CN103170778B - Welded joint automatic tracking system - Google Patents
Welded joint automatic tracking system Download PDFInfo
- Publication number
- CN103170778B CN103170778B CN201310111811.4A CN201310111811A CN103170778B CN 103170778 B CN103170778 B CN 103170778B CN 201310111811 A CN201310111811 A CN 201310111811A CN 103170778 B CN103170778 B CN 103170778B
- Authority
- CN
- China
- Prior art keywords
- image
- weld
- frame
- seam
- sequence
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a welded joint automatic tracking system which is characterized in that a camera unit (1) collects a welded joint image, an image identification device (3) obtains the welded image through an image collection card (2), a welded joint central current position is searched and matched in the welded joint image through a self-template method, a control signal is generated through the matched welded joint central current position and output to an automatic control device (4), and the automatic control device (4) regulates the position of an arm of a welding gun according to the given control signal so that the welding gun is enabled to move to the welded joint central current position. Due to adoption of the self-template method, the welded joint automatic tracking system improves interference resistant capacity of the identification process, and therefore an identification success rate is improved.
Description
Technical field
The invention belongs to soldering joint automatic tracking technical field in Automation of Welding, particularly a kind of seam tracking system utilizing image-recognizing method.
Background technology
In the production of large-scale steel structure body (as: petroleum gas steel pipe, hull, pressure vessel etc.), extensive use automatic welding technique.In order to realize automatic welding, need to make the relative position of welding gun and weld seam stably keep accurate by automatic seam tracking method.
Existing automatic seam tracking method comprises: Master Plate Tracker method, eddy current tracking, electromagnetic tracking method, photoelectric tracking method, image tracking method etc.Image tracking method wherein utilizes video camera Real-time Obtaining weld image, the Theories and methods of Land use models identification finds weld seam in present image, weld seam current location is transferred to automatic control link, then send adjustment position order, realize position while welding from motion tracking.
Image tracking method comprises again two large classes: a class is structure based light method.Its principle is, send a branch of fillet light moving towards into vertical direction with weld seam, keep the sight line of video camera in one plane, this plane is vertical with the section comprising axis of a weld, sight line also will be moved towards at an angle with weld seam, thus obtaining the image of reflection seam cross-section shape, this shape generally has V-shaped and U-shaped two kinds.So V-shaped or U-shaped are just as the feature identified.Because feature is simply obvious, be convenient to identify, under the condition that process conditions ensure, reliable tracking can be ensured.But this method is easily interfered.When weld seam is irregular, or when larger extraneous light disturbs, when cannot present obvious V-shaped or U-shaped, recognition failures will be caused.Another kind of is based on non-structural light method.Its principle directly obtains weld image, extracts the various features of weld seam, constitutive characteristic vector, then build complicated recognition methods, realizes the searching of weld seam.The difficult point of this method is the extraction of characteristic vector, the simplest feature extraction mode is, gets a window comprising weld seam as template, in present image, carry out video in window coupling in the weld image of a width standard, when matching error is minimum, thinks and find weld seam.Due to face of weld and out-of-flatness and clean, add the interference of extraneous light, be in fact difficult to obtain suitable " a standard weld image ", get template image and present image usually has larger difference, thus make recognition failures.
Summary of the invention
For above-mentioned existing two class image tracking method Problems existing, the present invention proposes a kind of seam tracking system, this system, equally based on image tracking method, utilizes self form and images themselves to do the current location of mating and finding weld seam, to improve the antijamming capability of tracing process.
The technological means that the present invention adopts is: a kind of seam tracking system, is characterized in that comprising: be placed in the shooting unit on welding gun arm, image pick-up card, image display and recognition device, automaton;
Shooting unit gathers weld image, image display and recognition device obtain this weld image by image pick-up card, self form method is utilized to search in this weld image and mate Weld pipe mill current location, generate control signal according to the Weld pipe mill current location matched and export to automaton, automaton adjusts the position of welding gun arm, so that welding gun is moved to Weld pipe mill position according to given control signal.
System of the present invention utilizes self form method to realize the centralized positioning of weld seam; Specifically, in the image sequence that shooting unit obtains, ensure that adjacent two two field pictures in weld seam direction of advance have only about half of region overlapping, that is, move to the first half at current frame image in the latter half of branch of previous frame image; So, a video in window comprising certain length weld seam is got and as template previous frame image latter half of, mate at the first half when previous frame image, just have found the image of template itself when preferably mating, then the lateral shift value of weld seam is just determined; Because the image in template acquired in front piece image is bound to repeat in the first half image when former frame, consider that the time interval of two continuous frames is very little, external environment does not have too large change, therefore done is self and self images match, improve the antijamming capability of identifying, thus improve recognition success rate.
Accompanying drawing explanation
Below in conjunction with drawings and Examples, the present invention is further elaborated:
Fig. 1 is seam tracking system schematic diagram of the present invention;
Fig. 2 extracts the schematic diagram that self form carries out searching for coupling;
Fig. 3 is data transmission scheme;
Fig. 4 is the schematic diagram of embodiment.
Detailed description of the invention
As shown in Figure 1, seam tracking system of the present invention comprises: be placed in the shooting unit 1 on welding gun arm, image pick-up card 2, pattern recognition device 3, automaton 4.Shooting unit 1 gathers weld image, pattern recognition device 3 obtains this weld image by image pick-up card 2, self form method is utilized to search in this weld image and mate Weld pipe mill current location, generate control signal according to the Weld pipe mill current location matched and servomechanism control principle and export to automaton 4, automaton 4 adjusts the position of welding gun arm, so that welding gun is moved to Weld pipe mill position according to given control signal.
Unit of wherein making a video recording can comprise one or two, if a frame weld image visual field is of a size of F
v=a × b, a is the length perpendicular to weld seam direction of advance, and its coordinate direction is decided to be x, and b is the length along weld seam direction of advance, and its coordinate direction is decided to be y, and the principle of selection one or two video camera is: establish weld movement speed S
w, a frame weld image duration T
v, work as S
w, b and T
vbetween meet relation
time, in the duration T of a frame weld image
vin, weld seam only in visual field displacement be less than b/2, can be only therefore that self form method provides enough image laps with a video camera.Work as S
w, b and T
vbetween meet relation
time, need employing two video cameras to gather the adjacent weld image of two frames respectively, and in order to apply self form method, need two adjacent frame weld images to have the overlap of half at least, for this reason, two camera field of view centers are L in the weld seam distance walked upwards
vF, then have: L
vF≤ T
v× S
w.
T wherein
vthere are two kinds of standards: under PAL-system, T
v=40ms; Under NTS system, T
v=30ms.The occurrence of a, b, x and y is determined by resolution ratio and television system; Resolution definition is that in weld image, pixel represents the size on actual weldment, and its occurrence is by the requirement of production technology to control accuracy; If px is the resolution ratio of pixel in x direction, the maximum departure of permission is ± Er
c, then px≤Er
cthe span of/2, x is x=1 ..., N; If py is the resolution ratio of pixel in y direction, the value of y is y=1 ..., M.Conventional image pick-up card is N=768 pixel, M=572 pixel; So a=Npx, b=Mpy; The aspect ratio of the picture of conventional video camera is 4.8/3.6,4.8/3.6 ≈ 768/572 ≈ 1.3, so py=px/1.3; Way when determining physical location is at the scene flexibly, the method that the present invention recommends is: the actual value calculating a, b according to mentioned above principle, the length ruler of a standard is placed on below video camera, the position of adjustment video camera and amasthenic lens, observe the image in display screen, make a be full of the width of whole video display as far as possible.
The process utilizing self form method to search in weld image to pattern recognition device 3 below and to mate Weld pipe mill current location is described in detail as follows:
Step one: three image sequence variablees are set, are designated as Q
i, C
iwith
i=1,2 ..., U, U=L/b, L be in continuous welding process weld seam through the total length of visual field.Wherein, when unit 1 of making a video recording is a video camera, Q
i(i=1 ..., U) sequentially arrange the image sequence obtained successively for the successive frame weld image of this shooting unit collection; When unit 1 of making a video recording is two video cameras, the odd-numbered frame in this image sequence is the weld image of a camera acquisition, and the even frame of this image sequence is the weld image of another camera acquisition, uses Q
i(x, y) (i=1 ..., U) and represent Q
iin be arranged in pixel grey scale value sequence in weld image (x, y) coordinate points; C
i(i=1 ..., U) and be cursor glyph sequence, C
i(x, y) (i=1 ..., U) and represent the cursor gray value sequence of (x, y) coordinate points in weld image, the shape of cursor is determined by designer, conventional cross;
represent the image sequence adding cursor,
represent
in be positioned at pixel grey scale value sequence in weld image (x, y) coordinate points.For all x and y in a two field picture, perform
computing can by cursor its superimposition in weld image frame, and ⊕ represents XOR.Below be referred to as β
iand β
i+1be respectively present frame and next frame, β difference expression place can represent respectively Q,
and C; It should be noted that, in the angle of algorithm, the number of plies of above-mentioned three sequences can be arranged according to the work period of welding system by designer, and also namely the maximum occurrences U of i can have a greater change.In practical implementations, if realized with computer, video camera is connected with image card, and computer fetches data from the frame memory read of image card by arranging multithreading, three above-mentioned sequence composition queue modes; If make the processor be made up of DSP and FPGA by oneself, then in order to save storage space, above-mentioned image sequence can only establish two layers, i.e. U=2, and adopt the frame memory access of traditional binary chi structure to realize the compression of storage space, and when a video camera, the structure of intersecting with the binary of two memories, when two video cameras, form the structure (having introduction in the document of multiple memorizers structure) of two pairs of binarys intersections with 4 frame access reservoirs;
Step 2: undertaken initializing location by operator's butt welded seam center: operator moves cursor to Weld pipe mill position by the manual input unit of such as keyboard, mouse and so on a video display, pattern recognition device 3 reads the position of cursor instruction, the initialization of this position as Weld pipe mill is located.
Operator does initialization location can two kinds of rules: the first, is located, be divided into again: 1. the operator with long period of operation experience by an operator, and by observing the image on video display, mobile cursor is to seen Weld pipe mill; 2. while mobile cursor, the movable signal of cursor is sent to automaton 4 by pattern recognition device 3, controls welding gun aim at Weld pipe mill by automaton 4; Or be put near welding gun by the video display of pattern recognition device 3, adjust welding torch position by operator's one side, make it aim at Weld pipe mill, one side is observed and mobile cursor position on a video display, completes the work of original cursor location; Second, located by the cooperation of two operators: first operator directly measures the relative deviation of welding gun and Weld pipe mill near welding gun, it is close to Weld pipe mill that second operator simultaneously before Command Display Unit moves cursor, when first operator determines that Weld pipe mill aimed at by welding gun, command second operator to stop mobile cursor simultaneously, then complete the initial alignment of cursor, and current cursor value is stored.Certainly, namely automatic tracking system is started after completing original cursor location.Initial alignment frame is as the weld image Q of the present frame of first in subsequent algorithm
j.
Step 3: at the weld image Q of present frame
jlatter half get a region as self form P
j.P
jcenter fix on Weld pipe mill, this centre coordinate is recorded in intermediate variable C
oin; P
jthe determination principle of size be: establish d
pxfor walking template length upwards perpendicular to weld seam, meet: d
px>=d
w+ 2 (d'
w-d
w), d
wfor the mean breadth of weld seam, d'
wfor the Breadth Maximum of weld seam; d
pyfor walking template length upwards along weld seam, the speed stability of its value weld seam, the present invention recommends
Step 4: use P
jat next frame weld image Q
j+1in first half, by use search and matching algorithm, complete Q
j+1the location of the current location of Weld pipe mill in frame weld image.Search wherein refers to mobile self form P
j, coupling refers to uses self form P
jwith Q
j+1subgraph on the position searched in frame weld image is made comparisons and is recorded comparative result.Self form P
jmovement to be undertaken by certain path order, different designers can adopt different path order; Most employing is first mobile in the x-direction moves a line more in y-direction, so repeats, forms "the" shape path order; Also the order that x and y exchanges can be adopted; Or design special path order by designer, the present invention is for "the" shape path order; The concrete steps that step 4 comprises again:
Step 1: at next frame weld image Q
j+1first half zone line, determine a search Matching band, if search Matching band starting point coordinate is (x
s0, y
s0), terminal point coordinate (x
s0+ L
sx, y
s0+ L
sy).The size of this search Matching band is determined by following principle: search Matching band is defined as rectangle, its size should ensure to be search in search matching process, said search excessively refers to, also to continue to do several times search coupling after obtaining optimum Match, then by backtracking, correct matched position is really found; If the length in x direction is L
sx, in order to ensure to be search, then require L
sx>=d'
w+ k
x1Δ d
wm+ k
x2, Δ d
wm>0 is the maximum weld seam deviation allowed in whole tracing process, if k
x1and k
x2arrange in man-machine interface on a video display as adjustable parameter or adjustment key is set especially, adjusting k by operator
x1and k
x2value; k
x1arrange for bucking-out system job stability, under the situation that system works is stable, k
x1can be little, under the situation of system works less stable, k
x1get large value; k
x2for ensure realized search for and arrange, as long as make k in theory
x2=py, but consider noise jamming, require k
x2>py, by trying out k on real system
x2adjust to rational value; The length in y direction is L
sy, L
syvalue is relevant with the stability of weld movement speed, suggestion L
sy>=d
py+ 2 Δ S
wt
v+ 3py, Δ S
w>=0 is the maximum deviation of weld movement speed, and this value is given in robot welding system design and running.
Step 2: aray variable M in the middle of arranging two
1(z) and M
2(z), wherein, z=1 ..., Z, record matching computing sum, and Q respectively
j+1the current position coordinates matched on frame weld image.
Step 3: by self form P
jthe upper left corner be placed on (x
s0, y
s0) on, from z=1, first x direction again y direction with a pixel for step pitch moves self form P
j, often moving moves a step is z=z+1 until Z only goes cyclically to do following matching operation:
M
1(z)=P
j(x,y)-Q
j+1(x,y)
This computing produces a data sequence M
1(z), z=1 ..., the value of Z, Z will ensure self form P
jin moving process, cover whole search Matching band, namely will ensure self form P
jthe lower right corner can finally arrive coordinate (x
s0+ L
sx, y
s0+ L
sy), L
sxand L
sybe respectively the length in x and y direction.
Wherein, self form P
jthe step pitch of movement has two kinds: one is minimum step, and namely a pixel is as a step pitch unit; Another kind is large step pitch, be several pixels is a step pitch unit.The general minimum step when operation time can meet the demands, wants when operation time is long large step pitch and minimum step to combine, is called combination algorithm.Combination algorithm first does coarse positioning by large step pitch to do accurate location with minimum step again.The length of operation time is determined by the ability of production technology and image processing apparatus; When the computing of image processing apparatus can complete once search matching operation with minimum step in the interval of twice control command required by manufacturing technique requirent, just with minimum step, otherwise use combination algorithm.
Step 4: try to achieve sequence M
1(z), z=1 ..., the minimum of a value of Z, this minimum of a value corresponds to accurate match, simultaneously the coupling coordinate points of record this minimum of a value corresponding, i.e. M
2z ()=(x, y) provides weld image Q
j+1the position of upper accurate match, then self form P on this position
jcenter be exactly the center of weld seam.
Step 5: be weld seam deviation computing Δ C=C
o-M
2z (), changes C according to the coordinate value that Δ C records
j+1in (x, y), the position of cursor sends the control command of mobile welding gun simultaneously to control device, then does computing
Afterwards will
Display is sent to show.
Step 5: checked whether and ceased and desisted order, if any, then stop above-mentioned computing and operation, otherwise get back to step 3, delivery plate of reforming, search mate, rewrite cursor position, send
show and send the operations such as control command.Send two kinds of situations of having ceased and desisted order: (1) system sends fault alarm, and (2) people is for quitting work.
Extract as Fig. 2 gives the schematic diagram that self form carries out searching for coupling, weld seam advances along Y direction, there will be deviation in X direction.Depict three coordinate planes in figure, they are designated as x (t respectively
j-1), y (t
j-1), x (t
j), y (t
j), x (t
j+1), y (t
j+1), spatially, in fact in one plane, just at time shaft, (be designated as t) and be expressed as three planes, its time coordinate note is respectively (t to these three planes
j-1), (t
j), (t
j+1).Three coordinate planes depict three frame weld image Q respectively
j-1, Q
j, Q
j+1; In figure, weld edge is divided into multistage in y-direction, each section represents by different line styles, this is the corresponding relation (being pointed out further by solid line double-head arrow) in order to show change in location in weld movement process, as can be seen from this corresponding relation, between three frame weld images, adjacent two frames half image of having an appointment is overlapping; P
jand P
j-1represent Q respectively
jand Q
j-1template in frame,
character number represents coupling center, wherein P
j-1at Q
j-1latter half ofly to obtain, at Q
jfirst half do search coupling, P
jat Q
jlatter half ofly to obtain, at Q
j+1first half do search coupling, this search coupling variation relation pointed out by dashed double.In figure, x
j-1, x
j, x
j+1represent the coordinate of Weld pipe mill respectively.
Pattern recognition device 3 of the present invention can be ordered or customize, and generally has the structure of two kinds of forms at present, and a kind of is universal architecture based on industrial computer, and another kind is the application specific architecture of the composition based on devices such as DSP, FPGA, single-chip microcomputers.The concrete constituted mode of application specific architecture is varied, but the acquisition of image and MPTS and aforesaid universal architecture basically identical.Fig. 1 illustrates the schematic diagram of universal architecture, in structural representation, only depict the parts relevant with the present invention.
In FIG, the image pick-up card with twin camera interface ordered or customize is inserted in the corresponding slot of industrial computer.Pattern recognition device 3 comprises again: internal memory, video display and processor; Internal memory comprises again core buffer 1, core buffer 2 and core buffer 3.Core buffer 1 stores the weld image frame Q of a camera acquisition
2n+1, core buffer 2 stores the weld image frame Q of another camera acquisition
2n, core buffer 3 stores cursor glyph, and the renewal of figure is determined by algorithm; Processor utilizes above-mentioned self form method search in this weld image and mate Weld pipe mill current location, generates control signal, and export to automaton 4 according to the Weld pipe mill current location matched.As pointed in describing above, the number of plies of core buffer 1,2,3 to reduce to two layers.Above-mentioned relevant parameter can be placed on remaining memory space.
Fig. 3 shows the transmission path of the control signal of view data and recognition result generation.
Fig. 4 is the demonstration graph of an embodiment.Wherein left figure is former frame weld image, and right figure is an adjacent rear two field picture.In right figure, the Lower Half of visible previous frame image has been moved on to the first half by the weld image of template institute frame in an adjacent rear two field picture, and in order to indicate this point, we stick a white triangles on weld seam side artificially.As can be seen from the figure, Weld pipe mill offset by approximately left 4 millimeters (note: this is demonstration graph, and do not add control feedback to adjust deviation.Adjacent two frame weld excursions can not be allowed in reality so many.）
Claims (4)
1. a seam tracking system, is characterized in that comprising: be placed in the shooting unit (1) on welding gun arm, image pick-up card (2), image display and recognition device (3), automaton (4);
Shooting unit (1) gathers weld image, image display and recognition device (3) obtain this weld image by image pick-up card (2) and are stored in image display and recognition device (3), in image display and recognition device (3), utilize self form method search in this weld image and mate Weld pipe mill current location, generate control signal according to the Weld pipe mill current location matched and export to automaton (4), automaton (4) is according to the position adjusting welding gun arm to control signal, so that welding gun is moved to Weld pipe mill position, image display and recognition device (3) utilize self form method to search in weld image and the step of mating Weld pipe mill current location comprises:
Step one: three image sequence variablees are set, are designated as Q
i, C
iwith
wherein i=1,2 ..., U, U=L/b, L be in continuous welding process weld seam through the total length of visual field; Wherein, when unit (1) of making a video recording is a video camera, Q
i(i=1,2 ..., U) sequentially arrange the image sequence obtained successively for the successive frame weld image of this shooting unit collection; When unit (1) of making a video recording is two video cameras, the odd-numbered frame in this image sequence is the weld image of a camera acquisition, and the even frame of this image sequence is the weld image of another camera acquisition, uses Q
i(x, y) (i=1,2 ..., U) represent Q
iin be arranged in pixel grey scale value sequence in weld image coordinate points, (x, y), x=1,2 ..., N, y=1,2 ..., M, N=a/px, M=b/py, px and py are the image resolution ratios in x and y direction, also the size of i.e. pixel; C
ifor cursor glyph sequence, C
i(x, y) (i=1,2 ..., U) represent C
ithe cursor gray value sequence of (x, y) coordinate points in middle weld image;
(i=1,2 ..., U) represent the image sequence adding cursor,
represent
in be positioned at pixel grey scale value sequence in weld image (x, y) coordinate points; For x and y all in a two field picture, perform
computing is by cursor its superimposition in weld image frame, and ⊕ represents XOR;
Step 2: carry out initialization by operator's butt welded seam center and locate;
Step 3: at the weld image Q of present frame
jlatter half get a rectangular area as self form P
j, wherein Q
jbelong to image sequence Q
iin piece image, formula is expressed as Q
j∈ Q
i, j ∈ i, P
j(x, y) x, y represents template P
jin coordinate value, P
jcenter fix on Weld pipe mill, this centre coordinate is recorded in intermediate variable C
oin; P
jthe determination principle of size be: establish d
pxfor walking template length upwards perpendicular to weld seam, meet: d
px>=d
w+ 2 (d'
w-d
w), d
wfor the mean breadth of weld seam, d'
wfor the Breadth Maximum of weld seam, d
pyfor walking template length upwards along weld seam,
Step 4: use P
jat next frame weld image frame Q
j+1in first half, by use search and matching algorithm, complete butt welded seam picture frame Q
j+1the location of middle Weld pipe mill current location;
Step 5: checked whether and ceased and desisted order; If any, then stop above-mentioned computing and operation, otherwise get back to step 3, delivery plate of reforming, search mate, rewrite cursor position, send
showing and send control command operation, sending two kinds of situations of having ceased and desisted order: (1) system sends fault alarm, and (2) people is for quitting work.
2. system according to claim 1, is characterized in that a frame weld image visual field is of a size of F
v=a × b, a is the length perpendicular to weld seam direction of advance, and its coordinate direction is decided to be x, and b is the length along weld seam direction of advance, and its coordinate direction is decided to be y, a frame weld image duration T
v, weld seam is at y direction translational speed S
w, when meeting relation
time, shooting unit is a video camera, and now adjacent two two field pictures just have the overlap of more than 1/2; And work as
time, also will ensure that adjacent two two field pictures have the overlap of more than 1/2, then require that shooting unit is 2 video cameras, these 2 video cameras gather the weld image of two consecutive frames respectively; The distance along weld seam trend at these two camera field of view centers is L
vF, require L
vF≤ T
v× S
w; Wherein, T under PAL-system
v=40ms, under NTS system, T
v=30ms.
3. system according to claim 1, is characterized in that step 4 also comprises:
Step 1: at Q
j+1the zone line of frame weld image the first half determines a search Matching band, if search
Rope Matching band is rectangle, and the coordinate of the upper left corner in picture frame is (x
s0, y
s0), lower right corner coordinate
(x
s0+ L
sx, y
s0+ L
sy), L
sxand L
sybe respectively the length in x and y direction;
Step 2: aray variable M in the middle of arranging two
1(z) and M
2(z), wherein, z=1 ..., Z, record matching operation result, and Q respectively
j+1the current position coordinates matched on frame weld image;
Step 3: by self form P
jthe upper left corner be placed on Q
j+1(the x of frame weld image
s0, y
s0) on, from z=1, first x direction again y direction with a pixel for step pitch moves self form P
j, often moving moves a step is z=z+1 until z=Z only goes cyclically to do following matching operation:
M
1(z)＝P
j(x,y)-Q
j+1(x,y)
Operation result is recorded in M
1z, in (), the value of Z will ensure self form P
jthe lower right corner can finally arrive coordinate (x
s0+ L
sx, y
s0+ L
sy), this computing produces a data sequence M
1(z), z=1 ..., Z;
Step 4: try to achieve sequence M
1(z), z=1 ..., the minimum of a value of Z, simultaneously the coupling coordinate points of record this minimum of a value corresponding, and use M
2z () record, even M
2(z)=(x, y), thus provide Q
j+1the position of accurate match on frame weld image, then self form P on this position
jcenter be exactly the center of weld seam;
Step 5: be weld seam deviation computing Δ C=C
o-M
2z (), changes C according to the deviate that Δ C records
j+1in (x, y), the position of cursor sends the control command of mobile welding gun simultaneously to control device, then does computing
afterwards will
display is sent to show.
4. system according to claim 3, the size and the position that it is characterized in that searching in step 1 Matching band are determined by following principle: search Matching band meets L
sx>=d'+k
x1Δ d
wm+ k
x2and L
sy>=d
py+ 2 Δ S
wt
v+ 3py, Δ d
wm>0 is the maximum weld seam deviation allowed in whole tracing process, k
x1and k
x2for arrange adjustable parameter, its position in the central authorities of the first half of a two field picture, Δ S
w>=0 is the maximum deviation of weld movement speed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310111811.4A CN103170778B (en) | 2013-04-01 | 2013-04-01 | Welded joint automatic tracking system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310111811.4A CN103170778B (en) | 2013-04-01 | 2013-04-01 | Welded joint automatic tracking system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103170778A CN103170778A (en) | 2013-06-26 |
| CN103170778B true CN103170778B (en) | 2015-06-24 |
Family
ID=48631202
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310111811.4A Expired - Fee Related CN103170778B (en) | 2013-04-01 | 2013-04-01 | Welded joint automatic tracking system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103170778B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104416305B (en) * | 2013-08-30 | 2016-08-10 | 中国石油天然气集团公司 | Pipeline Multipoint synchronous automatic welding prescan formula seam laser tracking and device |
| CN104526213B (en) * | 2015-01-24 | 2016-01-06 | 南通润雅机电科技有限公司 | Follow the tracks of the welding torch displacement control system of position while welding |
| CN105479029B (en) * | 2015-12-30 | 2018-05-25 | 中北大学 | A kind of automatic Tracking monitoring system of straight welding machine |
| CN107248170B (en) * | 2017-04-24 | 2020-06-26 | 辽宁师范大学 | Weld joint tracking method based on image matching |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1511669A (en) * | 2002-12-27 | 2004-07-14 | 中国科学院自动化研究所 | Arc welding robot control platform with visual welding seam automatic tracing function |
| CN201264138Y (en) * | 2008-08-01 | 2009-07-01 | 蔡杨波 | Device for automatically recognizing and tracking weld joint by video camera imaging |
| CN201300288Y (en) * | 2008-10-31 | 2009-09-02 | 鞍山荣邦电力电子有限公司 | Automatic seam tracking system |
| CN101961819A (en) * | 2009-07-22 | 2011-02-02 | 中国科学院沈阳自动化研究所 | Device for realizing laser welding and seam tracking and control method thereof |
| CN202278307U (en) * | 2011-08-19 | 2012-06-20 | 广州有色金属研究院 | An adjustable double line structured light weld tracking visual sensing system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8365602B2 (en) * | 2009-10-09 | 2013-02-05 | Olympus Ndt, Inc. | Weld seam tracking system using phased array ultrasonic devices |
-
2013
- 2013-04-01 CN CN201310111811.4A patent/CN103170778B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1511669A (en) * | 2002-12-27 | 2004-07-14 | 中国科学院自动化研究所 | Arc welding robot control platform with visual welding seam automatic tracing function |
| CN201264138Y (en) * | 2008-08-01 | 2009-07-01 | 蔡杨波 | Device for automatically recognizing and tracking weld joint by video camera imaging |
| CN201300288Y (en) * | 2008-10-31 | 2009-09-02 | 鞍山荣邦电力电子有限公司 | Automatic seam tracking system |
| CN101961819A (en) * | 2009-07-22 | 2011-02-02 | 中国科学院沈阳自动化研究所 | Device for realizing laser welding and seam tracking and control method thereof |
| CN202278307U (en) * | 2011-08-19 | 2012-06-20 | 广州有色金属研究院 | An adjustable double line structured light weld tracking visual sensing system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103170778A (en) | 2013-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11245806B2 (en) | Method and apparatus for scanning and printing a 3D object | |
| US11443444B2 (en) | Interior photographic documentation of architectural and industrial environments using 360 panoramic videos | |
| US11043026B1 (en) | Systems and methods for processing 2D/3D data for structures of interest in a scene and wireframes generated therefrom | |
| CN112771539B (en) | Employing three-dimensional data predicted from two-dimensional images using neural networks for 3D modeling applications | |
| CN105096386B (en) | A wide range of complicated urban environment geometry map automatic generation method | |
| Zollmann et al. | Augmented reality for construction site monitoring and documentation | |
| CN111209915B (en) | Three-dimensional image synchronous recognition and segmentation method based on deep learning | |
| Kim et al. | Improvement of realism of 4D objects using augmented reality objects and actual images of a construction site | |
| CN107357427A (en) | A kind of gesture identification control method for virtual reality device | |
| Zollmann et al. | Interactive 4D overview and detail visualization in augmented reality | |
| CN103170778B (en) | Welded joint automatic tracking system | |
| CN105069843A (en) | Rapid extraction method for dense point cloud oriented toward city three-dimensional modeling | |
| CN103838437A (en) | Touch positioning control method based on projection image | |
| US20210304430A1 (en) | System and method of operation for remotely operated vehicles for simultaneous localization and mapping | |
| CN110130987A (en) | A kind of tunnel convergence deformation monitoring method based on image analysis | |
| CN107330980A (en) | A kind of virtual furnishings arrangement system based on no marks thing | |
| US11395102B2 (en) | Field cooperation system and management device | |
| Al-Sabbag et al. | Enabling human–machine collaboration in infrastructure inspections through mixed reality | |
| CN114663473A (en) | Personnel target positioning and tracking method and system based on multi-view information fusion | |
| Shah et al. | Condition assessment of ship structure using robot assisted 3D-reconstruction | |
| Vacchetti et al. | A stable real-time AR framework for training and planning in industrial environments | |
| Amer et al. | Decentralized visual 3D mapping of scattered work locations for high-frequency tracking of indoor construction activities | |
| Yang et al. | Cost-Efficient Image Semantic Segmentation for Indoor Scene Understanding Using Weakly Supervised Learning and BIM | |
| Ma et al. | WeldNet: A deep learning based method for weld seam type identification and initial point guidance | |
| Meng et al. | Distortion-aware room layout estimation from a single fisheye image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150624 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |