CN110415304A - A kind of vision calibration method and system - Google Patents
A kind of vision calibration method and system Download PDFInfo
- Publication number
- CN110415304A CN110415304A CN201910701453.XA CN201910701453A CN110415304A CN 110415304 A CN110415304 A CN 110415304A CN 201910701453 A CN201910701453 A CN 201910701453A CN 110415304 A CN110415304 A CN 110415304A
- Authority
- CN
- China
- Prior art keywords
- coordinate
- target
- calibration
- image
- quadrangle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06037—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking multi-dimensional coding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06046—Constructional details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The embodiment of the present application discloses a kind of vision calibration method and system, it include: that polyphaser obtains target image respectively, determine the first coordinate and the second coordinate of calibration point in each target image, that is coordinate of the index fixed point in image coordinate system and the coordinate in local target co-ordinates system, third coordinate is obtained, third coordinate refers to coordinate of the two dimensional code in image coordinate system in target image；Extract two-dimensional barcode information in target image, it include 4-coordinate in two-dimensional barcode information, 4-coordinate refers to coordinate of the current two-dimension in global target co-ordinates system, is Five Axis by the second coordinate modification, and Five Axis index pinpoints the coordinate in global target co-ordinates system；According to the first coordinate and Five Axis, the mapping relations between global target co-ordinates system and each image coordinate system are established.The target that two dimensional code is had using one is corrected the local target co-ordinates of calibration point automatically by two-dimensional barcode information as global target co-ordinates, realizes full automatic calibration, the calibration of polyphaser vision system.
Description
Technical field
This application involves vision calibration technical field, in particular to a kind of vision calibration method and system.
Background technique
With the continuous improvement of factory automation degree, a large amount of production and detection device require installation image vision system
It unites to improve the degree of automation of equipment.In production equipment, Image Vision System often be used to guide the movement control of equipment
Automated system operation processed, to adapt to the upper discharge position of processed product, situations such as size is inconsistent, such as in LCD industry
Cog, fog equipment etc..
In detection device, Image Vision System is often used for the dimensional measurement of converted products, assembly precision measures, outer
Detection etc. is seen, to ensure the qualification rate of final products.High-precision industrial production and detection device propose Image Vision System
High-precision and wide-field requirement, and one camera vision system is difficult to meet high-precision and wide-field requirement simultaneously, multiphase
Machine vision system can be realized simultaneously high-precision and wide-field requirement.
Polyphaser vision system generally requires by high-precision calibrating target the unification for realizing polyphaser coordinate system, with same
When meet high-precision and wide-field requirement.Each camera is established and the coordinate transformation relation of target co-ordinates system respectively, it can
Realize that multiple camera coordinates systems are unified to unique target co-ordinates system.
But existing polyphaser scaling method, it needs to extract X-comers coordinate first, is then manually entered calibration
Target co-ordinates corresponding to label in image establish the mapping relations of image coordinate system Yu target co-ordinates system.This calibration side
Method efficiency is lower, complicated for operation, needs manually to participate in, can not achieve the full automatic calibration of polyphaser vision system.
Summary of the invention
The application discloses a kind of vision calibration method and system, lower to solve scaling method efficiency in the prior art,
It is complicated for operation, need the problem of manually participating in, can not achieve the full automatic calibration of polyphaser vision system.
The application's in a first aspect, disclosing a kind of vision calibration method, comprising:
Polyphaser obtains target image respectively, wherein the target is equipped at least one two dimensional code, each camera
A complete two dimensional code is included at least within sweep of the eye；
Determine that the first coordinate and the second coordinate of calibration point in each target image, first coordinate refer to described
Coordinate of the calibration point in image coordinate system, second coordinate refer to seat of the calibration point in local target co-ordinates system
Mark, wherein the part target co-ordinates system is the coordinate system where the local target within the scope of presently described camera fields of view, described
Image coordinate system is the coordinate system where presently described camera；
Third coordinate is obtained, the third coordinate refers to that two dimensional code is in described image coordinate system in the target image
Coordinate；
Two-dimensional barcode information in the target image is extracted, includes 4-coordinate in the two-dimensional barcode information, the described 4th sits
Mark refers to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation target co-ordinates system is complete target
Coordinate system where marking；
According to first coordinate, second coordinate, the third coordinate and the 4-coordinate, described second is sat
Mark is modified to Five Axis, and the Five Axis refers to coordinate of the calibration point in the global target co-ordinates system；
According to first coordinate and the Five Axis, establishes the global target co-ordinates system and each described image is sat
Mapping relations between mark system.
Further, in each target image of the determination calibration point the first coordinate and the second coordinate, comprising:
Extract the first coordinate of all calibration points in each target image；
According to first coordinate of extraction, the topological relation in each target image between calibration point is determined；
According to the topological relation, the second coordinate of calibration point in each target image is determined.
Further, described to extract all calibration in each target image when the target is gridiron pattern target
First coordinate of point, comprising:
The pixel coordinate of the calibration point is calculated by Hessian matrix, wherein the calibration point is the angle point；
The subpixel coordinates of the calibration point are calculated using Taylor expansion, wherein the subpixel coordinates are described the
One coordinate.
Further, it when the target is gridiron pattern target, according to first coordinate of extraction, determines each described
Topological relation in target image between calibration point, comprising:
According to first coordinate of extraction, delaunay triangle is generated, wherein the calibration point is the angle point；
Adjacent and gray average is merged into quadrangle closest to the delaunay triangle；
The quadrangle is screened, wherein the quadrangle that is obtained after screening while meeting: the quadrangle
Diagonal line intersection point is inside presently described quadrangle, and the opposite side length ratio of the quadrangle is between section (0.75,1.25), institute
The adjacent side length ratio of quadrangle is stated between section (0.75,1.25)；
According to the quadrangle after screening, topological network is organized the formation of；
According to the quadrangle in the topological network position, determine second coordinate.
It further, is Five Axis by second coordinate modification, comprising:
Correct the X-axis and Y direction of the calibration point coordinate；
Correct the X-axis positive direction of the calibration point coordinate and the positive direction of Y-axis；
Correct the initial value of the calibration point coordinate.
Further, the target can be gridiron pattern target, cable lattice target or circular array target.
In the second aspect of the application, a kind of vision calibration system is disclosed, comprising:
Image collection module obtains target image for polyphaser respectively, wherein the target be equipped at least one two
Code is tieed up, includes at least a complete two dimensional code within the scope of each camera fields of view；
First determining module, for determining the first coordinate and the second coordinate of calibration point in each target image, institute
It states the first coordinate and refers to that coordinate of the calibration point in image coordinate system, second coordinate refer to the calibration point in part
Coordinate in target co-ordinates system, wherein the part target co-ordinates system is the local target within the scope of presently described camera fields of view
The coordinate system at place, described image coordinate system are the coordinate system where presently described camera；
Module is obtained, for obtaining third coordinate, the third coordinate refers to that two dimensional code is described in the target image
Coordinate in image coordinate system；
First extraction module includes for extracting two-dimensional barcode information in the target image, in the two-dimensional barcode information
4-coordinate, the 4-coordinate refer to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation target
Marking coordinate system is the coordinate system where complete target；
Correction module, for being sat according to first coordinate, second coordinate, the third coordinate and the described 4th
Mark, is Five Axis by second coordinate modification, and the Five Axis refers to the calibration point in the global target co-ordinates
Coordinate in system；
Establish module, for according to first coordinate and the Five Axis, establish the global target co-ordinates system with
Mapping relations between each described image coordinate system.
Further, first determining module, specifically includes:
Second extraction module, for extracting the first coordinate of all calibration points in each target image；
Second determining module determines calibration point in each target image for first coordinate according to extraction
Between topological relation；
Third determining module, for determining second of calibration point in each target image according to the topological relation
Coordinate.
Further, when the target is gridiron pattern target, the second extraction module is specifically included:
First computing module, for calculating the pixel coordinate of the calibration point by Hessian matrix, wherein the mark
Fixed point is the angle point；
Second computing module, for calculating the subpixel coordinates of the calibration point using Taylor expansion, wherein the Asia
Pixel coordinate is first coordinate.
Further, when the target is gridiron pattern target, the second determining module is specifically included:
Generation module generates delaunay triangle, wherein the calibration point for first coordinate according to extraction
For the angle point；
Merging module, for adjacent and gray average to be merged into quadrangle closest to the delaunay triangle；
Screening module, for being screened to the quadrangle, wherein the quadrangle that is obtained after screening while full
Foot: the diagonal line intersection point of the quadrangle is inside presently described quadrangle, and the opposite side length ratio of the quadrangle is between area
Between (0.75,1.25), the adjacent side length ratio of the quadrangle is between section (0.75,1.25)；
Molded tissue block, for organizing the formation of topological network according to the quadrangle after screening；
4th determining module, for according to the quadrangle in the topological network position, determine second coordinate.
Vision calibration method disclosed in the present application and system are had the target of two dimensional code using one, are believed by two dimensional code
The local target co-ordinates of the automatic amendment calibration point of breath are global target co-ordinates, realize polyphaser vision system full automatic calibration,
Calibration, in addition, in the application each camera within sweep of the eye include a two dimensional code, that is to say, that each camera
The visual field do not need to design it is excessive, to ensure that one camera pixel precision.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without any creative labor,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of workflow schematic diagram of vision calibration method disclosed in the embodiment of the present application；
Fig. 2 is the schematic diagram of target disclosed in the embodiment of the present application；
Fig. 3 is the schematic diagram that polyphaser disclosed in the embodiment of the present application obtains target image；
Fig. 4 is the workflow schematic diagram of another vision calibration method disclosed in the embodiment of the present application；
Fig. 5 is the workflow schematic diagram of another vision calibration method disclosed in the embodiment of the present application；
Fig. 6 is the workflow schematic diagram of another vision calibration method disclosed in the embodiment of the present application；
Fig. 7 is the disclosed result schematic diagram for generating delaunay triangle of the embodiment of the present application；
Fig. 8, which is that the embodiment of the present application is disclosed, merges the result schematic diagram that adjacent triangle is quadrangle；
Fig. 9 is the result schematic diagram after the disclosed screening quadrangle of the embodiment of the present application；
Figure 10 is the result schematic diagram of the second coordinate determined disclosed in the embodiment of the present application
Figure 11 is the disclosed result schematic diagram for extracting two-dimensional barcode information of the embodiment of the present application；
Figure 12 is the result schematic diagram that the second coordinate modification is Five Axis disclosed in the embodiment of the present application；
Figure 13 is the schematic diagram that mapping relations are established disclosed in the embodiment of the present application；
Figure 14 is a kind of structural block diagram of vision calibration system disclosed in the embodiment of the present application；
Figure 15 is the structural block diagram of another vision calibration system disclosed in the embodiment of the present application；
Figure 16 is the structural block diagram of another vision calibration system disclosed in the embodiment of the present application；
Figure 17 is the structural block diagram of another vision calibration system disclosed in the embodiment of the present application.
Specific embodiment
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with the application reality
The attached drawing in example is applied, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described implementation
Example is merely a part but not all of the embodiments of the present application.Based on the embodiment in the application, this field is common
The application protection all should belong in technical staff's every other embodiment obtained without making creative work
Range.
Existing polyphaser scaling method needs to extract X-comers coordinate first, is then manually entered uncalibrated image
In label corresponding to target co-ordinates, establish the mapping relations of image coordinate system Yu target co-ordinates system.This scaling method effect
Rate is lower, complicated for operation, needs manually to participate in, can not achieve the full automatic calibration of polyphaser vision system.It is above-mentioned in order to solve
Technical problem, the embodiment of the present application disclose a kind of vision calibration method and system.
The embodiment of the present application one discloses a kind of vision calibration method, referring to Fig.1, includes the following steps:
Step S10, polyphaser obtains target image respectively, wherein the target is equipped at least one two dimensional code, each
A complete two dimensional code is included at least within the scope of the camera fields of view.
Target in the application is the target with two dimensional code, as shown in Fig. 2, target can be gridiron pattern target, cable
Lattice target or circular array target, the two dimensional code on target can be arranged according to array type as shown in Figure 2, can also be according to it
He arranges mode, and the application is not construed as limiting this.
In design target, by taking gridiron pattern target as an example, two dimensional code is rectangle, and the center of two dimensional code accurately covers chessboard
Some angle point in lattice.Preferably, two dimensional code occupies four rectangle positions in gridiron pattern target.
Wherein, two dimensional code can be DM two dimensional code or QR two dimensional code, include current two-dimension in each two dimensional code complete
Location information (coordinate) in chessboard target, for example first, upper left corner two dimensional code includes location information (5,5) in Fig. 2, that is, is worked as
Preceding two dimensional code center corresponds to the column of the 5th row the 5th in complete chessboard target.
As shown in figure 3, four cameras obtain target image respectively, a part of each camera shooting target is each described
A complete two dimensional code is all included at least in the image that camera obtains.
Step S11, the first coordinate and the second coordinate of calibration point in each target image, first coordinate are determined
Refer to that coordinate of the calibration point in image coordinate system, second coordinate refer to the calibration point in local target co-ordinates system
In coordinate, wherein it is described part target co-ordinates system be presently described camera fields of view within the scope of local target where coordinate
System, described image coordinate system are the coordinate system where presently described camera.
Target can be gridiron pattern target, cable lattice target or circular array target, corresponding if target is gridiron pattern target
Calibration point be X-comers, if target be cable lattice target, corresponding calibration point be cable crosspoint, if target
For circular array target, corresponding calibration point is dot center.
In order to guarantee the high-precision of shooting, a part of each camera shooting target, local target co-ordinates system is current phase
Coordinate system where the local target of machine shooting.
Wherein, as shown in figure 4, determining the specific of the first coordinate of calibration point and the second coordinate in each target image
Implementation can be with S110-S112 according to the following steps:
Step S110, the first coordinate of all calibration points in each target image is extracted.
By taking four cameras obtain target image respectively as an example, all calibration points in each target image are extracted respectively and are existed
Coordinate in corresponding image coordinate system.Such as, current target image is first camera shooting, then extracts current target figure
Coordinate of all calibration points in the corresponding image coordinate system of first camera as in.
When the target is gridiron pattern target, as shown in figure 5, described extract all calibration in each target image
The specific implementation of first coordinate of point, can be with S1100-S1101 according to the following steps:
Step S1100, the pixel coordinate of the calibration point is calculated by Hessian matrix, wherein the calibration point is institute
State angle point.
The Hessian matrix of image pixel is calculated by gaussian filtering:
Wherein, fxx、fxy、fyyRespectively second-order partial differential coefficient of the present image gray scale relative to x, y.The two of Hessian matrix
A eigenvalue λ1And λ2It respectively corresponds as the gradient value in the gradient value and its normal direction on the most violent direction of grey scale change.
For the corresponding angle point of gridiron pattern, judgment criterion are as follows:
Wherein, t is Grads threshold.
Step S1101, the subpixel coordinates of the calibration point are calculated using Taylor expansion, wherein the sub-pix is sat
It is designated as first coordinate.
The sub-pixel location of X-comers is further calculated by Taylor expansion, Taylor expansion can be expressed as follows:
Wherein, (x0, y0) be the corner pixels coordinate that is calculated by Hessian matrix, (s, t) be angle point relative to
(x0, y0) sub-pix offset, f0For current pixel gray value, fx、fy、fxx、fxy、fyyThe respectively single order local derviation of current pixel
Several and second-order partial differential coefficient.The extreme point of neighborhood image intensity profile is corresponded to according to the subpixel coordinates of angle point, can obtain following property
Matter
Above-mentioned equation group is solved, can be obtained
The subpixel coordinates of X-comers are (x0+ s, y0+t)。
Above-mentioned steps S1100-S1101, extract all calibration points in each target image first sit calibration method,
On the one hand corner extraction relative to traditional Harris, does not need interpolation and surface fitting, and extraction process is simple, speed
Fastly；On the other hand the method that is found intersection relative to edge fitting, only using the image grayscale in angle point close region in the application,
Therefore, even if throwing away extraction accuracy with higher in the case where image has distortion and larger noise.
Step S111, according to first coordinate of extraction, opening up between calibration point in each target image is determined
Flutter relationship.
According to the first coordinate of calibration points all in the target image of extraction, calibration point in each target image is determined
Between topological relation, that is, the neighbouring relations of each calibration point Yu other calibration points have been determined.
Step S112, according to the topological relation, the second coordinate of calibration point in each target image is determined.
The neighbouring relations of each calibration point Yu other calibration points have been determined, also just can determine that each target image acceptance of the bid
Pinpoint the coordinate in local target co-ordinates system.
When the target be gridiron pattern target when, as shown in fig. 6, in each target image of the determination calibration point it
Between topological relation specific implementation, can be with S1110-S1114 according to the following steps:
Step S1110, according to first coordinate of extraction, delaunay triangle is generated, as shown in Figure 7, wherein described
Calibration point is the angle point.
Delaunay triangle Shape definition: if the circumscribed circle of a triangle does not include other any points in plane, this
A triangle is known as delaunay triangle.
Step S1111, adjacent and gray average is merged into quadrangle closest to the delaunay triangle, such as Fig. 8 institute
Show.
It is as follows to merge rule: selecting a delaunay triangle, searches for all adjacent morals of selected delaunay triangle
Lip river interior triangular takes and merges therewith with the immediate delaunay triangle of its gray average.
As shown in figure 8, due in two dimensional code also containing white area after black, after above-mentioned merging compatible rule merging, meeting
There are some quadrangles for being unsatisfactory for requiring, so needing further to screen quadrangle.
Step S1112, the quadrangle is screened, as shown in Figure 9, wherein the quadrangle obtained after screening
Meet simultaneously: the diagonal line intersection point of the quadrangle is inside presently described quadrangle, the opposite side length ratio of the quadrangle
Between section (0.75,1.25), the adjacent side length ratio of the quadrangle is between section (0.75,1.25).
Step S1113, according to the quadrangle after screening, topological network is organized the formation of.
Quadrangle after screening links together to form topological network.
Step S1114, according to the quadrangle in the topological network position, determine second coordinate, such as Figure 10
It is shown.
It is sat according to the X-comers that each quadrangle vertex correspondence can be generated in position of the quadrangle in topological network
Mark.
Step S12, third coordinate is obtained, the third coordinate refers to that two dimensional code is sat in described image in the target image
Coordinate in mark system.
Determine two-dimension code area, the position of positioning two-dimension code area center in the picture indicates two dimension with a quadrangle
Code region, obtains third coordinate.
Step S13, two-dimensional barcode information in the target image is extracted, includes 4-coordinate, institute in the two-dimensional barcode information
It states 4-coordinate and refers to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation target co-ordinates system
For the coordinate system where complete target.
Two dimensional code in target image is decoded, two-dimensional barcode information is extracted, includes presently described two in two-dimensional barcode information
Tie up coordinate of the code in global target co-ordinates system.As shown in figure 11, coordinate of the two dimensional code in global target co-ordinates system is
(0.00,10.00), two dimensional code are seat of the two dimensional code center in global target co-ordinates system in the coordinate in global target co-ordinates system
Mark.
4-coordinate is recorded in two dimensional code, without manually marking, is solved corresponding two dimensional code when needing
Code.
Step S14, according to first coordinate, second coordinate, the third coordinate and the 4-coordinate, by institute
Stating the second coordinate modification is Five Axis, and the Five Axis refers to seat of the calibration point in the global target co-ordinates system
Mark.
According to third coordinate and the 4-coordinate, can obtain two dimensional code in image coordinate system coordinate in global target
The corresponding relationship for marking the coordinate in coordinate system, further according to the first coordinate and second coordinate, correcting the second coordinate is the 5th seat
The coordinate of mark, i.e. acquisition calibration point in the global target co-ordinates system.
Wherein, the second coordinate of amendment is Five Axis, comprising: corrects the X-axis and Y direction of the calibration point coordinate；It repairs
The X-axis positive direction of the just described calibration point coordinate and the positive direction of Y-axis；Correct the initial value of the calibration point coordinate.
As shown in figure 11, coordinate of the two dimensional code of acquisition in global target co-ordinates system is (0.00,10.00), will be such as figure
Coordinate modification of all angle points shown in 10 in local target co-ordinates system is all angle points as shown in figure 12 in global target
Coordinate in coordinate system.Such as: in target image, coordinate of the upper left corner angle point in local target co-ordinates system be (0.00,
0.00), after amendment, coordinate of the upper left corner angle point in global target co-ordinates system is (- 4.00,14.00).
Step S15, according to first coordinate and the Five Axis, the global target co-ordinates system and each institute are established
State the mapping relations between image coordinate system.
The corresponding image coordinate system of each camera unifies the corresponding image coordinate system of each camera and global target
The corresponding mapping relations of establishment of coordinate system, to realize polyphaser calibration, calibration.
Wherein, the mapping relations between global target co-ordinates system and image coordinate system, i.e. solution mapping matrix are established.
Mapping relations between two spaces plane can be indicated with homography matrix, should be singly general in projective geometry
It reads, also known as projective transformation.It is mapped to the point (three-dimensional homogeneous vector) on a projective plane on another projective plane,
And straight line is mapped as straight line, has and protects linear matter.
Coordinate of the same calibration point in image coordinate system is (u, v), coordinate in global target co-ordinates system be (x,
Y), the homography matrix of two coordinate systems is H, then has
Wherein, h11、h12、h21、h22Indicate the matrix of a linear transformation, h13、、h23Indicate translation vector, h31、h32Indicate perspective
Converted quantity.
It is found that homography matrix H there are 8 freedom degrees from above-mentioned formula, therefore, 4 groups of corresponding marks are theoretically at least only needed
Coordinate is pinpointed, homography matrix H can be solved.
As shown in figure 13, corresponding first coordinate of tetra- calibration points of A, B, C, D is obtained respectively and Five Axis, utilization are above-mentioned
Homography matrix formula is calculated, homography matrix H can be found out.
Vision calibration method disclosed in the embodiment of the present application, first polyphaser obtain target image respectively, wherein the target
It puts on and is equipped at least one two dimensional code, include at least a complete two dimensional code within the scope of each camera fields of view；Then really
The first coordinate and the second coordinate of calibration point, first coordinate refer to that the calibration point is being schemed in fixed each target image
As the coordinate in coordinate system, second coordinate refers to coordinate of the calibration point in local target co-ordinates system, wherein described
Local target co-ordinates system is the coordinate system where the local target within the scope of presently described camera fields of view, and described image coordinate system is
Coordinate system where presently described camera；Third coordinate is obtained again, and the third coordinate refers to two dimensional code in the target image
Coordinate in described image coordinate system；Two-dimensional barcode information in the target image is extracted again, includes in the two-dimensional barcode information
4-coordinate, the 4-coordinate refer to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation
Target co-ordinates system is the coordinate system where complete target；Then it is sat according to first coordinate, second coordinate, the third
Mark and the 4-coordinate, are Five Axis by second coordinate modification, and the Five Axis refers to the calibration point in institute
State the coordinate in global target co-ordinates system；Finally according to first coordinate and the Five Axis, the global target is established
Mapping relations between coordinate system and each described image coordinate system.
Vision calibration method disclosed in the present application has the target of two dimensional code using one, automatic by two-dimensional barcode information
The local target co-ordinates for correcting calibration point are global target co-ordinates, realize full automatic calibration, the calibration of polyphaser vision system, separately
Outside, in the application each camera within sweep of the eye include a two dimensional code, that is to say, that the visual field of each camera is not required to
Design it is excessive, to ensure that one camera pixel precision.
Correspondingly, referring to Fig.1 4, in an alternative embodiment of the invention, a kind of vision calibration system is also disclosed, comprising:
Image collection module 110 obtains target image for polyphaser respectively, wherein the target is equipped at least one
A two dimensional code includes at least a complete two dimensional code within the scope of each camera fields of view；
First determining module 120, for determining the first coordinate and the second coordinate of calibration point in each target image,
First coordinate refers to that coordinate of the calibration point in image coordinate system, second coordinate refer to the calibration point in office
Coordinate in target co-ordinates system, portion, wherein the part target co-ordinates system is the local target within the scope of presently described camera fields of view
Coordinate system where marking, described image coordinate system are the coordinate system where presently described camera；
Module 130 is obtained, for obtaining third coordinate, the third coordinate refers to that two dimensional code is in institute in the target image
State the coordinate in image coordinate system；
First extraction module 140 includes in the two-dimensional barcode information for extracting two-dimensional barcode information in the target image
4-coordinate, the 4-coordinate refer to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation
Target co-ordinates system is the coordinate system where complete target；
Correction module 150, for according to first coordinate, second coordinate, the third coordinate and the described 4th
Coordinate, is Five Axis by second coordinate modification, and the Five Axis refers to that the calibration point is sat in the global target
Coordinate in mark system；
Module 160 is established, for establishing the global target co-ordinates system according to first coordinate and the Five Axis
With the mapping relations between each described image coordinate system.
Further, referring to Fig.1 5, first determining module 120 specifically includes:
Second extraction module 210, for extracting the first coordinate of all calibration points in each target image；
Second determining module 220 is determined and is demarcated in each target image for first coordinate according to extraction
Topological relation between point；
Third determining module 230 determines the of calibration point in each target image for according to the topological relation
Two coordinates.
Further, referring to Fig.1 6, when the target is gridiron pattern target, the second extraction module 210 is specifically included:
First computing module 310, for calculating the pixel coordinate of the calibration point by Hessian matrix, wherein described
Calibration point is the angle point；
Second computing module 320, for calculating the subpixel coordinates of the calibration point using Taylor expansion, wherein institute
Stating subpixel coordinates is first coordinate.
Further, referring to Fig.1 7, when the target is gridiron pattern target, the second determining module 220 is specifically included:
Generation module 410 generates delaunay triangle, wherein the calibration for first coordinate according to extraction
Point is the angle point；
Merging module 420, for adjacent and gray average to be merged into quadrangle closest to the delaunay triangle；
Screening module 430, for being screened to the quadrangle, wherein the quadrangle obtained after screening is simultaneously
Meet: the diagonal line intersection point of the quadrangle inside presently described quadrangle, the opposite side length ratio of the quadrangle between
Section (0.75,1.25), the adjacent side length ratio of the quadrangle is between section (0.75,1.25)；
Molded tissue block 440, for organizing the formation of topological network according to the quadrangle after screening；
4th determining module 450, for according to the quadrangle in the topological network position, determine it is described second sit
Mark.
Same and similar part may refer to each other between each embodiment in this specification.Especially for the reality of system
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
In explanation.
Combine detailed description and exemplary example that the application is described in detail above, but these explanations are simultaneously
It should not be understood as the limitation to the application.It will be appreciated by those skilled in the art that without departing from the application spirit and scope,
A variety of equivalent substitution, modification or improvements can be carried out to technical scheme and embodiments thereof, these each fall within the application
In the range of.The protection scope of the application is determined by the appended claims.
In the specific implementation, the embodiment of the present application also provides a kind of computer readable storage medium, wherein this is computer-readable
Storage medium can be stored with program, which may include in each embodiment of vision calibration method provided by the present application when executing
Part or all of step.The storage medium can for magnetic disk, CD, read-only memory (read-only memory,
) or random access memory (random access memory, RAM) etc. ROM.
It is required that those skilled in the art can be understood that the technology in the embodiment of the present application can add by software
The mode of general hardware platform realize.Based on this understanding, the technical solution in the embodiment of the present application substantially or
Say that the part that contributes to existing technology can be embodied in the form of software products, which can deposit
Storage is in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that computer equipment (can be with
It is personal computer, server or the network equipment etc.) execute certain part institutes of each embodiment of the application or embodiment
The method stated.
Above-described the application embodiment does not constitute the restriction to the application protection scope.
Claims (10)
1. a kind of vision calibration method characterized by comprising
Polyphaser obtains target image respectively, wherein the target is equipped at least one two dimensional code, each camera fields of view
A complete two dimensional code is included at least in range；
Determine that the first coordinate and the second coordinate of calibration point in each target image, first coordinate refer to the calibration
Coordinate of the point in image coordinate system, second coordinate refer to coordinate of the calibration point in local target co-ordinates system,
In, the part target co-ordinates system is the coordinate system where the local target within the scope of presently described camera fields of view, described image
Coordinate system is the coordinate system where presently described camera；
Third coordinate is obtained, the third coordinate refers to seat of the two dimensional code in described image coordinate system in the target image
Mark；
Two-dimensional barcode information in the target image is extracted, includes 4-coordinate in the two-dimensional barcode information, the 4-coordinate is
Refer to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation target co-ordinates system is complete target institute
Coordinate system；
According to first coordinate, second coordinate, the third coordinate and the 4-coordinate, second coordinate is repaired
It is just Five Axis, the Five Axis refers to coordinate of the calibration point in the global target co-ordinates system；
According to first coordinate and the Five Axis, the global target co-ordinates system and each described image coordinate system are established
Between mapping relations.
2. vision calibration method according to claim 1, which is characterized in that each target image acceptance of the bid of determination
The first coordinate and the second coordinate of fixed point, comprising:
Extract the first coordinate of all calibration points in each target image；
According to first coordinate of extraction, the topological relation in each target image between calibration point is determined；
According to the topological relation, the second coordinate of calibration point in each target image is determined.
3. vision calibration method according to claim 2, which is characterized in that when the target is gridiron pattern target, institute
State the first coordinate for extracting all calibration points in each target image, comprising:
The pixel coordinate of the calibration point is calculated by Hessian matrix, wherein the calibration point is the angle point；
The subpixel coordinates of the calibration point are calculated using Taylor expansion, wherein the subpixel coordinates are first seat
Mark.
4. vision calibration method according to claim 2, which is characterized in that when the target is gridiron pattern target, root
According to first coordinate of extraction, the topological relation in each target image between calibration point is determined, comprising:
According to first coordinate of extraction, delaunay triangle is generated, wherein the calibration point is the angle point；
Adjacent and gray average is merged into quadrangle closest to the delaunay triangle；
The quadrangle is screened, wherein the quadrangle that is obtained after screening while meeting: the quadrangle it is diagonal
Line intersection point is inside presently described quadrangle, and the opposite side length ratio of the quadrangle is between section (0.75,1.25), and described four
The adjacent side length ratio of side shape is between section (0.75,1.25)；
According to the quadrangle after screening, topological network is organized the formation of；
According to the quadrangle in the topological network position, determine second coordinate.
5. vision calibration method according to claim 1, which is characterized in that sit second coordinate modification for the 5th
Mark, comprising:
Correct the X-axis and Y direction of the calibration point coordinate；
Correct the X-axis positive direction of the calibration point coordinate and the positive direction of Y-axis；
Correct the initial value of the calibration point coordinate.
6. vision calibration method according to claim 1, which is characterized in that the target is gridiron pattern target, cable lattice
Target or circular array target.
7. a kind of vision calibration system characterized by comprising
Image collection module obtains target image for polyphaser respectively, wherein the target is equipped at least one two dimension
Yard, a complete two dimensional code is included at least within the scope of each camera fields of view；
First determining module, for determining the first coordinate and the second coordinate of calibration point in each target image, described
One coordinate refers to that coordinate of the calibration point in image coordinate system, second coordinate refer to the calibration point in local target
Coordinate in coordinate system, wherein the part target co-ordinates system is the local target place within the scope of presently described camera fields of view
Coordinate system, described image coordinate system be presently described camera where coordinate system；
Module is obtained, for obtaining third coordinate, the third coordinate refers to that two dimensional code is in described image in the target image
Coordinate in coordinate system；
First extraction module includes the 4th seat in the two-dimensional barcode information for extracting two-dimensional barcode information in the target image
Mark, the 4-coordinate refer to coordinate of the presently described two dimensional code in global target co-ordinates system, wherein the overall situation target is sat
Mark system is the coordinate system where complete target；
Correction module is used for according to first coordinate, second coordinate, the third coordinate and the 4-coordinate, will
Second coordinate modification is Five Axis, and the Five Axis refers to the calibration point in the global target co-ordinates system
Coordinate；
Module is established, for according to first coordinate and the Five Axis, establishing the global target co-ordinates system and each
Mapping relations between described image coordinate system.
8. vision calibration system according to claim 7, which is characterized in that first determining module specifically includes:
Second extraction module, for extracting the first coordinate of all calibration points in each target image；
Second determining module determines in each target image between calibration point for first coordinate according to extraction
Topological relation；
Third determining module, for determining the second coordinate of calibration point in each target image according to the topological relation.
9. vision calibration system according to claim 8, which is characterized in that when the target is gridiron pattern target, the
Two extraction modules, specifically include:
First computing module, for calculating the pixel coordinate of the calibration point by Hessian matrix, wherein the calibration point
For the angle point；
Second computing module, for calculating the subpixel coordinates of the calibration point using Taylor expansion, wherein the sub-pix
Coordinate is first coordinate.
10. vision calibration system according to claim 8, which is characterized in that when the target is gridiron pattern target, the
Two determining modules, specifically include:
Generation module generates delaunay triangle, wherein the calibration point is institute for first coordinate according to extraction
State angle point；
Merging module, for adjacent and gray average to be merged into quadrangle closest to the delaunay triangle；
Screening module, for being screened to the quadrangle, wherein the quadrangle that is obtained after screening while meeting: institute
The diagonal line intersection point of quadrangle is stated inside presently described quadrangle, the opposite side length ratio of the quadrangle is between section
(0.75,1.25), the adjacent side length ratio of the quadrangle is between section (0.75,1.25)；
Molded tissue block, for organizing the formation of topological network according to the quadrangle after screening；
4th determining module, for according to the quadrangle in the topological network position, determine second coordinate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910701453.XA CN110415304B (en) | 2019-07-31 | 2019-07-31 | Vision calibration method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910701453.XA CN110415304B (en) | 2019-07-31 | 2019-07-31 | Vision calibration method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110415304A true CN110415304A (en) | 2019-11-05 |
| CN110415304B CN110415304B (en) | 2023-03-03 |
Family
ID=68364685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910701453.XA Active CN110415304B (en) | 2019-07-31 | 2019-07-31 | Vision calibration method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110415304B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179360A (en) * | 2020-04-13 | 2020-05-19 | 杭州利珀科技有限公司 | High-precision automatic calibration plate and calibration method |
| CN111986267A (en) * | 2020-08-20 | 2020-11-24 | 佛山隆深机器人有限公司 | Coordinate system calibration method of multi-camera vision system |
| CN112381893A (en) * | 2021-01-13 | 2021-02-19 | 中国人民解放军国防科技大学 | Three-dimensional calibration plate calibration method for annular multi-camera system |
| CN112651261A (en) * | 2020-12-30 | 2021-04-13 | 凌云光技术股份有限公司 | Calculation method for conversion relation between high-precision 2D camera coordinate system and mechanical coordinate system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101419709A (en) * | 2008-12-08 | 2009-04-29 | 北京航空航天大学 | Plane target drone characteristic point automatic matching method for demarcating video camera |
| US9230326B1 (en) * | 2012-12-31 | 2016-01-05 | Cognex Corporation | System, method and calibration plate employing embedded 2D data codes as self-positioning fiducials |
| CN107270810A (en) * | 2017-04-28 | 2017-10-20 | 深圳大学 | The projector calibrating method and device of multi-faceted projection |
| CN108571971A (en) * | 2018-05-17 | 2018-09-25 | 北京航空航天大学 | A kind of AGV vision positioning systems and method |
| CN109099883A (en) * | 2018-06-15 | 2018-12-28 | 哈尔滨工业大学 | The big visual field machine vision metrology of high-precision and caliberating device and method |
| US20190122388A1 (en) * | 2017-04-17 | 2019-04-25 | Cognex Corporation | High-accuracy calibration system and method |
| CN109920009A (en) * | 2019-03-13 | 2019-06-21 | 武汉汉宁轨道交通技术有限公司 | Control point detection and management method and device based on two dimensional code mark |
-
2019
- 2019-07-31 CN CN201910701453.XA patent/CN110415304B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101419709A (en) * | 2008-12-08 | 2009-04-29 | 北京航空航天大学 | Plane target drone characteristic point automatic matching method for demarcating video camera |
| US9230326B1 (en) * | 2012-12-31 | 2016-01-05 | Cognex Corporation | System, method and calibration plate employing embedded 2D data codes as self-positioning fiducials |
| US20190122388A1 (en) * | 2017-04-17 | 2019-04-25 | Cognex Corporation | High-accuracy calibration system and method |
| CN107270810A (en) * | 2017-04-28 | 2017-10-20 | 深圳大学 | The projector calibrating method and device of multi-faceted projection |
| CN108571971A (en) * | 2018-05-17 | 2018-09-25 | 北京航空航天大学 | A kind of AGV vision positioning systems and method |
| CN109099883A (en) * | 2018-06-15 | 2018-12-28 | 哈尔滨工业大学 | The big visual field machine vision metrology of high-precision and caliberating device and method |
| CN109920009A (en) * | 2019-03-13 | 2019-06-21 | 武汉汉宁轨道交通技术有限公司 | Control point detection and management method and device based on two dimensional code mark |
Non-Patent Citations (3)
| Title |
|---|
| MARK FIALA: "ARTag,a fiducial marker system using digital techniques", 《2005 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITON》 * |
| SERGIO LEANDRO ALVES DA SILVA等: "Using ARUCO coded targets for camera calibration automation", 《HTTP://DS.DOI.ORG/10.1590/S1982-21702014000300036》 * |
| 谭海曙等: "摄像机标定中特征点的一种自动对应方法", 《光电子.激光》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179360A (en) * | 2020-04-13 | 2020-05-19 | 杭州利珀科技有限公司 | High-precision automatic calibration plate and calibration method |
| CN111986267A (en) * | 2020-08-20 | 2020-11-24 | 佛山隆深机器人有限公司 | Coordinate system calibration method of multi-camera vision system |
| CN111986267B (en) * | 2020-08-20 | 2024-02-20 | 佛山隆深机器人有限公司 | Coordinate system calibration method of multi-camera vision system |
| CN112651261A (en) * | 2020-12-30 | 2021-04-13 | 凌云光技术股份有限公司 | Calculation method for conversion relation between high-precision 2D camera coordinate system and mechanical coordinate system |
| CN112651261B (en) * | 2020-12-30 | 2023-11-10 | 凌云光技术股份有限公司 | Calculation method for conversion relation between high-precision 2D camera coordinate system and mechanical coordinate system |
| CN112381893A (en) * | 2021-01-13 | 2021-02-19 | 中国人民解放军国防科技大学 | Three-dimensional calibration plate calibration method for annular multi-camera system |
| CN112381893B (en) * | 2021-01-13 | 2021-04-20 | 中国人民解放军国防科技大学 | Three-dimensional calibration plate calibration method for annular multi-camera system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110415304B (en) | 2023-03-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110415304A (en) | A kind of vision calibration method and system | |
| JP4531069B2 (en) | Method and system for determining a data model designed to be superimposed with an image of a real object | |
| US7596468B2 (en) | Method for measuring a selected portion of a curved surface of an object | |
| CN110047109A (en) | A kind of camera calibration plate and its recognition detection method based on self-identifying label | |
| CN106251353A (en) | Weak texture workpiece and the recognition detection method and system of three-dimensional pose thereof | |
| Huang et al. | A fast and flexible projector-camera calibration system | |
| CN101751672A (en) | Image processing system | |
| CN110096920A (en) | A kind of high-precision high-speed positioning label and localization method towards visual servo | |
| CN105303615A (en) | Combination method of two-dimensional stitching and three-dimensional surface reconstruction of image | |
| Atsushi et al. | System for reconstruction of three-dimensional micro objects from multiple photographic images | |
| CN110246161B (en) | Method for seamless splicing of 360-degree panoramic images | |
| CN109754436B (en) | Camera calibration method based on lens partition area distortion function model | |
| JP2012507067A (en) | Join feature boundaries | |
| CN106952262B (en) | Ship plate machining precision analysis method based on stereoscopic vision | |
| JP2012085026A (en) | Image processing device and image processing method | |
| CN104537627A (en) | Depth image post-processing method | |
| CN110619661A (en) | Method for measuring volume of outdoor stock ground raw material based on augmented reality | |
| CN108898679A (en) | A kind of method of component serial number automatic marking | |
| CN111968182B (en) | Calibration method for nonlinear model parameters of binocular camera | |
| CN115423746B (en) | Image processing method for calculating skin hole site and aperture | |
| CN116402904A (en) | Combined calibration method based on laser radar inter-camera and monocular camera | |
| CN112541943A (en) | Robot positioning method based on visual road signs | |
| CN111260723B (en) | Barycenter positioning method of bar and terminal equipment | |
| Gu et al. | A new optimal seam selection method for airborne image stitching | |
| CN109242910A (en) | A kind of monocular camera self-calibrating method based on any known flat shape |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | 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 |