CN106514064A - Weld verification method - Google Patents
Weld verification method Download PDFInfo
- Publication number
- CN106514064A CN106514064A CN201510582953.8A CN201510582953A CN106514064A CN 106514064 A CN106514064 A CN 106514064A CN 201510582953 A CN201510582953 A CN 201510582953A CN 106514064 A CN106514064 A CN 106514064A
- Authority
- CN
- China
- Prior art keywords
- verified
- weld
- straight line
- image
- weld seam
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K37/00—Auxiliary devices or processes, not specially adapted to a procedure covered by only one of the preceding main groups
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K33/00—Specially-profiled edge portions of workpieces for making soldering or welding connections; Filling the seams formed thereby
- B23K33/004—Filling of continuous seams
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Abstract
The invention relates to a verification method for butt weld detection results. The verification method includes the following steps that a weld image and a weld straight line to be verified in the weld image are acquired; q pixel points are selected at the two sides of each row of intersection points with the weld straight line to be verified to form an image subset, the direction of an original image row is generally perpendicular to the weld straight line to be verified, and q is a positive integer which is larger than or equal to 1; all the rows of the image subset are added to obtain an array, the array is subjected to first-order difference, and the positions of the maximum value and the minimum value are found out; and if the maximum value and the minimum value are located at the two sides of the weld straight line to be verified, it is judged that the weld position to be verified is reliable. By the adoption of the method, butt welds (including the welds without gaps and with certain gaps) of sheets under various conditions can be reliably verified, and the phenomenon of recognition failure can be avoided.
Description
Technical field
A kind of a kind of the present invention relates to visual identity in robot or automatic welding to metal thin plate structure butt weld, more particularly to weld seam verification method.
Background technology
Using very extensively in aerospace industry, the automatic identification and tracer request to such joint is also very urgent for the tight butt joint of thin-slab structure.People employ various methods and this are studied, such as laser scanning and ranging method and Intensity Analysis method, image analysis method and electromagnetic transducer method etc..These research work also have relatively large distance with practical application at present.Because these methods almost assume that joint has certain gap, misalignment or slant characteristic, and joint situation in practice is difficult to meet and requires.For example, existing weld seam has the situation of a variety of butt welds, and such as, some places have little gap, and some places gap is slightly bigger, in addition where it is completely very close to each other.So existing method can not meet actual demand.
The content of the invention
In order to overcome the drawbacks described above of prior art, the technical problem to be solved to be to provide a kind of weld seam verification method, its can solve the above problems in any one.
For reaching above-mentioned purpose, the technical scheme is that：
A kind of weld seam verification method, it comprises the following steps：
Obtain weld image and weld seam straight line to be verified in weld image；
Q pixel composition image subset is chosen on the both sides of every a line and the point of weld seam straight line intersection to be verified, generally perpendicular to weld seam straight line to be verified, wherein q is the positive integer more than or equal to 1 in the direction of original image row；
All rows of described image subset are added and obtain an array, do first-order difference to this array and the position of maximizing and minima；
If maximum and minima are located at the both sides of weld seam straight line to be verified, judge that position while welding to be verified is reliable.
Preferably, it includes：Choosing in q pixel composition image subset step with the both sides of the point of weld seam straight line intersection to be verified in every a line, q=32.
Preferably, it is included in and this array is done in first-order difference step, and difference formula is D (i)=F (i) F (i-1).
Preferably, it includes：If maximum and minima are not located at the both sides of weld seam straight line to be verified, judge that position while welding to be verified is insecure.
Preferably, it includes：When the distance of gained maxima and minima position is more than weld seam straight line to be verified, the distance based on maxima and minima position is adjusted to the weld seam straight line of gained.
The present invention adopts said method, can be to the butt weld of the thin plate of various situations（Including gapless and the weld seam for having certain interval）Reliable authentication, without the situation that authentication failed occurs.
Description of the drawings
Accompanying drawing described here is only used for task of explanation, and is not intended to by any way limit scope disclosed by the invention.In addition, shape and proportional sizes of each part in figure etc. are only schematic, the understanding of the present invention are used to help, are not the shape and proportional sizes for specifically limiting each part of the invention.Those skilled in the art under the teachings of the present invention, can select various possible shapes and proportional sizes to implement the present invention as the case may be.
Fig. 1 is the device in the application using the weld seam verification method.
Fig. 2 does not have the image of active illumination when being and welding using the aluminium sheet butt weld that welding robot of the prior art is obtained.
Fig. 3 is the image for having active illumination when the aluminium sheet butt weld of welding robot acquisition in the application is welded.
Fig. 4 is the image of the 3mm thickness aluminium sheet unspaced butted weld testing results of welding robot acquisition in the application.
Fig. 5 is the image of the 3mm steel plate unspaced butted weld testing results of welding robot acquisition in the application.
Fig. 6 is the image that the 3mm steel plate gaplesss of welding robot acquisition in the application have artificial cut butt weld testing result.
Fig. 7 is the image that the 3mm steel plate gaplesss of welding robot acquisition in the application have tack weld butt weld testing result.
Fig. 8 is the image of the 3mm thickness aluminium sheets gap 0.02mm butt weld testing results of welding robot acquisition in the application.
Fig. 9 is the image of the 3mm thickness aluminium sheets gap 0.05mm butt weld testing results of welding robot acquisition in the application.
Figure 10 is the image of the 3mm steel plates gap 0.3mm butt weld testing results of welding robot acquisition in the application.
Figure 11 is the image of the 3mm steel plates gap 0.5mm butt weld testing results of welding robot acquisition in the application.
The reference of the figures above is：1st, welding gun；2nd, video camera；3rd, active illumination.
Specific embodiment
With reference to the description of accompanying drawing and the specific embodiment of the invention, the details of the present invention can be clearly understood.But, the specific embodiment of invention described herein is only used for explaining the purpose of the present invention, and it is limitation of the present invention that can not be understood as by any way.Under the teachings of the present invention, technical staff is contemplated that any possible deformation based on the present invention, and these are regarded as belonging to the scope of the present invention.
The invention discloses a kind of weld inspection method, it comprises the following steps：
Obtain weld image；
In the weld image for obtaining, the line direction of initial pictures is approximately perpendicular to weld seam, chooses a line every n rows and is processed, and chooses m rows altogether.Here m is the positive integer more than 8, and n is the integer more than or equal to 0；F (i, j) illustrates the grey scale pixel value of i-th point of jth row；
Row to choosing is smoothed；
First-order difference is done to the data after smoothing processing；
The position of k maximum pixel and the position of k minimum pixel is chosen in differentiated data, and wherein k is 2 to 4；
In to previous step, the minimizing position of one gray scale of searching in the neighborhood of each of 2k position, obtains 2k new position；
To 2k point on the row of all selections（Common 2k*m）, it is Hao Er（Hough Transform）Change, obtains
ρ=Xcos θ+Ysin θ, wherein distance of the ρ origins to straight line, θ are the angle of the normal with X-axis of the straight line；
Specifically, in the present embodiment, after thin plate image is obtained, the upper left corner of image is defined as into origin, Y-direction is defined as downwards by origin；X-direction is defined as to the right by origin；Capable be oriented parallel to X-direction, row are oriented parallel to Y-direction.
For speed up processing, a line can be chosen every m rows and sampled along Y-direction.In the present embodiment, m can be the arbitrary number in 0 to 10.For example, a width gray level image has 262144 points if being sampled as the matrix of 512 512 points of X, and each point has a gray value, or is brightness value.Coordinate is followed the example of：, in the upper left corner of image, for level to the right, Y-direction is for vertically downward for X-direction for zero.Here f (i, j) has meant that the grey scale pixel value of i-th point of jth row.If a line being taken every 1 row, be exactly finally to obtain 256 rows, if taking a line every 3 rows, be exactly finally to obtain 128 rows.
The row that sampling is obtained is smoothed, smooth window is 3X1.
Wherein, it is smoothed in the row to choosing, the formula of smoothing processing is f (i)=(f (i-1)+f (i)+f (i+1))/3.
Data after smoothing processing are done with first-order difference, d (i, j)=f (i, j) f (i-1, j), d (i, j) is pixel (i here, j) place's gray scale difference score value, (i-1, j) is pixel (i respectively for f (i, j) and f, j) with (i-1, j) gray value at place.The position of k maximum pixel and the position of k minimum pixel is chosen in differentiated data.In the present embodiment, k=for any one positive integer in 2 to 4.
In to previous step, the minimizing position of one gray scale of searching in the neighborhood of each of 2k position, obtains 2k new position.Here the corresponding pixel count of maximal clearance width half that the selection treating excess syndrome border of the radius of neighbourhood is assumed.
To 2k point on the row of all selections（Common 2k*m）It is Hao Er（Hough Transform）ρ = Xcosθ + Ysinθ.Parameter ρ and θ uniquely determine straight line, the distance of ρ origins to straight line, and θ is the angle of the normal with X-axis of the straight line.
Maximum is searched in array after the conversion, the corresponding straight line of this maximum is exactly the weld seam straight line of our candidates（The corresponding physical region of image is probably in 11.2mm X
11.2mm, assume in the range of the whole image weld seam be straight line be to set up）.
Above weld seam detection algorithm can not only reliably detect out gapless and closely spaced butt weld, and in the case where there is strong interference such as surface scratch, tack weld, testing result is also very reliable.
After weld seam detection, the identifying system that the present invention is adopted can also be verified to weld seam.
A kind of weld seam verification method, it comprises the following steps：
Obtain weld image and weld seam straight line to be verified in weld image；
Q pixel composition image subset is chosen on every a line of original image and the both sides of the point of weld seam straight line intersection to be verified, the line direction of initial pictures is approximately perpendicular to weld seam, wherein q=32；
All rows of described image subset are added and obtain an array, do first-order difference to this array and the position of maximizing and minima, difference formula is D (i)
=f(i)–f(i-1)；
If the position of maximum and minima is located at the both sides of weld seam straight line to be verified, judge that position while welding to be verified is reliable；
If the position of maximum and minima is not located at the both sides of weld seam straight line to be verified, judge that position while welding to be verified is insecure.
During the weld gap width that the distance between maxima and minima position when gained is adjusted more than the need being previously set, the weld seam straight line of gained is adjusted with the distance between maxima and minima position.Specifically, in our actual experiments, if weld gap is more than 0.2mm, testing result just adjusts.Here the position of maximum corresponds to the left hand edge in gap, the right hand edge in the position correspondence gap of minima.The difference correspondence of maximum and minima position and gap width.In our experiment, the image of 512X512 corresponds to physical size 11.2mmX12.2mm, so 0.2mm corresponds to 9 pixels.The gap of 0.5mm then corresponds to 22 pixels.After the location gap for calculating maximum and minima pixel, it is possible to which according to above asking, image subset and image pixel are counter with the relation of physical size to extrapolate new value.
With reference to shown in Fig. 1, disclosed herein as well is a kind of plate sheet welding machine people, it includes：
Robotic arm, which has multiple degree of freedom；
Welding gun 1, the welding gun 1 are arranged on the robot arm, and so as to move with the robotic arm, the welding gun 1 is for welding to the thin plate；
Video camera 2, which includes the imageing sensor for obtaining the thin plate and/or weld image and obtains the electronic shutter of image for controlling described image sensor；
Active illumination 3, the active illumination 3 include the light emitting diode for providing illumination to the video camera 2, light emitting diode action based on the electronic shutter action.
Specifically, in the present embodiment, robotic arm can have six even more degree of freedom, carry out precise control with the running orbit of Butt welding gun 1.
Welding gun 1 is controlled by robotic arm, such that it is able to optionally be manipulated to thin plate, for example, welds or stops welding.Welding gun can other be similar to the devices that can be used for welding for argon-arc welding gun etc..
The high resolution of video camera 2, small volume are adapted to robot and use.
Video camera 2 includes high-precision imageing sensor, and imageing sensor is controlled by electronic shutter to carry out thin plate the acquisition of image.
In the present embodiment, active illumination 3 adopts powerful light emitting diode.As light emitting diode can be controlled by the signal of telecommunication, control so as to light emitting diode can be based on electronic shutter.For example, when electronic shutter manipulates imageing sensor obtains image, light emitting diode can provide light to the region of thin plate, so as to give thin plate with the illumination of certain brightness.
And the active illumination adopted in prior art, for traditional illuminating lamp, its on-off control usually has a certain degree of delay, thus is not suitable for high-precision, quick occasion.
Contrary, the plate sheet welding machine people in the application can quickly be processed to thin plate, be considerably increased its precision and navigability.
Preferably, the light emitting diode is large-power light-emitting diodes.
Preferably, the light emitting diode causes described image sensor to obtain image based on the electronic shutter and opens.
Preferably, the light emitting diode causes described image sensor not obtain image and close based on the electronic shutter.
Preferably, the light emitting diode is obtained based on described image sensor image and adjust its light intensity for sending.
Preferably, the video camera 2, the light emitting diode and the welding gun are along linear arrangement.
Preferably, the video camera 2 and the light emitting diode are located at the front of the moving direction of the welding gun.
Compare shown in Fig. 2, illustrate with reference to Fig. 3 to Figure 11, the image definition obtained using the welding robot in the application is higher.
In the present embodiment, active illumination 3 adopts powerful light emitting diode.As light emitting diode can be controlled by the signal of telecommunication, control so as to light emitting diode can be based on electronic shutter.For example, when electronic shutter manipulates imageing sensor obtains image, light emitting diode can provide illumination to the region of thin board welding seam, so that the thin board welding seam image for obtaining has suitable brightness.
And the active illumination adopted in prior art, for traditional illuminating lamp, its on-off control usually has a certain degree of delay, thus is not suitable for high-precision, quick occasion.
Contrary, the plate sheet welding machine people in the application can quickly be processed to thin plate, be considerably increased its precision and navigability.
Preferably, the light emitting diode is large-power light-emitting diodes.
Preferably, the light emitting diode causes described image sensor to obtain image based on the electronic shutter and opens.
Preferably, the light emitting diode causes described image sensor not obtain image and close based on the electronic shutter.
Preferably, the light emitting diode is obtained based on described image sensor image and adjust its light intensity for sending.
Preferably, the video camera 2, the light emitting diode and the welding gun are along linear arrangement.
Preferably, the video camera 2 and the light emitting diode are located at the front of the moving direction of the welding gun.
Compare shown in Fig. 2, illustrate with reference to Fig. 3 to Figure 11, the image definition obtained using the welding robot in the application is higher.
Each embodiment in this specification is described by the way of progressive, and what each embodiment was stressed is the difference with other embodiment, between each embodiment identical similar part mutually referring to.
Above-described embodiment technology design only to illustrate the invention and feature, its object is to allow person skilled in the art will appreciate that present disclosure and implement according to this, can not be limited the scope of the invention with this.All equivalence changes made according to spirit of the invention or modification, should all be included within the scope of the present invention.
Claims (5)
1. a kind of weld seam verification method, it is characterised in that it comprises the following steps：
Obtain weld image and weld seam straight line to be verified in weld image；
Q pixel composition image subset is chosen on the both sides of every a line and the point of weld seam straight line intersection to be verified, generally perpendicular to weld seam straight line to be verified, wherein q is the positive integer more than or equal to 1 in the direction of original image row；
All rows of described image subset are added and obtain an array, do first-order difference to this array and the position of maximizing and minima；
If maximum and minima are located at the both sides of weld seam straight line to be verified, judge that position while welding to be verified is reliable.
2. weld seam verification method according to claim 1, it is characterised in that it includes：Choosing in q pixel composition image subset step with the both sides of the point of weld seam straight line intersection to be verified in every a line, q=32.
3. weld seam verification method according to claim 1, it is characterised in that it is included in, difference formula are D (i)=F (i) F (i-1).
4. weld seam verification method according to claim 1, it is characterised in that it includes：If maximum and minima are not located at the both sides of weld seam straight line to be verified, judge that position while welding to be verified is insecure.
5. weld seam verification method according to claim 1, it is characterised in that it includes：When the distance of gained maxima and minima position is more than weld seam straight line to be verified, the distance based on maxima and minima position is adjusted to the weld seam straight line of gained.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510582953.8A CN106514064A (en) | 2015-09-15 | 2015-09-15 | Weld verification method |
| PCT/CN2015/099556 WO2017045305A1 (en) | 2015-09-15 | 2015-12-29 | Weld seam verification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510582953.8A CN106514064A (en) | 2015-09-15 | 2015-09-15 | Weld verification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106514064A true CN106514064A (en) | 2017-03-22 |
Family
ID=58288138
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510582953.8A Pending CN106514064A (en) | 2015-09-15 | 2015-09-15 | Weld verification method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106514064A (en) |
| WO (1) | WO2017045305A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106925924A (en) * | 2017-05-17 | 2017-07-07 | 山东双源印铁制罐有限公司 | A kind of weld seam applies outward benefit protection intensive |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2603916A1 (en) * | 2005-04-04 | 2006-10-12 | Fisher-Rosemount Systems, Inc. | Statistical processing methods used in abnormal situation detection |
| CN102175700A (en) * | 2011-01-20 | 2011-09-07 | 山东大学 | Method for detecting welding seam segmentation and defects of digital X-ray images |
| CN103759648A (en) * | 2014-01-28 | 2014-04-30 | 华南理工大学 | Complex fillet weld joint position detecting method based on laser binocular vision |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2465231C (en) * | 2001-11-15 | 2011-08-09 | Elpatronic Ag | Method and device for evaluation of jointing regions on workpieces |
| JP2005014027A (en) * | 2003-06-24 | 2005-01-20 | Enshu Ltd | Weld zone image processing method, welding management system, feedback system for welding machine, and butt line detection system |
| CN101526480B (en) * | 2009-04-27 | 2011-06-29 | 清华大学 | Real-time detection method of butt weld of thin plates based on visual sense |
| CN101559512B (en) * | 2009-05-21 | 2011-05-04 | 山东大学 | Welding track detection and control method of plate butt weld based on laser ranging |
| CN104636760B (en) * | 2015-03-11 | 2017-09-05 | 西安科技大学 | A kind of localization method of weld seam |
-
2015
- 2015-09-15 CN CN201510582953.8A patent/CN106514064A/en active Pending
- 2015-12-29 WO PCT/CN2015/099556 patent/WO2017045305A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2603916A1 (en) * | 2005-04-04 | 2006-10-12 | Fisher-Rosemount Systems, Inc. | Statistical processing methods used in abnormal situation detection |
| CN102175700A (en) * | 2011-01-20 | 2011-09-07 | 山东大学 | Method for detecting welding seam segmentation and defects of digital X-ray images |
| CN103759648A (en) * | 2014-01-28 | 2014-04-30 | 华南理工大学 | Complex fillet weld joint position detecting method based on laser binocular vision |
Non-Patent Citations (1)
| Title |
|---|
| 沈祯杰: "《波形检测法》", 《基于机器视觉的金属制罐焊缝缺陷检测系统研究》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106925924A (en) * | 2017-05-17 | 2017-07-07 | 山东双源印铁制罐有限公司 | A kind of weld seam applies outward benefit protection intensive |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2017045305A1 (en) | 2017-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105320399B (en) | Laser patterning skew correction | |
| CN105511123B (en) | A kind of High Precision Automatic Systems for optical inspection and method based on mechanical arm | |
| US8244040B2 (en) | Object position recognition system, object positioning system, and system and method for adjoining objects | |
| CN103759648B (en) | A kind of complicated angle welding method for detecting position based on Binocular stereo vision with laser | |
| CN111192307B (en) | Self-adaptive deviation rectifying method based on laser cutting of three-dimensional part | |
| CN107256567B (en) | Automatic calibration device and calibration method for hand-eye camera of industrial robot | |
| CN106378514B (en) | The non-homogeneous subtle more weld seam vision detection systems of stainless steel and method based on machine vision | |
| CN101033958A (en) | Mechanical vision locating method | |
| Liu et al. | Precise initial weld position identification of a fillet weld seam using laser vision technology | |
| CN108907526A (en) | A kind of weld image characteristic recognition method with high robust | |
| CN109709102A (en) | Die-cutting machine lithium battery pole slice physical imperfection detection system, method and device | |
| CN103499297A (en) | CCD (Charge Coupled Device)-based high-accuracy measuring method | |
| JPS60231208A (en) | Control method of automatic machine | |
| CN106931881A (en) | A kind of polyphaser visible detection method | |
| CN103954625A (en) | Traceable damage threshold measurement technology facing laser film internal defects | |
| CN106530269A (en) | Weld detection method | |
| CN109146866A (en) | The method and device that robot handles weld seam | |
| CN113538583A (en) | Method for accurately positioning position of workpiece on machine tool and vision system | |
| CN114581368B (en) | Bar welding method and device based on binocular vision | |
| CN106514063A (en) | Thin plate welding robot | |
| JP2017173142A (en) | Image processing device, image processing method and micro joint cutting system | |
| CN106514064A (en) | Weld verification method | |
| KR20170142379A (en) | Apparatus for detect dimensional welding line of welding robot using image processing | |
| US9636824B2 (en) | Transfer system | |
| CN108346138A (en) | A kind of detection method of surface flaw and system based on image procossing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170322 |