CN109658456A - Tank body inside fillet laser visual vision positioning method - Google Patents
Tank body inside fillet laser visual vision positioning method Download PDFInfo
- Publication number
- CN109658456A CN109658456A CN201811271177.XA CN201811271177A CN109658456A CN 109658456 A CN109658456 A CN 109658456A CN 201811271177 A CN201811271177 A CN 201811271177A CN 109658456 A CN109658456 A CN 109658456A
- Authority
- CN
- China
- Prior art keywords
- weld seam
- point
- laser
- image
- tank body
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000000007 visual effect Effects 0.000 title claims abstract description 23
- 238000003466 welding Methods 0.000 claims abstract description 46
- 230000003287 optical effect Effects 0.000 claims abstract description 39
- 230000008569 process Effects 0.000 claims abstract description 13
- 238000001914 filtration Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 230000017105 transposition Effects 0.000 claims description 7
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000009792 diffusion process Methods 0.000 claims description 4
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000003672 processing method Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 15
- 238000003860 storage Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000011218 segmentation Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000004807 localization Effects 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 238000013519 translation Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000002893 slag Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 238000005476 soldering Methods 0.000 description 1
- 239000003351 stiffener Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- G06T7/85—Stereo camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Laser Beam Processing (AREA)
Abstract
The invention discloses a kind of tank body inside fillet laser visual vision positioning methods, it include: S1, the optical plane spatially projected to the laser that CCD camera and laser issue is demarcated, and obtains calibrating parameters comprising join outside camera internal reference, Camera extrinsic and optical plane;S2 obtains the characteristic point two-dimensional coordinate in characteristics of weld seam region;S3 obtains the three-dimensional world coordinate of weld seam by the characteristic point two-dimensional coordinate in the obtained characteristics of weld seam region the obtained calibrating parameters of step S1 and step S2, realizes the positioning of weld seam.Tank body inside fillet laser visual vision positioning method provided by the invention, it is pre-processed by the two dimensional image of the fillet weld obtained to CCD camera, the three-dimensional world coordinate in three-dimensional world coordinate system is converted by the characteristic area of weld seam, to realize the precise positioning to weld seam, its process flow is quick, and accurate positioning, it can be widely applied to vision welding image processing method technical field.
Description
Technical field
The present invention relates to vision welding image processing method technical fields.It is more particularly related to a kind of tank body
Inside fillet laser visual vision positioning method.
Background technique
In petroleum chemical enterprise, the storage and transport of stoste can be usually carried out with large-scale storage tank.In the manufacturing process of storage tank
In, soldering need to be carried out to internal a large amount of stiffening ring.And the welding of stiffening ring at present is mostly artificial weldering, a large amount of repeatability
Not only there is security risks by human weld, and working efficiency is not also high, therefore automatic welding is essential.
With the development of modern intelligent Manufacturing Technology, the application of automatic welding is also more and more extensive, and visual sensor
It is largely used in automatic guidance welding due to the features such as its is non-contact, high-precision.Actual welding positioning at this stage is still more
For two-dimensional localization, i.e., two dimensional character information is extracted by laser stripe, obtain its horizontal deviator, though this method is easy to operate, and
Horizontal accuracy is higher, but due to that can not obtain accurate depth information, thus limitation is very big.And three-dimensional localization method is studied in part
More complex calibration mode and detection algorithm are then used, although positioning accuracy is higher, processing speed is slower, is not suitable for industry
It is practical.
Summary of the invention
It is an object of the invention to solve at least the above problems, and provide the advantages of at least will be described later.
It is a still further object of the present invention to provide a kind for the treatment of processes quickly, accurately tank body inside fillet swashs locating effect
Light 3 D visual vision positioning method.
In order to realize these purposes and other advantages according to the present invention, a kind of tank body inside fillet laser vision is provided
Three-dimensional visual positioning method, comprising:
S1, the optical plane spatially projected to the laser that CCD camera and laser issue are demarcated, are demarcated
Parameter comprising join outside camera internal reference, Camera extrinsic and optical plane;
S2 obtains the characteristic point two-dimensional coordinate in characteristics of weld seam region;
S3 is sat by the characteristic point two dimension in the obtained characteristics of weld seam region the obtained calibrating parameters of step S1 and step S2
Mark, obtains the three-dimensional world coordinate of weld seam, realizes the positioning of weld seam.
Preferably, the step S2 is specifically included:
S21, CCD camera obtain the original image of commissure;
S22 carries out just welded seam area using the template matching based on gray scale according to the original image that step S21 is obtained
Step positioning, finds rough characteristics of weld seam region;
S23 pre-processes weld seam characteristic area comprising smoothed image is denoised by median filtering;
S24, using improved Canny operator extraction characteristics of weld seam region edge and carry out straight line fitting, then to warp
It crosses pretreated template area to be repositioned, and obtains the two-dimensional coordinate of the characteristic point in characteristics of weld seam region.
Preferably, the step S24 is,
S241 carries out the edge extracting in characteristics of weld seam region using improved Canny operator:
The gaussian filtering that script in Canny operator is replaced using anisotropic diffusion filtering, is avoided the occurrence of false edge or lost
The case where losing part local edge, it may be assumed that
I is image in formula (1), and t is the number of iterations, four divergence expression formulas Respectively to the local derviation of current pixel on four direction;
S242 obtains high-low threshold value using Otsu algorithm automatically, avoids manually adjusting high-low threshold value, improves the adaptive of algorithm
Ying Xing;
S243 carries out straight line fitting using least square method in conjunction with the marginal information extracted, will be on extracted edge
Each point is used as input point, and fitting obtains:
Y=a0+a1x (2)
Wherein: in formula (2)By two of fitting
Straight line simultaneous obtains:
And the solution of equation (3)As it is fitted resulting intersection point, that is, characteristics of weld seam region
Characteristic point;
Wherein, (x, y) is the pixel coordinate of each point on the edge in extracted characteristics of weld seam region.
Preferably, the step S1 is specifically included:
Picture and Light-plane calibration are demarcated according to 16 width of the focal length of CCD camera, the parameter of scaling board, camera calibration acquisition
Two width pictures calibrating parameters are obtained using laser triangulation.
Preferably, the step S1 is specifically included:
S11 selects the plane dot target of 60mm*60mm according to the visual field size of CCD camera, and acquires 16 differences
The uncalibrated image of pose carries out camera calibration, the 16 width uncalibrated image its need to cover entire CCD camera visual field;
S12, choose welding gun theory offset highs and lows two positions come define a world coordinate system and
One reference frame;In order to determine the pose of optical plane, the scaling board image and laser image of two kinds of height are taken respectively;
S13 records all points of the light of two kinds of height, defines it in three-dimensional world coordinate system Z=0 plane projection point set
Respectively minimum point Pa(Xa,Ya, 0), highest point Pb(Xb,Yb,Zb);
S14 finds out the mass center P of all the pointsc(Xc,Yc,Zc), and system homogeneous equation matrix M is established, its transposition MT is sought, and
Carry out singular value decomposition:
Wherein, MTIt is the transposition of the homogeneous equation matrix M of the mass center of all the points in optical plane, UM,SM,VT MIt indicates unusual
It is worth the matrix decomposed;
In conjunction with the normal vector [α, beta, gamma] for the optical plane that these three values are fitted, and find out the expression formula of optical plane: α xw
+βyw+λzw=0 (5);
Wherein, (XW,YW,ZW) indicate optical plane on each point three-dimensional coordinate.
Preferably, the highest point of the welding gun theory offset is 5cm, and the minimum point of the welding gun theory offset is 0cm.
The present invention is include at least the following beneficial effects:
Tank body inside fillet laser visual vision positioning method provided by the invention, passes through what is obtained to CCD camera
The two dimensional image of fillet weld is pre-processed, and constructs three-dimensional world coordinate system, converts three-dimensional generation for the characteristic area of weld seam
Three-dimensional world coordinate in boundary's coordinate system, to realize the precise positioning to weld seam, process flow is simple, and positioning accurate
Standard can be widely applied to vision welding image processing method technical field.
Further advantage, target and feature of the invention will be partially reflected by the following instructions, and part will also be by this
The research and practice of invention and be understood by the person skilled in the art.
Detailed description of the invention
Fig. 1 is the line laser mathematical model figure that the embodiment of the present invention 1 constructs;
Fig. 2 is the scaling board image of two kinds of height of the embodiment of the present invention 1 and the schematic diagram of laser image;
Fig. 3 is the image processing flow schematic diagram of the embodiment of the present invention 1;
Fig. 4 is the template image of the embodiment of the present invention 1 and the comparison diagram of original image;
Fig. 5 is that the characteristics of weld seam region of the embodiment of the present invention 1 does not pre-process and its respectively by intermediate value and two kinds of mean value filters
Wave treated effect contrast figure;
Fig. 6 is side of the tradition Canny operator with improved Canny operator to weld seam characteristic area in the embodiment of the present invention 1
Edge extraction effect comparison diagram;
Fig. 7 is the process that the embodiment of the present invention 1 carries out straight line fitting to the marginal information in the characteristics of weld seam region extracted
Schematic diagram;
Fig. 8 is the embodiment of the present invention using welding effect figure.
Description of symbols: 1, laser, 2, CCD camera, 3, optical plane, 4, image planes.
Specific embodiment
Present invention will be described in further detail below with reference to the accompanying drawings, to enable those skilled in the art referring to specification text
Word can be implemented accordingly.
It should be noted that experimental method described in following embodiments is unless otherwise specified conventional method, institute
Reagent and material are stated, unless otherwise specified, is commercially obtained;In the description of the present invention, term " transverse direction ", " vertical
To ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", the instructions such as "outside" side
Position or positional relationship are to be based on the orientation or positional relationship shown in the drawings, and are merely for convenience of description of the present invention and simplification of the description,
It is not that the device of indication or suggestion meaning or element must have a particular orientation, be constructed and operated in a specific orientation, because
This is not considered as limiting the invention.
As shown in figures 1-8, the present invention provides a kind of tank body inside fillet laser visual vision positioning method, comprising:
S1, the optical plane spatially projected to the laser that CCD camera and laser issue are demarcated, are demarcated
Parameter comprising join outside camera internal reference, Camera extrinsic and optical plane;
S2 obtains the characteristic point two-dimensional coordinate in characteristics of weld seam region;
S3 is sat by the characteristic point two dimension in the obtained characteristics of weld seam region the obtained calibrating parameters of step S1 and step S2
Mark, obtains the three-dimensional world coordinate of weld seam, realizes the positioning of weld seam.
In this kind of technical solution, the accurate three-dimensional information of stiffening ring fillet weld in order to obtain need to be to camera and optical plane
Establish line laser mathematical model.The laser that laser issues spatially projects an optical plane, and Fig. 2 is its geometry
The mathematical model of relationship.
It is a laser rays that stiffening ring fillet weld intersects with optical plane, and each point on laser rays is required spy
Levy region.With the point O on optical planesFor origin, reference coordinate system O needed for establishing optical planes-XsYsZs, wherein Os-XsYs
The plane at place is optical plane.The mathematic(al) representation of the line laser model can be obtained by several picture are as follows:
Zs=0 (1)
WhereinFor direction vector,For translation vector, R and t determine optical plane relative to taking the photograph
The pose of camera coordinate system.
In another technical solution, the step S2 is specifically included:
S21, CCD camera obtain the original image of commissure;
S22 carries out just welded seam area using the template matching based on gray scale according to the original image that step S21 is obtained
Step positioning, finds rough characteristics of weld seam region;
S23 pre-processes weld seam characteristic area comprising smoothed image is denoised by median filtering;
S24, using improved Canny operator extraction characteristics of weld seam region edge and carry out straight line fitting, then to warp
It crosses pretreated template area to be repositioned, and obtains the two-dimensional coordinate of the characteristic point in characteristics of weld seam region.
In another technical solution, the step S24 is,
S241 carries out the edge extracting in characteristics of weld seam region using improved Canny operator:
The gaussian filtering that script in Canny operator is replaced using anisotropic diffusion filtering, is avoided the occurrence of false edge or lost
The case where losing part local edge, it may be assumed that
I is image in formula (1), and t is the number of iterations, four divergence expression formulas Respectively to the local derviation of current pixel on four direction;
S242 obtains high-low threshold value using Otsu algorithm automatically, avoids manually adjusting high-low threshold value, improves the adaptive of algorithm
Ying Xing;
S243 carries out straight line fitting using least square method in conjunction with the marginal information extracted, will be on extracted edge
Each point is used as input point, and fitting obtains:
Y=a0+a1x (2)
Wherein: in formula (2)By two of fitting
Straight line simultaneous obtains:
And the solution of equation (3)As it is fitted resulting intersection point, that is, characteristics of weld seam region
Characteristic point;
Wherein, (x, y) is the pixel coordinate of each point on the edge in extracted characteristics of weld seam region.
In another technical solution, the step S1 is specifically included:
Picture and Light-plane calibration are demarcated according to 16 width of the focal length of CCD camera, the parameter of scaling board, camera calibration acquisition
Two width pictures calibrating parameters are obtained using laser triangulation.
In another technical solution, the step S1 is specifically included:
S11 selects the plane dot target of 60mm*60mm according to the visual field size of CCD camera, and acquires 16 differences
The uncalibrated image of pose carries out camera calibration, the 16 width uncalibrated image its need to cover entire CCD camera visual field;
S12, choose welding gun theory offset highs and lows two positions come define a world coordinate system and
One reference frame;In order to determine the pose of optical plane, the scaling board image and laser image of two kinds of height are taken respectively;
S13 records all points of the light of two kinds of height, defines it in three-dimensional world coordinate system Z=0 plane projection point set
Respectively minimum point Pa(Xa,Ya, 0), highest point Pb(Xb,Yb,Zb);
S14 finds out the mass center P of all the pointsc(Xc,Yc,Zc), and system homogeneous equation matrix M is established, its transposition MT is sought, and
Carry out singular value decomposition:
Wherein, MTIt is the transposition of the homogeneous equation matrix M of the mass center of all the points in optical plane, UM,SM,VT MIt indicates unusual
It is worth the matrix decomposed;
In conjunction with the normal vector [α, beta, gamma] for the optical plane that these three values are fitted, and find out the expression formula of optical plane: α xw
+βyw+λzw=0 (5);
Wherein, (XW,YW,ZW) indicate optical plane on each point three-dimensional coordinate.
In another technical solution, the highest point of the welding gun theory offset is 5cm, and the welding gun theory offset is most
Low spot is 0cm.
Embodiment 1
Step 1: the calibration of camera and optical plane
In order to convert space three-dimensional information for two-dimensional pixel characteristic point accessed on image, need herein respectively to phase
Machine and optical plane are demarcated.According to the visual field size of camera, the plane dot target of 60mm*60mm is selected.In order to improve mark
Fixed precision, the uncalibrated image that need to acquire 16 different positions and poses demarcate and need to cover entire viewing field of camera.Then, this system
Two positions of highs and lows of welding gun theory offset are chosen to define a world coordinate system and a reference frame.
In order to determine the pose of optical plane, the scaling board image and laser image of two kinds of height are taken respectively, as shown in Fig. 2, Fig. 2 a) figure
Indicate minimum point laser image, Fig. 2 b) indicate minimum point scaling board image, Fig. 2 c) indicate highest point laser image, figure
2d) indicate the scaling board image of highest point.
All points of two kinds of height light are recorded, are respectively minimum point point in world coordinate system Z=0 plane projection point set
Pa(Xa,Ya, 0), highest point Pb(Xb,Yb,Zb).Find out the mass center P of all the pointsc(Xc,Yc,Zc), and establish system homogeneous equation square
Battle array M, seeks its transposition MT, and carry out singular value decomposition:
And the characteristic of the unusual decomposition of matrix be its preceding 10% or even 1% singular value and be whole singular values it
99% or more of sum, therefore need to only extract SMIn first three value, in conjunction with the normal direction for the optical plane that these three values are fitted
It measures [α, beta, gamma], and finds out the expression formula of optical plane:
αxw+βyw+λzw=0 (5)
By calibrating procedure operation, the inside and outside parameter of camera and optical plane is obtained:
Camera internal reference: effective focal length f=0.0086039;Distortion factor K1=-1214.45.
Camera extrinsic: X, Y, Z axis rotation parameter is 359.09,0.146949,359.892;X, Y, Z axis translation parameters be-
0.00296825,0.00188242,0.194793.
Join outside optical plane: X, Y, Z axis rotation parameter is 317.5,359.6,359.7;X, Y, Z axis translation parameters be-
0.00276,0.0136,0.0164.
Step 2: image procossing
Accurate characteristic point coordinate is obtained using the method that two steps position herein.The first step uses the mould based on gray scale
Plate matching carries out Primary Location to welded seam area, finds rough characteristic area.By the collected figure of institute in the welding process
As upper inevitably there is the interference that the reasons such as many welding slag, dense smokes penetrated by four occur, stiffening ring angle can be all caused
There is error in the extraction of seam laser striped.Thus before being repositioned, it is necessary to pre-process, reject to template part
The interference of noise.It completes finally by using improved Canny operator extraction edge feature and carrying out straight line fitting to weld seam spy
The repositioning for levying region, obtains accurate characteristics of weld seam information, Fig. 3 is image processing process.
Primary Location is carried out to image using template matching, i.e., selectes stiffening ring fillet weld characteristic area from piece image
It is found and the most similar part in the region as template, and in other images then obtained.In this way at subsequent image
The region being matched to only just need to be handled when reason, reduces the interference of irrelevant information, Fig. 4 is selected Prototype drawing, and Fig. 4 is left
Side e) indicates original image, f) indicate template area, which only has the 10.66% of original image size, can substantially shorten calculating
Time improves efficiency.
The template matching method based on correlation is used herein, its matching degree is determined with R (r, c), it may be assumed that
Wherein: I (x, y) and I (x+r, y+c) respectively (x, y), (x+r, y+c) pixel gray value,WithRespectively T (x, y), the corresponding region I (x, y) average value, R (r, c) i.e. representative image (r, c) Region Matching coefficient,
It indicates more to match when its numerical value is closer to 1.
The image that is acquired in welding scene there is largely because of welding slag, dense smoke etc. due to the noise that generates.These are all
Meeting generates interference to the extraction of subsequent fillet weld laser stripe.Therefore, institute's matching area need to be pre-processed, smoothed image,
Partial noise interference is removed, Fig. 5 is the figure that camera is intercepted during actual welding, wherein containing due to splashing, Qiang Guang
The much noise of generation is handled to obtain effect picture as shown in Figure 5, g with intermediate value and two kinds of mean value filtering separately below),
H) it is respectively effect picture after Threshold segmentation of original untreated figure and the figure, i), j) is respectively that original graph is filtered by mean value
After wave processing and effect picture of the upper figure after Threshold segmentation;K), l) be respectively original graph after median filter process and on
Effect picture of the figure after Threshold segmentation;Comparison can be seen that median filtering denoising effect is more preferable.By multiple repetition test, this hair
Clearly original graph is carried out after removing dryness processing using median filtering surely, then carry out Threshold segmentation effect picture it is best.
Then, characteristic area is repositioned.It is inevitable although thering is barn door to block arc light in actual welding
Will appear in Fig. 5 the case where partial arc leaks out, the arc light of this randomness causes the extraction of edge feature largely dry
It disturbs, carries out edge extracting used here as a kind of improved Canny operator, to adapt to complicated weld seam picture.
Firstly, replacing the gaussian filtering of script in Canny operator using anisotropic diffusion filtering, false edge is avoided the occurrence of
Or the case where lost part local edge, it may be assumed that
Then high-low threshold value is obtained using Otsu algorithm automatically, avoids manually adjusting high-low threshold value, improves the adaptive of algorithm
Ying Xing.
Fig. 6 is the comparing result of tradition Canny operator and improved Canny operator, 6m) it is traditional operator to characteristics of weld seam
The extraction effect figure in region, 6n) it is to improve Canny operator to the extraction effect figure of weld seam characteristic area.Improved Canny operator
Effect is more preferable in the edge extracting of complicated weld image, precision is higher, and the two is more as shown in Figure 6.
The marginal information extracted is finally combined to carry out straight line fitting, by point each on extracted edge as input
Point, fitting obtain:
Y=a0+a1x (2)
Wherein:
Two straight line simultaneous of fitting are obtained:
And non trivial solutionIt is as fitted resulting intersection point, that is, required characteristic point,
Process is as shown in Figure 7.Fig. 7 o) indicate edge extracting, p) indicate straight line fitting, q) indicate intersecting point coordinate.
Determine finally, converting resulting characteristic point two-dimensional coordinate to three-dimensional world coordinate and realizing by calibrating parameters above
Position.
Application effect
A kind of creeping-type welding robot based on line laser carries out weld image with the template matching based on correlation
Just positioning substantially reduces ROI region, improves image processing efficiency, and use improved Canny operator extraction edge feature, most
It combines calibrating parameters to complete the three-dimensional localization to the characteristic area of stiffening ring fillet weld afterwards, realizes high efficiency smart guidance welding.
Laser-vision sensing system is mainly by CCD camera, line-structured light laser, computer, welding robot of creeping
And its composition such as controller, vision positioning sensor (camera and laser) are placed on the cantilever on front side of welding climbing robot
On, at 50mm before welding gun, separated by light barrier.
Creeping-type welding robot is placed in tank inside when welding, and storage tank drives uniform rotation, machine by rolling stand
Device people then reversely creeps at the same rate and carries out rate controlling using angular position pick up, it is made to be in tank body lowest part always, with
Ground keeps opposing stationary.The working principle of above system are as follows:
After robot motion is tended towards stability, laser projects a laser striation, obtains image by CCD camera, most
Afterwards by computer disposal laser stripe image, tested face of weld three-dimensional coordinate is obtained, and calculates weld seam real offset, is completed
Real-time deviation correcting.
The specific workflow of above system are as follows:
1, camera and optical plane are demarcated;
2, world three dimensional coordinate system is established;
3, it scans and records weld seam initial position;
4, formally start welding tractor, welding tractor is scanned again after stablizing, and determines welding seam position information;
5, the intelligent processing system of computer, the weldering that the welding seam position information obtained in verification step 4 obtains in step 5
Whether seam initial position matches:
If matching, issues instruction, so that welding tractor normal weld;
If mismatching, weld seam real offset is calculated, and controls the position of the welding gun adjusting mechanism adjustment welding gun of welding tractor
It sets;
Judge whether welding gun is in correct position:
It is, so that welding tractor normal weld, normal weld;
No, alarm is shut down.
In order to verify the stability and accuracy of this laser vision guidance welding system, now a storage tank inner stiffener ring is carried out
GMAW welding test.Welding effect by driving welding robot along inner support circle as shown in figure 8, creeped welding, wherein storage tank
Rotation speed and robot crawling speed are 230mm/min, keep matching by angular position pick up, weldingvoltage 22.3V, weldering
Meet electric current 220A.
Although the embodiments of the present invention have been disclosed as above, but its is not only in the description and the implementation listed
With it can be fully applied to various fields suitable for the present invention, for those skilled in the art, can be easily
Realize other modification, therefore without departing from the general concept defined in the claims and the equivalent scope, the present invention is simultaneously unlimited
In specific details and legend shown and described herein.
Claims (6)
1. a kind of tank body inside fillet laser visual vision positioning method characterized by comprising
S1, the optical plane spatially projected to the laser that CCD camera and laser issue are demarcated, and calibration ginseng is obtained
Number comprising join outside camera internal reference, Camera extrinsic and optical plane;
S2 obtains the characteristic point two-dimensional coordinate in characteristics of weld seam region;
S3 is obtained by the characteristic point two-dimensional coordinate in the obtained characteristics of weld seam region the obtained calibrating parameters of step S1 and step S2
To the three-dimensional world coordinate of weld seam, the positioning of weld seam is realized.
2. tank body inside fillet laser visual vision positioning method as described in claim 1, which is characterized in that the step
Rapid S2 is specifically included:
S21, CCD camera obtain the original image of commissure;
S22, according to the original image that step S21 is obtained, it is tentatively fixed to be carried out using the template matching based on gray scale to welded seam area
Position, finds rough characteristics of weld seam region;
S23 pre-processes weld seam characteristic area comprising smoothed image is denoised by median filtering;
S24, using improved Canny operator extraction characteristics of weld seam region edge and carry out straight line fitting, then to by pre-
The template area of processing is repositioned, and obtains the two-dimensional coordinate of the characteristic point in characteristics of weld seam region.
3. tank body inside fillet laser visual vision positioning method as claimed in claim 2, which is characterized in that the step
Rapid S24 is,
S241 carries out the edge extracting in characteristics of weld seam region using improved Canny operator:
The gaussian filtering that script in Canny operator is replaced using anisotropic diffusion filtering, avoids the occurrence of false edge or loss portion
The case where dividing local edge, it may be assumed that
I is image in formula (1), and t is the number of iterations, four divergence expression formula λ cNx,y▽N(It)、cSx,y▽S(It)、cEx,y▽E
(It)、cWx,y▽W(It) it is respectively local derviation on four direction to current pixel;
S242 obtains high-low threshold value using Otsu algorithm automatically, avoids manually adjusting high-low threshold value, improves the adaptive of algorithm
Property;
S243 carries out straight line fitting using least square method in conjunction with the marginal information extracted, will be each on extracted edge
Point is used as input point, and fitting obtains:
Y=a0+a1x (2)
Wherein: in formula (2)By two straight lines of fitting
Simultaneous obtains:
And the solution of equation (3)As it is fitted resulting intersection point, that is, the spy in characteristics of weld seam region
Sign point;
Wherein, (x, y) is the pixel coordinate of each point on the edge in extracted characteristics of weld seam region.
4. tank body inside fillet laser visual vision positioning method as claimed in claim 3, which is characterized in that the step
Rapid S1 is specifically included:
According to the two of the focal length of CCD camera, the parameter of scaling board, the 16 width calibration picture of camera calibration acquisition and Light-plane calibration
Width picture obtains calibrating parameters using laser triangulation.
5. tank body inside fillet laser visual vision positioning method as claimed in claim 4, which is characterized in that the step
Rapid S1 is specifically included:
S11 selects the plane dot target of 60mm*60mm according to the visual field size of CCD camera, and acquires 16 different positions and poses
Uncalibrated image carry out camera calibration, the 16 width uncalibrated image its need to cover entire CCD camera visual field;
S12 chooses two positions of the highs and lows of welding gun theory offset to define a world coordinate system and one
Reference frame;In order to determine the pose of optical plane, the scaling board image and laser image of two kinds of height are taken respectively;
S13 records all points of the light of two kinds of height, defines it and distinguishes in three-dimensional world coordinate system Z=0 plane projection point set
For minimum point Pa(Xa,Ya, 0), highest point Pb(Xb,Yb,Zb);
S14 finds out the mass center P of all the pointsc(Xc,Yc,Zc), and system homogeneous equation matrix M is established, its transposition MT is sought, and carry out
Singular value decomposition:
Wherein, MTIt is the transposition of the homogeneous equation matrix M of the mass center of all the points in optical plane, UM, SM, VT MIndicate singular value point
The matrix of solution;
In conjunction with the normal vector [α, beta, gamma] for the optical plane that these three values are fitted, and find out the expression formula of optical plane: α xw+βyw+
λzw=0 (5);
Wherein, (XW,YW,ZW) indicate optical plane on each point three-dimensional coordinate.
6. tank body inside fillet laser visual vision positioning method as claimed in claim 5, which is characterized in that the weldering
The highest point of rifle theory offset is 5cm, and the minimum point of the welding gun theory offset is 0cm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811271177.XA CN109658456A (en) | 2018-10-29 | 2018-10-29 | Tank body inside fillet laser visual vision positioning method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811271177.XA CN109658456A (en) | 2018-10-29 | 2018-10-29 | Tank body inside fillet laser visual vision positioning method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109658456A true CN109658456A (en) | 2019-04-19 |
Family
ID=66110429
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811271177.XA Pending CN109658456A (en) | 2018-10-29 | 2018-10-29 | Tank body inside fillet laser visual vision positioning method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109658456A (en) |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110052755A (en) * | 2019-05-23 | 2019-07-26 | 哈尔滨焊接研究院有限公司 | The system and method for minister's right angle welding seam automatic capturing and positioning in workpiece space based on collision short-circuit sensor |
CN110208278A (en) * | 2019-07-09 | 2019-09-06 | 电子科技大学 | The apparent slight crack vision measurement system in road surface |
CN111055021A (en) * | 2019-12-28 | 2020-04-24 | 深圳市诺亚云谷科技有限公司 | Visual positioning method of laser marking software |
CN111105463A (en) * | 2019-12-31 | 2020-05-05 | 东北大学 | Label welding and positioning method for end faces of bundled rods |
CN111612848A (en) * | 2020-04-30 | 2020-09-01 | 重庆见芒信息技术咨询服务有限公司 | Automatic generation method and system for arc welding track of robot |
CN111982016A (en) * | 2020-08-27 | 2020-11-24 | 科锐特(辽宁)智能装备有限责任公司 | Two-dimensional and three-dimensional integrated visual detection sensor and image processing algorithm thereof |
CN112276300A (en) * | 2020-09-30 | 2021-01-29 | 上海新时达机器人有限公司 | Welding seam scanning method for corrugated plate |
CN112508932A (en) * | 2020-12-21 | 2021-03-16 | 华南理工大学 | Welding seam positioning method and system based on simulation template matching and storage medium |
CN112529858A (en) * | 2020-12-02 | 2021-03-19 | 南京理工大学北方研究院 | Welding seam image processing method based on machine vision |
CN112749732A (en) * | 2020-12-15 | 2021-05-04 | 华南理工大学 | Multi-template included angle resolution calculation method for structured light welding seam positioning |
CN112861228A (en) * | 2021-02-18 | 2021-05-28 | 深圳辰视智能科技有限公司 | Large-scale steel structure positioning method and system |
CN112935562A (en) * | 2021-01-19 | 2021-06-11 | 中国科学院云南天文台 | Laser precision machining method based on paraxial offline measurement |
CN113192029A (en) * | 2021-04-29 | 2021-07-30 | 南通大学 | Welding seam identification method based on ToF |
CN113319411A (en) * | 2021-03-04 | 2021-08-31 | 湖南大学 | Visual positioning method and system and computing equipment |
CN114043045A (en) * | 2021-11-29 | 2022-02-15 | 苏州全视智能光电有限公司 | Round hole automatic plug welding method and device based on laser vision |
CN116579955A (en) * | 2023-07-13 | 2023-08-11 | 厦门微图软件科技有限公司 | New energy battery cell weld reflection point denoising and point cloud complement method and system |
CN117146727A (en) * | 2023-10-30 | 2023-12-01 | 北京通泰恒盛科技有限责任公司 | Tower tube welding seam monitoring method and system based on machine vision |
CN117773334A (en) * | 2024-02-28 | 2024-03-29 | 无锡丰毅汽车配件有限公司 | Automatic welding device for automobile skylight frame |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105844622A (en) * | 2016-03-16 | 2016-08-10 | 南京工业大学 | V-shaped groove welding seam detection method based on laser visual sense |
CN108709499A (en) * | 2018-04-28 | 2018-10-26 | 天津大学 | A kind of structured light vision sensor and its quick calibrating method |
-
2018
- 2018-10-29 CN CN201811271177.XA patent/CN109658456A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105844622A (en) * | 2016-03-16 | 2016-08-10 | 南京工业大学 | V-shaped groove welding seam detection method based on laser visual sense |
CN108709499A (en) * | 2018-04-28 | 2018-10-26 | 天津大学 | A kind of structured light vision sensor and its quick calibrating method |
Non-Patent Citations (5)
Title |
---|
岳建锋等: "基于激光单目视觉系统的焊缝提取", 《天津工业大学学报》 * |
张洁等: "基于各向异性扩散方程的Canny边缘检测算法", 《计算机应用》 * |
张鹏贤等: "坡口及焊缝表面三维轮廓的激光视觉测量", 《焊接学报》 * |
甘文龙等: "管道焊接激光视觉跟踪的定位方法研究", 《激光与红外》 * |
郭伟等: "基于多尺度自适应扩散方程的边缘检测方法", 《计算机工程与科学》 * |
Cited By (28)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110052755B (en) * | 2019-05-23 | 2020-09-22 | 哈尔滨焊接研究院有限公司 | System and method for automatically aligning and positioning long right-angle welding seam in workpiece space based on collision short-circuit sensor |
CN110052755A (en) * | 2019-05-23 | 2019-07-26 | 哈尔滨焊接研究院有限公司 | The system and method for minister's right angle welding seam automatic capturing and positioning in workpiece space based on collision short-circuit sensor |
CN110208278A (en) * | 2019-07-09 | 2019-09-06 | 电子科技大学 | The apparent slight crack vision measurement system in road surface |
CN111055021A (en) * | 2019-12-28 | 2020-04-24 | 深圳市诺亚云谷科技有限公司 | Visual positioning method of laser marking software |
CN111105463A (en) * | 2019-12-31 | 2020-05-05 | 东北大学 | Label welding and positioning method for end faces of bundled rods |
CN111105463B (en) * | 2019-12-31 | 2024-04-16 | 东北大学 | Label welding positioning method for end faces of bundled bars |
CN111612848B (en) * | 2020-04-30 | 2023-10-10 | 湖北煌朝智能自动化装备有限公司 | Automatic generation method and system for arc welding track of robot |
CN111612848A (en) * | 2020-04-30 | 2020-09-01 | 重庆见芒信息技术咨询服务有限公司 | Automatic generation method and system for arc welding track of robot |
CN111982016A (en) * | 2020-08-27 | 2020-11-24 | 科锐特(辽宁)智能装备有限责任公司 | Two-dimensional and three-dimensional integrated visual detection sensor and image processing algorithm thereof |
CN111982016B (en) * | 2020-08-27 | 2022-04-01 | 科锐特(辽宁)智能装备有限责任公司 | Two-dimensional and three-dimensional integrated visual detection sensor and image processing algorithm thereof |
CN112276300A (en) * | 2020-09-30 | 2021-01-29 | 上海新时达机器人有限公司 | Welding seam scanning method for corrugated plate |
CN112529858A (en) * | 2020-12-02 | 2021-03-19 | 南京理工大学北方研究院 | Welding seam image processing method based on machine vision |
CN112749732A (en) * | 2020-12-15 | 2021-05-04 | 华南理工大学 | Multi-template included angle resolution calculation method for structured light welding seam positioning |
CN112749732B (en) * | 2020-12-15 | 2023-08-22 | 华南理工大学 | Multi-template pinch angle resolution calculation method for structured light weld positioning |
CN112508932A (en) * | 2020-12-21 | 2021-03-16 | 华南理工大学 | Welding seam positioning method and system based on simulation template matching and storage medium |
CN112508932B (en) * | 2020-12-21 | 2023-08-22 | 华南理工大学 | Weld joint positioning method, system and storage medium based on simulation template matching |
CN112935562A (en) * | 2021-01-19 | 2021-06-11 | 中国科学院云南天文台 | Laser precision machining method based on paraxial offline measurement |
CN112861228B (en) * | 2021-02-18 | 2024-02-13 | 深圳辰视智能科技有限公司 | Positioning method and system for large-scale steel structure |
CN112861228A (en) * | 2021-02-18 | 2021-05-28 | 深圳辰视智能科技有限公司 | Large-scale steel structure positioning method and system |
CN113319411A (en) * | 2021-03-04 | 2021-08-31 | 湖南大学 | Visual positioning method and system and computing equipment |
CN113192029A (en) * | 2021-04-29 | 2021-07-30 | 南通大学 | Welding seam identification method based on ToF |
CN114043045A (en) * | 2021-11-29 | 2022-02-15 | 苏州全视智能光电有限公司 | Round hole automatic plug welding method and device based on laser vision |
CN116579955A (en) * | 2023-07-13 | 2023-08-11 | 厦门微图软件科技有限公司 | New energy battery cell weld reflection point denoising and point cloud complement method and system |
CN116579955B (en) * | 2023-07-13 | 2023-10-20 | 厦门微图软件科技有限公司 | New energy battery cell weld reflection point denoising and point cloud complement method and system |
CN117146727A (en) * | 2023-10-30 | 2023-12-01 | 北京通泰恒盛科技有限责任公司 | Tower tube welding seam monitoring method and system based on machine vision |
CN117146727B (en) * | 2023-10-30 | 2024-01-30 | 北京通泰恒盛科技有限责任公司 | Tower tube welding seam monitoring method and system based on machine vision |
CN117773334A (en) * | 2024-02-28 | 2024-03-29 | 无锡丰毅汽车配件有限公司 | Automatic welding device for automobile skylight frame |
CN117773334B (en) * | 2024-02-28 | 2024-05-14 | 无锡丰毅汽车配件有限公司 | Automatic welding device for automobile skylight frame |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109658456A (en) | Tank body inside fillet laser visual vision positioning method | |
CN210046133U (en) | Welding seam visual tracking system based on laser structured light | |
CN105894499B (en) | A kind of space object three-dimensional information rapid detection method based on binocular vision | |
CN107824940A (en) | Welding seam traking system and method based on laser structure light | |
CN105783726B (en) | A kind of curved welding seam three-dimensional rebuilding method based on line-structured light vision-based detection | |
Shah et al. | Butt welding joints recognition and location identification by using local thresholding | |
CN104331896B (en) | A kind of system calibrating method based on depth information | |
CN106181162B (en) | A kind of real-time weld joint tracking detection method based on machine vision | |
He et al. | Autonomous detection of weld seam profiles via a model of saliency-based visual attention for robotic arc welding | |
CN107907048A (en) | A kind of binocular stereo vision method for three-dimensional measurement based on line-structured light scanning | |
CN106392304B (en) | A kind of laser assisted weld seam Intelligent tracing system and method | |
CN106182004A (en) | The method of the industrial robot automatic pin hole assembling that view-based access control model guides | |
Nguyen et al. | 3D scanning system for automatic high-resolution plant phenotyping | |
CN109822216B (en) | Welding bead track and posture real-time tracking detection method, electronic equipment and medium | |
CN113894481B (en) | Welding pose adjusting method and device for complex space curve welding seam | |
CN110425998B (en) | Three-dimensional measurement method for component with height of coupling characteristic points of gray level image | |
CN108032011B (en) | Initial point guiding device and method are stitched based on laser structure flush weld | |
CN111982921A (en) | Hole defect detection method and device, conveying platform and storage medium | |
CN110930368B (en) | Method for extracting characteristics of real-time welding image of sheet lap weld | |
CN110097516B (en) | Method, system and medium for correcting distortion of image on inner hole wall surface | |
Zou et al. | Conditional generative adversarial network-based training image inpainting for laser vision seam tracking | |
CN106735749A (en) | A kind of laser assisted weld seam Intelligent tracing system | |
Zou et al. | Research on a real-time pose estimation method for a seam tracking system | |
CN116228652A (en) | Pipeline weld joint surface defect detection method based on three-dimensional point cloud curvature | |
CN102519387A (en) | Visual inspection method of electron beam welding pool shape parameter |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190419 |
|
RJ01 | Rejection of invention patent application after publication |