CN107160046B - A kind of camera calibration method of vision auxiliary large format lathe plate cutting - Google Patents
A kind of camera calibration method of vision auxiliary large format lathe plate cutting Download PDFInfo
- Publication number
- CN107160046B CN107160046B CN201710253805.0A CN201710253805A CN107160046B CN 107160046 B CN107160046 B CN 107160046B CN 201710253805 A CN201710253805 A CN 201710253805A CN 107160046 B CN107160046 B CN 107160046B
- Authority
- CN
- China
- Prior art keywords
- camera
- plate
- machine coordinates
- lathe
- laser cutting
- 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.)
- Active
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
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/36—Removing material
- B23K26/38—Removing material by boring or cutting
-
- 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
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/08—Devices involving relative movement between laser beam and workpiece
- B23K26/082—Scanning systems, i.e. devices involving movement of the laser beam relative to the laser head
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Abstract
The present invention relates to laser cutting field, specifically a kind of camera calibration method of vision auxiliary large format lathe plate cutting.Include the following steps: 1) step 1, obtains source figure from camera；2) step 2, according to camera distortion calibration as a result, carrying out the distortion correction of image；3) step 3 puts the plate of at least two pieces different-thickness respectively on laser cutting machine tool, records the thickness value of plate, and chooses the pixel coordinate and machine coordinates of at least four mark points respectively to every block of plate；4) step 4 selects the plate thickness actually cut；5) step 5, the machine coordinates shown by host computer record the machine coordinates of camera at this time；6) step 6 is calculated by perspective transform, obtains perspective transformation matrix M.The present invention compared with the existing technology, allows operator without grasping a large amount of professional knowledge, it is only necessary to which the calibration of camera and lathe can be completed in the data for acquiring several groups of pixel coordinates and machine coordinates.
Description
Technical field
The present invention relates to laser cutting field, specifically a kind of camera of vision auxiliary large format lathe plate cutting
Scaling method.
Background technique
In the plate cutting of large format lathe, the stroke range of X-direction is more normal, and the stroke range of Y-direction is very big,
In order to improve processing efficiency with the utilization rate of clout plate, one monitoring camera can be installed in the top of lathe gantry beam,
Both it can be used to monitor lathe, vision also may be implemented and seek the functions such as side, vision clout stock layout, and it is wherein most basic most crucial
, it is how rapidly and accurately to determine the corresponding relationship between pixel coordinate and machine coordinates.
In traditional laser cutting parameter, before the sheet fabrication of whole picture face, requires cutting head and scan back and forth entire lathe width
Face determines the edge of plate by the greatest gradient value of capacitance sensor, to establish the corresponding relationship with image to be processed.
This capacitor seeks the mode on side, and due to needing scanning back and forth for entire lathe breadth, the consuming time is longer, influences processing efficiency.
Meanwhile capacitance sensor be easy to cause cutting head to prick head, causes the damage of cutting head and lathe.
At the same time, for the utilization of clout plate, traditional method is, manual operation lathe in a manner of walking frame,
Determine which clout region can also be cut into desired workpiece.The mode of this plate waste utilization, place one's entire reliance upon operation
The experience of worker, and precision is not high, low efficiency.
Summary of the invention
The present invention in order to overcome the deficiencies of the prior art, designs a kind of camera mark of vision auxiliary large format lathe plate cutting
Determine method, it is characterised in that include the following steps: 1) step 1, obtain source figure from camera；2) step 2 is demarcated according to camera distortion
As a result, carry out image distortion correction；3) step 3 puts at least two pieces of different-thickness respectively on laser cutting machine tool
Plate records the thickness value of plate, and chooses the pixel coordinate and machine coordinates of at least four mark points respectively to every block of plate；
4) step 4 selects the plate thickness actually cut；5) step 5, the machine coordinates shown by host computer record camera at this time
Machine coordinates；6) step 6 is calculated by perspective transform, obtains perspective transformation matrix M, and obtains relational expression M '=M1(R
T), the inverse matrix that wherein M ' is perspective transformation matrix M, R is the rotating part of Camera extrinsic, and T is the translating sections of Camera extrinsic.
Further include following steps before the step 1: camera being mounted on above the gantry beam of laser cutting machine tool, and
And camera and horizontal plane inclination are installed, it is ensured that the visual field of camera it can be seen that lathe regional area X-direction complete breadth, simultaneously
The distortion calibration of camera is completed, and obtains the internal reference of camera:
WhereinFor the internal reference of camera.
Pixel coordinate in the step 3 includes the following steps: that 1) in the plate breadth of camera fields of view, any selection is big
In the mark point for being equal to four；2) one mark point of every selection, host computer record the pixel coordinate (u, v) of this mark point, and
Mark the selecting sequence of this mark point.
Machine coordinates in the step 3 include the following steps: to press laser cutting head 1) with element marking point for reference
The sequence for the plain mark point that takes pictures is moved to corresponding points one by one；2) by by laser cutting head alignment mark point and recording machine at this time
Bed coordinate (x, y).
According to one group of at least four pixel coordinate (u, v) and one group of at least four machine coordinates (x, y) in the step 6,
3 × 3 perspective transformation matrix M can be calculated:Wherein,
The present invention compared with the existing technology, by under different plates thickness twice, the pixel coordinate of minimum of four point and
Corresponding machine coordinates, to determine under any plate thickness and any camera position, the standard of pixel coordinate and machine coordinates
True corresponding relationship allows operator without grasping a large amount of professional knowledge, it is only necessary to acquire several groups of pixel coordinates and machine coordinates
The calibration of camera and lathe can be completed in data.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Specific embodiment
Embodiment one:
As shown in Figure 1, the present invention includes the following steps: 1) to obtain source figure from camera；2) knot demarcated according to camera distortion
Fruit carries out the distortion correction of image；3) plate of at least two pieces different-thickness is put on laser cutting machine tool, one piece point every time
Acquisition coordinate information is opened, chooses the pixel coordinate and machine coordinates of at least four mark points respectively to every block of plate；4) selection is real
The plate thickness of border cutting；5) machine coordinates shown by host computer record the machine coordinates of camera；6) pass through perspective transform
It calculates, obtains calibration result.
Wherein before obtaining source figure from camera, camera is mounted on first above the gantry beam of laser cutting machine tool,
And camera and horizontal plane inclination are installed, it is ensured that the visual field of camera it can be seen that lathe regional area X-direction complete breadth, together
When complete the distortion calibration of camera, and obtain the internal reference of camera:WhereinFor the internal reference of camera,
R is the rotating part in outer ginseng, and T is the translating sections in outer ginseng.In camera calibration, there are multiple to demarcate photo, just have phase
Join outside the multiple groups answered, and internal reference is the attribute of camera, is fixed and invariable value.
When obtaining pixel coordinate, in the plate breadth first within the scope of camera fields of view, four mark points are arbitrarily selected
Or it is more, and one mark point of every label, the pixel coordinate (u, v) of this mark point is recorded by the host computer of lathe,
The flag sequence of each mark point is recorded simultaneously.
When obtaining machine coordinates, with pixel coordinate be reference, by laser cutting head according to element marking point sequence by
One is moved to the corresponding points of element marking point, and a mode for aligned pixel coordinate is then cut by laser and collects and records this
The machine coordinates (x, y) of element marking point.
Finally according to the pixel coordinate (u, v) of same sequence and machine coordinates (x, y), one 3 can be obtained by calculation
× 3 perspective transformation matrix M:Wherein,
For process different-thickness plate the case where, the Z value being equivalent in machine coordinates changes, therefore is reacted to
In outer ginseng, linear variation is had occurred in only T.When at least two different-thickness plate of calibration, so that it may calculate in outer ginseng
The linear variability law of translating sections T, at the same time, camera internal reference A and spin matrix R are constant.In later use process,
Any thickness plate is inputted, translation matrix T can be calculated, and then calculate perspective transformation matrix M.
The principle of camera calibration is substantially to ask homography matrix, that is, perspective transformation matrix.Available such as ShiShimonoseki
System: M '=M1(R T), wherein M ' is the inverse matrix of perspective transformation matrix M.For the plate of the different-thickness in the present invention, only
There are the translating sections in outer ginseng to change, and is a kind of linear variation.Therefore, two groups of different-thickness plates are at least being demarcated
In the case where material, so that it may obtain the perspective transformation matrix M of different-thickness plate.
The case where changing for camera position, since camera is mounted in the top of lathe gantry beam, being can be with
Gantry move together, and camera can only be translatable in the biggish Y-direction of range, do not occur in X-direction and Z-direction
Movement, therefore when the camera is moving, it is only necessary on the basis of the perspective transformation matrix calculated, increase the benefit in a Y-direction
It repays, formula is
Claims (1)
1. a kind of camera calibration method of vision auxiliary large format lathe plate cutting, it is characterised in that include the following steps:
1) camera is mounted on above the gantry beam of laser cutting machine tool by step 1, and camera and horizontal plane inclination are installed,
Ensure the visual field of camera it can be seen that lathe regional area X-direction complete breadth, be completed at the same time the distortion calibration of camera, and obtain
Obtain the internal reference of camera:WhereinFor the internal reference of camera, R is the rotating part in outer ginseng, and T is the translating sections in outer ginseng；2) step
Rapid 2, source figure is obtained from camera；3) step 3, according to camera distortion calibration as a result, carrying out the distortion correction of image；4) step 4,
The plate for putting at least two pieces different-thickness respectively on laser cutting machine tool records the thickness value of plate, and to every block of plate
The pixel coordinate and machine coordinates of at least four mark points are chosen respectively, and the acquisition modes of the pixel coordinate are first in camera
In the plate breadth in the visual field, any selection is more than or equal to four mark points, then one mark point of every selection, host computer record
This time pixel coordinate (u, v) of mark point, and the selecting sequence of this mark point is marked, headed by the acquisition modes of the machine coordinates
It is first reference with element marking point, laser cutting head is moved to corresponding points according to the sequence of element marking point one by one, is then led to
It crosses laser cutting head alignment mark point and the machine coordinates (x, y) of record at this time；5) step 5 selects the plate actually cut
Thickness；6) step 6, the machine coordinates shown by host computer record the machine coordinates of camera at this time；7) step 7 passes through perspective
Transformation calculations obtain perspective transformation matrix M, and obtain relational expression M '=M1(R T), wherein M ' is the inverse of perspective transformation matrix M
Matrix, R are the rotating parts of Camera extrinsic, and T is the translating sections of Camera extrinsic；According to one group at least four in the step 6
Pixel coordinate (u, v) and one group of at least four machine coordinates (x, y), can be calculated 3 × 3 perspective transformation matrix M:Wherein,
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710253805.0A CN107160046B (en) | 2017-04-18 | 2017-04-18 | A kind of camera calibration method of vision auxiliary large format lathe plate cutting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710253805.0A CN107160046B (en) | 2017-04-18 | 2017-04-18 | A kind of camera calibration method of vision auxiliary large format lathe plate cutting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107160046A CN107160046A (en) | 2017-09-15 |
| CN107160046B true CN107160046B (en) | 2019-08-02 |
Family
ID=59812197
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710253805.0A Active CN107160046B (en) | 2017-04-18 | 2017-04-18 | A kind of camera calibration method of vision auxiliary large format lathe plate cutting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107160046B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230400B (en) * | 2017-12-26 | 2021-10-19 | 常州固高智能控制技术有限公司 | Self-adaptive coordinate reconstruction method suitable for laser cutting machine |
| CN108637469B (en) * | 2018-04-10 | 2020-07-17 | 上海柏楚电子科技股份有限公司 | Visual edge finding system for laser cutting and image processing method thereof |
| CN108620745B (en) * | 2018-04-10 | 2020-07-31 | 上海柏楚电子科技股份有限公司 | Coil stock continuous cutting device for short-stroke cutting of long parts and cutting method thereof |
| CN110936026B (en) * | 2018-09-21 | 2022-06-03 | 大族激光科技产业集团股份有限公司 | Cutting method, cutting device, cutting equipment and storage medium for plate excess material |
| CN109773582B (en) * | 2018-12-27 | 2021-02-05 | 大族激光科技产业集团股份有限公司 | Method for cutting excess material of plate |
| CN110136204B (en) * | 2019-03-19 | 2021-08-03 | 浙江大学山东工业技术研究院 | Sound film dome assembly system based on calibration of machine tool position of bilateral telecentric lens camera |
| CN110977154A (en) * | 2019-12-05 | 2020-04-10 | 苏州领鹿智能科技有限公司 | Positioning marking method suitable for large breadth |
| CN111476841B (en) * | 2020-03-04 | 2020-12-29 | 哈尔滨工业大学 | Point cloud and image-based identification and positioning method and system |
| CN113618253B (en) * | 2020-05-08 | 2023-06-23 | 大族激光科技产业集团股份有限公司 | Laser processing method, system, laser processing controller and storage medium |
| CN111901569A (en) * | 2020-08-11 | 2020-11-06 | 上海柏楚电子科技股份有限公司 | Monitoring processing method, device, equipment and medium for dynamically tracking cutting head |
| CN113538283A (en) * | 2021-07-22 | 2021-10-22 | 浙江赫千电子科技有限公司 | Distortion correction method for images shot by redundant fisheye cameras |
| CN116038155B (en) * | 2023-03-17 | 2023-06-16 | 深圳市睿达科技有限公司 | System and method for splicing and cutting super-breadth non-deformable material by laser cutting |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1342527A1 (en) * | 2002-03-01 | 2003-09-10 | Koike Sanso Kogyo Co., Ltd | Cutting apparatus |
| CN1511677A (en) * | 2002-12-27 | 2004-07-14 | 中国科学院自动化研究所 | Method for identifying work piece characteristic and coordinate measurement using laser strucutre vision |
| CN104439698A (en) * | 2014-11-26 | 2015-03-25 | 北京凌云光技术有限责任公司 | Calibration method and device used for laser processing system |
| CN105157609A (en) * | 2015-09-01 | 2015-12-16 | 大连理工大学 | Two-sets-of-camera-based global morphology measurement method of large parts |
| CN105345254A (en) * | 2015-12-04 | 2016-02-24 | 湖北工业大学 | Calibration method for positional relation between paraxial type visual system and laser vibrating mirror machining system |
| CN106066185A (en) * | 2016-05-24 | 2016-11-02 | 华南理工大学 | A kind of line laser sensor automatic calibration device towards weld joint tracking and method |
-
2017
- 2017-04-18 CN CN201710253805.0A patent/CN107160046B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1342527A1 (en) * | 2002-03-01 | 2003-09-10 | Koike Sanso Kogyo Co., Ltd | Cutting apparatus |
| CN1511677A (en) * | 2002-12-27 | 2004-07-14 | 中国科学院自动化研究所 | Method for identifying work piece characteristic and coordinate measurement using laser strucutre vision |
| CN104439698A (en) * | 2014-11-26 | 2015-03-25 | 北京凌云光技术有限责任公司 | Calibration method and device used for laser processing system |
| CN105157609A (en) * | 2015-09-01 | 2015-12-16 | 大连理工大学 | Two-sets-of-camera-based global morphology measurement method of large parts |
| CN105345254A (en) * | 2015-12-04 | 2016-02-24 | 湖北工业大学 | Calibration method for positional relation between paraxial type visual system and laser vibrating mirror machining system |
| CN106066185A (en) * | 2016-05-24 | 2016-11-02 | 华南理工大学 | A kind of line laser sensor automatic calibration device towards weld joint tracking and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107160046A (en) | 2017-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107160046B (en) | A kind of camera calibration method of vision auxiliary large format lathe plate cutting | |
| US11042146B2 (en) | Automated 360-degree dense point object inspection | |
| CN105538345B (en) | A kind of puma manipulator and positioning assembly method based on many camera lenses | |
| CN103292695B (en) | A kind of single eye stereo vision measuring method | |
| CN102135417B (en) | Full-automatic three-dimension characteristic extracting method | |
| JPH05123997A (en) | Clothing cutting equipment for setting pattern by computor | |
| CN105222727B (en) | Linear array CCD camera imaging plane and the measuring method and system of the workbench depth of parallelism | |
| CN110595392B (en) | Cross line structured light binocular vision scanning system and method | |
| CN110987971B (en) | Crystal bubble detection device and method based on machine vision | |
| WO2017080294A1 (en) | Board card image splicing method and processing apparatus and system | |
| CN104551865A (en) | Image measuring system and method | |
| CN105989588B (en) | Special-shaped material cutting image correction method and system | |
| CN108637469A (en) | One kind patrolling edge system and its image processing method for being cut by laser vision | |
| CN103615980A (en) | Method and system for measuring parameters of round holes in plate | |
| CN109903342A (en) | A kind of laser in-situ process unit and method based on scanning galvanometer | |
| CN106204560B (en) | Colony picker automatic calibration method | |
| US20100017012A1 (en) | Multi model registration (mmr) for a galvanometer and laser system | |
| CN111624203B (en) | Relay contact point alignment non-contact measurement method based on machine vision | |
| WO2020093603A1 (en) | High-intensity multi-directional fdm 3d printing method based on stereoscopic vision monitoring | |
| CN106952262B (en) | Ship plate machining precision analysis method based on stereoscopic vision | |
| CN113538583A (en) | Method for accurately positioning position of workpiece on machine tool and vision system | |
| CN114274260A (en) | Cutting machine and machine-readable carrier | |
| CN110640303B (en) | High-precision vision positioning system and positioning calibration method thereof | |
| CN116188544A (en) | Point cloud registration method combining edge features | |
| CN105243659A (en) | Image positioning method of three-dimensional precise engraving machine |
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 | ||
| CB02 | Change of applicant information |
Address after: 200240 No. 953 lane, Jianchuan Road, Minhang District, Shanghai 322 Applicant after: SHANGHAI FRIENDESS ELECTRONIC TECHNOLOGY Co.,Ltd. Address before: 200240 west two floor, 2 building, 940 Jianchuan Road, Minhang District, Shanghai. Applicant before: FRIENDESS CO.,LTD. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: 200241 No. 1000, Lanxiang Hunan Road, Minhang District, Shanghai Patentee after: SHANGHAI FRIENDESS ELECTRONIC TECHNOLOGY Co.,Ltd. Address before: 200240 No. 953 lane, Jianchuan Road, Minhang District, Shanghai 322 Patentee before: SHANGHAI FRIENDESS ELECTRONIC TECHNOLOGY Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder |