapid 125, if above-mentioned two aspect has one of them to be unsatisfactory for, the two lines section currently judged is not " L " shape line segment, then chooses
Two other line segment performs step 124 and is judged again, until filtering out whole " L " shape line segments.Every two " L " shape line segments can
It can correspond to a Quick Response Code.
Step 125, the length of " L " shape line segment and the position coordinates on three vertex are obtained.Three vertex on " L " shape side point
(x is not denoted as iti,yi) (i=1,2,3), the length of two lines section is denoted as len1,len2And len1＞ len2.So that it is determined that go out this
The approximate location of two solid line outer boundaries of all Quick Response Codes in candidate region.
Step 126, empty side extraction.In the present embodiment, sat according to the position on the length of " L " shape line segment and three vertex
The position relationship of solid line outer boundary and dotted line outer boundary determines the square where two dotted line outer boundaries in mark, and positioning figure
Shape region, specifically includes following steps：
As shown in figure 8, according to " L " shape side information, the rectangular area that certain distance obtains empty side, rectangular area are deviated respectively
Information is represented with the upper left point and lower-right most point of rectangle respectively, it is assumed that the upper apex coordinate of the vertical edge on " L " shape side is (x1,y1), " L "
The horizontal edge on shape side is (x by apex coordinate3,y3), the intersecting point coordinate on both sides is (x2,y2), offset distance τ, then in Fig. 8 on
The rectangular area of side is represented by：
regionupper={ x1-τ,y1-τ,x1+len1+τ,y1+len1-len2+τ} (10)
The rectangular area on right side can be expressed as in Fig. 8：
regionright={ x1+len2-len1-τ,y1-len1-τ,x3+τ,y3+τ} (11)
Wherein τ is the parameter of user setting., can be with so as to obtain approximate region where two dotted line outer boundaries of Quick Response Code
It was initially believed that dotted line outer boundary is located at the midline position of rectangular area along its length.
In other embodiments, the approximate location of two dotted line outer boundaries can also be determined by other methods.
According to above-mentioned steps, the two solid line outer boundaries and two dotted line outer boundaries that can extract, outside two solid lines
The region that boundary and two dotted line outer boundaries enclose is as candidate's two-dimension code area.
In a preferred embodiment, step 127 can also be included after step 126, respectively along the short of rectangular area
Edge direction searches angle point, and the angle point refers to the intersection point of two edge lines；Judge the face of the rectangular area of angle point quantity and dotted line
Whether product ratio is more than the 4th given threshold, if it is, performing step 128 retains candidate's two-dimension code area to be carried out to it
It is accurately positioned, otherwise performs step 129, abandon candidate's two-dimension code area, no longer it is accurately positioned.
Fig. 3 is refer again to, step 130 is performed to candidate's two-dimension code area, Quick Response Code is accurately positioned.
In step 130, the exact boundary of Quick Response Code is determined according to the approximate location of Quick Response Code outer boundary, so as to obtain
Quick Response Code is accurately positioned, its specific process flow is as shown in figure 9, comprise the following steps：
Step 131, a pixel of Δ is offset to both sides from the approximate location of solid line outer boundary and dotted line outer boundary respectively, it is raw
Into scanning area.As shown in Figure 10 a, region 17 folded by region 16 folded by two horizontal lines and two vertical lines is respectively two realities
The scanning area of line outer boundary.As shown in fig. lob, region 19 folded by region 18 folded by two horizontal lines and two vertical lines is distinguished
For the scanning area of two dotted line outer boundaries.
Step 132, scan marginal point along the direction of arrow in the figure in scanning area, and the marginal point of acquisition is grouped
P is stored in respectivelyleft,pright,pupper,plowerIn.
Step 133, the marginal point of each group is subjected to fitting a straight line (such as least square method), can obtain through iteration optimal straight
Line, so as to respectively obtain the exact position of solid line outer boundary and dotted line outer boundary.
Step 134, the position coordinates of four intersection points of two solid line outer boundaries and two dotted line outer boundaries is calculated.
After above-mentioned steps, the exact position of the four edges of Quick Response Code can obtain.
It will be appreciated by those skilled in the art that in other specific embodiments, other methods can be also used in solid line
The approximate location of outer boundary and dotted line outer boundary accurately determines the position of two solid line outer boundaries and two dotted line outer boundaries.
Embodiment two：
In the present embodiment, as shown in figure 11, alignment system includes figure acquiring unit 431, prescreening unit 432, rough
Positioning unit 433 and it is accurately positioned unit 434.
Figure acquiring unit 431 is used for input image data.
Prescreening unit 432 is used to carry out prescreening to view data using the marginal information in two-dimensional code data area, determines
The image-region of Quick Response Code may be included as candidate region, the marginal information refers to that gray value has the region of mutation.
In one instantiation, prescreening unit 432 includes piecemeal subelement 4321, edge detection subelement 4322, edge parameters and calculates
Subelement 4323, the first judgment sub-unit 4324, connection 4325 and second judgment sub-unit 4326 of subelement.
Piecemeal subelement 4321 is used to partition image data into several data blocks.
Edge detection subelement 4322 is used to detect each data marginal information in the block.In a kind of specific embodiment, side
Edge detection sub-unit 4322 detects marginal information in the following manner：Calculate ladder of each pixel in the first dimension and the second dimension
Spend component, the gradient magnitude of each pixel calculated according to gradient component, by the gradient magnitude of pixel and the second given threshold into
Row compares, if gradient magnitude is more than the second given threshold, then it is assumed that the pixel is a marginal information.
Edge parameters computation subunit 4323 is used to calculate each data edge parameters in the block according to marginal information, in a tool
In body embodiment, for edge parameters including at least one in marginal frequency and edge angle difference, marginal frequency is data block manhole ladder
Spend the number that amplitude is more than the marginal information of the second given threshold；Edge angle difference obtains in the following manner：According to gradient point
Amount calculates the edge angle of each pixel, the histogram of the edge angle of statistics block, calculate in histogram obvious wave crest it
Between differential seat angle, the differential seat angle for edge angle it is poor.
First judgment sub-unit 4324 be used for according to the edge parameters of each data block judge the data block whether be and two dimension
The relevant data block of code.In one embodiment, the first judgment sub-unit 4324 is used to the pixel in data block carrying out two
Value is handled, and the ratio of monochrome pixels, obtains the first ratio in statistics block；The marginal frequency of statistics block and division
The ratio of all data numbers of blocks, obtains the second ratio；Edge angle difference and 90 degree of the ratio of statistics block, obtain the 3rd
Ratio；The weighted sum of the first ratio, the second ratio and the 3rd ratio is calculated, obtains the Relevance scores of the data block；By correlation
Property score compared with the 3rd given threshold, according to comparative result mark the data block whether be and the relevant data of Quick Response Code
Block.
Connection subelement 4325 is used to carry out connecting processing with the relevant data block of Quick Response Code by all, obtains connected region
Domain.
Second judgment sub-unit 4326 is used to judge whether connected region size meets the first given threshold, if it is,
The connected region is that possible include the candidate region of Quick Response Code.
Coarse localization unit 433 is used for each candidate region, and two dimension is determined using the positioning figure of Quick Response Code to be positioned
The approximate location of code outer boundary, so as to obtain the coarse localization of Quick Response Code.It is as shown in figure 11, rough fixed in an instantiation
Bit location 433 includes line segments extraction subelement 4331, solid line coarse positioning subelement 4332 and dotted line coarse positioning subelement 4333, line
Section extraction subelement 4331 is used to extract line segment from candidate region.Solid line coarse positioning subelement 4332 is used to be schemed according to positioning
Shape, filters out the line segment that length meets solid line outer boundary length range；Judgement meets the line segment of solid line outer boundary length range
Whether quantity is less than two, if it is, abandoning the candidate region, " L " shape line segment is otherwise carried out in the line segment of candidate region
Matching, and the length of " L " shape line segment and the position coordinates on three vertex are obtained, so that it is determined that going out all two in the candidate region
Tie up the approximate location of two solid line outer boundaries of code.Dotted line coarse positioning subelement 4333 be used for according to the length of " L " shape line segment with
And the position relationship of solid line outer boundary and dotted line outer boundary determines two in the position coordinates on three vertex, and positioning figure
Rectangular area where dotted line outer boundary.The region that two solid line outer boundaries and two dotted line outer boundaries enclose is candidate's Quick Response Code
Region.
In another instantiation, as shown in figure 11, coarse localization unit 433 further includes and screens subelement 4334 again, then
Subelement 4334 is screened to be used for behind the rectangular area where determining two dotted line outer boundaries respectively along the rectangular area
Short side direction search angle point, angle point refers to the intersection point of two edge lines, judges angle point quantity and the area ratio of rectangular area
Whether it is more than the 4th given threshold, if it is, retaining candidate's two-dimension code area to be accurately positioned to it, otherwise abandons
Candidate's two-dimension code area, is no longer accurately positioned it.
The exact boundary that unit 434 is used to determine Quick Response Code according to the approximate location of Quick Response Code outer boundary is accurately positioned,
So as to obtain being accurately positioned for Quick Response Code.In an instantiation, as shown in figure 11, being accurately positioned unit 434 includes offset
Unit 4341, scanning subelement 4342, fitting subelement 4343 and intersecting point coordinate computation subunit 4344.Deviate subelement 4341
For being offset to both sides intended pixel from the rough position of solid line outer boundary and dotted line outer boundary respectively, scanning area is generated.Sweep
Retouch subelement 4342 to be used in scanning area scan marginal point along predetermined direction, and the marginal point of acquisition is grouped and is preserved.Intend
Zygote unit 4343 is used to the marginal point of each group carrying out fitting a straight line, respectively obtains the essence of solid line outer boundary and dotted line outer boundary
True position.Intersecting point coordinate computation subunit 4344 is used for four intersection points for calculating two solid line outer boundaries and two dotted line outer boundaries
Position coordinates.
It will be understood by those skilled in the art that all or part of step of various methods can pass through in the above embodiment
Program instructs related hardware to complete, which can be stored in a computer-readable recording medium, storage medium can wrap
Include：Read-only storage, random access memory, disk or CD etc..
Use above specific case is illustrated the present invention, is only intended to help and understands the present invention, not limiting
The system present invention.For those skilled in the art, according to the thought of the present invention, can also make some simple
Deduce, deform or replace.
Claims (10)
- A kind of 1. Quick Response Code localization method, it is characterised in that including：Input image data；Prescreening is carried out to view data using the marginal information in two-dimensional code data area, determines that the image district of Quick Response Code may be included As candidate region, the marginal information refers to that gray value has the region of mutation in domain；To each candidate region, the approximate location of Quick Response Code outer boundary is determined using the positioning figure of Quick Response Code to be positioned, so that Obtain the coarse localization of Quick Response Code；The exact boundary of Quick Response Code is determined according to the approximate location of Quick Response Code outer boundary, so as to obtain the accurate fixed of Quick Response Code Position；The marginal information using two-dimensional code data area carries out prescreening to view data to be included：Partition image data into several data blocks；Each data marginal information in the block is detected, is specifically included：Each pixel is calculated in the first dimension and the gradient of the second dimension Component, the gradient magnitude of each pixel is calculated according to gradient component, gradient magnitude and the second given threshold are compared, if The gradient magnitude of pixel is more than the second given threshold, then it is assumed that the pixel is a marginal information；Each data edge parameters in the block are calculated according to marginal information, the edge parameters include marginal frequency and edge angle Difference, the marginal frequency refer to the number of marginal information in data block, and the edge angle difference is calculated by following steps：According to The gradient component of pixel calculates the edge angle of each pixel, and the edge angle of statistics block simultaneously generates histogram, calculates Differential seat angle in histogram between obvious wave crest, the differential seat angle are poor for edge angle；According to the edge parameters of each data block judge the data block whether be with the relevant data block of Quick Response Code, specifically include：Will Pixel in data block carries out binary conversion treatment, and the ratio of monochrome pixels, obtains the first ratio in statistics block；Statistics The ratio of the marginal frequency of block and all data numbers of blocks of division, obtains the second ratio；The edge angle of statistics block is poor With 90 degree of ratio, the 3rd ratio is obtained；The weighted sum of the first ratio, the second ratio and the 3rd ratio is calculated, obtains the data The Relevance scores of block；By Relevance scores compared with the 3rd given threshold, the data block is marked to be according to comparative result It is no to be and the relevant data block of Quick Response Code；Carry out connecting processing with the relevant data block of Quick Response Code by all, obtain connected region；Judge whether connected region size meets the first given threshold, if it is, the connected region is that possible include Quick Response Code Candidate region.
- 2. the method as described in claim 1, it is characterised in that the Quick Response Code is Data Matrix Quick Response Codes, the two dimension The positioning figure of code includes two solid line outer boundary information to intersect vertically and two dotted line outer boundary information to intersect vertically, profit Determine that the approximate location of Quick Response Code outer boundary includes with the positioning figure of Quick Response Code to be positioned：Line segment is extracted from candidate region；According to positioning figure, the line segment that length meets solid line outer boundary length range is filtered out；Judge whether the quantity for meeting the line segment of solid line outer boundary length range is less than two, if it is, abandoning the candidate regions Domain, otherwise carries out the candidate region following processing；" L " shape line match is carried out in the line segment of candidate region, and obtains the length of " L " shape line segment and the position on three vertex Coordinate is put, so that it is determined that going out the approximate location of two solid line outer boundaries of all Quick Response Codes in the candidate region；According to solid line outer boundary and dotted line in the length of " L " shape line segment and the position coordinates on three vertex, and positioning figure The position relationship of outer boundary determines the rectangular area where two dotted line outer boundaries, outside two solid line outer boundaries and two dotted lines The region that border encloses is candidate's two-dimension code area.
- 3. method as claimed in claim 2, it is characterised in that line segment is extracted from candidate region by region growth method, is had Body includes：Seed point is found out according to marginal information in candidate region；Search meets that the point of angle conditions adds growth region in the eight neighborhood of seed point；The point of new growth is continued to increase as seed point, until terminating to increase；Fitting a straight line is carried out to the point in growth region, so as to extract line segment.
- 4. method as claimed in claim 2, it is characterised in that behind the rectangular area where determining two dotted line outer boundaries Further include：Angle point is searched respectively along the short side direction of the rectangular area, the angle point refers to the intersection point of two edge lines；Judge whether the area ratio of the rectangular area of angle point quantity and dotted line is more than the 4th given threshold, if it is, retaining Otherwise candidate's two-dimension code area abandons candidate's two-dimension code area to be accurately positioned to it, no longer it is carried out accurately Positioning.
- 5. method as claimed in claim 2, it is characterised in that Quick Response Code is determined according to the approximate location of Quick Response Code outer boundary Exact boundary include：Intended pixel is offset to both sides from the approximate location of solid line outer boundary and dotted line outer boundary respectively, generates scanning area；Marginal point is scanned along predetermined direction in scanning area, and the marginal point of acquisition is grouped and is preserved；The marginal point of each group is subjected to fitting a straight line, respectively obtains the exact position of solid line outer boundary and dotted line outer boundary；Calculate the position coordinates of four intersection points of two solid line outer boundaries and two dotted line outer boundaries.
- A kind of 6. Quick Response Code alignment system, it is characterised in that including：Figure acquiring unit, for input image data；Prescreening unit, prescreening is carried out for the marginal information using two-dimensional code data area to view data, and determining may bag For image-region containing Quick Response Code as candidate region, the marginal information refers to that gray value has the region of mutation；The pre-sifted Menu member includes：Piecemeal subelement, edge detection subelement, edge parameters computation subunit, the first judgment sub-unit, connection Unit and the second judgment sub-unit, wherein,Piecemeal subelement is used to partition image data into several data blocks；Edge detection subelement is used to detect each data marginal information in the block in the following manner：Each pixel is calculated first The gradient component of dimension and the second dimension；The gradient magnitude of each pixel is calculated according to gradient component；By the gradient width of pixel Value and the second given threshold are compared, if gradient magnitude is more than the second given threshold, then it is assumed that the pixel is a side Edge information；Edge parameters computation subunit is used to calculate each data edge parameters in the block, the edge parameters bag according to marginal information Include marginal frequency and edge angle is poor, the marginal frequency refers to the number of marginal information in data block；Edge parameters calculate son Unit is used for the marginal information in statistics block, obtains marginal frequency, and the edge angle of each pixel is calculated according to gradient component Degree, the histogram of the edge angle of statistics block, calculates the differential seat angle between obvious wave crest in histogram, which is side Edge differential seat angle；First judgment sub-unit is used to judge whether the data block is relevant with Quick Response Code according to the edge parameters of each data block Data block, specifically includes：Pixel in data block is carried out binary conversion treatment by the first judgment sub-unit, black and white in statistics block The ratio of pixel, obtains the first ratio；The ratio of the marginal frequency of statistics block and all data numbers of blocks of division, obtains Second ratio；Edge angle difference and 90 degree of the ratio of statistics block, obtain the 3rd ratio；Calculate the first ratio, the second ratio The weighted sum of example and the 3rd ratio, obtains the Relevance scores of the data block；Relevance scores and the 3rd given threshold are carried out Compare, according to comparative result mark the data block whether be and the relevant data block of Quick Response Code；Connection subelement is used to carry out connecting processing with the relevant data block of Quick Response Code by all, obtains connected region；Second judgment sub-unit is used to judge whether connected region size meets the first given threshold, if it is, the connected region Domain is the candidate region that possible include Quick Response Code；Coarse localization unit, for each candidate region, being determined using the positioning figure of Quick Response Code to be positioned outside Quick Response Code The approximate location on boundary, so as to obtain the coarse localization of Quick Response Code；Unit is accurately positioned, for determining the exact boundary of Quick Response Code according to the approximate location of Quick Response Code outer boundary, so as to obtain Obtain being accurately positioned for Quick Response Code.
- 7. system as claimed in claim 6, it is characterised in that the Quick Response Code is Data Matrix Quick Response Codes, the two dimension The positioning figure of code includes two solid line outer boundary information to intersect vertically and two dotted line outer boundary information to intersect vertically, slightly Slightly positioning unit includes：Line segments extraction subelement, for extracting line segment from candidate region；Solid line coarse positioning subelement, for according to positioning figure, filtering out the line segment that length meets solid line outer boundary length range； Judge whether the quantity for meeting the line segment of solid line outer boundary length range is less than two, if it is, the candidate region is abandoned, it is no " L " shape line match is then carried out in the line segment of candidate region, and obtains the length of " L " shape line segment and the position on three vertex Coordinate, so that it is determined that going out the approximate location of two solid line outer boundaries of all Quick Response Codes in the candidate region；Dotted line coarse positioning subelement, for according to the length of " L " shape line segment and the position coordinates on three vertex, and positioning figure The position relationship of solid line outer boundary and dotted line outer boundary determines the rectangular area where two dotted line outer boundaries in shape；Two realities The region that line outer boundary and two dotted line outer boundaries enclose is candidate's two-dimension code area.
- 8. system as claimed in claim 7, it is characterised in that line segments extraction subelement is by region growth method from Quick Response Code Line segment is extracted in approximate region.
- 9. system as claimed in claim 7, it is characterised in that coarse localization unit further includes and screens subelement again, it is described again Subelement is screened to be used for behind the rectangular area where determining two dotted line outer boundaries respectively along the short of the rectangular area Edge direction searches angle point, and the angle point refers to the intersection point of two edge lines, judges angle point quantity and the area ratio of rectangular area Whether it is more than the 4th given threshold, if it is, retaining candidate's two-dimension code area to be accurately positioned to it, otherwise abandons Candidate's two-dimension code area, is no longer accurately positioned it.
- 10. system as claimed in claim 7, it is characterised in that being accurately positioned unit includes：Subelement is deviated, for being offset to both sides intended pixel from the rough position of solid line outer boundary and dotted line outer boundary respectively, Generate scanning area；Subelement is scanned, for scanning marginal point along predetermined direction in scanning area, and the marginal point of acquisition is grouped and is preserved；Subelement is fitted, for the marginal point of each group to be carried out fitting a straight line, respectively obtains solid line outer boundary and dotted line outer boundary Exact position；Intersecting point coordinate computation subunit, the position of four intersection points for calculating two solid line outer boundaries and two dotted line outer boundaries Coordinate.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610556584X | 2016-07-14 | ||
| CN201610556584 | 2016-07-14 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106485183A CN106485183A (en) | 2017-03-08 |
| CN106485183B true CN106485183B (en) | 2018-05-08 |
Family
ID=58268902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610857210.1A Active CN106485183B (en) | 2016-07-14 | 2016-09-27 | A kind of Quick Response Code localization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106485183B (en) |
Families Citing this family (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107301366B (en) * | 2017-05-12 | 2020-07-31 | 赵毅 | Decoding method and device for graphic code in embedded code video |
| CN109543484B (en) * | 2017-09-21 | 2023-02-07 | 杭州海康机器人股份有限公司 | Position information acquisition method and device based on two-dimensional code and electronic equipment |
| CN107908996B (en) * | 2017-10-25 | 2020-12-15 | 福建联迪商用设备有限公司 | Method and terminal for extracting one-dimensional bar code information |
| CN108174217B (en) * | 2017-12-06 | 2021-01-19 | 广州索答信息科技有限公司 | Image code generation and analysis method, electronic equipment and storage medium |
| CN108573184B (en) * | 2018-03-12 | 2021-04-27 | 青岛乾程科技股份有限公司 | Two-dimensional code positioning method, module and computer readable storage medium |
| CN108805033B (en) * | 2018-05-22 | 2020-12-18 | 中国科学院自动化研究所 | Shot image optimization method and device based on local gradient distribution |
| CN109101854A (en) * | 2018-06-25 | 2018-12-28 | 华南理工大学 | A kind of multiple barcode localization method |
| CN109086644B (en) * | 2018-07-27 | 2021-05-25 | 广东奥普特科技股份有限公司 | Positioning method of sub-pixel-precision DataMatrix two-dimensional code |
| CN108931982A (en) * | 2018-09-04 | 2018-12-04 | 赵凯雅 | Vision navigation system and method for robot moving equipment |
| CN109492562B (en) * | 2018-10-30 | 2021-05-25 | 河南科技大学 | Non-contact livestock identification system |
| CN109492453A (en) * | 2018-11-12 | 2019-03-19 | 凌云光技术集团有限责任公司 | A kind of DM code localization method and device based on geometry location |
| CN109409163B (en) * | 2018-11-12 | 2022-04-01 | 凌云光技术股份有限公司 | Quick QR code positioning method based on texture characteristics |
| CN110046528B (en) * | 2018-11-20 | 2022-10-28 | 维库（厦门）信息技术有限公司 | Dot-shaped DataMatrix two-dimensional code identification method |
| CN109740402B (en) * | 2019-01-02 | 2022-04-22 | 苏州天准科技股份有限公司 | Two-dimensional code direction and position obtaining method based on frequency spectrum information |
| CN109977714B (en) * | 2019-01-04 | 2022-06-10 | 上海大学 | Multi-QR-code integrated visual positioning method for warehoused goods |
| CN109858304B (en) * | 2019-01-04 | 2022-02-01 | 广州广电研究院有限公司 | Method and device for detecting two-dimensional code position detection graph and storage medium |
| CN110490020B (en) * | 2019-07-31 | 2023-01-13 | 珠海市杰理科技股份有限公司 | Bar code identification method and device, computer equipment and storage medium |
| CN110866578B (en) * | 2019-10-29 | 2021-07-06 | 上海交通大学 | Two-dimensional texture code and encoding and decoding methods thereof |
| CN111337011A (en) * | 2019-12-10 | 2020-06-26 | 亿嘉和科技股份有限公司 | Indoor positioning method based on laser and two-dimensional code fusion |
| CN111563930B (en) * | 2020-04-29 | 2023-07-07 | 达闼机器人股份有限公司 | Positioning method, device, medium, electronic equipment and auxiliary positioning module |
| CN111639507B (en) * | 2020-05-29 | 2023-06-09 | 天津维创微智能科技有限公司 | Positioning method and device for bar codes in image, code scanning equipment and storage medium |
| CN113947097B (en) * | 2020-07-15 | 2024-04-09 | 花瓣云科技有限公司 | Two-dimensional code identification method and electronic equipment |
| CN111815725B (en) * | 2020-07-29 | 2024-03-08 | 苏州中科全象智能科技有限公司 | QR code region positioning method |
| CN112101060B (en) * | 2020-10-28 | 2023-06-23 | 山东青橙视联信息科技有限公司 | Two-dimensional code positioning method based on translation invariance and small-area template matching |
| CN112598100A (en) * | 2020-12-23 | 2021-04-02 | 深兰人工智能芯片研究院（江苏）有限公司 | Method for determining specification of dot matrix type DM two-dimensional code and related device |
| CN112598652B (en) * | 2020-12-25 | 2024-01-30 | 凌云光技术股份有限公司 | Gradient transformation-based liquid crystal display edge broken line detection method |
| CN112733565A (en) * | 2020-12-29 | 2021-04-30 | 美的集团股份有限公司 | Two-dimensional code coarse positioning method, equipment and storage medium |
| CN112651259A (en) * | 2020-12-29 | 2021-04-13 | 芜湖哈特机器人产业技术研究院有限公司 | Two-dimensional code positioning method and mobile robot positioning method based on two-dimensional code |
| CN112800797B (en) * | 2020-12-30 | 2023-12-19 | 凌云光技术股份有限公司 | Region positioning method and system for DM code |
| CN113139400B (en) * | 2021-06-21 | 2021-10-15 | 浙江华睿科技股份有限公司 | QR code positioning method and device and computer readable medium |
| CN113536822B (en) * | 2021-07-28 | 2024-05-03 | 中移（杭州）信息技术有限公司 | Two-dimensional code correction method and device and computer readable storage medium |
| CN113688649A (en) * | 2021-08-16 | 2021-11-23 | 江苏博赛孚医疗科技有限公司 | Quick QR code positioning method |
| 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 |
| CN115131545A (en) * | 2022-07-14 | 2022-09-30 | 深圳市朗派电子科技有限公司 | Rapid rectangle positioning algorithm based on image recognition |
| CN114911435B (en) * | 2022-07-18 | 2022-11-11 | 北京中宏立达科技发展有限公司 | Copying control method and system based on two-dimensional bar code |
| CN116151290B (en) * | 2023-02-21 | 2023-10-24 | 广州玺明机械科技有限公司 | Information acquisition system and method for intelligent code scanning and milk tea modulation robot |
| CN117197422B (en) * | 2023-11-07 | 2024-03-26 | 深圳优艾智合机器人科技有限公司 | Identification code positioning method, electronic equipment and storage medium |
| CN117218328B (en) * | 2023-11-07 | 2024-03-08 | 深圳优艾智合机器人科技有限公司 | Two-dimensional code positioning method, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930268A (en) * | 2012-08-31 | 2013-02-13 | 西北工业大学 | Accurate positioning method for data matrix code under pollution and multi-view situation |
| CN104517089A (en) * | 2013-09-29 | 2015-04-15 | 北大方正集团有限公司 | Two-dimensional code decoding system and method |
| CN104636701A (en) * | 2014-12-12 | 2015-05-20 | 浙江工业大学 | Laser two-dimension code identification method based on image restoration |
-
2016
- 2016-09-27 CN CN201610857210.1A patent/CN106485183B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930268A (en) * | 2012-08-31 | 2013-02-13 | 西北工业大学 | Accurate positioning method for data matrix code under pollution and multi-view situation |
| CN104517089A (en) * | 2013-09-29 | 2015-04-15 | 北大方正集团有限公司 | Two-dimensional code decoding system and method |
| CN104636701A (en) * | 2014-12-12 | 2015-05-20 | 浙江工业大学 | Laser two-dimension code identification method based on image restoration |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106485183A (en) | 2017-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106485183B (en) | A kind of Quick Response Code localization method and system | |
| CN110717489B (en) | Method, device and storage medium for identifying text region of OSD (on Screen display) | |
| CN109711284A (en) | A kind of test answer sheet system intelligent recognition analysis method | |
| CN110348264B (en) | QR two-dimensional code image correction method and system | |
| CN102136058B (en) | Bar code image identification method | |
| CN102360421B (en) | Face identification method and system based on video streaming | |
| US20110290882A1 (en) | Qr code detection | |
| CN103543277B (en) | A kind of blood group result recognizer based on gray analysis and category identification | |
| CN109658424B (en) | Improved robust two-dimensional OTSU threshold image segmentation method | |
| US20080193020A1 (en) | Method for Facial Features Detection | |
| CN110443128A (en) | One kind being based on SURF characteristic point accurately matched finger vein identification method | |
| US20120269443A1 (en) | Method, apparatus, and program for detecting facial characteristic points | |
| CN104537376B (en) | One kind identification platform calibration method and relevant device, system | |
| CN110765992B (en) | Seal identification method, medium, equipment and device | |
| CN107945111A (en) | A kind of image split-joint method based on SURF feature extraction combination CS LBP descriptors | |
| CN110490913A (en) | Feature based on angle point and the marshalling of single line section describes operator and carries out image matching method | |
| CN108491786A (en) | A kind of method for detecting human face based on hierarchical network and Cluster merging | |
| WO2018040948A1 (en) | Digital object unique identifier (doi) recognition method and device | |
| CN108241860A (en) | A kind of QR codes detection localization method under complex environment | |
| CN108665495A (en) | Image processing method and device, mobile terminal | |
| CN114970590A (en) | Bar code detection method | |
| CN113435219B (en) | Anti-counterfeiting detection method and device, electronic equipment and storage medium | |
| CN104268550A (en) | Feature extraction method and device | |
| CN110210467A (en) | A kind of formula localization method, image processing apparatus, the storage medium of text image | |
| CN109977714A (en) | A kind of integrated vision positioning method of the more QR codes of warehoused cargo |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |