CN108788544A - A kind of weld seam starting point detecting method based on structured light vision sensor - Google Patents
A kind of weld seam starting point detecting method based on structured light vision sensor Download PDFInfo
- Publication number
- CN108788544A CN108788544A CN201810547763.6A CN201810547763A CN108788544A CN 108788544 A CN108788544 A CN 108788544A CN 201810547763 A CN201810547763 A CN 201810547763A CN 108788544 A CN108788544 A CN 108788544A
- Authority
- CN
- China
- Prior art keywords
- area
- vision sensor
- weld seam
- light vision
- structured light
- 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
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
Abstract
The invention discloses a kind of, and the weld seam based on structured light vision sensor originates point detecting method, including step：Robot motion to structured light vision sensor is started detection zone by step 1；Step 2, structured light vision sensor are along the direction of weld seam starting point with speed S1It is mobile and in real time in detection image in dynamic area-of-interest connected region quantity and area；Step 3, when detecting the quantity and sharply changing section of connected region in dynamic area-of-interest in image, negative direction driving structure light vision sensor is moved；Step 4, structured light vision sensor are moved in reverse direction, when in image in area-of-interest connected region quantity and area and start detection when state consistency when, then identify the initial point position of weld seam；Step 5 stops scanning motion, and detection terminates.The present invention keeps the detection of weld seam starting point quick and accurate, extends the application range of structured light vision sensor in such a way that short scan and slow scanning are combined.
Description
Technical field
The invention belongs to intelligent robot welding fields, are related to a kind of weldering of the structured light vision sensor for robot
Seam starting point detecting method.
Background technology
A critical issue is the positioning of workpiece in intelligent robot welding.Due to the foozle of workpiece to be welded and fixed
Position error, the simple consistency that can not ensure weldquality using robot welding, can not meet the quality requirement of production.Structure
Light vision sensor is increasingly used in machine due to the feature that its is untouchable, measurement accuracy is higher, measuring speed is fast
In people's welding field.Structured light vision sensor by acquire laser line image, image procossing and weld seam recognition and etc. after, will
Weld seam recognition point in image is converted to the weld seam spatial point under robot coordinate system, and robot controller is by controlling robot
The functions such as the weld joint tracking of robot are realized in end.For some open loop weld seams, need robot that the starting of welding is accurately positioned
It directly welding the phenomenon that causing to be partially welded, is detected and is welded using structured light vision sensor after position deviation occurs in workpiece in position
The starting welding position of seam very it is necessary to.
The detection of weld seam starting point there are two main classes detection mode：Detection mode based on passive vision and based on actively regarding
The detection mode of feel.Passive vision does not use laser rays, and is to rely on natural light, is easy by workpiece shapes and workpiece table
The influence of face quality, and need by the way of passive vision the mode for constructing stereoscopic vision that could obtain the space bits of spot welds
It sets, weld seam starting point is shot simultaneously so generally requiring two cameras, not only equipment occupation space bigger, but also shoot
Image need carry out characteristic point matching, precision be easy influenced by picture quality, processing time and precision are all relatively low, no
Meet the requirement of actual welding.By the way of active vision, such as structured light vision detection, as a result of laser rays, light intensity
Greatly, strong antijamming capability, it is only necessary to which the position of laser stripe in detection image does not have to carry out matching operation, and precision is higher, can
Meet the requirement of actual welding.
Invention content
The present invention proposes a kind of weld seam starting point detecting method based on structured light vision sensor, it is intended that carrying
Solved in intelligent robot welding process for a kind of simple and effective mode, workpiece to be welded due to geomery itself it is inconsistent with
And frock clamp positioning accuracy it is not high caused by weld seam initial point position deviation the case where.
The technical scheme is that：
A kind of weld seam starting point detecting method based on structured light vision sensor, includes the following steps：
Robot motion to structured light vision sensor is started detection zone by step 1, ensures structured light vision sensor
When starting detection, spot welds region is located at image intermediate region；
Step 2 and then robotically-driven structured light vision sensor are moved along the direction of weld seam starting point with speed S1, together
When structured light vision sensor detect the quantity and area of connected region in dynamic area-of-interest in real time；
Step 3, when detecting in image that the quantity of connected region and area mutate in dynamic area-of-interest,
Then structured light vision sensor has been detected by the starting point region of weld seam, and host computer sends a signal to robot at this time, control
Robot is moved according to negative direction driving structure light vision sensor；
Step 4, robot are according to speed S2Driving structure light vision sensor is moved in reverse direction, S2<S1, meanwhile,
Structured light vision sensor is in real time detected weld seam, when in image in area-of-interest connected region quantity and area with
When starting state consistency when detection, then the initial point position that weld seam is used as in current spot welds position is detected by weld seam recognition；
Step 5, host computer send a signal to robot, and robot is made to stop scanning motion, and detection terminates.
Further, in the step 1, after robot motion is started detection zone to structured light vision sensor, also
Including step：
Spot welds are identified by image procossing, set initial area-of-interest, and initial area-of-interest is rectangle region
Domain, rectangular centre are weld seam recognition point position.
Further, in step 1, the structured light vision sensor is connected on the robotic gun.
Further, the step 2 specifically includes：
Step 21 starts robot and structured light vision sensor, the robotically-driven structured light vision sensor edge
The direction of weld seam starting point with speed S1It is mobile；
Step 22, the structured light vision sensor acquire laser line image in real time in moving process；
Step 23 handles the laser stripe in acquired laser line image in dynamic area-of-interest and is felt
The quantity and area of connected region in interest region.
Further, in the step 22, the area-of-interest of each frame laser line image acquired be it is dynamic,
Center changes with the weld seam recognition point of previous frame laser line image, and the size of area-of-interest remains unchanged：
Wherein [Xmin,Xmax] and [Ymin,Ymax] indicate dynamic area-of-interest X and Y-direction range, Δ w and Δ h
It is the width and height of dynamic area-of-interest, (ui-1,vi-1) indicate previous frame image in weld seam recognition point coordinate.
Further, in the step 3, the quantity of connected region and area are sent out in dynamic area-of-interest in image
Life is mutated：
In weld seam starting point inside region, the quantity of the connected region in dynamic area-of-interest remains 1, and should
The area of connected region is also maintained at maximum value；When laser scanning is to weld seam starting point lateral area, dynamic region of interest
Connected region quantity in domain will not be 1, and connected region when the area in its largest connected region can also be significantly less than initial detecting
The area in domain.
Further, the step 4 specifically includes：
Step 41, the robot are according to speed S2Driving structure light vision sensor is moved in reverse direction, S2<S1；
Step 42, the structured light vision sensor acquire laser line image in real time in moving process；
Step 43 handles the laser stripe in acquired laser line image in dynamic area-of-interest and is felt
The quantity and area of connected region in interest region；
Step 44 becomes 1, and the area of its connected region when the connected region quantity in dynamic area-of-interest again
When being less than setting value with the difference of the area of connected region when initial detecting, robot location is preserved, and identify in laser line image
Spot welds position；
Step 45, by coordinate transform, the weld seam recognition point in laser line image is transformed under robot coordinate system
Spot welds, the position while welding are the initial point position as weld seam.
Further, at the laser stripe in acquired laser line image in dynamic area-of-interest
Reason obtains the quantity of connected region and area in area-of-interest and specifically includes：
Step 101 carries out image binaryzation processing to acquired laser line image：
Wherein, Iuv indicates that the image pixel value under (u, v) coordinate position, Ie are the pixel threshold of setting；
It after step 102, binary conversion treatment, is operated using closing operation of mathematical morphology, tiny segmentation is combined；
Step 103, the holes filling for carrying out region are 255 to the hole setting pixel value inside laser stripe region, protect
Demonstrate,prove the connection of interior zone；
Step 104, the label that connected region is carried out using 8 neighborhood descriptions, mark identical region as a whole,
Then the connected region quantity in dynamic area-of-interest and the area in largest connected region are calculated.
Further, in the step 4, the movement velocity S of robot2It is set as：
Further, in the step 3, the PC control robot is according to negative direction driving structure light vision
Before device is moved, the robot movement velocity is first reduced to 0.
Compared with prior art, the beneficial effects of the invention are as follows：
1, the present invention passes through coarse scanning and close scanning two in such a way that structured light vision sensor is scanned formula detection
A process analyzes the quantity of connected region in dynamic area-of-interest region in image according to T-type corner connection weld shape feature
And the situation of change of area, the initial point position of tack weld.
2, method of the present invention is simple and effective, in such a way that coarse scanning and close scanning combine, has both improved detection
Rapidity, have the precision that disclosure satisfy that detection, can solve workpiece to be welded it is inconsistent caused by welding starting point location it is inaccurate
True problem provides effective solution method for intelligent robot welding.
3, the starting point location of T-type corner connection weld seam may be implemented in the present invention, may be implemented to docking by being suitably modified
The starting point of type weld seam and lap jointing type weld seam detects.
Description of the drawings
Fig. 1 is that two kinds of starting points of T-type corner connection weld seam define schematic diagram.
Fig. 2 is weld seam starting point overhaul flow chart.
Fig. 3 is connected region change procedure schematic diagram.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below in conjunction with drawings and examples pair
The present invention is further elaborated.
Workpiece is due to foozle and the position error of frock clamp, and when leading to welding, the starting point of weld seam deviates reason
The position thought.The present invention carries out T-type fillet welding tailor's part using structured light vision sensor the detection of starting point, due to structure
Light vision sensor can only export a spot welds position per frame image, therefore the detection of starting point is needed to use scan-type
Detection mode.
Be the starting dot shape of typical T-type corner connection weld seam shown in Fig. 1 leads to laser irradiation due to the difference of its structure
When to weld seam starting point near zone, the laser stripe shape in image is also different.Fig. 2 is weld seam starting point overhaul flow chart,
Whole flow process includes coarse scanning and close scanning process, and the guarantee of coarse scanning process is quickly found out weld seam starting point region, and close scanning makes
It is more accurate to obtain weld seam starting point location.
As shown in Fig. 2, a kind of weld seam based on structured light vision sensor originates point detecting method, include the following steps：
Robot motion to structured light vision sensor is started detection zone by step 1, ensures structured light vision sensor
When starting detection, spot welds region is located at image intermediate region, and the structured light vision sensor is connected in the robot
On welding gun；
Step 2 and then robotically-driven structured light vision sensor are along the direction of weld seam starting point with speed S1It is mobile, together
When structured light vision sensor detect the quantity and area of dynamic area-of-interest (ROI) interior connected region in real time；
Step 3, when detecting in image that the quantity of connected region and area mutate in dynamic area-of-interest,
Then structured light vision sensor has been detected by the starting point region of weld seam, and host computer sends a signal to robot at this time, control
Robot is moved according to negative direction driving structure light vision sensor, and since robot motion has inertia, robot is not
Meeting counter motion immediately, therefore, the PC control robot is transported according to negative direction driving structure light vision sensor
Before dynamic, the robot movement velocity is first reduced to 0；
Step 4, robot are according to speed S2Driving structure light vision sensor is moved in reverse direction, S2<S1, meanwhile,
Structured light vision sensor is in real time detected weld seam, when in image in area-of-interest connected region quantity and area with
When starting state consistency when detection, then the initial point position that weld seam is used as in current spot welds position is detected by weld seam recognition；
Step 5, host computer send a signal to robot, and robot is made to stop scanning motion, and detection terminates.
Specifically, in the step 1, after robot motion is started detection zone to structured light vision sensor, also
Including step：
Spot welds are identified by image procossing, set initial area-of-interest, and initial area-of-interest is rectangle region
Domain, rectangular centre are weld seam recognition point position.
Specifically, in step 1, the structured light vision sensor is connected on the robotic gun.
Specifically, the step 2 specifically includes：
Step 21 starts robot and structured light vision sensor, the robotically-driven structured light vision sensor edge
The direction of weld seam starting point with speed S1It is mobile；
Step 22, the structured light vision sensor acquire laser line image in real time in moving process；
Step 23 handles the laser stripe in acquired laser line image in dynamic area-of-interest and is felt
The quantity and area of connected region in interest region.
Specifically, in the step 22, the area-of-interest of each frame laser line image acquired be it is dynamic,
Center changes with the weld seam recognition point of previous frame laser line image, and the size of area-of-interest remains unchanged：
Wherein [Xmin,Xmax] and [Ymin,Ymax] indicate area-of-interest X and Y-direction range, Δ w and Δ h are dynamics
Area-of-interest width and height, (ui-1,vi-1) indicate previous frame image in weld seam recognition point coordinate.
Specifically, in the step 3, the quantity of connected region and area are sent out in dynamic area-of-interest in image
Life is mutated：
In weld seam starting point inside region, the quantity of the connected region in dynamic area-of-interest remains 1, and should
The area of connected region is also maintained at maximum value；When laser scanning is to weld seam starting point lateral area, dynamic region of interest
Connected region quantity in domain will not be 1, and connected region when the area in its largest connected region can also be significantly less than initial detecting
The area in domain.
Specifically, the step 4 specifically includes：
Step 41, the robot are according to speed S2Driving structure light vision sensor is moved in reverse direction, S2<S1；
Step 42, the structured light vision sensor acquire laser line image in real time in moving process；
Step 43 handles the laser stripe in acquired laser line image in dynamic area-of-interest and is felt
The quantity and area of connected region in interest region；
Step 44 becomes 1, and the area of its connected region when the connected region quantity in dynamic area-of-interest again
When being less than setting value with the difference of the area of connected region when initial detecting, robot location is preserved, and identify in laser line image
Spot welds position；
Step 45, by coordinate transform, the weld seam recognition point in laser line image is transformed under robot coordinate system
Spot welds, the position while welding are the initial point position as weld seam.
Specifically, in order to identify the mutation of laser stripe, it is described to dynamically feeling emerging in acquired laser line image
Laser stripe in interesting region is handled to obtain the quantity of connected region and area in area-of-interest and is specifically included：
Step 101 carries out image binaryzation processing to acquired laser line image：
Wherein, Iuv indicates that the image pixel value under (u, v) coordinate position, Ie are the pixel threshold of setting；
It after step 102, binary conversion treatment, is operated using closing operation of mathematical morphology, tiny segmentation is combined；
Step 103, the holes filling for carrying out region are 255 to the hole setting pixel value inside laser stripe region, protect
Demonstrate,prove the connection of interior zone；
Step 104, the label that connected region is carried out using 8 neighborhood descriptions, mark identical region as a whole,
Then the area of the connected region quantity and largest connected region in dynamic area-of-interest is calculated.
In the robot according to speed S2When driving structure light vision sensor is moved in reverse direction, speed S1
Meet：
Under low-speed situations, the precision of structured light vision sensor test point can be improved.
Fig. 3 is reflected in weld seam starting point detection process, the connected region in different moments image in area-of-interest
Situation of change：
Fig. 3 (a) moment laser rays is located on the inside of weld seam；Fig. 3 (b) moment detects that connected region is mutated, Fig. 3 (c) moment
Due to robot inertia, division mouth continues to expand, and Fig. 3 (d) moment connected regions recombine into an entirety；Quickly scanning rank
Section, connected region is from entirety to division, the slow scanning stage, and connected region is from splitting into entirety.By identifying connected region
Situation of change completes the positioning of weld seam starting point.
The object, technical solutions and advantages of the present invention are further described in detail in present embodiment, should say
Bright, the foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all spirit in the present invention
Within principle, made by all any modification, equivalent and improvement etc., should be included within the scope of the present invention.
Claims (10)
1. a kind of weld seam based on structured light vision sensor originates point detecting method, which is characterized in that include the following steps：
Robot motion to structured light vision sensor is started detection zone by step 1, ensures that structured light vision sensor is being opened
Spot welds region is located at image intermediate region when beginning to detect；
Step 2 and then robotically-driven structured light vision sensor are along the direction of weld seam starting point with speed S1It is mobile, it ties simultaneously
Structure light vision sensor detects the quantity and area of connected region in dynamic area-of-interest in real time；
Step 3, when detecting in image that the quantity of connected region and area mutate in dynamic area-of-interest, then tie
Structure light vision sensor has been detected by the starting point region of weld seam, and host computer sends a signal to robot at this time, controls machine
People moves according to negative direction driving structure light vision sensor；
Step 4, robot are according to speed S2Driving structure light vision sensor is moved in reverse direction, S2<S1, meanwhile, structure
Light vision sensor is in real time detected weld seam, when in image in area-of-interest connected region quantity and area and beginning
When detection when state consistency, then the initial point position that weld seam is used as in current spot welds position is detected by weld seam recognition；
Step 5, host computer send a signal to robot, and robot is made to stop scanning motion, and detection terminates.
2. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, which is characterized in that
Further include step after robot motion is started detection zone to structured light vision sensor in the step 1：
Spot welds are identified by image procossing, set initial area-of-interest, and initial area-of-interest is rectangular area, square
Shape center is weld seam recognition point position.
3. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, which is characterized in that
In step 1, the structured light vision sensor is connected on the robotic gun.
4. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, which is characterized in that
The step 2 specifically includes：
Step 21 starts robot and structured light vision sensor, and the robotically-driven structured light vision sensor is along weldering
The direction of starting point is stitched with speed S1It is mobile；
Step 22, the structured light vision sensor acquire laser line image in real time in moving process；
Step 23, the laser stripe in acquired laser line image in dynamic area-of-interest is handled to obtain it is interested
The quantity and area of connected region in region.
5. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, which is characterized in that
In the step 22, the area-of-interest of each frame laser line image acquired is dynamic, and center is swashed with previous frame
The weld seam recognition point of light image changes, and the size of area-of-interest remains unchanged：
Wherein [Xmin,Xmax] and [Ymin,Ymax] indicate dynamic area-of-interest X and Y-direction range, Δ w and Δ h are
The width and height of the area-of-interest of state, (ui-1,vi-1) indicate previous frame image in weld seam recognition point coordinate.
6. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, which is characterized in that
In the step 3, the quantity of connected region and area mutation are specially in dynamic area-of-interest in image：
In weld seam starting point inside region, the quantity of the connected region in dynamic area-of-interest remains 1, and the connection
The area in region is also maintained at maximum value；When laser scanning is to weld seam starting point lateral area, in dynamic area-of-interest
Connected region quantity will not be 1, and connected region when the area in its largest connected region can also be significantly less than initial detecting
Area.
7. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, which is characterized in that
The step 4 specifically includes：
Step 41, the robot are according to speed S2Driving structure light vision sensor is moved in reverse direction, S2<S1；
Step 42, the structured light vision sensor acquire laser line image in real time in moving process；
Step 43, the laser stripe in acquired laser line image in dynamic area-of-interest is handled to obtain it is interested
The quantity and area of connected region in region；
Step 44 becomes 1 when the connected region quantity in dynamic area-of-interest again, and the area of its connected region with just
When the difference of the area of connected region is less than setting value when beginning to detect, robot location is preserved, and identify the weldering in laser line image
Seam point position；
Step 45, by coordinate transform, the weld seam recognition point in laser line image is transformed into the weld seam under robot coordinate system
Point, the position while welding are the initial point position as weld seam.
8. the weld seam based on structured light vision sensor according to claim 4 or 7 originates point detecting method, feature exists
In the laser stripe in acquired laser line image in dynamic area-of-interest is handled to obtain region of interest
The quantity of connected region and area specifically include in domain：
Step 101 carries out image binaryzation processing to acquired laser line image：
Wherein, Iuv indicates that the image pixel value under (u, v) coordinate position, Ie are the pixel threshold of setting；
It after step 102, binary conversion treatment, is operated using closing operation of mathematical morphology, tiny segmentation is combined；
Step 103, the holes filling for carrying out region are 255 to the hole setting pixel value inside laser stripe region, in guarantee
The connection in portion region；
Step 104, the label that connected region is carried out using 8 neighborhood descriptions, mark identical region as a whole, then
Calculate the connected region quantity in dynamic area-of-interest and the area in largest connected region.
9. the weld seam according to claim 1 or claim 7 based on structured light vision sensor originates point detecting method, feature exists
In, in the step 4, the movement velocity S of robot2It is set as：
10. the weld seam according to claim 1 based on structured light vision sensor originates point detecting method, feature exists
In in the step 3, the PC control robot carries out moving it according to negative direction driving structure light vision sensor
Before, the robot movement velocity is first reduced to 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810547763.6A CN108788544B (en) | 2018-05-31 | 2018-05-31 | Welding seam initial point detection method based on structured light vision sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810547763.6A CN108788544B (en) | 2018-05-31 | 2018-05-31 | Welding seam initial point detection method based on structured light vision sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108788544A true CN108788544A (en) | 2018-11-13 |
| CN108788544B CN108788544B (en) | 2020-09-22 |
Family
ID=64089720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810547763.6A Active CN108788544B (en) | 2018-05-31 | 2018-05-31 | Welding seam initial point detection method based on structured light vision sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108788544B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110375675A (en) * | 2019-08-30 | 2019-10-25 | 易思维（杭州）科技有限公司 | Binocular optical grating projection measurement method based on space phase expansion |
| CN112122775A (en) * | 2020-09-22 | 2020-12-25 | 成都卡诺普自动化控制技术有限公司 | Transformer oil tank welding line and cooling fin identification method based on laser welding line tracking |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105844655A (en) * | 2016-04-19 | 2016-08-10 | 南京工程学院 | Laser line stripe edge extraction method |
| CN105957023A (en) * | 2016-04-19 | 2016-09-21 | 南京工程学院 | Laser stripe image reinforcing and de-noising method based on color space conversion |
-
2018
- 2018-05-31 CN CN201810547763.6A patent/CN108788544B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105844655A (en) * | 2016-04-19 | 2016-08-10 | 南京工程学院 | Laser line stripe edge extraction method |
| CN105957023A (en) * | 2016-04-19 | 2016-09-21 | 南京工程学院 | Laser stripe image reinforcing and de-noising method based on color space conversion |
Non-Patent Citations (2)
| Title |
|---|
| NIANFENG WANG ET AL.: "Recognition of Initial Welding Position Based on Structured-Light for Arc Welding Robot", 《INTELLIGENT ROBOTICS AND APPLICATIONS （PART Ⅱ）》 * |
| 石小东: "基于结构光视觉传感的弧焊机器人焊缝纠偏系统的设计与研究", 《网络资源：HTTPS://KNS.CNKI.NET/KNS/BRIEF/DEFAULT_RESULT.ASPX》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110375675A (en) * | 2019-08-30 | 2019-10-25 | 易思维（杭州）科技有限公司 | Binocular optical grating projection measurement method based on space phase expansion |
| CN112122775A (en) * | 2020-09-22 | 2020-12-25 | 成都卡诺普自动化控制技术有限公司 | Transformer oil tank welding line and cooling fin identification method based on laser welding line tracking |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108788544B (en) | 2020-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lü et al. | Feature extraction of welding seam image based on laser vision | |
| Lei et al. | A review of vision-aided robotic welding | |
| US20200269340A1 (en) | Active Laser Vision Robust Weld Tracking System and Weld Position Detection Method | |
| CN108907526A (en) | A kind of weld image characteristic recognition method with high robust | |
| CN108817613A (en) | A kind of arc welding robot weld seam deviation-rectifying system and method | |
| Luo et al. | Laser visual sensing for seam tracking in robotic arc welding of titanium alloys | |
| CN109702293B (en) | Welding penetration quality real-time control method based on visual detection | |
| Liu et al. | Precise initial weld position identification of a fillet weld seam using laser vision technology | |
| JPS60231208A (en) | Control method of automatic machine | |
| CN108344693B (en) | Automatic welding-oriented visual measurement method for misalignment of sheet welding seam | |
| CN102284769A (en) | System and method for initial welding position identification of robot based on monocular vision sensing | |
| CN108788544A (en) | A kind of weld seam starting point detecting method based on structured light vision sensor | |
| Chen et al. | Automatic weld type classification, tacked spot recognition and weld ROI determination for robotic welding based on modified YOLOv5 | |
| CN112238292A (en) | Method for tracking space curve track of friction stir welding robot based on vision | |
| JPH0798214A (en) | Method and device for three dimensional position and attitude recognition method based on sense of sight | |
| KR20170142379A (en) | Apparatus for detect dimensional welding line of welding robot using image processing | |
| CN109671059A (en) | A kind of battery case image processing method and system based on OpenCV | |
| CN112620926A (en) | Welding spot tracking method and device and storage medium | |
| CN104835156B (en) | A kind of non-woven bag automatic positioning method based on computer vision | |
| CN206952357U (en) | Weld seam tracking sensor | |
| WO2017045303A1 (en) | Weld seam testing method | |
| CN205129209U (en) | Automatically, seek mark welding system based on robot welding | |
| Li et al. | Narrow weld joint recognition method based on laser profile sensor | |
| Huang et al. | Weld line detection using realsense depth camera based on depth map | |
| JPH091338A (en) | Method and device for automatically recognizing weld seam |
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 |