CN103398660A - Structured light visual sensor parameter calibration method for acquiring height information of welded joint - Google Patents

Abstract

The invention provides a structured light visual sensor parameter calibration method for acquiring height information of a welded joint, relates to the measurement non-specially applied to specific variable, and is a structured light visual system parameter calibration method based on a wedge-shaped target, which processes a two-dimensional image to obtain offset of structured light by calibrating obtained parameters so as to compute and obtain the height information of the welded joint and then to compute three-dimensional coordinate of an intersection point of a laser belt and the welded joint on the target, wherein the structured light visual sensor parameter calibration method specifically comprises the following steps of firstly recognizing horizontal deviation, secondly recognizing vertical deviation, thirdly calibrating structured light visual system parameters by utilizing a camera, and fourthly acquiring the height information of the welded joint. According to the structured light visual sensor parameter calibration method disclosed by the invention, the calibration process doest not involve a large-scale matrix transformation, the computation amount is smaller, a blockage problem of the target doest not exist, and the problem that the misconvergence of globally optimal solution generates in a solving process of a constrained nonlinear equation also does not exist.

Description

For obtaining the structured light vision sensor parameter calibration method of weld bead height information

Technical field

Technical scheme of the present invention relates to the non-measurement that is exclusively used in particular variables, specifically be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information.

Background technology

The demarcation of vision system is a major issue in robot vision, the image that vision system gathers, certain one-to-one relationship between a bit of the image coordinate location of every bit and space object surface needs the imaging model of vision system to decide, and the parameters of imaging model need to be determined by the demarcation of vision system, so the demarcation of vision system is the key that realizes the conversion from two-dimensional image information to three-dimensional environment information.In the calibration process of vision system, obtaining of three-dimensional elevation information is a more thorny difficult problem always.

The structured light vision detection method has the advantages such as wide range, noncontact, large visual field and system flexibility are good, in recent years in industrial environment especially welding field, is widely used.Demarcation to the parameter of structured light vision sensor mainly comprises the calibration of camera of video camera and the structural parameters calibration of line structured light vision sensor.Intrinsic parameters of the camera mainly refers to geometry and the optical characteristics of video camera inside, as optical centre, focal length, position etc., the structural parameters of sensor refer to the transformation relation between image coordinate system and three-dimensional camera coordinate system, the purpose that sensor parameters is demarcated makes, by two dimensional image coordinate reconstruct three-dimensional world coordinate, to realize three-dimensional measurement.

At welding technology field, traditional structured light vision sensor parameter calibration method, be about to dot structure light or linear structure light and project surface of the work, by the triangulation method, calculate weld bead height information (referring to " Wu Lin, Chen Shanben, intelligent solder technology; 210th～211 pages; 215th～217 pages, Beijing: National Defense Industry Press, 1999 years "; Referring to " Lee is former for Xu De, Tan Min, and robot vision is measured and controlled, and the 133rd page, Beijing: National Defense Industry Press, 2008 ").The calibrated and calculated process is based upon on laser instrument and the lucky basis of crossing the lens optical center of surface level intersection point, and still, in practical operation, we are difficult to determine the optical centre shaft position of video camera, when using especially at the scene, are difficult to Accurate Calibration.Therefore, above-mentioned triangulation method need to further be improved perfect for actual conditions.

CN03142658.1 discloses " a kind of scaling method of structured light vision sensor based on the plane target ", the method is mainly to adopt the two dimensional surface target that is distributed with the black square that can move freely to demarcate structured light vision sensor, in calibration process, using the summit of black square as unique point, by image coordinate system and the conversion between world coordinate system that utilizes unique point, complete the demarcation to intrinsic parameters of the camera, the method relates to extensive matrixing, calculate the complexity of deriving, need high-precision two-dimensional plane target, the target difficulty of processing is large, and there is the problem of mutually blocking.

CN200710121397.X has disclosed " a kind of method for standardizing structural parameter of structure optical vision sensor ", the method is mainly by the plane target that a plurality of nonlinear characteristic points are arranged, by the position of plane of motion target repeatedly, obtain after each movement four or more nonlinear characteristic point coordinate on target image, and transform under camera coordinate system and then simulate high order ternary structural light equation.The method calibration process complex steps can affect precision, and coordinate transform and fit procedure complexity.

CN200910089307.2 has reported " based on the structured light vision sensor scaling method of one dimension target ", the method is utilized at least three unique points with known spatial constraint of one dimension target, in conjunction with the perspective projection equation, according to the video camera of the length constraint of unique point and direction constrain calculated characteristics point, be coordinate and carry out match and obtain optic plane equations.The method need to solve the constrained nonlinear equation of tool, has introduced nonlinear optimization, and method is more numerous, and speed is slow, and is more responsive to Initial value choice and noise ratio, and can not guarantee that parameter converges to global optimum.

Summary of the invention

technical matters to be solved by this invention is: the structured light vision sensor parameter calibration method that is provided for obtaining weld bead height information, it is a kind of parameter calibration method of Constructed Lighting Vision System based on the wedge shape target, weld bead height parameter in target is accurately measured in advance, can calculate thus the three-dimensional coordinate of laser band and weld seam intersection point on target, in calibration process, the accurate parameter of Fast Calibration vision system still when the optical centre shaft position of uncertain video camera, and calibration process does not relate to extensive matrixing, calculated amount is less, there do not is the occlusion issue of target, do not exist yet and solve in constrained nonlinear equation a difficult problem that does not converge to globally optimal solution that occurs.

The present invention solves this technical problem the technical scheme that adopts: be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information, be a kind of parameter calibration method of Constructed Lighting Vision System based on the wedge shape target, step is:

Described wedge shape target adopts the triangle wedge, it is triangle wedge, weldment, electronics wedge shape vernier scale and the servomotor of 20 °～60 ° that major equipment used has scrambler, one of them acute angle in video camera, laser instrument, motion, in described video camera, relates generally to camera lens;

The first step, identify horizontal departure

By controlling servomotor, video camera is moved to distance the survey record of 5mm～25mm along the parallel direction of structure light belt, then calculate the pixel value that in two dimensional image, weld seam moves up at upper and lower, and then calculate the actual range size of each pixel representative, can obtain the Proportional coefficient K between pixel value and actual distance value;

Second step, identify the vertical direction deviation

2-I step, be placed on the video camera below by the triangle wedge, guarantees that the laser instrument band just beats in triangle wedge edge;

2-II step, move forward distance y by the triangle wedge _X1The structure light belt that laser instrument produces can move up by the hypotenuse along the triangle wedge along with the movement of triangle wedge, until the height A C=10mm that this structure light belt is risen on surface level stops, structure light belt mobile distance on the picture plane is △ y ₁, the scrambler in moving process in motion can record the distance y that the triangle wedge moves forward _X1, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₁

2-III step, repeat the operating process of 2-II step, just continues the triangle wedge is moved forward to distance y again _X2, until the structure light belt is risen to apart from the height A on surface level ' and C '=20mm place stops, and structure light belt this moment mobile distance on the picture plane is △ y ₂, the scrambler in moving process in motion can record the distance y that the triangle wedge moves forward _X2, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₂

The 3rd step, carry out the Constructed Lighting Vision System parameter calibration with video camera

The position mark of the residing surface level 0mm in acute angle summit of triangle wedge that will be arranged in the triangle wedge position of 2-I step operating process is made D, the position mark of acute angle summit surface level 0mm of living in of triangle wedge that will be arranged in the triangle wedge position of 2-II step operating process is made D1, the position mark of acute angle summit surface level 0mm of living in of triangle wedge that will be arranged in the triangle wedge position of 2-III step operating process is made D2, and the distance between D and D1 is y _X1, the distance between D1 and D2 is y _X2,

If the line laser that laser instrument sends is A with the hypotenuse intersection point of the triangle wedge of the triangle wedge position that is arranged in the operating process of 2-II step, the camera lens center is designated as O, the extended line of camera lens center O and A hands over the surface level of height 0mm in a B, make the vertical line of A to the surface level of height 0mm, hand over surface level in a C, 3 form △ ABC; Make the vertical line of camera lens central point O to the surface level of height 0mm, hand over surface level in an O ', 3 O, B, O ' form triangle △ OBO ', and the distance between note AC is h ₁,

According to △ ABC ∽ △ OBO ' is similar, obtain:

$\frac{{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{BC}}{{\mathrm{BO}}^{\&prime;}}---\left(1\right)$

If the line laser that laser instrument sends is A ' with the hypotenuse intersection point of the triangle wedge of the position of the triangle wedge that is arranged in the operating process of 2-III step, the extended line of camera lens center O and A ' hands over the surface level of height 0mm in a B ', make the vertical line of A ' to the surface level of height 0mm, hand over surface level in a C ', 3 form △ A ' B ' C '; Make camera lens 1 central point O and hand over surface level and some O ' to the vertical line of the surface level of height 0mm, 3 O, B ', O ' form triangle △ OB ' O '.Distance between note A ' C ' is h ₂.

According to △ A ' B ' C ' ∽ △ OB ' O ' is similar, obtain:

$\frac{{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2}{{\mathrm{OO}}^{\&prime;}}=\frac{B^{\&prime;}{C}^{\&prime;}}{B^{\&prime;}{O}^{\&prime;}}---\left(2\right)$

Laser instrument central shaft and vertical direction angle are designated as to θ, note BD=y ₁, B ' B=y ₂,

BC=y ₁-(y _X1-h ₁/ tan θ), B ' C '=y ₁+ y ₂-(y _X1+ y _X2-h ₁/ tan α), note optical center O be H to height 0mm's apart from OO ', 2-I step intermediate cam shape wedge acute angle summit D to camera optics central shaft OO ' apart from being Y, i.e. OO '=H, DO '=Y, can obtain BO '=Y+y ₁, B ' O '=Y+y ₁+ y ₂, bring (1) into and can obtain with (2):

$\left\{\begin{array}{c}\frac{h_1}{H}=\frac{y_1-(y_{x1}-h_1/\tan \mathrm{\&alpha;})}{y_1+Y}\\ \frac{h_2}{H}=\frac{y_1+y_2-(y_{x1}+y_{x2}-h_2/\tan \mathrm{\&alpha;})}{y_1+y_2+Y}\end{array}\right.---\left(3\right)$

H wherein ₁=10mm, h ₂=20mm, α=20 °～60 °, bring known quantity into, can obtain:

By (4) formula, can be obtained the value of Parameter H, Y, in addition, by geometric relationship in figure, can be obtained:

y _x1-AC/tanα＝AC*tanθ

That is:

$\tan \mathrm{\&theta;}=\frac{y_{x1}-\mathrm{AC}/\tan \mathrm{\&alpha;}}{\mathrm{AC}}---\left(5\right)$

Known h ₁=AC=10mm, α=20 °～60 ° of substitutions (5),

By following formula, namely can calculate tan θ.

So far, with video camera, carry out the Constructed Lighting Vision System parameter calibration and complete, obtained respectively the numerical value of parameter K, H, Y, tan θ;

The 4th step, the obtaining of weld bead height information

In the weld joint tracking situation, what measure is the weld bead height information at weld seam and structure light belt intersection point place, the weldment that will measure weld bead height information moves, until will measure weld seam solder joint place and the structure light belt of weld bead height, intersect, the note intersection point is M, make the extended line of OM and hand over the 0mm plane in a N, the distance of note N point and 2-I step intermediate cam shape wedge acute angle summit D is y ₃=ND, visible y ₃Be the horizontal range of intersection point M and the 2-I step intermediate cam shape wedge acute angle summit D of structure light belt and the weld seam solder joint that will measure weld bead height, as the vertical line of M point to the 0mm plane, the point that hangs down is P, remembers that some M and the distance of some P are h ₃=MP,

According to △ MNP ∽ △ ONO ' is similar, obtain:

$\frac{\mathrm{MP}}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{NP}}{{\mathrm{NO}}^{\&prime;}}$

Be:

$\frac{h_3}{H}=\frac{y_3-\tan \mathrm{\&theta;}*h_3}{y_3+Y}---\left(6\right)$

That is: $h_3=\frac{y_3*H}{y_3+Y+\tan \mathrm{\&theta;}*H}=f\left(y_3\right)---\left(7\right)$

By following formula, can see h ₃About y ₃Function,

Weld seam is moved along vertical stratification light belt direction, until the structure light belt intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 ', △ y3=△ y3 '-Y/K, namely obtain the size of side-play amount △ y3.the size of y3 is y3=K* △ y3, the elevation of water of setting the welding initial point is 0mm, when there is the deviation on vertical direction in video camera with respect to weld seam, will cause the laser stripe in image to be moved on left and right directions, this moment, weld seam did not move, when the image generation so changes, by image, process and record weld seam and structure light belt intersection point shift offset △ y3, and then learn y3=K* △ y3, in conjunction with (7) formula, can identify the deviation that welding gun exists with actual welds in vertical direction, so far, weld seam and structure light belt intersection point place weld bead height information have been obtained, realized the reconstruct of three-dimensional camera coordinate.

Above-mentioned be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information, institute's laser instrument is laser line generator.

Above-mentioned be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information, described triangle wedge, one of them acute angle are 30 °.

Above-mentioned be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information, wherein related equipment is by commercially available.

The invention has the beneficial effects as follows:

The outstanding substantive distinguishing features of the inventive method is: the inventive method is a kind of parameter calibration method of Constructed Lighting Vision System based on the wedge shape target, utilize and demarcate parameters obtained, two dimensional image is processed to the side-play amount that obtains structured light, weld bead height information can be calculated, and then the three-dimensional coordinate of laser band and weld seam intersection point on target can be calculated.

The marked improvement of the inventive method is:

(1) there is certain difficulty in monocular vision when obtaining weld bead height information, binocular vision need to merge the information that two video cameras obtain when obtaining weld bead height information, calculated amount is large, affect system real time, algorithms to improve proposed by the invention this present situation, not only avoided matrixing loaded down with trivial details when obtaining weld bead height information, coarse problem while also having avoided definite camera optics central shaft position, greatly saved and demarcated and computing time, significantly improved the real-time of system.

(2) in structured light vision sensor parameter calibration process, the accurate parameter of Fast Calibration vision system still when the optical centre shaft position of uncertain video camera, and calibration process does not relate to extensive matrixing, calculated amount is less, there do not is the occlusion issue of target, do not exist yet and solve in constrained nonlinear equation a difficult problem that does not converge to globally optimal solution that occurs.

(3) the inventive method has been avoided setting up complicated camera imaging model on the one hand, has strengthened on the other hand precision, robustness and the adaptability of camera calibration.

The accompanying drawing explanation

The present invention is further described below in conjunction with drawings and Examples.

Fig. 1 (a) is total schematic diagram of the inventive method calibration process intermediate cam shape wedge change in location.

Fig. 1 (b) identifies the schematic diagram of intermediate cam shape wedge position in the inventive method, demarcating 2-I step to the vertical direction deviation.

Fig. 1 (c) identifies the schematic diagram of triangle wedge position in the inventive method, demarcating 2-II step to the vertical direction deviation.

The schematic diagram of the triangle wedge position of the operating process of repetition 2-II step during Fig. 1 (d) identifies the vertical direction deviation for demarcation 2-III in the inventive method walks.

Fig. 1 (e) utilizes the calibration result butt-welding fitting to carry out the schematic diagram of weld bead height acquisition of information in the inventive method.

There is the variation relation schematic diagram of deviation and image thereof in the horizontal direction in Fig. 2 (a) for video camera in the inventive method and weld seam.

There is the variation relation schematic diagram of deviation and image thereof in Fig. 2 (b) for video camera in the inventive method and weld seam in the vertical direction.

Fig. 3 is the test findings data fitting result curve figure of the inventive method.

In figure, 1. camera lens, 2. laser instrument, 3-1. the position of the triangle wedge in the 2-III step operating process during the vertical direction deviation is identified of the position of the triangle wedge in the 2-II step operating process during the vertical direction deviation is identified of the position of the triangle wedge in the 2-I step operating process of second step during the vertical direction deviation is identified, 3-2. second step, 3-3. second step, 4. weldment, 5. video camera, 6. laser stripe, 7. weld seam.

Embodiment

Fig. 1 (a) illustrated embodiment has shown the changing condition of the position of the inventive method calibration process intermediate cam shape wedge.in figure, indicated camera lens 1, laser instrument 2, the 2-I that second step is identified the vertical direction deviation walks the position 3-1 of operating process intermediate cam shape wedge, second step is identified the position 3-2 of intermediate cam shape wedge to the vertical direction deviation, the 3rd step is identified the position 3-3 of intermediate cam shape wedge to the vertical direction deviation, the acute angle of triangle wedge is α, the position that the acute angle summit of the triangle wedge of the position 3-1 of the triangle wedge in the 2-I step operating process during second step is identified the vertical direction deviation is in surface level 0mm is denoted as D, the D point is designated as to Y to the distance of the optical centre axle OO ' of camera lens 1, the position that the acute angle summit of the triangle wedge of the position 3-2 of the triangle wedge in the operating process of 2-II step is in to surface level 0mm is denoted as D1, the position that the acute angle summit of the triangle wedge of the position 3-2 of the triangle wedge in the operating process of 2-III step is in to surface level 0mm is denoted as D2, the distance that the triangle wedge moves forward is y _x1the time structured light mobile apart from △ y on the picture plane ₁, the distance that the triangle wedge moves forward is y _x2the time structured light mobile apart from △ y on the picture plane ₂the hypotenuse intersection point of the triangle wedge during the position 3-2 of the triangle wedge in the 2-II step operating process during the vertical direction deviation is identified of line laser and second step is A, the extended line of the line OA of camera lens 1 center O and A hands over the surface level of height 0mm in a B, make A and hand over surface level in a C to the vertical line of the surface level of height 0mm, 3 form △ ABC, making camera lens 1 central point O hands over surface level in an O ' to the vertical line of the surface level of height 0mm, 3 O, B, O ' form triangle △ OBO ', the hypotenuse intersection point of line laser and the triangle wedge of the position 3-3 of the triangle wedge that is arranged in the 2-III step operating process that second step identifies the vertical direction deviation is for being A ', the extended line of camera lens 1 center O and A ' hands over the surface level of height 0mm in a B ', make the vertical line of A ' to the surface level of height 0mm, hand over surface level in a C ', 3 form △ A ' B ' C ', BD=y ₁, B ' B=y ₂.

For the ease of clearly demonstrating, this figure is decomposed into to following Fig. 1 (b), Fig. 1 (c) and Fig. 1 (d).

Fig. 1 (b) illustrated embodiment shows, in the 2-I step operating process that the vertical direction deviation is identified, first the triangle wedge being placed on to camera lens 1 below is the position 3-1 of the triangle wedge in the 2-I step operating process during second step is identified the vertical direction deviation, guarantee that the laser instrument band of laser instrument 2 just beats in triangle wedge edge, the position that the acute angle summit of triangle wedge is in to surface level 0mm is denoted as D, camera lens 1 center is O, it is O ' that camera lens 1 central point O hands over the intersection point of surface level to the vertical line of the surface level of height 0mm, the acute angle of triangle wedge is α, laser instrument 2 central shafts and vertical direction angle are θ.

Fig. 1 (c) illustrated embodiment shows, laser instrument 2 central shafts and vertical direction angle are θ, the acute angle of triangle wedge is α, the position that the acute angle summit of the triangle wedge of the position 3-1 of the triangle wedge in the 2-I step operating process during second step is identified the vertical direction deviation is in surface level 0mm is denoted as D, the position that the acute angle summit of the triangle wedge of the position 3-2 of the triangle wedge in the operating process of 2-II step is in to surface level 0mm is denoted as D1, and the distance that the triangle wedge moves forward is y _X1The time structured light mobile apart from △ y on the picture plane ₁.In the 2-II step operating process that the vertical direction deviation is identified, line laser is A with the hypotenuse intersection point of the triangle wedge of the position 3-2 of the triangle wedge that is arranged in the 2-II step operating process that second step identifies the vertical direction deviation, the extended line of the line OA of camera lens 1 center O and A hands over the surface level of height 0mm in a B, make A and meet at a C to vertical line and the surface level of surface level of height 0mm, 3 composition △ ABC; Make camera lens 1 central point O and hand over surface level in an O ' to the vertical line of the surface level of height 0mm, 3 O, B, O ' form triangle △ OBO '.According to △ ABC ∽ △ OBO ' is similar, obtain:

$\frac{{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{BC}}{{\mathrm{BO}}^{\&prime;}}---\left(1\right)$

Wherein, h ₁For the distance between AC.

Fig. 1 (d) illustrated embodiment shows, laser instrument 2 central shafts and vertical direction angle are θ, the acute angle of triangle wedge is α, the position that the acute angle summit of the triangle wedge of the position 3-1 of the triangle wedge in the 2-I step operating process during second step is identified the vertical direction deviation is in surface level 0mm is denoted as D, the position that the acute angle summit of the triangle wedge of the position 3-2 of the triangle wedge in the operating process of 2-III step is in to surface level 0mm is denoted as D2, and the distance that the triangle wedge moves forward is y _X2The time structured light mobile apart from △ y on the picture plane ₂.In the 2-III step operating process that the vertical direction deviation is identified, the hypotenuse intersection point of line laser and the triangle wedge of the position 3-3 of the triangle wedge that is arranged in the 2-III step operating process that second step identifies the vertical direction deviation is for being A ', the extended line of camera lens 1 center O and A ' hands over the surface level of height 0mm in a B ', make the vertical line of A ' to the surface level of height 0mm, hand over surface level in a C ', 3 form △ A ' B ' C '; Make camera lens 1 central point O and hand over surface level and some O ' to the vertical line of the surface level of height 0mm, 3 O, B ', O ' form triangle △ OB ' O '.According to △ A ' B ' C ' ∽ △ OB ' O ' is similar, obtain:

$\frac{{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2}{{\mathrm{OO}}^{\&prime;}}=\frac{B^{\&prime;}{C}^{\&prime;}}{B^{\&prime;}{O}^{\&prime;}}---\left(2\right)$

Wherein, h ₂For the distance between A ' C '.

BD=y ₁, B ' B=y ₂, BC=y ₁-(y _X1-h ₁/ tan θ), B ' C '=y ₁+ y ₂-(y _X1+ y _X2-h ₁/ tan α).Suppose camera lens 1 center O to height 0mm apart from OO ', be H, first step intermediate cam shape calibrating block 3 acute angle summits to the camera optics central shaft apart from being Y, OO '=H, BO '=Y+y ₁, B ' O '=Y+y ₁+ y ₂, bring (1) into and can obtain with (2):

$\left\{\begin{array}{c}\frac{h_1}{H}=\frac{y_1-(y_{x1}-{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+Y}\\ \frac{h_2}{H}=\frac{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+y_2-(y_{x1}+y_{x2}-{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_2+Y}\end{array}\right.---\left(3\right)$

H wherein ₁=10mm, h ₂=20mm, α=20 °～60 °, bring known quantity into, can obtain:

By (4) formula, can be obtained the value of Parameter H, Y, in addition, by geometric relationship in figure, can be obtained:

y _x1-AC/tanα＝AC*tanθ

That is:

$\tan \mathrm{\&theta;}=\frac{y_{x1}-\mathrm{AC}/\tan \mathrm{\&alpha;}}{\mathrm{AC}}---\left(5\right)$

Known h ₁=AC=10mm, α=20 °～60 ° of substitutions (5),

By following formula, namely can calculate tan θ.

So far, camera parameters is demarcated and to be completed, and has obtained respectively the size of parameter K, H, Y, tan θ.

In Fig. 1 (e), camera lens 1 center is O, and it is O ' that camera lens 1 central point O hands over the intersection point of surface level to the vertical line of the surface level of height 0mm, and laser instrument 2 central shafts and vertical direction angle are θ.

Fig. 1 (e) illustrated embodiment shows, in the weld joint tracking situation, what measure is the weld bead height information at weld seam and structure light belt intersection point place, the weldment 4 that will measure weld bead height information moves, until will measure weld seam solder joint place and the structure light belt of weld bead height, intersect, the note intersection point is M, makes the extended line of OM and hands over the 0mm plane in a N, and the distance of note N point and 2-I step intermediate cam shape wedge acute angle summit D is y ₃=ND, visible y ₃Be the horizontal range of intersection point M and the 2-I step intermediate cam shape wedge acute angle summit D of structure light belt and the weld seam solder joint that will measure weld bead height, as the vertical line of M point to the 0mm plane, the point that hangs down is P, remembers that some M and the distance of some P are h ₃=MP,

According to △ MNP ∽ △ ONO ' is similar, obtain:

$\frac{\mathrm{MP}}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{NP}}{{\mathrm{NO}}^{\&prime;}}$

Be:

$\frac{h_3}{H}=\frac{y_3-\tan \mathrm{\&theta;}*h_3}{y_3+Y}---\left(6\right)$

That is: $h_3=\frac{y_3*H}{y_3+Y+\tan \mathrm{\&theta;}*H}=f\left(y_3\right)---\left(7\right)$

By following formula, can see h ₃About y ₃Function.

Weld seam is moved along vertical stratification light belt direction, until the structure light belt intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 ', △ y3=△ y3 '-Y/K, namely obtain the size of side-play amount △ y3.Take y1 as example, in Fig. 1 (a), by image, processing the side-play amount that obtains is △ y1, and the size of y1 is K* △ y1.In like manner as can be known, y3=K* △ y3, therefore as long as process and obtain the value that side-play amount △ y3 can obtain y3 by image, and then can calculate the weld bead height information h of this point ₃.The elevation of water of setting the welding initial point is 0mm, when there is the deviation on vertical direction in video camera with respect to weld seam, will cause the laser stripe in image to be moved on left and right directions, this moment, weld seam did not move, when image produces and so to change, in conjunction with above-mentioned scaling method, can identify the deviation that welding gun exists with actual welds in vertical direction.So far, weld seam and structure light belt intersection point place weld bead height information have been obtained.

The demonstration of Fig. 2 (a) illustrated embodiment, weldment 4 is shown in (1) in Fig. 2 (a) with the position relationship of video camera 5, weld image is shown in (2) in Fig. 2 (a).When there is the deviation on horizontal direction in video camera 5 with respect to weld seam 7, will cause the weld seam 7 in its image, at above-below direction, the movement of position as shown in phantom in FIG. occurs, this moment, laser stripe 6 did not move.

The demonstration of Fig. 2 (b) illustrated embodiment, weldment 4 is shown in (1) in Fig. 2 (b) with the position relationship of video camera 5, weld image is shown in (2) in Fig. 2 (b).When there is the deviation on vertical direction in video camera 5 with respect to weld seam 7, will cause the laser stripe 6 in image, at left and right directions, the movement of position as shown in phantom in FIG. occurs, this moment, weld seam 7 did not move.

Embodiment 1

In the present embodiment, acute angle of wedge shape target employing is the triangle wedge of 30 °, major equipment used has the scrambler in video camera, laser instrument, motion, triangle wedge, weldment, electronics wedge shape vernier scale and the servomotor that one of them acute angle is 30 °, in described video camera, relates generally to camera lens;

The first step, carry out horizontal departure identification

By controlling motor, video camera is moved to the distance of 10mm along the parallel direction of structure light belt, then calculate the pixel value that in two dimensional image, weld seam moves up at upper and lower, and then calculate the actual range size of each pixel representative, can obtain the Proportional coefficient K between pixel value and actual value, K=83;

Second step, identify the vertical direction deviation

2-I step, operate as above-mentioned Fig. 1 (b) illustrated embodiment, and the triangle wedge is placed on to the video camera below, guarantees that the laser instrument band just beats in triangle wedge edge; The position that the acute angle summit of triangle wedge is in to surface level 0mm is denoted as D,

2-II step, operate and calculate as above-mentioned Fig. 1 (c) illustrated embodiment, and wherein the triangle wedge moves forward y _X1, the height A C=10mm place that the structure light belt is risen on surface level stops, and the scrambler in motion records the triangle wedge and moves forward distance y _X1=20.118mm, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₁=0.0291mm;

2-III step, repeat the operating process of 2-II step, just continues the triangle wedge is moved forward to distance y again _X2, the structure light belt is risen to apart from the height A on surface level ' and C '=20mm place stops, and the scrambler in motion records the triangle wedge and moves forward distance y _X2=20.118mm, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₂=0.03497mm;

The 3rd step, carry out the Constructed Lighting Vision System parameter calibration with video camera

Description as above-mentioned Fig. 1 (d) illustrated embodiment obtains:

$\frac{{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{BC}}{{\mathrm{BO}}^{\&prime;}}---\left(1\right)$

$\frac{{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2}{{\mathrm{OO}}^{\&prime;}}=\frac{B^{\&prime;}{C}^{\&prime;}}{B^{\&prime;}{O}^{\&prime;}}---\left(2\right)$

$\left\{\begin{array}{c}\frac{h_1}{H}=\frac{y_1-(y_{x1}-{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+Y}\\ \frac{h_2}{H}=\frac{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+y_2-(y_{x1}+y_{x2}-{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_2+Y}\end{array}\right.---\left(3\right)$

According to (3) formula, wherein,

h ₁=AC=10mm,h ₂=A’C’=20mm，

Y as can be known again ₁=K* △ y ₁=83*0.0291=2.4153, y ₂=K* △ y ₂=83*0.03497=2.9024;

The substitution known quantity obtains

Calculate H=120.1, Y=-6.8379

By

Obtain tan θ=0.2787

So far, with video camera, carry out the Constructed Lighting Vision System parameter calibration and complete, the numerical value that has obtained respectively parameter K, H, Y, tan θ is: K=83, and H=120.1, tan θ=0.2787, Y=-6.8379, wherein negative sign does not represent size, only represents directional information.

The 4th step, the obtaining of weld bead height information

In the weld joint tracking situation, what measure is the weld bead height information at weld seam and structure light belt intersection point place, the weldment 4 that will measure weld bead height information moves, until will measure weld seam solder joint place and the structure light belt of weld bead height, intersect, the note intersection point is M, make the extended line of OM and hand over the 0mm plane in a N, the distance of note N point and 2-I step intermediate cam shape wedge acute angle summit D is y ₃=ND, visible y ₃Be the horizontal range of intersection point M and the 2-I step intermediate cam shape wedge acute angle summit D of structure light belt and the weld seam solder joint that will measure weld bead height, as the vertical line of M point to the 0mm plane, the point that hangs down is P, remembers that some M and the distance of some P are h ₃=MP,

According to △ MNP ∽ △ ONO ' is similar, obtain:

$\frac{\mathrm{MP}}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{NP}}{{\mathrm{NO}}^{\&prime;}}$

Be:

$\frac{h_3}{H}=\frac{y_3-\tan \mathrm{\&theta;}*h_3}{y_3+Y}---\left(6\right)$

That is: $h_3=\frac{y_3*H}{y_3+Y+\tan \mathrm{\&theta;}*H}=f\left(y_3\right)---\left(7\right)$

Weld seam is moved along vertical stratification light belt direction, until the structure light belt intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 '=-0.05168mm, according to △ y3=△ y3 '-Y/K=-0.05168-(6.8379)/83=0.0307mm, side-play amount △ y3 can be tried to achieve, and then y3=K* △ y3=83*0.0307=2.5481 can be obtained.Above-mentioned negative sign does not all represent size, only represents directional information.

According to (7) formula $h_3=\frac{y_3*H}{y_3+Y+\tan \mathrm{\&theta;}*H},$

Can obtain weld bead height

$h_3=\frac{2.5481*120.1}{2.5481+(-6.8379)+0.2787*120.1}=10.50\mathrm{mm}$

Use electronics wedge shape vernier scale butt welded seam true altitude to measure weld bead height and be 10.45mm.

By the product of two dimensional image coordinate and ratio K, can determine the two-dimensional coordinate (181,130.5) at weld seam solder joint to be measured place, realize the reconstruct (181,130.5,10.50) of weld seam three-dimensional camera coordinate.

Embodiment 2

Except by controlling motor, video camera being moved along the vertical stratification light direction, the triangle wedge moves forward and makes the structure light belt rise to h on surface level ₁The At The Height of=AC=5mm stops, distance y _X1=10.079mm and structured light are mobile apart from △ y on the picture plane ₁=0.01424mm; The triangle wedge moves forward the structure light belt is risen to apart from h on surface level ₂The At The Height of=A ' C '=15mm stops, distance y _X2=20.157mm and structured light are mobile apart from △ y on the picture plane ₂=0.0325mm; In addition, other are with embodiment 1,

The camera calibration result is K=82.8, H=121.3, and tan θ=0.2826, Y=-6.8612, negative sign does not represent size herein, only represents directional information.

Weld seam is moved along the vertical stratification light direction, until structured light intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 '=-0.05176mm, according to △ y3=△ y3 '-Y/K=-0.05176-(6.8612)/82.8=0.0311mm, side-play amount can be tried to achieve, and then y3=K* △ y3=82.8*0.0311=2.5751 can be obtained.Above-mentioned negative sign does not all represent size, only represents directional information.

According to (7) formula

Can obtain weld bead height

$h_3=\frac{2.5751*121.3}{2.5751+(-6.8612)+0.2826*121.3}=10.31\mathrm{mm}$

Use electronics wedge shape vernier scale butt welded seam true altitude to measure weld bead height and be 10.31mm.

By the product of two dimensional image coordinate and ratio K, can determine the two-dimensional coordinate (181.5,130.2) at weld seam solder joint to be measured place, realize the reconstruct (181.5,130.2,10.31) of weld seam three-dimensional camera coordinate.

Embodiment 3

Except by controlling motor, video camera being moved to 15mm along the parallel direction of structure light belt, the triangle wedge moves forward and makes the structure light belt rise to h on surface level ₁The At The Height of=AC=15mm stops, distance y _X1=28.737mm and structured light are mobile apart from △ y on the picture plane ₁=0.02594; The triangle wedge moves forward the structure light belt is risen to apart from h on surface level ₂The At The Height of=A ' C '=25mm stops, distance y _X2=21.443mm and structured light are mobile apart from △ y on the picture plane ₂=0.05701mm; In addition, other are with embodiment 1,

The camera calibration result is K=83.1, H=119.8, and tan θ=0.2741, Y=-6.8856, negative sign does not represent size herein, only represents directional information.

Weld seam is moved along the vertical stratification light direction, until structured light intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 '=-0.05316mm, according to △ y3=△ y3 '-Y/K=-0.05316-(6.8856)/83.1=0.0297mm, side-play amount can be tried to achieve, and then y3=K* △ y3=83.1*0.0297=2.4681 can be obtained.Above-mentioned negative sign does not all represent size, only represents directional information.

According to (7) formula Can obtain weld bead height

$h_3=\frac{2.4681*119.8}{2.4681+(-6.8856)+0.2741*119.8}=10.40\mathrm{mm}$

Use electronics wedge shape vernier scale butt welded seam true altitude to measure weld bead height and be 10.45mm.

The two-dimensional coordinate (181.2,130.2) that can determine weld seam solder joint to be measured place by the product of two dimensional image coordinate and ratio K has been realized the reconstruct (181.2,130.3,10.40) of weld seam three-dimensional camera coordinate.

Embodiment 4

It except the acute angle of triangle wedge shape target employing that uses, is the triangle wedge of 20 °; In addition, other are with embodiment 1, and resulting Different Results is as follows in calibration process:

In the first step, carry out in the horizontal departure identification step, by controlling motor, video camera is moved to the distance of 10mm along the parallel direction of structure light belt, then calculate the pixel value that in two dimensional image, weld seam moves up at upper and lower, and then calculate the actual range size of each pixel representative, can obtain the Proportional coefficient K between pixel value and actual value, K=83.2;

In second step carried out identification step to the vertical direction deviation, in 2-II step, wherein the triangle wedge moved forward y _X1, the height A C=10mm place that the structure light belt is risen on surface level stops, and the scrambler in motion records the triangle wedge and moves forward distance y _X1=30.306mm, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₁=0.0297mm; In 2-III step, repeat the operating process of 2-II step, just continue the triangle wedge is moved forward to distance y again _X2, the structure light belt is risen to apart from the height A on surface level ' and C '=20mm place stops, and the scrambler in motion records the triangle wedge and moves forward distance y _X2=30.403mm, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₂=0.0369mm;

In the 3rd step is carried out Constructed Lighting Vision System parameter calibration process with video camera,

By formula (3)

$\left\{\begin{array}{c}\frac{h_1}{H}=\frac{y_1-(y_{x1}-{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+Y}\\ \frac{h_2}{H}=\frac{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+y_2-(y_{x1}+y_{x2}-{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_2+Y}\end{array}\right.$

Wherein,

h ₁=AC=10mm,h ₂=A’C’=20mm，α=20°

Y as can be known again ₁=K* △ y ₁=83.2*0.0297=2.4680, y ₂=K* △ y ₂=83.2*0.0369=3.0729;

The substitution known quantity obtains

Calculate H=121.1, Y=-6.8643

By

Obtain tan θ=0.2831

So far, with video camera, carry out the Constructed Lighting Vision System parameter calibration and complete, the numerical value that has obtained respectively parameter K, H, Y, tan θ is: K=83.2, and H=121.1, tan θ=0.2831, Y=-6.8643, wherein negative sign does not represent size, only represents directional information.

In the obtaining step of the 4th step weld bead height information, weld seam is moved along the vertical stratification light direction, until structured light intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 '=-0.05221mm, according to △ y3=△ y3 '-Y/K=-0.05221-(6.8643)/83.2=0.0303mm, side-play amount can be tried to achieve, and then y3=K* △ y3=83.2*0.0303=2.5210 can be obtained.Above-mentioned negative sign does not all represent size, only represents directional information.

According to (7) formula

Can obtain weld bead height

$h_3=\frac{2.5210*121.1}{2.5210+(-6.8643)+0.2831*121.1}=10.20\mathrm{mm}$

Use electronics wedge shape vernier scale butt welded seam true altitude to measure weld bead height and be 10.43mm.

By the product of two dimensional image coordinate and ratio K, can determine the two-dimensional coordinate (179.5,131.2) at weld seam solder joint to be measured place, realize the reconstruct (179.5,131.2,10.20) of weld seam three-dimensional camera coordinate.

Embodiment 5

It except the acute angle of triangle wedge shape target employing that uses, is the triangle wedge of 60 °; In addition, other are with embodiment 1, and resulting Different Results is as follows in calibration process:

In the first step, carry out in the horizontal departure identification step, by controlling motor, video camera is moved to the distance of 10mm along the parallel direction of structure light belt, then calculate the pixel value that in two dimensional image, weld seam moves up at upper and lower, and then calculate the actual range size of each pixel representative, can obtain the Proportional coefficient K between pixel value and actual value, K=83.7;

In second step carried out identification step to the vertical direction deviation, in 2-II step, wherein the triangle wedge moved forward y _X1, the height A C=10mm place that the structure light belt is risen on surface level stops, and the scrambler in motion records the triangle wedge and moves forward distance y _X1=8.5675mm, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₁=0.0289mm; In 2-III step, repeat the operating process of 2-II step, just continue the triangle wedge is moved forward to distance y again _X2, the structure light belt is risen to apart from the height A on surface level ' and C '=20mm place stops, and the scrambler in motion records the triangle wedge and moves forward distance y _X2=8.5694mm, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₂=0.0348mm;

In the 3rd step is carried out Constructed Lighting Vision System parameter calibration process with video camera,

By formula (3)

$\left\{\begin{array}{c}\frac{h_1}{H}=\frac{y_1-(y_{x1}-{\mathrm{<figure-callout\; id="1"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_1/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+Y}\\ \frac{h_2}{H}=\frac{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+y_2-(y_{x1}+y_{x2}-{\mathrm{<figure-callout\; id="2"\; label="h"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">h</figure-callout>}}_2/\tan \mathrm{\&alpha;})}{{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="HDA00003620396800011.png,HDA00003620396800021.png"\; state="\{\{state\}\}">y</figure-callout>}}_2+Y}\end{array}\right.$

Wherein,

h ₁=AC=10mm,h ₂=A’C’=20mm，α=60°

Y as can be known again ₁=K* △ y ₁=83.7*0.0289=2.4204, y ₂=K* △ y ₂=83.7*0.0348=2.9108;

The substitution known quantity obtains

Calculate H=119.2, Y=-6.8731

By

Obtain tan θ=0.2794

So far, with video camera, carry out the Constructed Lighting Vision System parameter calibration and complete, the numerical value that has obtained respectively parameter K, H, Y, tan θ is: K=83.7, and H=119.2, tan θ=0.2794, Y=-6.8731, wherein negative sign does not represent size, only represents directional information.

In the obtaining step of the 4th step weld bead height information, weld seam is moved along the vertical stratification light direction, until structured light intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 '=-0.05315mm, according to △ y3=△ y3 '-Y/K=-0.05315-(6.8731)/83.7=0.0289mm, side-play amount can be tried to achieve, and then y3=K* △ y3=83.7*0.0289=2.4189 can be obtained.Above-mentioned negative sign does not all represent size, only represents directional information.

According to (7) formula Can obtain weld bead height

$h_3=\frac{2.4189*119.2}{2.4189+(-6.8731)+0.2794*119.2}=9.99\mathrm{mm}$

Use electronics wedge shape vernier scale butt welded seam true altitude to measure weld bead height and be 10.23mm.

By the product of two dimensional image coordinate and ratio K, can determine the two-dimensional coordinate (179.1,130.2) at weld seam solder joint to be measured place, realize the reconstruct (179.1,130.2,9.99) of weld seam three-dimensional camera coordinate.

In above-described embodiment, laser instrument used is laser line generator.

The weld seam of choosing weldment diverse location height carries out test of many times, in test, with electronics wedge shape vernier scale butt welded seam height situation, has carried out actual measurement, obtains weld bead height information.Actual welds three-dimensional coordinate information and the weld seam three-dimensional coordinate information that demarcation obtains are carried out respectively to match, and the fitting result curve is shown in Fig. 3.

Experimental result shown in Figure 3 shows, the absolute error of directions X is in 0.5mm, and the absolute error that height Z measures is in 1mm, and the mean square deviation result is 0.707, and experiment shows, this arithmetic accuracy and stability are higher, meet the engineering precision demand.

Equipment related in above-described embodiment is by commercially available.

Claims (3)

1. be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information, it is characterized in that: be a kind of parameter calibration method of Constructed Lighting Vision System based on the wedge shape target, step is:

Described wedge shape target adopts the triangle wedge, it is triangle wedge, weldment, electronics wedge shape vernier scale and the servomotor of 20 °～60 ° that major equipment used has scrambler, one of them acute angle in video camera, laser instrument, motion, in described video camera, relates generally to camera lens;

The first step, identify horizontal departure

By controlling servomotor, video camera is moved to distance the survey record of 5mm～25mm along the parallel direction of structure light belt, then calculate the pixel value that in two dimensional image, weld seam moves up at upper and lower, and then calculate the actual range size of each pixel representative, can obtain the Proportional coefficient K between pixel value and actual distance value;

Second step, identify the vertical direction deviation

2-I step, be placed on the video camera below by the triangle wedge, guarantees that the laser instrument band just beats in triangle wedge edge;

2-II step, move forward distance y by the triangle wedge _X1The structure light belt that laser instrument produces can move up by the hypotenuse along the triangle wedge along with the movement of triangle wedge, until the height A C=10mm that this structure light belt is risen on surface level stops, structure light belt mobile distance on the picture plane is △ y ₁, the scrambler in moving process in motion can record the distance y that the triangle wedge moves forward _X1, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₁

2-III step, repeat the operating process of 2-II step, just continues the triangle wedge is moved forward to distance y again _X2, until the structure light belt is risen to apart from the height A on surface level ' and C '=20mm place stops, and structure light belt this moment mobile distance on the picture plane is △ y ₂, the scrambler in moving process in motion can record the distance y that the triangle wedge moves forward _X2, in two dimensional image, process and the interrecord structure light belt mobile apart from △ y on the picture plane ₂

The 3rd step, carry out the Constructed Lighting Vision System parameter calibration with video camera

The position mark of the residing surface level 0mm in acute angle summit of triangle wedge that will be arranged in the triangle wedge position of 2-I step operating process is made D, the position mark of acute angle summit surface level 0mm of living in of triangle wedge that will be arranged in the triangle wedge position of 2-II step operating process is made D1, the position mark of acute angle summit surface level 0mm of living in of triangle wedge that will be arranged in the triangle wedge position of 2-III step operating process is made D2, and the distance between D and D1 is y _X1, the distance between D1 and D2 is y _X2,

If the line laser that laser instrument sends is A with the hypotenuse intersection point of the triangle wedge of the triangle wedge position that is arranged in the operating process of 2-II step, the camera lens center is designated as O, the extended line of camera lens center O and A hands over the surface level of height 0mm in a B, make the vertical line of A to the surface level of height 0mm, hand over surface level in a C, 3 form △ ABC; Make the vertical line of camera lens central point O to the surface level of height 0mm, hand over surface level in an O ', 3 O, B, O ' form triangle △ OBO ', and the distance between note AC is h ₁,

According to △ ABC ∽ △ OBO ' is similar, obtain:

$\frac{h_1}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{BC}}{{\mathrm{BO}}^{\&prime;}}---\left(1\right)$

If the line laser that laser instrument sends is A ' with the hypotenuse intersection point of the triangle wedge of the position of the triangle wedge that is arranged in the operating process of 2-III step, the extended line of camera lens center O and A ' hands over the surface level of height 0mm in a B ', make the vertical line of A ' to the surface level of height 0mm, hand over surface level in a C ', 3 form △ A ' B ' C '; Make camera lens 1 central point O and hand over surface level and some O ' to the vertical line of the surface level of height 0mm, 3 O, B ', O ' form triangle △ OB ' O '.Distance between note A ' C ' is h ₂.

According to △ A ' B ' C ' ∽ △ OB ' O ' is similar, obtain:

$\frac{h_2}{{\mathrm{OO}}^{\&prime;}}=\frac{B^{\&prime;}{C}^{\&prime;}}{B^{\&prime;}{O}^{\&prime;}}---\left(2\right)$

Laser instrument central shaft and vertical direction angle are designated as to θ, note BD=y ₁, B ' B=y ₂,

BC=y ₁-(y _X1-h ₁/ tan θ), B ' C '=y ₁+ y ₂-(y _X1+ y _X2-h ₁/ tan α), note optical center O be H to height 0mm's apart from OO ', 2-I step intermediate cam shape wedge acute angle summit D to camera optics central shaft OO ' apart from being Y, i.e. OO '=H, DO '=Y, can obtain BO '=Y+y ₁, B ' O '=Y+y ₁+ y ₂, bring (1) into and can obtain with (2):

$\left\{\begin{array}{c}\frac{h_1}{H}=\frac{y_1-(y_{x1}-h_1/\tan \mathrm{\&alpha;})}{y_1+Y}\\ \frac{h_2}{H}=\frac{y_1+y_2-(y_{x1}+y_{x2}-h_2/\tan \mathrm{\&alpha;})}{y_1+y_2+Y}\end{array}\right.---\left(3\right)$

H wherein ₁=10mm, h ₂=20mm, α=20 °～60 °, bring known quantity into, can obtain:

By (4) formula, can be obtained the value of Parameter H, Y, in addition, by geometric relationship in figure, can be obtained:

y _x1-AC/tanα＝AC*tanθ

That is:

$\tan \mathrm{\&theta;}=\frac{y_{x1}-\mathrm{AC}/\tan \mathrm{\&alpha;}}{\mathrm{AC}}---\left(5\right)$

Known h ₁=AC=10mm, α=20 °～60 ° of substitutions (5),

By following formula, namely can calculate tan θ.

So far, with video camera, carry out the Constructed Lighting Vision System parameter calibration and complete, obtained respectively the numerical value of parameter K, H, Y, tan θ;

The 4th step, the obtaining of weld bead height information

In the weld joint tracking situation, what measure is the weld bead height information at weld seam and structure light belt intersection point place, the weldment that will measure weld bead height information moves, until will measure weld seam solder joint place and the structure light belt of weld bead height, intersect, the note intersection point is M, make the extended line of OM and hand over the 0mm plane in a N, the distance of note N point and 2-I step intermediate cam shape wedge acute angle summit D is y ₃=ND, visible y ₃Be the horizontal range of intersection point M and the 2-I step intermediate cam shape wedge acute angle summit D of structure light belt and the weld seam solder joint that will measure weld bead height, as the vertical line of M point to the 0mm plane, the point that hangs down is P, remembers that some M and the distance of some P are h ₃=MP,

According to △ MNP ∽ △ ONO ' is similar, obtain:

$\frac{\mathrm{MP}}{{\mathrm{OO}}^{\&prime;}}=\frac{\mathrm{NP}}{{\mathrm{NO}}^{\&prime;}}$

Be:

$\frac{h_3}{H}=\frac{y_3-\tan \mathrm{\&theta;}*h_3}{y_3+Y}---\left(6\right)$

That is: $h_3=\frac{y_3*H}{y_3+Y+\tan \mathrm{\&theta;}*H}=f\left(y_3\right)---\left(7\right)$

By following formula, can see h ₃About y ₃Function,

Weld seam is moved along vertical stratification light belt direction, until the structure light belt intersects with the solder joint place that will measure weld bead height, by two dimensional image, record between the optical centre axle OO ' of weld seam and structure light belt intersection point place and camera lens apart from △ y3 ', △ y3=△ y3 '-Y/K, namely obtain the size of side-play amount △ y3.the size of y3 is y3=K* △ y3, the elevation of water of setting the welding initial point is 0mm, when there is the deviation on vertical direction in video camera with respect to weld seam, will cause the laser stripe in image to be moved on left and right directions, this moment, weld seam did not move, when the image generation so changes, by image, process and record weld seam and structure light belt intersection point shift offset △ y3, and then learn y3=K* △ y3, in conjunction with (7) formula, can identify the deviation that welding gun exists with actual welds in vertical direction, so far, weld seam and structure light belt intersection point place weld bead height information have been obtained, realized the reconstruct of three-dimensional camera coordinate.

2. said be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information according to claim 1, it is characterized in that: described laser instrument is laser line generator.

3. said be used to obtaining the structured light vision sensor parameter calibration method of weld bead height information according to claim 1, it is characterized in that: described triangle wedge, one of them acute angle are 30 °.

Images