CN109685758A - Weld seam vision positioning method based on multi-template matching - Google Patents
Weld seam vision positioning method based on multi-template matching Download PDFInfo
- Publication number
- CN109685758A CN109685758A CN201811271176.5A CN201811271176A CN109685758A CN 109685758 A CN109685758 A CN 109685758A CN 201811271176 A CN201811271176 A CN 201811271176A CN 109685758 A CN109685758 A CN 109685758A
- Authority
- CN
- China
- Prior art keywords
- template
- image
- weld seam
- camera
- fixed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- 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
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30244—Camera pose
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses the weld seam vision positioning methods based on multi-template matching, comprising the following steps: S1. fixes welding gun and vision system, keeps the two relative position constant；S2. to weld seam centralized positioning: S21. chooses the clear and apparent weld image of target signature as reference picture；S22. two rectangular areas comprising weld seam turning, i.e. ROI region are created in a reference image, are passed through Threshold segmentation and are extracted ROI region laser stripe, generate template area image；S23. parameter selection is carried out in the image of template area successively create multi-template；S24. real-time image acquisition in the welding process, determines the position at two turnings of groove or so in image, then the position of Weld pipe mill is determined by two corner location centers.The present invention reduces the interference of means of chaotic signals between two inflection points, greatly improve matching accuracy rate.
Description
Technical field
The present invention relates to vision positioning fields.It is more particularly related to a kind of weld seam based on multi-template matching
Vision positioning method.
Background technique
Pipeline welding as the important productive closely related with manufacturing industry, with industrial modernization development by
Step deeply, faces unprecedented challenge in all various aspects, proposes new want to welding technique level and welding production mode
It asks, it would be highly desirable to improve the intelligence and automatization level in welding.And pipeline-weld automatic tracking system is automatic as welding is embodied
An importance for changing technical level, is just playing increasingly important role.The automatic welding machine people that the country uses at present
Most of to be designed using guide rail type welding, robot advances along guide rail, achieves certain effect.But it is in use
Accurately guide rail installation and debugging must be completed in advance, and the guide rail of certain special materials is expensive, this is also to a certain extent
Limit its popularization and application.
With in recent years, the artificial intelligence technologys such as machine vision, image procossing rapid development, based on machine vision
Weld and HAZ technology has obtained the extensive concern of more and more domestic and foreign scholars, realizes automatic welding so that guiding using visual sensing
It is connected into order to possible.The present invention is based on machine vision principles, design vision system, propose a kind of vision based on multi-template matching
Localization method is used for the localization process of Weld pipe mill and torch swinging center, and the deviation of the two position can be used as guidance welding gun
The important reference indicator of automatic welding.Template matching algorithm there has been related introduction and application in the prior art, but it is usually
It is to choose single template, when carrying out vision positioning in the case where pipeline is irregular or not perfectly aligned, usually will appear matching
Unsuccessful phenomenon.
Summary of the invention
The object of the present invention is to provide the weld seam vision positioning methods based on multi-template matching, reduce between two inflection points not
The interference of regular signal, on the other hand greatly improves matching accuracy rate, can guarantee that pipeline automatic welding welding system can be more stable
Operation.
In order to realize these purposes and other advantages according to the present invention, the weld seam vision based on multi-template matching is provided
Localization method, comprising the following steps:
S1. welding gun and vision system are fixed, keeps the two relative position constant；
S2. to weld seam centralized positioning:
S21. the clear and apparent weld image of target signature is chosen as reference picture；
S22. two rectangular areas comprising weld seam turning, i.e. ROI region are created in a reference image, pass through Threshold segmentation
ROI region laser stripe is extracted, template area image is generated；
S23. parameter selection is carried out in the image of template area successively create multi-template；
S24. real-time image acquisition in the welding process, determines the position at two turnings of groove or so in image, then by two
A corner location center determines the position of Weld pipe mill.
Preferably, the reference picture in the S21 is removed the pretreatment of arc light interference.
Preferably, the S23 carries out parameter selection in the image of template area and successively creates multi-template, specifically includes:
One first area template comprising turning left a little of creation；
One second area template comprising right knee region of creation；
Parameter setting is carried out to first area template and second area template, parameter setting passes through observation first area template
Or second area template the pyramid number of plies and include in first area template or second area template object-point minimum it is right
Influence than two parameters of angle value to created first area template and second area template determines parameter.
Preferably, the parameter include: first area template and the rotatable angular range of second area template and
Step-length in rotation angle range.
Preferably, in the S24 specifically:
Real-time image acquisition in the welding process defines the rectangle ROI region of a position comprising left and right inflection point first,
It determines the search range of template, the Template Information created is then called to find template position in the picture, determine image mesoslope
The position at mouth or so two turnings, so that it is determined that the position of Weld pipe mill.
Preferably, the camera in the vision system is fixed as an entirety by attachment device and reflecting mirror；
The attachment device includes:
First fixed plate fixes reflecting mirror, the top surface of first fixed plate is connected with disc-shaped for cylinder
Barn door；
Two root posts are respectively perpendicular the bottom surface for being set to the first fixed plate, and the line of two root posts passes through described first
The center of circle of fixed plate bottom surface；
Camera inner seat, camera is fixed, and the outside of the camera inner seat is cube structure, opposite first side
It is symmetrically arranged with horizontal shaft, the gear of same model is symmetrically connected on a pair of of shaft；
The outer seat of camera, the square cavity with top surface opening are vertically arranged with a pair of of rack gear in the cavity, described
Rack gear can be engaged with the gear, and the rack gear is along the vertical direction by extending to the middle part of cavity at the top of cavity；The camera
It is provided with socket outside outer seat, the socket is fixed with the column；
Four groups of gravity centre adjustment mechanisms are respectively arranged at cavity bottom, and four groups of gravity centre adjustment mechanisms are located in camera
On the middle line on seat four side of bottom surface, the gravity centre adjustment mechanism includes:
Movable block, top surface is horizontal and has card slot, and middle line of the camera inner seat bottom surface along four sides extends to the midline
Convex block is formed, the convex block is sticked in the card slot, and the bottom surface of the movable block is inclined-plane, and bottom surface has and inclined bottom surface
Spend identical skewback；
Fixed block is fixed in the outer seat of camera, and the top surface of the fixed block is inclined-plane, and top surface has to be cooperated with skewback
Skewed slot, the bottom surface of the movable block is bonded with the top surface of the fixed block；
Electric putter, the side that telescopic end is fixed on movable block are fixed, and drive movable block to slide on fixed block, the electricity
The pedestal of push rod is slideably positioned in block set, offers the first opening on the outer seat of the camera so that electric putter moves up and down；
Four groups of infrared distance measuring sensors are fixed on the outer seat of camera and respectively with respect to four angles of camera inner seat bottom surface；
Controller, the signal that receiving infrared-ray distance measuring sensor issues, and signal drives each electricity to push away based on the received
Bar stretches.
Preferably, the fixed block its be made of jacking block, middle piece and sole piece three parts, the jacking block top surface be inclined-plane,
Bottom surface is plane, and described middle piece and sole piece are rectangular parallelepiped structure；
Wherein, first component is provided on the sole piece, the first component lower part is located at the top surface of sole piece hereinafter, described
First component rotatably connects first body of rod, and the other end of first body of rod rotatably connects second component, and described
The top of two components is located at the bottom surface of jacking block or more, and the first slot to match with first component top is provided on the jacking block
Body is provided with the second groove body to match with second component lower part on the sole piece, and the top surface of the sole piece is provided with receiving
The third groove body of one body of rod；Middle piece of the side has the second opening, not interfered when moving by first body of rod.
Preferably, the column is made of transparent material.
The present invention is include at least the following beneficial effects: the method applicability of multi-template matching tack weld is stronger, robustness
It is higher.Matching accuracy rate can be stably held in 90% or more substantially, reduce the interference of means of chaotic signals between two inflection points, another
Aspect greatly improves matching accuracy rate, can guarantee the operation that pipeline automatic welding welding system can be more stable.Even if there is arc light interference,
In the case where not covering inflection point, template matching positioning can also play good function and effect.
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 flow diagram of the weld seam vision positioning method based on multi-template matching；
Fig. 2 is first embodiment weld groove schematic surface；
Fig. 3 left template area image；
Fig. 4 right mould plate area image；
Fig. 5 is Weld pipe mill positioning result schematic diagram；
Fig. 6 is automatic welding effect picture；
Fig. 7 is the structural schematic diagram of attachment device；
Fig. 8 is the structural schematic diagram of fixed block；
Fig. 9 is the connection schematic diagram of jacking block and sole piece in fixed block.
1 barn door, 2 first fixed plates, 3 reflecting mirrors, 4 columns, 5 cameras, 6 camera inner seats, 7 sockets, 8 rack gears, 9 shafts,
10 gears, 11 movable blocks, 12 electric putters, 13 block sets, 14 fixed blocks, 15 sliding rails, 16 convex blocks, the outer seat of 17 cameras, 141 jacking blocks, 142
Middle piece, 143 sole pieces, 144 first bodies of rod, 145 first groove bodies, 146 second components, 147 first components, 148 second groove bodies, 149
Baffle,
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.
In the description of the present invention, term " transverse direction ", " longitudinal direction ", "upper", "lower", "front", "rear", "left", "right", " perpendicular
Directly ", the orientation or positional relationship of the instructions such as "horizontal", "top", "bottom", "inner", "outside" is orientation based on the figure or position
Relationship is set, is merely for convenience of description of the present invention and simplification of the description, is not that device or the element of indication or suggestion meaning are necessary
It with specific orientation, is constructed and operated in a specific orientation, therefore is not considered as limiting the invention.
As shown in figs. 1 to 6, first embodiment provided by the invention.
A kind of weld seam vision positioning method based on multi-template matching, comprising the following steps:
S1. welding gun and vision system are fixed, keeps the two relative position constant；
S2. to weld seam centralized positioning:
S21. the clear and apparent weld image of target signature is chosen as reference picture；
S22. in initial time, two rectangular areas comprising weld seam turning, i.e. ROI region are created in a reference image,
ROI region laser stripe is extracted by Threshold segmentation, generates template area image, as shown in Figure 3 and Figure 4.
S23. parameter selection is carried out in the image of template area successively create multi-template；
S24. real-time image acquisition in the welding process, determines the position at two turnings of groove or so in image, then by two
A corner location center determines the position of Weld pipe mill.
In the present embodiment, creation is respectively provided with the template in left and right corner position region, although more compared to single mode plate
Setting is primary, but is that of avoiding the means of chaotic signals interference in groove inflection center region, reduces difficulty of matching, increases positioning
Accuracy.Selection multi-template can be reduced noise jamming, improve positioning accuracy.
Secondly, the annoyance level using multi-template matching method compared to other weld joint tracking image processing method arc lights drops
It is low, as shown in figure 5, it is obvious that in the welding process, even if there is arc light interference, template matching positioning can also be played very
Good function and effect.
Further, in order to avoid arc light is possible to the case where flooding inflection point, the reference picture in the S21 is removed
The pretreatment of arc light interference, it is ensured that the positioning of inflection point.
Further, the S23 carries out parameter selection in the image of template area and successively creates multi-template, specifically includes:
One first area template comprising turning left a little of creation.
Creation simultaneously one includes the second area template of right knee region, two template establishment method phases in the picture
Together, only parameter is different.
Parameter setting is carried out to first area template and second area template, in order to check whether template includes description target
The relevant information of object passes through the pyramid number of plies of observation first area template or second area template and included in first first
Two parameters of the minimum contrast value of object-point are to created first area template and second in region template or second area template
The influence of region template determines suitable parameter.
Further, algorithm defines first area template and the rotatable angular range of second area template and rotation angle
Step-length in degree range limits template (first area template and second area template) object.
The method that the creation of template optimized and template is arranged in algorithm.
Algorithm is provided with matching template condition in the picture.
Finally, a suitable template creates successfully, template handle is returned, and the template created is stored in specified
In file.
Further, in the S24 specifically:
The rectangle ROI region for defining a position comprising left and right inflection point first, determines the search range of template, then adjusts
Template position is found in the picture with the Template Information created, and algorithm design is as follows:
Firstly, the start angle and angular range of algorithm setting search, setting value have intersection when must be with drawing template establishment.
Further, smallest match value is arranged in algorithm, and it is that template is found in image which, which defines in template matching,
Mass fraction.
Further, pyramidal number of plies when algorithm setting search.
Further, algorithm setting search greediness degree, avoids different location in the picture from finding the identical match with target
Region.
Finally, the row and column coordinate of output matching position, angle, score.
Determine that the position at two turnings of groove or so in image, then the row coordinate by exporting are put down by the row, column coordinate exported
Mean value determines the position of Weld pipe mill, and obtained angle information may determine that trolley wanders off direction, and obtained matching score can
Matching degree height is judged, score height illustrates matching degree height, otherwise it fails to match.
In the first embodiment, as shown in figure 5, big rectangle frame is the ROI that initial time is chosen, it is determined that image template
Region of search can reduce match time, small rectangle frame indicates to pass through the Weld pipe mill position that target search determines in welding process
It sets, the vertical line on small frame indicates base position (before starting welding, Weld pipe mill position when welding gun is directed at Weld pipe mill).This two
The deviation (being negative it is assumed herein that vertical line is to the left, be positive on the contrary) of a center has reacted the deviation of trolley path of welding,
If deviation is 0, as normally, trolley is not rectified a deviation, and direction of travel is constant；If deviation is positive (or negative), i.e., in torch swinging
The heart is (or right) to the left compared with Weld pipe mill, then after control system receives deviation, (or left) is complete to the right automatically for driving welding tractor
At correction process.
Meanwhile under other all conditions unanimous circumstances, proposed multi-template matching method and list are used respectively
Template matching algorithm carries out vision positioning, completes the automatic welding of pipelines of a whole circle.In two kinds of algorithm calculation amounts quite (i.e. equal energy
Meet soldering joint automatic tracking requirement of real-time) in the case where, (single mode plate refers to that creation one includes simultaneously two to single template matching
The template of a knee region) matching accuracy rate be only 70% or so, the matching accuracy rate of multi-template matching algorithm substantially can be steady
Surely it is maintained at 90% or more, there is great promotion compared with the former, on the one hand reduces means of chaotic signals between two inflection points
Interference, on the other hand greatly improves matching accuracy rate, can guarantee the operation that pipeline automatic welding welding system can be more stable.
Shown in Fig. 6, the actual effect that automatic welding of pipelines is used for using the method for first embodiment is illustrated.It is seen that
Under vision system correction effect, though slightly wander off in welding tractor (due to pipeline is non-round or welding process in magnet pulley
On be adsorbed with welding slag, cause trolley to wander off) in the case where, have good effect, further demonstrate the validity of the method
And stability.
In another technical solution, the camera 5 in the vision system is fixed as one by attachment device and reflecting mirror 3
A entirety；
As shown in figs. 7-9, the attachment device includes:
First fixed plate 2, for cylinder, reflecting mirror 3 is fixed, and the top surface of first fixed plate 2 is connected with circle
The barn door 1 of sheet；
Two root posts 4, are respectively perpendicular the bottom surface for being set to the first fixed plate 2, and the line of two root posts 4 passes through described the
The center of circle of one fixed plate, 2 bottom surface；
Camera inner seat 6, camera 5 is fixed, and the outside of the camera inner seat 6 is cube structure, opposite first
Side is symmetrically arranged with horizontal shaft 9, and the gear 10 of same model is symmetrically connected on a pair of of shaft 9；
The outer seat 17 of camera, the square cavity with top surface opening are vertically arranged with a pair of of rack gear 8 in the cavity,
The rack gear 8 can be engaged with the gear 10, and the rack gear 8 is along the vertical direction by extending to the middle part of cavity at the top of cavity；
Socket 7 is provided with outside the outer seat 17 of the camera, the socket 7 and the column 4 are fixed；
Four groups of gravity centre adjustment mechanisms are respectively arranged at cavity bottom, and four groups of gravity centre adjustment mechanisms are located in camera
On the middle line on seat 6 bottom surface, four side, the gravity centre adjustment mechanism includes:
Movable block 11, top surface is horizontal and has card slot, and middle line of 6 bottom surface of camera inner seat along four sides prolongs to the midline
Stretch to form convex block 16, the convex block 16 is sticked in the card slot, and the bottom surface of the movable block 11 is inclined-plane, and bottom surface have with
The identical skewback of bottom surface gradient；
Fixed block 14 is fixed in the outer seat 17 of camera, and the top surface of the fixed block 14 is inclined-plane, top surface have with it is oblique
The skewed slot of block cooperation, the bottom surface of the movable block 11 is bonded with the top surface of the fixed block 14；
Electric putter 12, the side that telescopic end is fixed on movable block 11 are fixed, and drive movable block 11 sliding on fixed block 14
Dynamic, the pedestal of the electric putter 12 is slideably positioned in block set 13 that (block set 13 is enclosed set on outside camera outside seat 17, for fixed electricity
Push rod 12 is provided with four groups of sliding rails 15 in block set 13, and sliding rail 15 is slidably connected the pedestal of four electric putters 12 respectively), the phase
The first opening is offered on the outer seat 17 of machine so that electric putter 12 moves up and down；Four groups of infrared distance measuring sensors, are fixed on camera
Outer seat 17 and respectively with respect to four angles of 6 bottom surface of camera inner seat；
Controller, the signal that receiving infrared-ray distance measuring sensor issues, and signal drives each electricity to push away based on the received
Bar 12 stretches.
In the above-mentioned technical solutions, it needs to guarantee camera 5 and reflecting mirror 3 on the same axis, first has to guarantee stationary phase
Position is accurately, could to improve precision, while also needing to adjust the focal length of camera 5, therefore to adjust camera inner seat 6 when machine 5
The distance of the outer seat 17 of opposite camera, the setting of rack gear 8 and gear 10 guarantee one when outer with respect to the camera movement of seat 17 of camera inner seat 6
Cause property, when four of 6 bottom surface four corners of camera inner seat of infrared distance measuring sensor detection apart from it is different when, controller control pair
The distance of stretch out and draw back of each electric putter 12 is finely adjusted, until the distance of detection is identical, to guarantee the opposite phase of entire camera inner seat 6
The sag of the outer seat 17 of machine.
In another technical solution, the fixed block 14 its by jacking block 141, middle piece 142 and 143 three parts structure of sole piece
At 141 top surface of jacking block is inclined-plane, bottom surface is plane, and described middle piece 142 and sole piece 143 are rectangular parallelepiped structure；
Wherein, first component 147 is provided on the sole piece 143,147 lower part of first component is located at sole piece 143
Hereinafter, the first component 147 rotatably connects first body of rod 144, the other end of first body of rod 144 be can be rotated for top surface
Connection second component 146, the top of the second component 146 is located at the bottom surface of jacking block 141 or more, sets on the jacking block 141
It is equipped with the first groove body 145 to match with 147 top of first component, is provided on the sole piece 143 and 146 lower part of second component
The second groove body 148 to match, the top surface of the sole piece 143 are provided with the third groove body for accommodating first body of rod 144；Described middle piece
142 side has the second opening, not interfered when moving by first body of rod 144.
In the above-mentioned technical solutions, the height of entire fixed block 14 is adjusted, by block 142 in setting or extraction to adapt to
Different cameras 5, is drawn off when not needing middle piece 142, rotates first body of rod 144, is taken in third groove body, make to push up
Block 141 and sole piece 143 are fastened togather, in addition when block 142 in needs, after middle piece 142 engaging is entered, the one of sole piece 143
Baffle 149 is arranged in side, and baffle 149 and jacking block 141 is fixed, to prevent middle piece 142 to move to the direction far from first body of rod 144
It is dynamic, the function of a limit is realized to it.
In another technical solution, the column 4 is made of transparent material.
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 (8)
1. the weld seam vision positioning method based on multi-template matching, which comprises the following steps:
S1. welding gun and vision system are fixed, keeps the two relative position constant；
S2. to weld seam centralized positioning:
S21. the clear and apparent weld image of target signature is chosen as reference picture；
S22. two rectangular areas comprising weld seam turning, i.e. ROI region are created in a reference image, are extracted by Threshold segmentation
ROI region laser stripe generates template area image；
S23. parameter selection is carried out in the image of template area successively create multi-template；
S24. real-time image acquisition in the welding process, determines the position at two turnings of groove or so in image, then turned by two
Angle Position center determines the position of Weld pipe mill.
2. the weld seam vision positioning method based on multi-template matching as described in claim 1, which is characterized in that in the S21
Reference picture be removed arc light interference pretreatment.
3. the weld seam vision positioning method based on multi-template matching as described in claim 1, which is characterized in that the S23 exists
Parameter selection is carried out in the image of template area and successively creates multi-template, is specifically included:
One first area template comprising turning left a little of creation；
One second area template comprising right knee region of creation；
Parameter setting carried out to first area template and second area template, parameter setting passes through observation first area template or the
The pyramid number of plies of two region templates and include object-point in first area template or second area template minimum contrast
Influence of two parameters of value to created first area template and second area template determines parameter.
4. the weld seam vision positioning method based on multi-template matching as claimed in claim 3, which is characterized in that the parameter packet
It includes: the step-length in first area template and the rotatable angular range of second area template and rotation angle range.
5. the weld seam vision positioning method based on multi-template matching as claimed in claim 4, which is characterized in that in the S24
Specifically:
Real-time image acquisition in the welding process defines the rectangle ROI region of a position comprising left and right inflection point first, determines
Then the search range of template calls the Template Information created to find template position in the picture, determine that groove is left in image
The position at right two turnings, so that it is determined that the position of Weld pipe mill.
6. the weld seam vision positioning method based on multi-template matching as described in claim 1, which is characterized in that the vision system
Camera in system is fixed as an entirety by attachment device and reflecting mirror；
The attachment device includes:
First fixed plate fixes reflecting mirror, the top surface of first fixed plate is connected with the screening of disc-shaped for cylinder
Tabula rasa；
Two root posts are respectively perpendicular the bottom surface for being set to the first fixed plate, and the line of two root posts passes through described first and fixes
The center of circle in board bottom face；
Camera inner seat, camera is fixed, and the outside of the camera inner seat is cube structure, and opposite first side is symmetrical
It is provided with horizontal shaft, the gear of same model is symmetrically connected on a pair of of shaft；
The outer seat of camera, the square cavity with top surface opening are vertically arranged with a pair of of rack gear, the rack gear in the cavity
It can be engaged with the gear, the rack gear is along the vertical direction by extending to the middle part of cavity at the top of cavity；The outer seat of the camera
It is provided with socket outside, the socket is fixed with the column；
Four groups of gravity centre adjustment mechanisms are respectively arranged at cavity bottom, and four groups of gravity centre adjustment mechanisms are located at camera inner seat bottom
On the middle line on four side of face, the gravity centre adjustment mechanism includes:
Movable block, top surface is horizontal and has card slot, and middle line of the camera inner seat bottom surface along four sides extends to form to the midline
Convex block, the convex block are sticked in the card slot, and the bottom surface of the movable block is inclined-plane, and bottom surface has and bottom surface gradient phase
Same skewback；
Fixed block is fixed in the outer seat of camera, and the top surface of the fixed block is inclined-plane, and top surface is oblique with cooperating with skewback
Slot, the bottom surface of the movable block are bonded with the top surface of the fixed block；
Electric putter, the side that telescopic end is fixed on movable block are fixed, and drive movable block to slide on fixed block, the electric putter
Pedestal be slideably positioned in block set, offer the first opening on the outer seat of the camera so that electric putter moves up and down；
Four groups of infrared distance measuring sensors are fixed on the outer seat of camera and respectively with respect to four angles of camera inner seat bottom surface；
Controller, the signal that receiving infrared-ray distance measuring sensor issues, and signal drives each electric putter based on the received
It is flexible.
7. the weld seam vision positioning method based on multi-template matching as claimed in claim 6, which is characterized in that the fixed block
It is made of jacking block, middle piece and sole piece three parts, and the jacking block top surface is inclined-plane, bottom surface is plane, and described middle piece equal with sole piece
For rectangular parallelepiped structure；
Wherein, first component is provided on the sole piece, the first component lower part is located at the top surface of sole piece hereinafter, described first
Component rotatably connects first body of rod, and the other end of first body of rod rotatably connects second component, second structure
The top of part is located at the bottom surface of jacking block or more, and the first groove body to match with first component top, institute are provided on the jacking block
The second groove body for being provided on sole piece and matching with second component lower part is stated, the top surface of the sole piece, which is provided with, accommodates first body of rod
Third groove body；Middle piece of the side has the second opening, not interfered when moving by first body of rod.
8. the weld seam vision positioning method based on multi-template matching as claimed in claim 6, which is characterized in that the column is
Transparent material is made.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811271176.5A CN109685758B (en) | 2018-10-29 | 2018-10-29 | Weld visual positioning method based on multi-template matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811271176.5A CN109685758B (en) | 2018-10-29 | 2018-10-29 | Weld visual positioning method based on multi-template matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109685758A true CN109685758A (en) | 2019-04-26 |
| CN109685758B CN109685758B (en) | 2020-10-16 |
Family
ID=66185743
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811271176.5A Active CN109685758B (en) | 2018-10-29 | 2018-10-29 | Weld visual positioning method based on multi-template matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109685758B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112785648A (en) * | 2021-04-12 | 2021-05-11 | 成都新西旺自动化科技有限公司 | Visual alignment method, device and equipment based on to-be-imaged area and storage medium |
| CN113639630A (en) * | 2021-04-01 | 2021-11-12 | 浙江大学台州研究院 | Dimension measuring instrument system based on multi-template matching and automatic focusing functions |
| CN113838137A (en) * | 2021-11-24 | 2021-12-24 | 中导光电设备股份有限公司 | Method and system for acquiring correction mark position based on TFT-LCD |
| CN114119456A (en) * | 2021-09-02 | 2022-03-01 | 中国化学工程第六建设有限公司 | Automatic pipeline centering method based on machine vision |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120165668A1 (en) * | 2010-08-02 | 2012-06-28 | Guided Therapy Systems, Llc | Systems and methods for treating acute and/or chronic injuries in soft tissue |
| CN103203760A (en) * | 2012-01-17 | 2013-07-17 | 上海未来伙伴机器人有限公司 | Camera fixing device and panoramic vision system with same |
-
2018
- 2018-10-29 CN CN201811271176.5A patent/CN109685758B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120165668A1 (en) * | 2010-08-02 | 2012-06-28 | Guided Therapy Systems, Llc | Systems and methods for treating acute and/or chronic injuries in soft tissue |
| CN103203760A (en) * | 2012-01-17 | 2013-07-17 | 上海未来伙伴机器人有限公司 | Camera fixing device and panoramic vision system with same |
Non-Patent Citations (3)
| Title |
|---|
| 王中任 等: "基于线结构光和运动平台的三维重构方法", 《工具技术》 * |
| 肖光润 等: "用于管道焊接的双模板匹配视觉定位方法", 《激光与红外》 * |
| 赵一一: "HALCON形状匹配讲解", 《CSDN》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113639630A (en) * | 2021-04-01 | 2021-11-12 | 浙江大学台州研究院 | Dimension measuring instrument system based on multi-template matching and automatic focusing functions |
| CN112785648A (en) * | 2021-04-12 | 2021-05-11 | 成都新西旺自动化科技有限公司 | Visual alignment method, device and equipment based on to-be-imaged area and storage medium |
| CN114119456A (en) * | 2021-09-02 | 2022-03-01 | 中国化学工程第六建设有限公司 | Automatic pipeline centering method based on machine vision |
| CN114119456B (en) * | 2021-09-02 | 2024-08-23 | 中国化学工程第六建设有限公司 | Automatic pipeline centering method based on machine vision |
| CN113838137A (en) * | 2021-11-24 | 2021-12-24 | 中导光电设备股份有限公司 | Method and system for acquiring correction mark position based on TFT-LCD |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109685758B (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109685758A (en) | Weld seam vision positioning method based on multi-template matching | |
| CN109604777A (en) | Welding seam traking system and method based on laser structure light | |
| CN107824940A (en) | Welding seam traking system and method based on laser structure light | |
| CN108857152B (en) | Full-space pose detection visual sensor device based on composite laser structure light | |
| CN102073148B (en) | High-accuracy visual coaxial optical alignment assembly system for miniature structural member | |
| CN109658456A (en) | Tank body inside fillet laser visual vision positioning method | |
| CN106182004A (en) | The method of the industrial robot automatic pin hole assembling that view-based access control model guides | |
| JPH03180292A (en) | Apparatus for adjusting laser beam path for laser beam machine | |
| CN104084669A (en) | Visual detecting method and device based on decoupling detection for circular seams | |
| CN102303190A (en) | Method for visually tracking plane abut-jointed weld beam by linear laser | |
| CN105303560A (en) | Robot laser scanning welding seam tracking system calibration method | |
| CN108015455A (en) | Multifunctional vision sensor device based on recombination laser structure light | |
| CN1600505A (en) | Servo binocular vision sensors on welding robot | |
| CN203973025U (en) | A kind of circumferential weld vision inspection apparatus detecting based on decoupling zero | |
| CN102179598B (en) | Method and device for centering initial welding positions of flat narrow welding lines and tracking welding lines | |
| CN209446480U (en) | Vehicle lacquer painting and apparent flaws detection device | |
| CN110315249A (en) | Space right-angle weld seams shaped zigzag line based on line laser structured light is fitted system | |
| CN117047237B (en) | Intelligent flexible welding system and method for special-shaped parts | |
| CN212493722U (en) | Five-axis gluing equipment | |
| CN102538707B (en) | Three dimensional localization device and method for workpiece | |
| CN109822194A (en) | A kind of weld tracker and welding method | |
| CN109447971A (en) | A kind of cycloid gear molded line error is accurately extracted and crucial accuracy of form and position rapid analysis method | |
| CN104006789A (en) | Device and method for measuring spatial dynamic angle | |
| CN203673356U (en) | Light spot indication robot | |
| CN101706665B (en) | Method for adjusting self-tracing pose of mobile welding robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |