CN103913131B - Free curve method vector measurement method based on binocular vision - Google Patents
Free curve method vector measurement method based on binocular vision Download PDFInfo
- Publication number
- CN103913131B CN103913131B CN201410149149.6A CN201410149149A CN103913131B CN 103913131 B CN103913131 B CN 103913131B CN 201410149149 A CN201410149149 A CN 201410149149A CN 103913131 B CN103913131 B CN 103913131B
- Authority
- CN
- China
- Prior art keywords
- point
- plane
- image
- spot center
- spot
- 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
- 239000013598 vector Substances 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000000691 measurement method Methods 0.000 title abstract 5
- 238000005259 measurement Methods 0.000 claims abstract description 20
- 230000005484 gravity Effects 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 9
- 241000287196 Asthenes Species 0.000 claims description 3
- 238000006467 substitution reaction Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000013519 translation Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Landscapes
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to a free curve method vector measurement method based on the binocular vision, and belongs to the field of computer vision measurement. According to the measurement method, a projection pattern from a laser projection device to a free curve is composed of two orthogonal straight lines and four round light spots on the two straight lines; the two straight lines are intersected at one point; the four round light spots are the same in size and are evenly distributed on the same circumference. The image of the projection pattern is collected by a binocular vision system, through the threshold judgment condition based on the distance, the curvature of one point of the curve within the small neighborhood range is estimated, and different normal vector measurement schemes are selected. The measurement method considers the curvature of the neighborhood of the point to be tested, the measurement flexibility is high, the adaptability is high, and online efficient measurement of a normal vector of any point of the free curve can be achieved. The measurement method is simple, and the algorithm is easy to implement.
Description
Technical field
The invention belongs to computer vision measurement field, is related to a kind of surface normal vector measuring based on binocular vision
Method.
Background technology
Curved surface parts such as turbo blade, loudspeaker, curved surface cavity, train covering etc. are increasingly becoming each application product
Indispensable important composition part.For these parts, surface normal vector becomes important measurement parameter.Covering is used as aircraft allusion quotation
Type composite curved surface part has that wall is thin, complex-shaped, random deformation big, material character anisotropy, size range are big etc.
Feature.Covering flexible drilling riveter skills have high requirements to boring riveting perpendicularity, and during the drilling of aircraft skin riveting parts, it is prefabricated by processing
Part lays, hot-press solidifying and the deformation caused by part self gravitation and positioning, coordination, when clamping the accumulation of various errors make
Obtain and actually treat that law vector produces deviation with theoretical law vector in three-dimensional digital model at riveting point.If riveting point normal direction precision exceeds
Scope, can make hole machined Quality Down, type of attachment to force connection, junction generation stress concentration and skin-surface not only
It is sliding, and then riveting quality and Aerodynamic Configuration of Aireraft are affected, and ultimately result in aircraft usage hydraulic performance decline.Therefore, how to realize flying
The online high accuracy of machine skin-surface any point law vector, high efficiency measurement become the important problem of urgent need to resolve.
" the method at one kind measurement free form surface arbitrfary point of Li Yuan, Patent No. CN 201120358775 of remaining sword cutting edge of a knife or a sword invention
To the device of vector " invent a kind of utilization ball intersecting curve is drawn on curved surface, ask for method using the arrow of cutting of curve
The measuring method of vector.This kind of method is contact type measurement, and for random deformation curved surface its law vector, to ask for ratio of precision relatively low.Yao
Strong, Hu Yongxiang inventions Patent No. CN102248450A of shaking " for larger radius of curvature curved surface normal vector quick detection side
Method " has been invented and has realized two imaginary orthogonal planes and the acquisition of surfaces intersection using optical measuring technique, is existed using this two intersections
The cross product of the law vector of tested point asks for the curved surface normal vector of a bit, but the method do not account for curved surface a little at curvature it is big
It is little, and a kind of method of single utilization asks for law vector, efficiency is low, flexible difference.
The content of the invention
The invention solves the problems that technical barrier be the defect for overcoming prior art, invent a kind of freedom based on binocular vision
Surface normal vector measuring method, carries out free form surface and appoints using the measuring system being made up of binocular vision system and projection pattern
The measurement of some law vectors.Different measuring side is chosen apart from size with tested point to fit Plane by relatively more set threshold value
Case completes the high-precision rapid survey of law vector.The size of the curvature of the vertex neighborhood to be measured that the measuring method considers, flexibility compares
Height, is capable of achieving online high accuracy and the high efficiency measurement of curved surface any point law vector.
The technical solution used in the present invention is a kind of free form surface law vector measuring method based on binocular vision, and it is special
Levying is：In measuring method, the projection pattern projected to free form surface from laser projection device is by two orthogonal straight lines L1、L2With position
Four circular light spot G in two lines are constituted；Two straight-line intersections are P, and four circular light spots are equal in magnitude and are distributed on same
On circumference；Using the image of binocular vision system acquired projections pattern；By the threshold determination condition based on distance, curved surface is estimated
The amount of curvature of some small neighbourhood scopes, to choose different law vector measurement schemes：As d≤ε, from the method for fit Plane
Vector approaches curved surface tested point law vector；As d ＞ ε, then using two space curve tangent vectors at tested point after fitting
The cross product law vector of asking at tested point；Measuring method is comprised the following steps that:
(1) spot center based on threshold grayscale gravity model appoach is extracted
The present invention carries out extracted with high accuracy, gray level image I to spot center using Canny operators with reference to grey scale centre of gravity method
The grey scale centre of gravity of target S is in (i, j):
In formula, (Xk,Yk) be k-th spot center point image coordinate；W (i, j) is set weights；Consider actual
Half-tone information situation between background and target, the present invention adopts threshold grayscale gravity model appoach, its weights W (i, j) to be defined as：
Wherein, T is the threshold value for distinguishing target and background；Grey scale centre of gravity takes W (i, j)=I (i, j)；
(2) matching and reconstruction of spot center point
The spot center point on the image of left and right cameras collection is matched after the extraction for completing spot center point
With reconstruction；Matching process is as follows：
8 normalization algorithms of improvement proposed initially with Hartley calculate the fundamental matrix F of left and right cameras, then
The first matching of spot center point is carried out by epipolar-line constraint relation between the two-dimensional digital image that left and right cameras are gathered, it is assumed that
Left image spot center point xi' and right image spot center point xi'' match, i.e., 2 spot center points meet limit restraint bar
Part, limit restraint equation can be expressed as follows：
Wherein, xi' be left camera acquisition image spot central point image coordinates；xi'' be and xi' match by the right side
Video camera gathers the image coordinates of image spot central point；F is the fundamental matrix between left and right cameras；
On this basis three-dimensional reconstruction is carried out to all spot center points for meeting limiting constraint in left images with
D coordinates value of the spot center point under world coordinate system is obtained, reconstruction formula is as follows：
Wherein, xi'=(Xi',Yi'), Xi', Yi' be respectively left camera acquisition image spot central point xi' in image planes
Horizontal stroke, ordinate under coordinate system；xi’'=(Xi’’,Yi’'), Xi’', Yi’' it is respectively the image spot center of right camera acquisition
Point xi‘' the horizontal stroke under image coordinates system, ordinate；(xi,yi,zi) it is by two matching spot center point xi’、xi‘' reconstruct what is come
The three-dimensional coordinate of free token point；f1、f2The respectively focal length of left and right cameras；To connect left and right cameras
The spin matrix of relation, [tx ty tz] it is translation matrix of the right video camera relative to left video camera；
(3) curvature judges
1) least square fitting space plane
By four spot center points reconstructing in the D coordinates value of world coordinate system based on, using least square method
Fitting space plane, step is as follows：
The general expression of plane equation is：
Wherein (A, B, C) is the normal vector of plane；D is distance of the origin to plane；NoteThen z=a0x+a1y+a2；
N point (n >=3) is utilized from least square method:(xi,yi,zi), i=0,1 ..., n-1 is fitted above-mentioned plane, then
Make：It is minimum；
Wherein S is quadratic sum of the point to the distance of straight line；
Make S obtain minimum of a value, should meet：K=0,1,2；I.e.：
By the 3 d space coordinate (x of four reconstruction spot centersi,yi,zi), i=0,1,2,3 brings above-mentioned equation group into and asks
Obtain a0,a1,a2；
That is the equation of fit Plane is：Z=a0x+a1y+a2；The normal vector of spatial fit plane is:
2) tested point P ' is sought to the distance of fit Plane
Space is a little represented by the range formula of plane：
Wherein, SPlaneFor the equation of spatial fit plane；D is distance of the tested point to plane；P '=(x ', y ', z ') is to treat
Coordinate of the measuring point under world coordinate system；Q=(xq,yq,zq) for any point in fit Plane；ε is set threshold value；When d≤
During ε, it is believed that Curvature varying is little in the range of tested point P ' small neighbourhood curved surfaces；Tested point P ' small neighbourhood curved surfaces are then thought as d ＞ ε
In the range of Curvature varying it is larger；
(4) law vector is solved
Based on the law vector measurement scheme selection criterion that distance threshold is constrained, situation one：If the tested point on curved surface
P ' to the distance of spatial fit plane meets d≤ε, then it is assumed that Curvature varying is little in the range of tested point P ' small neighbourhood curved surfaces, this
When think the normal vector of planeIt is exactly the law vector of tested point on curved surface, i.e.,
Situation two：If on curved surface tested point P ' to spatial fit plane apart from d ＞ ε, then it is assumed that tested point P ' small neighbourhoods
Curvature varying is larger in the range of curved surface, and its Proximal surface is probably other quadratic surfaces such as sphere, parabola, saddle camber, now
The solution of law vector is carried out from two space curves projected on curved surface；Its step is as follows：
1) extraction of laser stripe centerline points, matching and reconstruction
The present invention is adopted based on the laser stripe center line detecting method of direction template, respectively in level, vertical, left bank
45 °, arrange the variable template of size fixed-direction on 45 ° of directions of right bank, K is designated as respectively0、K1、K2、K3, with this four templates
Two-dimensional digital image is respectively processed per a line；As a example by the process of i rows, for K0Template has：
Wherein M is the line number corresponding to template；N is the corresponding columns of template；K0[s][t]≥0；RepresentThe gray value of point；Accordingly for template K1、K2、K3Have
Hg1、Hg2、Hg3；Ask for Hg=max (Hg0,Hg1,Hg2,Hg3), then there is the position of central point of the i-th row laser stripe at point g；With
The method is detected pixel-by-pixel line by line the extraction that can complete laser stripe center line to two-dimensional digital image；Complete laser strip
On the basis of line central line pick-up, laser stripe is carried out using spot center Point matching in step (2) and reconstruction identical method
The matching and reconstruction of central point, obtains D coordinates value of the laser stripe centerline points under world coordinate system；
2) B-spline Curve is fitted two space curves
The present invention is fitted two space laser striped curves, B-spline curves piecewise function expression formula using B-spline Curve
For：
Wherein Pi(i=0,1 ... 5) represents respectively control vertex；Nij(i=1 ... 3, j=0,1 ... 4) represents basic function；If
There is discrete point on the curve that two camera rebuildings go out to be b1,b2,…,bn；Wherein front i point is located at c1Duan Shang, k-i point is located at c2
Duan Shang, n-j point is located at c3Duan Shang, then obtain some above-mentioned equation group of substitution：
The coefficient matrix that M is the left side is made, P is the vector for being constituted of required control vertex, and p is the laser of three-dimensional reconstruction
Stripe centerline point, above-mentioned equation is abbreviated as：
MP=p (11)
Thus can obtain be fitted normal equation be：
M'MP=M'p (12)
To improve the fitting precision of near intersections curve, weights are introduced to above-mentioned equation；Equation after weighting is：
(M'H'MP)=(M'H') Mp (13)
The equation of two curves can be asked for by this weighted equation.Two curves are asked for respectively on this basis in tested point two
Arrow is cut in individual direction, is designated asThen required law vector is：
The invention has the beneficial effects as follows the measuring method invented is noncontact, flexible strong, real-time height.It is applicable to song
The online high efficiency measurement of face difference, and its method is simple, algorithm is easily achieved.
Description of the drawings：
Fig. 1 is Surface Method vector method for solving schematic diagram.Wherein, L1'-floor projection laser stripe, L2'-vertical projection swashs
Striations, P '-tested point, S- fit Planes,The normal vector of-fit Plane S, d- tested points P ' to fit Plane S distance,- curved surface point P ' law vector,- curve L1' point P ' tangent vector,- curve L2' in the tangent vector of point P '.
Fig. 2 is the projection pattern of invention.Wherein L1- floor projection line, L2- vertical projection line, the rectangular projection lines of P- two are handed over
Point, G1- the first circular light spot, G2- the second circular light spot, G3- the three circular light spot, G4- the four circular light spot.
Fig. 3 is that the law vector based on two CCD camera measure system solves flow chart.
Specific embodiment
Combination technology scheme of the present invention and accompanying drawing, in order to better illustrate law vector solution procedure, with aircraft skin as reality
Example is to its detailed narration.Idiographic flow as shown in accompanying drawing 3 is as follows：
(1) laser projection device is moved to into aircraft skin tested point P ' (x ', y ', z ') place using digital control system, is being protected
Two laser stripe intersection points are demonstrate,proved on the basis of tested point, the pattern of accompanying drawing 2 to be projected on aircraft skin surface to be measured, is projected as four
Individual highlighted hot spot G1′、G2′、G3′、G4' and two laser stripe L1′、L2', at the same time using the left and right of binocular vision system
Camera acquisition projection pattern image.
(2) based on threshold grayscale gravity model appoach spot center extraction
From Canny operators spot center in left and right cameras collection image is carried with reference to threshold values grey scale centre of gravity method
Take, complete the positioning of spot center.Obtain the image coordinates (X of four spot center points in left imagei,Yi) i=1,2,3,4 and the right side
The image coordinates of four spot center points of image
For (Xi′,Yi′) i '=1,2,3,4.
(3) matching and reconstruction of spot center point
Bring the image coordinates of the spot center point of left images into formula (3) and (4) obtain the spot center point that matches
D coordinates value under world coordinate system：
G1′(x1,y1,z1)、G2′(x2,y2,z2)、G3′(x3,y3,z3)、G4′(x4,y4,z4)。
(4) vertex neighborhood curvature judges
1) based on spot center discrete point least square fitting space plane
Using reconstruct four spot centers as space finite points, using least square fitting space plane S.By four
The three-dimensional coordinate of spot center point brings formula (6) into, obtains plane equation for z=a0x+a1y+a2, the normal vector for obtaining plane is
2) tested point P ' is sought to the distance of fit Plane
In point P ' (x ', y ', z ') generation, is asked for into tested point to fit Plane apart from d to the range formula (7) of space plane
And this distance and set threshold epsilon are compared obtain d ＞ ε.Now should choose two space curves tested point P ' (x ', y ',
Z ') the cross product for cutting arrow ask for law vector.
(5) solution of curved surface any point law vector
1) extraction of laser stripe centerline points, matching and reconstruction
Using the laser stripe Spot detection method based on direction template, the extraction of laser stripe centerline points is completed, it is such as public
Formula (8).Obtain the image coordinates (X of four spot center points in left imagei,Yi) i=1,2 ... n and right image four hot spots in
Image coordinates (the X of heart pointi′,Yi′) i '=1,.The image coordinates of the left images for extracting is substituted into into formula (3) and (4) to enter
The matching and reconstruction of row spot center line point, the D coordinates value for obtaining spot center line point is (xi,yi,zi) i=1,2 ... n.
2) it is fitted using B-spline Curve based on spatial point discrete points data
By the D coordinates value (x of the laser stripe centerline points for reconstructingi,yi,zi) i=1,2 ... n substitutions formula (10)
With weighted equation (13), the equation of two curves is asked for.Ask for two curves cutting in tested point both direction respectively on this basis
Vector, is designated asThen required law vector is：
The measuring method of the present invention is non-contact measurement, in the case of the curvature for taking into full account the vertex neighborhood of curved surface one, choosing
Take the measurement that different measurement schemes ask for free form surface any point law vector.Its method is simple, and flexible strong, real-time is high, calculation
Method is easily achieved, and improves law vector well under conditions of certainty of measurement requirement is met and asks for efficiency.
Claims (1)
1. a kind of free form surface law vector measuring method based on binocular vision, is characterized in that：In measuring method, by laser projection
The projection pattern that device is projected to free form surface is by two orthogonal straight lines L1、L2With four circular light spot structures in two lines
Into；Two straight-line intersections are P, and four circular light spots are equal in magnitude and are distributed on same circumference；Gathered using binocular vision system
The image of projection pattern；By the threshold determination condition based on distance, the amount of curvature of some small neighbourhood scopes of curved surface is estimated, with
Choose different law vector measurement schemes：As d≤ε, from the normal vector of fit Plane curved surface tested point law vector is approached；When
During d ＞ ε, then using fitting after two space curves at tested point the cross product of tangent vector ask at tested point method arrow
Amount；Measuring method is comprised the following steps that:
(1) spot center based on threshold grayscale gravity model appoach is extracted
Extracted with high accuracy, target S in gray level image I (i, j) are carried out to spot center with reference to grey scale centre of gravity method using Canny operators
Grey scale centre of gravity be:
In formula, (Xk,Yk) be k-th spot center point image coordinate；W (i, j) is set weights；Consider real background
Half-tone information situation and target between, using threshold grayscale gravity model appoach, its weights W (i, j) are defined as：
Wherein, T is the threshold value for distinguishing target and background；Grey scale centre of gravity takes W (i, j)=I (i, j)；
(2) matching and reconstruction of spot center point
After the extraction for completing spot center point, the spot center point on the image of left and right cameras collection is matched and weighed
Build；Matching process is as follows：
8 normalization algorithms of improvement proposed initially with Hartley calculate the fundamental matrix F of left and right cameras, then pass through
Epipolar-line constraint relation carries out the first matching of spot center point between the two-dimensional digital image that left and right cameras are gathered, it is assumed that left figure
As spot center point xi' and right image spot center point xi′' match, i.e., 2 spot center points meet limiting constraint, pole
Limit constraint equation can be expressed as follows：
Wherein, xi' for left camera acquisition image spot central point image coordinates；xi′' be and xi' match by right shooting
Machine gathers the image coordinates of image spot central point；F is the fundamental matrix between left and right cameras；
On this basis all spot center points for meeting limiting constraint in left images are carried out three-dimensional reconstruction to obtain
D coordinates value of the spot center point under world coordinate system, reconstruction formula is as follows：
Wherein, xi'=(Xi',Yi'), Xi', Yi' be respectively left camera acquisition image spot central point xi' in image coordinates
Horizontal stroke, ordinate under system；xi′'=(Xi′′,Yi′'), Xi′', Yi′' be respectively right camera acquisition image spot central point xi′′
Horizontal stroke, ordinate under image coordinates system；
(xi,yi,zi) it is by two matching spot center point xi'、xi′The three-dimensional coordinate of ' free token the point for reconstructing；f1、f2Point
Not Wei left and right cameras focal length；To connect the spin matrix of left and right cameras relation, [tx ty tz] it is right
Translation matrix of the video camera relative to left video camera；
(3) curvature judges
1) least square fitting space plane
By four spot center points reconstructing in the D coordinates value of world coordinate system based on, using least square fitting
Space plane, step is as follows：
The general expression of plane equation is：
Wherein (A, B, C) is the normal vector of plane；D is distance of the origin to plane；Note
Then z=a0x+a1y+a2；
N point (n >=3) is utilized from least square method:(xi,yi,zi), i=0,1 ..., n-1 is fitted above-mentioned plane, then make：It is minimum；
Wherein S is quadratic sum of the point to the distance of straight line；
Make S obtain minimum of a value, should meet：I.e.：
By the 3 d space coordinate (x of four reconstruction spot centersi,yi,zi), i=0,1,2,3 brings above-mentioned equation group into and tries to achieve a0,
a1,a2；
That is the equation of fit Plane is：Z=a0x+a1y+a2；The normal vector of spatial fit plane is:
2) tested point P ' is sought to the distance of fit Plane
Space is a little represented by the range formula of plane：
Wherein, SPlaneFor the equation of spatial fit plane；D is distance of the tested point to plane；P '=(x ', y ', z ') is tested point
Coordinate under world coordinate system；Q=(xq,yq,zq) for any point in fit Plane；ε is set threshold value；As d≤ε,
Think that Curvature varying is little in the range of tested point P ' small neighbourhood curved surfaces；Tested point P ' small neighbourhood curved surface scopes are then thought as d ＞ ε
Incurvature is changed greatly；
(4) law vector is solved
Based on the law vector measurement scheme selection criterion that distance threshold is constrained, situation one：If the tested point P ' on curved surface is arrived
The distance of spatial fit plane meets d≤ε, then it is assumed that Curvature varying less, is now recognized in the range of tested point P ' small neighbourhood curved surfaces
For the normal vector of planeIt is exactly the law vector of tested point on curved surface, i.e.,
Situation two：If on curved surface tested point P ' to spatial fit plane apart from d ＞ ε, then it is assumed that tested point P ' small neighbourhood curved surfaces
In the range of Curvature varying it is larger, its Proximal surface is sphere, parabola, saddle camber, now from project on curved surface two
Space curve carries out the solution of law vector；Its step is as follows：
1) extraction of laser stripe centerline points, matching and reconstruction
Using the laser stripe center line detecting method based on direction template, respectively in level, vertical, 45 ° of left bank, right bank
The variable template of size fixed-direction is arranged on 45 ° of directions, K is designated as respectively0、K1、K2、K3, with this four templates to two-dimensional digital
Image is respectively processed per a line；To i row K0Template has：
Wherein M is the line number corresponding to template；N is the corresponding columns of template；K0[s][t]≥0；RepresentThe gray value of point；Accordingly for template K1、K2、K3HaveAsk forThen there is the position of the central point of the i-th row laser stripe
At point g；The extraction that can complete laser stripe center line is detected pixel-by-pixel line by line to two-dimensional digital image with the method；
On the basis of completing laser stripe central line pick-up, entered using spot center Point matching in step (2) and reconstruction identical method
The matching and reconstruction of row laser stripe central point, obtains D coordinates value of the laser stripe centerline points under world coordinate system；
2) B-spline Curve is fitted two space curves
Two space laser striped curves are fitted using B-spline Curve, B-spline curves piecewise function expression formula is：
Wherein Pi(i=0,1 ... 5) represents respectively control vertex；Nij(i=1 ... 3, j=0,1 ... 4) represents basic function；It is provided with two
Discrete point is b on the curve that camera rebuilding goes out1,b2,…,bn；Wherein front i point is located at c1Duan Shang, k-i point is located at c2Section
On, n-j point is located at c3Duan Shang, then obtain some above-mentioned equation group of substitution：
The coefficient matrix that M is the left side is made, P is the vector for being constituted of required control vertex, and p is the laser stripe of three-dimensional reconstruction
Centerline points, above-mentioned equation is abbreviated as：
MP=p (11)
Thus can obtain be fitted normal equation be：
M'MP=M'p (12)
To improve the fitting precision of near intersections curve, weights are introduced to above-mentioned equation；Equation after weighting is：
(M'H'MP)=(M'H') Mp (13)
The equation of two curves can be asked for by this weighted equation, two curves are asked for respectively on this basis in two sides of tested point
To cut arrow, be designated asThen required law vector is：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410149149.6A CN103913131B (en) | 2014-04-14 | 2014-04-14 | Free curve method vector measurement method based on binocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410149149.6A CN103913131B (en) | 2014-04-14 | 2014-04-14 | Free curve method vector measurement method based on binocular vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103913131A CN103913131A (en) | 2014-07-09 |
| CN103913131B true CN103913131B (en) | 2017-04-12 |
Family
ID=51039000
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410149149.6A Active CN103913131B (en) | 2014-04-14 | 2014-04-14 | Free curve method vector measurement method based on binocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103913131B (en) |
Families Citing this family (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105678684B (en) * | 2014-11-18 | 2020-11-03 | 中兴通讯股份有限公司 | Method and device for intercepting image |
| CN104732553B (en) * | 2015-04-10 | 2017-07-11 | 大连理工大学 | A kind of Feature Points Extraction based on many laser assisted targets |
| CN105081884B (en) * | 2015-08-24 | 2023-08-04 | 成都飞机工业（集团）有限责任公司 | Normal measuring device for rotary scanning 3D molding |
| CN105043259B (en) * | 2015-08-25 | 2017-07-11 | 大连理工大学 | Digit Control Machine Tool rotary shaft error detection method based on binocular vision |
| CN105300316B (en) * | 2015-09-22 | 2017-10-13 | 大连理工大学 | Optical losses rapid extracting method based on grey scale centre of gravity method |
| CN105345600B (en) * | 2015-10-20 | 2020-02-21 | 上海拓璞数控科技股份有限公司 | Normal measurement and adjustment method for free-form surface drilling of aviation thin-walled part |
| CN105571518B (en) * | 2016-01-19 | 2017-10-17 | 大连理工大学 | Three-dimensional information vision measuring method based on refraction image deviation |
| CN105698699B (en) * | 2016-01-26 | 2017-12-19 | 大连理工大学 | A kind of Binocular vision photogrammetry method based on time rotating shaft constraint |
| CN105716539B (en) * | 2016-01-26 | 2017-11-07 | 大连理工大学 | A kind of three-dimentioned shape measurement method of quick high accuracy |
| CN107957234B (en) * | 2016-10-17 | 2023-09-29 | 中国航空制造技术研究院 | Method and device for measuring normal vector at any point of free-form surface |
| CN106553086B (en) * | 2016-10-27 | 2018-06-08 | 大连理工大学 | The complex-curved drilling point law vector measuring method of quick high accuracy |
| CN107121061B (en) * | 2017-03-15 | 2018-07-13 | 大连理工大学 | A kind of drilling point law vector method for solving based on elliptic cone constraint |
| CN107238378B (en) * | 2017-07-11 | 2019-06-25 | 武汉大学 | A kind of three-dimensional positioning device and positioning construction method of saddle camber building |
| CN108195305B (en) * | 2018-02-09 | 2020-03-31 | 京东方科技集团股份有限公司 | Binocular detection system and depth detection method thereof |
| CN110307806A (en) * | 2018-03-27 | 2019-10-08 | 广东电网有限责任公司 | A kind of cable curvature estimation method and detection device based on binocular stereo vision |
| CN110322502A (en) * | 2018-03-27 | 2019-10-11 | 广东电网有限责任公司 | A kind of cable curvature estimation method and system based on binocular stereo vision |
| CN108931186B (en) * | 2018-03-28 | 2020-06-12 | 中国航空制造技术研究院 | Curved surface measurement and reference positioning method and measurement equipment based on cross laser |
| CN109774197B (en) * | 2018-07-13 | 2021-05-07 | 中国航空工业集团公司济南特种结构研究所 | Method for determining position of laser projector of composite material curved surface layering |
| CN109048911B (en) * | 2018-08-31 | 2021-08-24 | 河南工程学院 | Robot vision control method based on rectangular features |
| CN109346849B (en) * | 2018-09-06 | 2021-01-08 | 中国科学院国家空间科学中心 | Device for generating millimeter wave Bessel wave beam |
| CN109285195B (en) * | 2018-10-18 | 2020-09-04 | 易思维（杭州）科技有限公司 | Monocular projection system pixel-by-pixel distortion correction method based on large-size target and application thereof |
| CN109541626B (en) * | 2018-12-12 | 2020-12-11 | 华南农业大学 | Target plane normal vector detection device and detection method |
| CN110223355B (en) * | 2019-05-15 | 2021-01-05 | 大连理工大学 | Feature mark point matching method based on dual epipolar constraint |
| CN110641947B (en) * | 2019-10-16 | 2021-02-09 | 山东中衡光电科技有限公司 | Intelligent inspection robot system for bulk cargo conveyor and detection method thereof |
| CN111008602B (en) * | 2019-12-06 | 2023-07-25 | 青岛海之晨工业装备有限公司 | Scribing feature extraction method combining two-dimensional vision and three-dimensional vision for small-curvature thin-wall part |
| CN112033307B (en) * | 2020-07-15 | 2021-08-03 | 成都飞机工业（集团）有限责任公司 | Farnet vector measuring device |
| CN111652947B (en) * | 2020-08-06 | 2020-11-17 | 上海上飞飞机装备制造有限公司 | Visual normal vector solving method for automatic hole making and automatic drilling and riveting of airplane wallboard |
| CN111964612B (en) * | 2020-08-25 | 2021-10-29 | 大连工业大学 | Drilling normal vector alignment method based on high-reflectivity imaging principle |
| CN112419254B (en) * | 2020-11-16 | 2024-02-09 | 重庆长安汽车股份有限公司 | Method for analyzing visual sense continuous sense between uniform parting curved surfaces based on human eye visual limit |
| CN114688967A (en) * | 2020-12-29 | 2022-07-01 | 北京机械工业自动化研究所有限公司 | Normal amount measuring method and device based on machine vision and operating equipment |
| CN112809037B (en) * | 2021-01-04 | 2022-04-29 | 中国航空制造技术研究院 | Method for drilling on curved surface structure |
| CN114565565A (en) * | 2022-02-11 | 2022-05-31 | 山西支点科技有限公司 | Method for positioning sub-pixels in center of vision measurement target |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101976336A (en) * | 2010-10-21 | 2011-02-16 | 西北工业大学 | Fuzzy enhancement and surface fitting-based image edge characteristic extraction method |
| CN103489222A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | Target body surface reconstruction method in three-dimensional image |
| CN103558808A (en) * | 2013-09-28 | 2014-02-05 | 大连理工大学 | Kinematics control method for complex-curved-surface five-axis numerical control machining cutter vectors |
| CN103592891A (en) * | 2013-09-28 | 2014-02-19 | 大连理工大学 | Method for cutter-axis vector fairing of complex curved surface five-axis numerical control machining based on kinematical constraints |
-
2014
- 2014-04-14 CN CN201410149149.6A patent/CN103913131B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101976336A (en) * | 2010-10-21 | 2011-02-16 | 西北工业大学 | Fuzzy enhancement and surface fitting-based image edge characteristic extraction method |
| CN103489222A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | Target body surface reconstruction method in three-dimensional image |
| CN103558808A (en) * | 2013-09-28 | 2014-02-05 | 大连理工大学 | Kinematics control method for complex-curved-surface five-axis numerical control machining cutter vectors |
| CN103592891A (en) * | 2013-09-28 | 2014-02-19 | 大连理工大学 | Method for cutter-axis vector fairing of complex curved surface five-axis numerical control machining based on kinematical constraints |
Non-Patent Citations (2)
| Title |
|---|
| 基于自由曲面点云的快速原型制作技术研究;孙玉文 等;《机械工程学报》;20030131;第39卷(第1期);全文 * |
| 磁研法在模具曲面中的抛光机理与技术研究;马东雄 等;《机械设计与制造》;20090331(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103913131A (en) | 2014-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103913131B (en) | Free curve method vector measurement method based on binocular vision | |
| CN109544456B (en) | Panoramic environment sensing method based on two-dimensional image and three-dimensional point cloud data fusion | |
| CN107392964B (en) | The indoor SLAM method combined based on indoor characteristic point and structure lines | |
| CN106651942B (en) | Three-dimensional rotating detection and rotary shaft localization method based on characteristic point | |
| CN104484648B (en) | Robot variable visual angle obstacle detection method based on outline identification | |
| US9942535B2 (en) | Method for 3D scene structure modeling and camera registration from single image | |
| CN105300316B (en) | Optical losses rapid extracting method based on grey scale centre of gravity method | |
| CN110230998B (en) | Rapid and precise three-dimensional measurement method and device based on line laser and binocular camera | |
| CN104463899B (en) | A kind of destination object detection, monitoring method and its device | |
| CN109784333A (en) | Based on an objective detection method and system for cloud bar power channel characteristics | |
| CN109000557B (en) | A kind of nuclear fuel rod pose automatic identifying method | |
| JP2016161569A (en) | Method and system for obtaining 3d pose of object and 3d location of landmark point of object | |
| CN109685886A (en) | A kind of distribution three-dimensional scenic modeling method based on mixed reality technology | |
| CN106446773B (en) | Full-automatic robust three-dimensional face detection method | |
| CN110009690A (en) | Binocular stereo vision image measuring method based on polar curve correction | |
| CN106709950A (en) | Binocular-vision-based cross-obstacle lead positioning method of line patrol robot | |
| CN108597009A (en) | A method of objective detection is carried out based on direction angle information | |
| CN111126116A (en) | Unmanned ship river channel garbage identification method and system | |
| CN107016697A (en) | A kind of height measurement method and device | |
| CN112085675B (en) | Depth image denoising method, foreground segmentation method and human motion monitoring method | |
| CN101794459A (en) | Seamless integration method of stereoscopic vision image and three-dimensional virtual object | |
| CN113393524B (en) | Target pose estimation method combining deep learning and contour point cloud reconstruction | |
| CN108010125A (en) | True scale three-dimensional reconstruction system and method based on line structure light and image information | |
| CN111008602B (en) | Scribing feature extraction method combining two-dimensional vision and three-dimensional vision for small-curvature thin-wall part | |
| CN108182705A (en) | A kind of three-dimensional coordinate localization method based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |