CN109934820A - Linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image - Google Patents
Linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image Download PDFInfo
- Publication number
- CN109934820A CN109934820A CN201910222868.9A CN201910222868A CN109934820A CN 109934820 A CN109934820 A CN 109934820A CN 201910222868 A CN201910222868 A CN 201910222868A CN 109934820 A CN109934820 A CN 109934820A
- Authority
- CN
- China
- Prior art keywords
- pixel
- sub
- line
- point
- edge point
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses linear edge sub-pixel detection methods in a kind of laser assembly solder part to be welded image, specifically include: choosing a pixel edge point first is that origin establishes coordinate system, and pixel in 7 ╳, 7 range is taken centered on origin；Grey scale centre of gravity is sought according to the sum of the grayscale values coordinate of each pixel, and the line of pixel edge dot center and center of gravity is the normal of ideal line；Then pixel in 7 ╳, 7 range is divided into two regions with ideal line, gray value is identical in each region；The parameter of straight line is acquired according to the gray value of each pixel in 7 region of the normal of ideal line and 7 ╳；The intersection point of last normal direction and ideal line is sub-pixel edge point position；It successively chooses each pixel edge point and repeats above procedure, the corresponding sub-pixel marginal point of each pixel edge point can be acquired.This method has model simple, and calculating process stablizes quick advantage.
Description
Technical field
The invention belongs to field of visual inspection, linear edge sub-pix in specifically a kind of laser assembly solder part to be welded image
Detection method.
Background technique
Laser welding is as a kind of connection metalwork technology with active area is small, energy density is high, welding efficiency is high
Feature is applied in the fields such as automobile, aerospace, shipbuilding.Currently, laser welding head is mounted in industrial production
On multiple degrees of freedom robotic weld arm or numerical control bonding machine bed, plumb joint realizes welding by setting track is mobile when welding.By
The influence of the factors such as error is clamped in fixture, part to be welded clamps rear welding position and desired guiding trajectory has deviation.When laser welding
The laser beam zone of action is smaller, and this deviation will affect welding quality, or even cause weld failure.To eliminate the practical weldering of part to be welded
It stitches position and sets the deviation of welding track, need to plan welding track again before welding.The welding track of part to be welded can answer
It is determined with visible sensation method by detecting the edge of weld seam.The present invention is mainly sub- to linear edge in laser assembly solder part to be welded image
Pixel detecting method is studied, and provides technical support for welding track planning.
Bo Zhong Seiko Science and Technology Co., Ltd. Wu Xiaojun et al., in number of patent application are as follows: 201610209158.9, patent
Adaptive high-low threshold value calculation method is used in a kind of entitled image sub-pixel edge extracting method with extensive adaptability
Gradient image is obtained, using any location of pixels as origin, establishes relative coordinate, taking eight neighborhood pixel around the point is local maximum
Central value selects data sample, obtains the comparison result of neighborhood according to gradient direction, determines whether current pixel position is boundary
Point position candidate reaches the progress edge inspection of this phenomenon of maximum value in marginal point to adjoint point intensity-weighted value using each of pixel
It surveys.If it is marginal point that the extreme value of partial gradient amplitude, which is greater than given threshold,.Using based on Steger curved surface fitting method
Hessian matrix method seeks the sub-pixel location of marginal point.The above method can complete sub-pixel edge inspection under given conditions
It surveys, but there is a problem of that calculating process is complex, computational efficiency is not high.
Summary of the invention
The application provides linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image, and this method has mould
Type is simple, and calculating process stablizes the advantages that quick.
To achieve the above object, the technical solution of the application are as follows: linear edge is sub- in a kind of laser assembly solder part to be welded image
Pixel detecting method specifically includes: choosing a pixel edge point first as origin and establishes coordinate system, and centered on origin
Take pixel in 7 ╳, 7 range；Seek grey scale centre of gravity coordinate according to the sum of the grayscale values coordinate of each pixel, pixel edge dot center with again
The coordinate line of the heart is the normal of ideal line；Then selected 7 region 7 ╳ is divided into two regions with ideal line, each
Gray value is identical in region；The parameter of straight line is acquired according to the gray value of each pixel in 7 region of the normal of ideal line and 7 ╳；
The intersection point of normal direction and ideal line is finally sought, which is sub-pixel edge point position；Successively choose each pixel edge
Above procedure is put and repeated, the corresponding sub-pixel edge point position of each pixel edge point is acquired.
Further, grey scale centre of gravity coordinate is sought method particularly includes: after part to be welded image gaussian filtering, using Sobel
Operator seeks the pixel edge point of part to be welded in image.Take Pixel-level marginal point (ik,jk), with (ik,jk) center be that coordinate is former
Point is that ordinate establishes coordinate system using horizontal direction as abscissa, vertical direction.With (ik,jk) centered on take picture in 7 ╳, 7 range
Element, the gray value f (i of each pixelk,jk) indicate, then grey scale centre of gravity coordinate (ikh,jkh) are as follows:
Further, seek ideal line slope: the straight line for crossing sub-pixel edge point is ideal line, the normal of ideal line
For pixel edge point (ik,jk) center and grey scale centre of gravity (ikh,jkh) line.Ideal line lkIt indicates are as follows:
Y=ax+b
Ideal line l in formulakSlope a can be acquired by normal.
Further, sub-pixel edge point calculating method are as follows: ideal line will be with pixel edge point (ik,jk) be
Pixel is divided into two regions in 7 ╳, 7 range of the heart, and gray value is identical in region；Gray value uses W respectively in two side areas1And W2Table
Show；W1And W2Value adjacent area in grey scale pixel value calculate；When grey scale centre of gravity is located at two, four-quadrant, calculation formula is such as
Under:
When grey scale centre of gravity is located at one, three quadrant, calculation formula is as follows:
Single pixel is square its unilateral length and is expressed as L, enables L=1, with pixel edge point (ik,jk) centered on
The gray value S of each column pixel in 7 ╳, 7 range1、S2、S3、S4、S5、S6And S7It indicates；Ideal line lkUnder from left to right respectively arrange
Area K1、K2、K3、K4、K5、K6And K7It indicates；Then:
With pixel edge point (ik,jk) centered on 7 ╳, 7 range in ideal line lkUnder area indicated with K, then:
It can thus be concluded that (L=1):
In formula, S=S1+S2+S3+S4+S5+S6+S7；
Ideal line lkIntersection point with its normal is sub-pixel edge point.
The present invention due to using the technology described above, can obtain following technical effect: by part to be welded curve in image
The sub-pixel edge point solution of weld seam is converted into using pixel edge point as the solution of ideal line in the rectangular pixel area of center.
Ideal line is divided to rectangular pixel area to be the identical part of two interior intensity values, utilizes grey scale centre of gravity in rectangular area and pixel
Gray value seeks the parameter of ideal line, and the intersection point of ideal line and its normal is sub-pixel edge.This method has model simple,
Calculating process stablizes quick advantage.
Detailed description of the invention
Fig. 1 is with (ik,jk) be 7 ╳ of center, 7 pixel region and ideal line figure (in figure grey scale centre of gravity be located at second as
Limit)；
Fig. 2 is sub-pixel edge calculation flow chart.
Specific embodiment
The present invention is described in further detail in the following with reference to the drawings and specific embodiments: doing as example to the application
Further description explanation.
Embodiment 1
The present embodiment provides linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image, used in this method
The plumb joint for projecting laser is mounted on numerically-controlled machine tool or multi-freedom robot arm.Industrial CCD camera is mounted on plumb joint
On.Pixel edge point (the i at part to be welded edge in image is acquired using Soble edge detection algorithmk,jk), k=1,2 ... n (n
For the number of pixel edge point).Take Pixel-level marginal point (ik,jk), seek the specific steps of its corresponding sub-pixel edge point such as
Under:
Step 1: calculating grey scale centre of gravity coordinate
With pixel edge point (ik,jk) centered on take pixel in 7 ╳, 7 range, with pixel edge point (ik,jk) in
The heart is coordinate origin O, establishes coordinate system as x-axis, by y-axis of vertical direction using horizontal direction, as shown in Figure 1.The ash of each pixel
Angle value f (ik,jk) indicate, then grey scale centre of gravity coordinate (ikh,jkh) are as follows:
Step 2: calculating the slope of ideal line
The straight line for crossing sub-pixel edge point is ideal line, and the normal of ideal line is pixel edge point (ik,jk) in
Heart O and grey scale centre of gravity (ikh,jkh) line.Ideal line lkIt indicates are as follows:
Y=ax+b (2)
Ideal line l in formulakSlope a acquired by normal.
Step 3: calculating sub-pixel edge point position
As shown in Figure 1, ideal line is with pixel edge point (ik,jk) it is that pixel is divided into the area Liang Ge in 7 ╳ of center, 7 range
Gray value is identical in the region of domain；Gray value uses W respectively in two side areas1And W2It indicates；W1And W2Value can use picture in adjacent area
Plain gray value calculates.When grey scale centre of gravity is located at two, four-quadrant, calculation formula is as follows:
When grey scale centre of gravity is located at one, three quadrant:
Single pixel is square its unilateral length and is expressed as L (can enable L=1), from left to right each column picture in 7 ╳, 7 range
The gray value S of element1、S2、S3、S4、S5、S6And S7It indicates；Ideal line lkUnder the area K that from left to right respectively arranges1、K2、K3、
K4、K5、K6And K7It indicates.Then:
Ideal line lkUnder area indicated with K, then:
(L=1) is obtained by formula (7) and formula (8):
In formula, S=S1+S2+S3+S4+S5+S6+S7。
Ideal line lkIntersection point with its normal is sub-pixel edge point.
The flow chart of sub-pixel edge point is calculated, as shown in Figure 2.Each pixel edge point is successively chosen, the above step is repeated
Suddenly the corresponding sub-pixel edge position of each pixel edge point can be obtained.
The preferable specific embodiment of the above, only the invention, but the protection scope of the invention is not
It is confined to this, anyone skilled in the art is in the technical scope that the invention discloses, according to the present invention
The technical solution of creation and its inventive concept are subject to equivalent substitution or change, should all cover the invention protection scope it
It is interior.
Claims (4)
1. linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image, which is characterized in that specifically include: first
Choosing a pixel edge point is that origin establishes coordinate system, and pixel in 7 ╳, 7 range is taken centered on origin；According to each picture
The sum of the grayscale values coordinate of element seeks grey scale centre of gravity coordinate, and the coordinate line of pixel edge dot center and center of gravity is the method for ideal line
Line；Then selected 7 region 7 ╳ is divided into two regions with ideal line, gray value is identical in each region；According to ideal straight
The gray value of each pixel acquires the parameter of straight line in 7 region of the normal of line and 7 ╳；Finally ask the friendship of normal direction and ideal line
Point, the intersection point are sub-pixel edge point position；It successively chooses each pixel edge point and repeats above procedure, acquire each Pixel-level
The corresponding sub-pixel edge point position of marginal point.
2. linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image according to claim 1, feature
It is, seeks grey scale centre of gravity coordinate method particularly includes: takes Pixel-level marginal point (ik,jk), with (ik,jk) center be that coordinate is former
Point is that ordinate establishes coordinate system using horizontal direction as abscissa, vertical direction；With pixel edge point (ik,jk) centered on take
Pixel in 7 ╳, 7 range, the gray value f (i of each pixelk,jk) indicate, then grey scale centre of gravity coordinate (ikh,jkh) are as follows:
3. linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image according to claim 1, feature
Be, seek the slope of ideal line: the straight line for crossing sub-pixel edge point is ideal line, and the normal of ideal line is Pixel-level side
Edge point (ik,jk) center and grey scale centre of gravity (ikh,jkh) line；Ideal line lkIt indicates are as follows:
Y=ax+b
Ideal line l in formulakSlope a can be acquired by normal.
4. linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image according to claim 1, feature
It is, the calculation method of sub-pixel edge point are as follows: ideal line will be with pixel edge point (ik,jk) centered on 7 ╳, 7 range
Interior pixel is divided into two regions, and gray value is identical in region；Gray value uses W respectively in two side areas1And W2It indicates；W1And W2's
Grey scale pixel value calculates in value adjacent area；When grey scale centre of gravity is located at two, four-quadrant, calculation formula is as follows:
When grey scale centre of gravity is located at one, three quadrant, calculation formula is as follows:
Single pixel is square its unilateral length and is expressed as L, enables L=1, with pixel edge point (ik,jk) centered on 7 ╳ 7
The gray value S of each column pixel in range1、S2、S3、S4、S5、S6And S7It indicates；Ideal line lkUnder from left to right each column area use
K1、K2、K3、K4、K5、K6And K7It indicates；Then:
With pixel edge point (ik,jk) centered on 7 ╳, 7 range in ideal line lkUnder area indicated with K, then:
It can thus be concluded that (L=1):
In formula, S=S1+S2+S3+S4+S5+S6+S7；
Ideal line lkIntersection point with its normal is sub-pixel edge point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910222868.9A CN109934820B (en) | 2019-03-22 | 2019-03-22 | Method for detecting linear edge sub-pixels in laser tailor-welded workpiece image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910222868.9A CN109934820B (en) | 2019-03-22 | 2019-03-22 | Method for detecting linear edge sub-pixels in laser tailor-welded workpiece image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109934820A true CN109934820A (en) | 2019-06-25 |
| CN109934820B CN109934820B (en) | 2020-10-09 |
Family
ID=66988008
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910222868.9A Active CN109934820B (en) | 2019-03-22 | 2019-03-22 | Method for detecting linear edge sub-pixels in laser tailor-welded workpiece image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109934820B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717502A (en) * | 2019-10-22 | 2020-01-21 | 大连大学 | Molten pool identification method in pulse laser tailor-welded image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102915535A (en) * | 2012-08-23 | 2013-02-06 | 深圳大学 | Method and system for correcting circle center deviation of round mark points during camera projection transformation |
| CN103136758A (en) * | 2013-03-25 | 2013-06-05 | 长春工业大学 | Rapid edge detecting method based on orthogonal polynomial fitting |
| CN103400399A (en) * | 2013-08-07 | 2013-11-20 | 长春工业大学 | Spatial moment based line structured light center extraction method |
| CN105761256A (en) * | 2016-02-05 | 2016-07-13 | 成都康烨科技有限公司 | Image sub-pixel edge line acquisition method and device |
| CN107301636A (en) * | 2017-05-17 | 2017-10-27 | 华南理工大学 | A kind of high density circuit board circular hole sub-pixel detection method based on Gauss curve fitting |
-
2019
- 2019-03-22 CN CN201910222868.9A patent/CN109934820B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102915535A (en) * | 2012-08-23 | 2013-02-06 | 深圳大学 | Method and system for correcting circle center deviation of round mark points during camera projection transformation |
| CN103136758A (en) * | 2013-03-25 | 2013-06-05 | 长春工业大学 | Rapid edge detecting method based on orthogonal polynomial fitting |
| CN103400399A (en) * | 2013-08-07 | 2013-11-20 | 长春工业大学 | Spatial moment based line structured light center extraction method |
| CN105761256A (en) * | 2016-02-05 | 2016-07-13 | 成都康烨科技有限公司 | Image sub-pixel edge line acquisition method and device |
| CN107301636A (en) * | 2017-05-17 | 2017-10-27 | 华南理工大学 | A kind of high density circuit board circular hole sub-pixel detection method based on Gauss curve fitting |
Non-Patent Citations (2)
| Title |
|---|
| GAO XIANG DONG 等: "Detection of micro gap weld using magneto-optical imaging during laser welding", 《THE INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY》 * |
| 王邦国 等: "焊接激光致熔池亚像素边缘提取方法研究", 《大连交通大学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717502A (en) * | 2019-10-22 | 2020-01-21 | 大连大学 | Molten pool identification method in pulse laser tailor-welded image |
| CN110717502B (en) * | 2019-10-22 | 2023-09-05 | 大连大学 | Method for identifying molten pool in pulse laser tailor-welded images |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109934820B (en) | 2020-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lei et al. | A review of vision-aided robotic welding | |
| Xu et al. | Visual sensing technologies in robotic welding: Recent research developments and future interests | |
| EP3863791B1 (en) | System and method for weld path generation | |
| CN106181162B (en) | A kind of real-time weld joint tracking detection method based on machine vision | |
| CN103759648B (en) | A kind of complicated angle welding method for detecting position based on Binocular stereo vision with laser | |
| CN109903279B (en) | Automatic teaching method and device for welding seam movement track | |
| Tsai et al. | Machine vision based path planning for a robotic golf club head welding system | |
| CN105598600B (en) | A kind of box part weld seam independently seeks position and track automatic generation method | |
| CN103231162A (en) | Device and method for visual detection of welding quality of robot | |
| CN112365439B (en) | Method for synchronously detecting forming characteristics of GMAW welding seam of galvanized steel and direction of welding gun in real time | |
| CN110475649B (en) | Operation program correction method, structure assembly method, medium, and welding robot system | |
| CN110431498B (en) | Method for acquiring weld bead information and welding robot system | |
| Mohd Shah et al. | Autonomous detection and identification of weld seam path shape position | |
| Liu et al. | Welding seam recognition and tracking for a novel mobile welding robot based on multi-layer sensing strategy | |
| CN108032011A (en) | Initial point guiding device and method are stitched based on laser structure flush weld | |
| Yang et al. | Multi-pass path planning for thick plate by DSAW based on vision sensor | |
| CN110831718A (en) | Apparatus and method for automatic seam welding of workpieces comprising a base plate with a pattern of upright profiles | |
| CN108098134A (en) | A kind of new pattern laser vision weld joint tracking system and method | |
| CN109934820A (en) | Linear edge sub-pixel detection method in a kind of laser assembly solder part to be welded image | |
| CN113385869B (en) | Robot welding equipment for large square lattice component based on machine vision and welding seam positioning method | |
| Yu et al. | Unified seam tracking algorithm via three-point weld representation for autonomous robotic welding | |
| Kiddee et al. | Visual recognition of the initial and end points of lap joint for welding robots | |
| Dinham et al. | Experiments on automatic seam detection for a MIG welding robot | |
| CN105606040B (en) | Large-scale liquefied natural gas module attitude detecting method based on three-dimensional laser detection | |
| CN108067725A (en) | A kind of new robotic laser vision weld joint detecting system and method |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211215 Address after: 116600 706d-50, free trade building, No. 1, Dongting Road, Dalian Free Trade Zone, Liaoning Province Patentee after: Dalian Ruitong Machinery Industry Technology Co.,Ltd. Address before: 116622 No. 10, Xuefu Avenue, Dalian economic and Technological Development Zone, Liaoning Patentee before: DALIAN University |
|
| TR01 | Transfer of patent right |