CN114160926A - Monoclinic line laser vision sensing method and system for welding seam tracking - Google Patents
Monoclinic line laser vision sensing method and system for welding seam tracking Download PDFInfo
- Publication number
- CN114160926A CN114160926A CN202111570543.3A CN202111570543A CN114160926A CN 114160926 A CN114160926 A CN 114160926A CN 202111570543 A CN202111570543 A CN 202111570543A CN 114160926 A CN114160926 A CN 114160926A
- Authority
- CN
- China
- Prior art keywords
- monoclinic
- image
- laser
- welding
- welding 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
Images
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
- B23K9/00—Arc welding or cutting
- B23K9/12—Automatic feeding or moving of electrodes or work for spot or seam welding or cutting
- B23K9/127—Means for tracking lines during arc welding or cutting
- B23K9/1272—Geometry oriented, e.g. beam optical trading
- B23K9/1274—Using non-contact, optical means, e.g. laser means
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Optics & Photonics (AREA)
- Geometry (AREA)
- Plasma & Fusion (AREA)
- Mechanical Engineering (AREA)
- Laser Beam Processing (AREA)
Abstract
The invention provides a monoclinic line laser vision sensing method and a monoclinic line laser vision sensing system for welding seam tracking, wherein the method comprises the following steps: extracting a target area image by adopting an ROI (region of interest) from an initial image of a monoclinic laser stripe projected on a workpiece, which is acquired by a camera; performing median filtering on the target area image, improving the spike interference effect, keeping the edge steep, and removing interference; carrying out graying processing and binarization processing on the image subjected to median filtering, and improving the image processing efficiency; processing the binarized weld image by adopting an opening operation; and calculating to obtain welding seam information and judging the spatial position of the welding seam according to the processed welding seam image. The monoclinic line laser vision sensing method and the monoclinic line laser vision sensing system for welding seam tracking are used for identifying the spatial position of the welding seam and tracking the welding seam during automatic welding, and can improve the welding quality.
Description
Technical Field
The invention relates to the technical field of weld joint tracking, in particular to a monoclinic line laser vision sensing method and a monoclinic line laser vision sensing system for weld joint tracking.
Background
At present, most of civil air defense engineering butt weld joints in China still adopt a manual welding method, the problems of bad working conditions and high working strength of workers exist, and meanwhile, the weld joint forming, the welding efficiency, the welding quality and the like are all required to be improved. The welding automation technology is widely applied to industrial production with excellent working efficiency, is an inevitable trend of welding technology development, and is an important means for realizing safer, efficient and intelligent production.
At present, sensors adopted in automatic welding are mainly arc sensors, mechanical sensors, visual sensors and the like. The arc sensor directly extracts a welding seam position deviation signal from a welding arc, and does not need to be additionally provided with any device, so that the arc sensor has good real-time performance, good accessibility and flexibility of welding gun movement, and particularly meets the requirement of low-cost automation in the welding process. The method utilizes the dynamic and static changes of the electric parameters of the electric arc as characteristic signals and realizes the tracking control in the high and low directions and the horizontal direction through a certain control strategy. However, the arc sensor is only suitable for V-shaped, single V-shaped, U-shaped and fillet welds, and can obtain a small amount of information, which has limitations and is not suitable for butt welding of thin plates.
The typical mechanical contact sensor detects the position of a weld joint in front of a welding torch according to a guide wheel or a guide rod, guides the weld joint through the forcible force of the shape of the weld joint on the guide rod or the guide wheel, and reflects the deviation information of the weld joint into a detector, thereby realizing the tracking of the weld joint. The mechanical contact type sensor has simple structure, convenient operation and low price and is not interfered by arc smoke, splashing and the like. However, because the amount of information of the mechanical sensor is small, the mechanical sensor is mainly matched with other sensors to complete the task of seam tracking in automatic welding and is not generally used independently.
The laser vision-based weld joint tracking technology has good development prospect in the welding manufacturing industry. The laser vision sensor adopts a non-contact sensing mode, can acquire high-quality welding seam images, and effectively improves the sensitivity and detection precision of welding seam tracking. The laser light source has the advantages of high brightness and good monochromaticity, can effectively weaken the influence of arc light on the image quality, well protect the characteristic information of a welding line and reduce the difficulty of subsequent image processing. However, when a single laser marker irradiates a workpiece, only the central feature point of the weld can be distinguished, and the spatial position of the weld cannot be detected. The external arrangement of the double-laser line marker is too cumbersome and is not as simple as a single line marker.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a monoclinic line laser sensing method and a monoclinic line laser sensing system which can improve the welding quality and are used for identifying the spatial position of a welding line and tracking the welding line in automatic welding.
In order to solve the problems, the technical scheme of the invention is as follows:
a monoclinic laser vision sensing method for weld seam tracking, the method comprising the steps of:
extracting a target area image by adopting an ROI (region of interest) from an initial image of a monoclinic laser stripe projected on a workpiece, which is acquired by a camera;
performing median filtering on the target area image, improving the spike interference effect, keeping the edge steep, and removing interference;
carrying out graying processing and binarization processing on the image subjected to median filtering, and improving the image processing efficiency;
processing the binarized weld image by adopting an opening operation; and
and calculating to obtain welding seam information according to the processed welding seam image and judging the spatial position of the welding seam.
Optionally, the formula of the two-dimensional median filtering is:
wherein A is the window size; { fijIs a two-dimensional data sequence.
Optionally, the binary function of the binarization processing is:
where T is a predetermined threshold, white pixels are larger than the gray level T and black pixels are smaller than T.
Optionally, the step of processing the binarized weld image by using an opening operation specifically includes: the on operation is represented as a fitting process:
wherein {. } represents the union of all sets in the parenthesis.
Optionally, the step of calculating to obtain the weld information and determining the spatial position of the weld according to the processed weld image specifically includes: and generating a monoclinic line laser stripe through a laser striper, calculating the position relation between the current welding gun and the welding seam through the distribution information of the monoclinic line laser stripe, and guiding the welding gun to automatically track the welding seam.
Optionally, the step of calculating to obtain the weld information and determining the spatial position of the weld according to the processed weld image specifically includes: and calculating the width, the depth and the deviation correction quantity of the butt weld by a computer according to the image of the monoclinic laser stripe weld.
The system comprises a CCD camera, a composite optical filter, a laser striper and a workpiece, wherein the laser striper is obliquely arranged in the horizontal direction to generate oblique laser broken line stripes, the emitted laser stripes are projected on the workpiece and collected on the CCD camera through the composite optical filter, and after image processing, welding seam information and offset are obtained, and the spatial position of the welding seam is judged.
Optionally, with reference to the welding direction, the laser striping machine is obliquely placed at the rear side of the CCD camera at an angle of 20 °, and the included angle between the laser striping machine and the vertical direction is 20 °.
Compared with the prior art, the laser marker is obliquely arranged in the horizontal direction, a single-line laser sensor is adopted to collect a welding line in a dark environment, the cross section of a welding groove is scanned in real time, the shape of the groove is detected, a welding line image is obtained simply and efficiently, the response speed is high, the real-time performance is strong, and the requirement of intelligent control can be met; the spatial position of the welding seam is judged by adopting the monoclinic line laser, and a folding line is generated when the laser stripe is projected to the joint of the wall surface and the plane, so that the position close to the boundary of the welding seam is judged, and the spatial position of the welding seam can be judged without a plurality of laser stripers.
In addition, the spatial position of welding can be automatically judged, manual intervention is reduced, the automation level of welding is improved, and the welding efficiency is improved; the method has the capability of continuously extracting the characteristic points of the welding line, reduces the cost of automatic welding equipment, reduces peripheral equipment, and improves the continuity and accuracy of welding.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic diagram of a monoclinic laser vision sensing system for seam tracking according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the position of a monoclinic boundary provided by the embodiment of the present invention;
FIG. 3 is a block diagram of a flow chart of a monoclinic laser vision sensing method for weld seam tracking according to an embodiment of the present invention;
FIG. 4 is a schematic view of a partial weld image provided by an embodiment of the present invention;
fig. 5 is a schematic diagram of weld information calculation provided in the embodiment of the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
Specifically, fig. 1 is a schematic diagram of a monoclinic laser vision sensing system for seam tracking according to an embodiment of the present invention, and fig. 2 is a schematic diagram of a monoclinic boundary position according to an embodiment of the present invention, as shown in fig. 1 and fig. 2, the system includes: the CCD camera 1, the composite optical filter 2, the laser line marker 3 and the workpiece 4 are obliquely irradiated by the laser line marker 3, the CCD camera 1 vertically shoots images, and welding seam information and offset are obtained after image processing, and the spatial position of a welding seam is judged.
The laser striping machine 3 is obliquely arranged at the rear side of the CCD camera 1 at an angle of 20 degrees, the laser striping machine 3 is obliquely arranged in the horizontal direction to generate oblique laser stripes, and the laser stripes are projected on a workpiece 4 and collected on the CCD camera 1 through the composite optical filter 2. And the laser striping machine 3 is positioned behind the CCD camera 1 by taking the welding direction as reference, and the included angle between the laser striping machine 3 and the vertical direction is 20 degrees and is obliquely arranged in the horizontal direction to generate oblique laser stripes.
The method comprises the steps of carrying out gray level processing on monoclinic laser stripes projected on a workpiece 1, taking a target area, carrying out binarization, carrying out morphological opening operation, extracting a central line of a welding line, solving the position of a characteristic point of the welding line by utilizing the intersection point of the central line in the horizontal direction and the stripes, and judging the spatial position of the welding line.
Fig. 3 is a flow chart of a monoclinic laser vision sensing method for seam tracking according to an embodiment of the present invention, as shown in fig. 3, the method includes the following steps:
s1: extracting a target area image by adopting an ROI (region of interest) from an initial image of a monoclinic laser stripe projected on a workpiece, which is acquired by a camera;
specifically, the laser striping machine is placed in a manner of inclining a laser striping machine to the horizontal direction to generate inclined laser broken line stripes, the laser stripes emitted by the laser striping machine are projected on a workpiece, and are collected to a CCD camera through a composite optical filter and then are subjected to subsequent image processing.
S2: performing median filtering on the target area image, improving the spike interference effect, keeping the edge steep, and removing interference;
specifically, the formula of the two-dimensional median filter is as follows:
in the formula: a is the window size; { fijIs a two-dimensional data sequence.
S3: carrying out graying processing and binarization processing on the image subjected to median filtering, and improving the image processing efficiency;
specifically, the threshold segmentation not only can compress a large amount of data and reduce the storage capacity, but also can greatly simplify the subsequent analysis and processing steps. When the gray difference between the welding seam and the surrounding is large and the level is clear, the method can be used for detecting the position of the welding seam well. The binary function of the binarization processing is as follows:
wherein T is a specified threshold. A picture pixel is white when it is larger than the grayscale T and is black when it is smaller than T. And in the binarization process, selecting an optimal threshold value by adopting a maximum variance threshold value method according to the statistical distribution property of the image.
S4: processing the binarized weld image by adopting an opening operation;
specifically, in order to eliminate fine objects, the object and the boundary of a larger object are separated at the fine part, so that the burrs at the edge of the fold line can be removed, and the connection between the fold lines can be separated to prepare for the extraction of the central line. The opening operation generally smoothes the contour of the object, breaking the narrower narrow neck and eliminating the thin protrusions. The open operation of the structural element B on the set A is represented as
Thus, the opening of B to A is the erosion of B to A, followed by the expansion of the result with B.
The on operation can be expressed as a fitting process:
wherein {. } represents the union of all sets in the parenthesis.
In this embodiment, in order to better detect the image stripes and calculate, it is particularly important to extract the center line of the polyline stripe, add the edge values of the same longitudinal coordinate of the stripe and divide by 2 to obtain the longitudinal coordinate of the center value, and then change all the points except the center point into black or background color.
S5: and calculating to obtain welding seam information according to the processed welding seam image and judging the spatial position of the welding seam.
The processing of the laser stripes of the butt welding seam is to calculate the position relation between the current welding gun and the welding seam according to the distribution information of the laser stripes and guide the welding gun to automatically track the welding seam. After the position of each feature point on the pixel plane is determined, the amount of correction is calculated by a computer.
Because the camera is always in the middle position, the taken images are symmetrical left and right. Therefore, the broken line part in the center of the image is taken for image processing, and the real-time performance of the image processing is improved. Fig. 4 is a schematic diagram of a cut-out part of a weld image, and as shown in fig. 4, the coordinate positions of 6 points 1, 2, 3, 4, 5 and 6 are firstly determined, and the basic idea is as follows: searching the images from top to bottom in the order of the positive direction of the y-axis, the black dots in the first and last columns are the starting point 1 and the ending point 6. Because only one pixel point in each column is black at most, the pixel points are found from the starting point, and if the pixel points of the left three neighborhoods or the right three neighborhoods are blank, the pixel points are characteristic points, and the broken line points 2, 3, 4, 5 and 6 are judged. This allows the determination of x, y values for 6 points.
When searching the characteristic points row by row, if the slope changes obviously and the characteristic points are not at the broken line, judging that the characteristic points reach the welding seam boundary, and stopping welding.
And then calculating the width, the depth and the deviation correction amount of the butt weld by a computer, wherein the width of the butt weld is x3-x 2.
Therefore, the centering deviation of the welding wire and the welding seam can be obtained by half of the difference of the two non-broken line light veins of the welding seam after the ROI, namely the offset
If Δ x is less than 0, the welding gun is inclined to the right; if delta x is larger than 0, welding is performed to the left; if Δ x is 0, the welding torch is aligned exactly to the center of the weld.
As shown in FIG. 5, the length of the dashed line in FIG. 5 reflects the depth information, let it be h1Then, then
h1Not the depth of the weld, but the relationship between the depth of the weld and the depth of the image plane, h2 the actual depth of the weld, laser sheetAnd the light beam is projected to the workpiece at an angle alpha to the optical axis of the CCD. Actual weld depth h2Projection angles alpha and h with laser1It is related.
Thus, h2＝h1And the width, the depth and the deviation correction quantity of the butt welding seam can be calculated by cot alpha.
Compared with the prior art, the laser marker is obliquely arranged in the horizontal direction, a single-line laser sensor is adopted to collect a welding line in a dark environment, the cross section of a welding groove is scanned in real time, the shape of the groove is detected, a welding line image is obtained simply and efficiently, the response speed is high, the real-time performance is strong, and the requirement of intelligent control can be met; the spatial position of the welding seam is judged by adopting the monoclinic line laser, and a folding line is generated when the laser stripe is projected to the joint of the wall surface and the plane, so that the position close to the boundary of the welding seam is judged, and the spatial position of the welding seam can be judged without a plurality of laser stripers.
In addition, the spatial position of welding can be automatically judged, manual intervention is reduced, the automation level of welding is improved, and the welding efficiency is improved; the method has the capability of continuously extracting the characteristic points of the welding line, reduces the cost of automatic welding equipment, reduces peripheral equipment, and improves the continuity and accuracy of welding.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (8)
1. A monoclinic laser vision sensing method for weld seam tracking, characterized in that the method comprises the following steps:
extracting a target area image by adopting an ROI (region of interest) from an initial image of a monoclinic laser stripe projected on a workpiece, which is acquired by a camera;
performing median filtering on the target area image, improving the spike interference effect, keeping the edge steep, and removing interference;
carrying out graying processing and binarization processing on the image subjected to median filtering, and improving the image processing efficiency;
processing the binarized weld image by adopting an opening operation; and
and calculating to obtain welding seam information according to the processed welding seam image and judging the spatial position of the welding seam.
2. The monoclinic laser vision sensing method for weld seam tracking of claim 1, characterized in that the formula of the two-dimensional median filter is:
wherein A is the window size; { fijIs a two-dimensional data sequence.
3. The monoclinic laser vision sensing method for weld seam tracking according to claim 1, characterized in that the binary function of the binarization process is:
where T is a predetermined threshold, white pixels are larger than the gray level T and black pixels are smaller than T.
4. The monoclinic laser vision sensing method for weld joint tracking according to claim 1, characterized in that the step of processing the binarized weld joint image by using an opening operation specifically comprises: the on operation is represented as a fitting process:
wherein {. } represents the union of all sets in the parenthesis.
5. The monoclinic laser vision sensing method for weld seam tracking according to claim 1, wherein the steps of calculating weld seam information according to the processed weld seam image and judging the spatial position of the weld seam specifically comprise: and generating a monoclinic line laser stripe through a laser striper, calculating the position relation between the current welding gun and the welding seam through the distribution information of the monoclinic line laser stripe, and guiding the welding gun to automatically track the welding seam.
6. The monoclinic laser vision sensing method for weld seam tracking according to claim 5, wherein the steps of calculating weld seam information according to the processed weld seam image and judging the spatial position of the weld seam specifically comprise: and calculating the width, the depth and the deviation correction quantity of the butt weld by a computer according to the image of the monoclinic laser stripe weld.
7. The utility model provides a monoclinic line laser vision sensing system for welding seam tracking, its characterized in that, the system includes CCD camera, combined filter, laser marker and work piece, the laser marker is inclined to the horizontal direction and is placed to produce oblique laser broken line stripe, and the laser stripe projection of transmission is on the work piece, gathers on the CCD camera through combined filter, obtains welding seam information and offset after image processing to judge the spatial position of welding seam.
8. The monoclinic laser vision sensing system for weld seam tracking according to claim 7, characterized in that the laser marker is obliquely placed at an angle of 20 ° on the rear side of the CCD camera with reference to the welding direction, and the angle between the laser marker and the vertical direction is 20 °.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111570543.3A CN114160926A (en) | 2021-12-21 | 2021-12-21 | Monoclinic line laser vision sensing method and system for welding seam tracking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111570543.3A CN114160926A (en) | 2021-12-21 | 2021-12-21 | Monoclinic line laser vision sensing method and system for welding seam tracking |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114160926A true CN114160926A (en) | 2022-03-11 |
Family
ID=80487647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111570543.3A Pending CN114160926A (en) | 2021-12-21 | 2021-12-21 | Monoclinic line laser vision sensing method and system for welding seam tracking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114160926A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5481085A (en) * | 1994-09-09 | 1996-01-02 | University Of Kentucky Research Foundation | Apparatus and method for measuring 3-D weld pool shape |
| CN1782659A (en) * | 2004-12-02 | 2006-06-07 | 中国科学院自动化研究所 | Welding seam tracking sight sensor based on laser structure light |
| CN102780845A (en) * | 2012-06-14 | 2012-11-14 | 清华大学 | Light source alternate strobe synchronous camera shooting method and vision detection system |
| CN107186319A (en) * | 2017-07-03 | 2017-09-22 | 江苏科技大学 | A kind of online tracking of welding robot cosmetic welding based on laser sensor |
| CN107824940A (en) * | 2017-12-07 | 2018-03-23 | 淮安信息职业技术学院 | Welding seam traking system and method based on laser structure light |
-
2021
- 2021-12-21 CN CN202111570543.3A patent/CN114160926A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5481085A (en) * | 1994-09-09 | 1996-01-02 | University Of Kentucky Research Foundation | Apparatus and method for measuring 3-D weld pool shape |
| CN1782659A (en) * | 2004-12-02 | 2006-06-07 | 中国科学院自动化研究所 | Welding seam tracking sight sensor based on laser structure light |
| CN102780845A (en) * | 2012-06-14 | 2012-11-14 | 清华大学 | Light source alternate strobe synchronous camera shooting method and vision detection system |
| CN107186319A (en) * | 2017-07-03 | 2017-09-22 | 江苏科技大学 | A kind of online tracking of welding robot cosmetic welding based on laser sensor |
| CN107824940A (en) * | 2017-12-07 | 2018-03-23 | 淮安信息职业技术学院 | Welding seam traking system and method based on laser structure light |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN210046133U (en) | Welding seam visual tracking system based on laser structured light | |
| CN105571502B (en) | The measurement method of weld gap in Friction Stir Welding | |
| CN110064819B (en) | Cylindrical surface longitudinal weld characteristic region extraction and weld tracking method and system based on structured light | |
| Dinham et al. | Autonomous weld seam identification and localisation using eye-in-hand stereo vision for robotic arc welding | |
| US8660349B2 (en) | Screen area detection method and screen area detection system | |
| CN105678776B (en) | Weld image feature point extraction method based on laser vision sensor | |
| CN110480127B (en) | Welding seam tracking system and method based on structured light vision sensing | |
| CN102175700B (en) | Method for detecting welding seam segmentation and defects of digital X-ray images | |
| CN112529858A (en) | Welding seam image processing method based on machine vision | |
| CN111127402A (en) | Visual detection method for welding quality of robot | |
| CN107424144A (en) | Weld joint tracking image processing algorithm based on laser vision | |
| CN114240944B (en) | Welding defect detection method based on point cloud information | |
| CN108907526A (en) | A kind of weld image characteristic recognition method with high robust | |
| Guo et al. | Weld deviation detection based on wide dynamic range vision sensor in MAG welding process | |
| CN105458462A (en) | Variable gap trapezoidal welding line multi-parameter synchronous visual inspection tracking method | |
| KR100677762B1 (en) | Machine vision apparatus and method for automatic data gathering do hull forming process | |
| Dinham et al. | Weld seam detection using computer vision for robotic arc welding | |
| CN111353993A (en) | Thread angle measuring method based on machine vision | |
| Wu et al. | Research on robust laser vision feature extraction method for fillet welds with different reflective materials under uncertain interference | |
| CN114066752B (en) | Line structure light skeleton extraction and burr removal method for weld tracking | |
| CN114309889B (en) | Laser vision sensing system and method for weld tracking | |
| CN114160926A (en) | Monoclinic line laser vision sensing method and system for welding seam tracking | |
| Naji et al. | Square groove detection based on Förstner with Canny edge operator using laser vision sensor | |
| Zhang et al. | A study on seam tracking in robotic GMA welding process | |
| KR101637977B1 (en) | Feature point detecting method of welding joint using laser vision system |
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 |