CN102478390A - Welding pool surface shape acquiring and sensing system on basis of binocular vision - Google Patents
Welding pool surface shape acquiring and sensing system on basis of binocular vision Download PDFInfo
- Publication number
- CN102478390A CN102478390A CN2010105662062A CN201010566206A CN102478390A CN 102478390 A CN102478390 A CN 102478390A CN 2010105662062 A CN2010105662062 A CN 2010105662062A CN 201010566206 A CN201010566206 A CN 201010566206A CN 102478390 A CN102478390 A CN 102478390A
- Authority
- CN
- China
- Prior art keywords
- module
- reflective mirror
- video camera
- ccd video
- mounting bracket
- 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.)
- Withdrawn
Links
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a welding pool surface shape acquiring and sensing system on the basis of binocular vision, which comprises a CCD (Charge Coupled Device) camera module, a composite optical path module and a real-time calculating module and has the innovations that the system consists of two optical path systems; images at both left and right sides of a welding pool are respectively acquired by the two optical path systems and are converged on the same target surface of the CCD camera module by the composite optical path module to form a simultaneous same-amplitude system; and the CCD camera module is used for transmitting the acquired images to the real-time calculating module to carry out image processing and welding pool surface height restore calculation by an image acquisition card. The welding pool surface shape acquiring and sensing system on the basis of binocular vision is suitable for detecting a surface shape of the welding pool in the arc welding process.
Description
Technical field
The invention belongs to welding process visual sensing technology, particularly a kind of welding pool surface configuration based on binocular vision is obtained and sensor-based system, and it is applicable to arc welding process welding pool surface shape detection.
Background technology
Welding has critical role as a kind of important advanced manufacturing technology, and robotize, welding intellectuality have become Development of Welding Technology trend.The welding robot intellectualized technology is meant the various sensors of assembling on welding robot; Make welding robot to external world environment had certain perception; Thereby welding robot can independently be conformed, according to the variation of environment, the autonomous adjustment of next step operation to self.Appearance of weld control is an important branch of robot welding intellectualized technology; Its objective is the appearance of weld that guarantees the reliable penetration of workpiece, obtains uniformity; Its main process is through detecting the situation of welding process, adjust welding parameter in real time, guaranteeing welding quality stable.The shape in molten bath and big or small shaping and the inherent quality that is directly determining weld seam.The shape at the back side, molten bath can't directly be observed in the welding process, but can predict the form parameter at the back side through the positive form parameter in observation molten bath.Skilled welder waits the shape and size parameter of predicting the back side through the joint form of observing positive molten bath characteristic parameter, workpiece, the shape and the mode of metal transfer of electric arc; Through regulating the control that welding parameter realizes penetration, guarantee the stable of welding quality.Therefore the extraction of molten bath front shape information has important effect for Control Welding Process.
The melting pool shape Parameter Extraction mainly concentrates on the two-dimensional signal in molten bath at present; But the surface configuration in molten bath more can reflect weld seam penetration and back-welding shape situation; With weld reinforcement certain corresponding relation is arranged, have important effect for the control of melting pool shape.But owing to exist sound, light, electricity, magnetic, high temperature in the welding process, splash and the interference of flue dust; Simultaneously welding pool has that volume is little, in light weight, temperature is high and characteristics such as skewness, the effect short, that cooling velocity is fast, the molten bath receives various power of high temperature life period are in the dynamic movement process, fusing and the process of setting of filling metal exists simultaneously, so the unusual difficulty of weld pool surface shape measure.Because the electromagnetic interference (EMI) of welding process and the high temperature characteristics in molten bath, weld pool surface can not adopt the method for machinery and electromagnetism to measure the weld pool surface height.Along with the development of machine vision and image processing techniques, the optical sensing technology has obtained great progress in the sensing of welding weld penetration information, can be used for measuring the height of weld pool surface.
Summary of the invention
To the above-mentioned deficiency of prior art, technical matters to be solved by this invention has been to propose a kind of welding pool surface configuration acquisition methods and sensor-based system based on the binocular vision sensing, is used for online detection welding pool surface configuration.
The present invention is based on that following technical scheme is achieved; A kind of welding pool surface configuration acquisition methods and sensor-based system based on the binocular vision sensing; Comprise ccd video camera module, composite light path module and real-time computing module; Its innovative point is: this system is made up of two light path systems, gathers the image of the left and right sides, molten bath respectively, converges in the same target surface of ccd video camera module through the composite light path module; Formation is simultaneously with width of cloth system, and the ccd video camera module is sent to real-time computing module with the image of gathering through image pick-up card and carries out Flame Image Process, weld pool surface height recovery calculating.
Preferably, in the above-mentioned welding pool surface configuration acquisition methods and sensor-based system based on the binocular vision sensing of the present invention, said ccd video camera module is made up of adjustable focus optical lens, ccd video camera.The optical lens focal length is that 5-50mm is adjustable, and ccd video camera is WAT-902H.Video camera and camera lens, neutral light damping plate, the coaxial installation of broad band pass filter.
Preferably; In the above-mentioned welding pool surface configuration acquisition methods and sensor-based system of the present invention based on the binocular vision sensing; Said composite light path is formed by isolating optical mirror slip, neutral colour filter, broad band pass filter and coated optics eyeglass; Said broad band pass filter wavelength coverage is 580nm-720nm, and peak transmittance is 25%, and said neutral light damping plate transmitance is 10%.
Preferably; In the above-mentioned welding pool surface configuration acquisition methods and sensor-based system of the present invention based on the binocular vision sensing; Said real-time computing module is made up of computing machine and image pick-up card, and image pick-up card adopts the CG400 image pick-up card, and computing machine is an industrial control computer.
Preferably; Above-mentioned welding pool surface configuration acquisition methods and the sensor-based system based on the binocular vision sensing of the present invention also comprises supplementary module, and said supplementary module is by sensor outer housing, isolation eyeglass mounting bracket, neutral colour filter mounting bracket, broad band pass filter support and coated optics eyeglass mounting bracket, ccd video camera mounting bracket and be connected support.Sensor outer housing, isolation eyeglass mounting bracket, neutral colour filter mounting bracket, broad band pass filter support and coated optics eyeglass mounting bracket, ccd video camera mounting bracket link to each other with welding gun through being connected support.Broad band pass filter support and neutral colour filter mounting bracket are made detachable state, are convenient to install debugging.
The present invention proposes based on the welding pool surface configuration of binocular vision and obtain and sensor-based system, can extract the weld pool surface shape, information is provided for welding modeling and Based Intelligent Control.
Description of drawings
Fig. 1 is that the welding pool surface configuration that the present invention is based on binocular vision is obtained the structural representation with sensor-based system.
Fig. 2 is that the surperficial melting pool shape that the present invention is based on binocular vision obtains software flow pattern.
Embodiment
Below in conjunction with accompanying drawing and specific embodiment, further set forth the present invention.These embodiment are interpreted as only being used to the present invention is described and are not used in restriction protection scope of the present invention.After the content of having read the present invention's record, those skilled in the art can do various changes or modification to the present invention, and these equivalences change and modify and fall into claim of the present invention institute restricted portion equally.
As shown in Figure 1; Welding pool surface configuration acquisition methods and sensor-based system that preferred embodiment of the present invention provides based on the binocular vision sensing; Comprise ccd video camera module, composite light path module, real-time computing module, supplementary module; It is characterized in that: this sensor is made up of two light path systems, gathers the image of the left and right sides, molten bath respectively, converges in the same target surface of video camera through light path; Formation is simultaneously with width of cloth system, and ccd video camera is sent to computing machine with the image of gathering through image pick-up card and carries out Flame Image Process, weld pool surface height recovery calculating.
Described ccd video camera module is made up of adjustable focus optical lens 10, ccd video camera 8.Adjustable focus optical lens 10 focal lengths are 5-50mm, and ccd video camera 8 is WAT-902H.But ccd video camera 8 and adjustment optical lens 10, neutral light damping plate 12, broad band pass filter 11 coaxial installations.
Said composite light path module is made up of neutral light damping plate 12, broad band pass filter 11 and two light paths; Left side optical routing is isolated optical mirror slip 5, coated optics reflective mirror 1; Coated optics reflective mirror 2 is formed; Right optical routing is isolated optical mirror slip 6, coated optics reflective mirror 3, coated optics reflective mirror 4 and is formed, and two light paths send the image of the welding pool left and right sides the same target surface of ccd video camera 8 to simultaneously, form simultaneously with width of cloth image.
Said broad band pass filter wavelength coverage is 580nm-720nm, and peak transmittance is 25%, and said neutral light damping plate transmitance is 10%.
Said real-time computing module is made up of industrial control computer 15 and image pick-up card 14; Image pick-up card 14 is the CG400 image pick-up card; Send industrial control computer 15 to the width of cloth image when image pick-up card 14 obtains ccd video camera, carry out Flame Image Process and welding pool surface elevation and calculate.
Said supplementary module is by sensor outer housing 7, isolation eyeglass mounting bracket, neutral colour filter mounting bracket, broad band pass filter support and coated optics eyeglass mounting bracket, ccd video camera mounting bracket and be connected support 13.Sensor outer housing 7, isolation eyeglass mounting bracket, neutral colour filter mounting bracket, broad band pass filter support and coated optics eyeglass mounting bracket, ccd video camera mounting bracket link to each other with welding gun 9 through being connected support 13.Broad band pass filter support and neutral colour filter mounting bracket are made detachable state, are convenient to install debugging.
Said sensor outer housing 7 adopts the 2B12 aluminium alloy, is used for shielding the electromagnetic field of welding process, weight reduction simultaneously, and case surface is handled through living bright black, reduces the repeatedly reflection of light path system, reduces the interference of light path.
Said reflective mirror 1 mounting center position is (40,120,90), and reflective mirror 1 surface normal is (126 °, 90 °, 108 °).
Said reflective mirror 2 installation sites are (40,120,90), and reflective mirror 2 surface normals are (126 °, 90 °, 108 °).
Said reflective mirror 3 installation sites are (40,10,90), and reflective mirror 3 surface normal positions are (90 °, 45 °, 45 °).
Said reflective mirror 4 installation sites are (40 ,-10,90), and reflective mirror 4 surface normal positions are (90 °, 135 °, 45 °).
Specifically describe a kind of surface elevation computing method below in conjunction with accompanying drawing 1-2, comprise the steps: based on binocular vision
1) crater image is carried out the image enhancement processing based on wavelet analysis;
Wherein, f (x, y), φ (x, y), (x y) is crater image, scaling function and wavelet function respectively to ψ, and j is the picture breakdown level, g
K, nBe gain coefficient, c
J, nBe wavelet coefficient
2) crater image is carried out rim detection;
The selection Gaussian function is a scaling function, and one of which, second derivative are wavelet function, constitutes multi-scale wavelet transformation.
Wherein, smooth function θ (x) is a scaling function.
3) same point on two width of cloth images is calculated the computed altitude value.
Wherein, (x, y z) are the coordinate of welding pool surface point P, and B is the distance between reflective mirror 1 and the reflective mirror 4, (X
Left, Y), (X
Right, be respectively the image coordinate of welding pool surface point P on ccd video camera 8 target surfaces Y), f is the focal length of ccd video camera 8.
Claims (7)
1. the welding pool surface configuration based on binocular vision is obtained and sensor-based system; Comprise ccd video camera module, composite light path module and real-time computing module; It is characterized in that: this system is made up of two light path systems; Gather the image of the left and right sides, molten bath respectively; Converge in the same target surface of ccd video camera module through the composite light path module, form simultaneously with width of cloth system, the ccd video camera module is sent to the image of gathering that the real-time computer module is carried out Flame Image Process, the weld pool surface height recovers to calculate through image pick-up card.
2. the welding pool surface configuration based on binocular vision according to claim 1 is obtained and sensor-based system; It is characterized in that: also comprise the supplementary module module, said supplementary module comprises sensor outer housing, isolation eyeglass mounting bracket, neutral colour filter mounting bracket, broad band pass filter support and coated optics eyeglass mounting bracket, ccd video camera mounting bracket and is connected support; Sensor outer housing, isolation eyeglass mounting bracket, neutral colour filter mounting bracket, broad band pass filter support and coated optics eyeglass mounting bracket, ccd video camera mounting bracket link to each other with welding gun through being connected support; Broad band pass filter support and neutral colour filter mounting bracket are made detachably.
3. the welding pool surface configuration based on binocular vision according to claim 1 is obtained and sensor-based system, it is characterized in that: said ccd video camera module is made up of adjustable focus optical lens, ccd video camera; The optical lens focal length is that 550mm is adjustable, and ccd video camera is WAT-902H; Ccd video camera and adjustable focus optical lens, neutral light damping plate, the coaxial installation of broad band pass filter.
4. the welding pool surface configuration based on binocular vision according to claim 1 is obtained and sensor-based system; It is characterized in that: said composite light path module is made up of neutral light damping plate, broad band pass filter and two light paths; Left side optical routing is isolated optical mirror slip, coated optics reflective mirror 1 and coated optics reflective mirror 2 and is formed; Right optical routing is isolated optical mirror slip, coated optics reflective mirror 3 and coated optics reflective mirror 4 and is formed; Two light paths send the image of the welding pool left and right sides the same target surface of ccd video camera to simultaneously, form simultaneously with width of cloth image.
5. the welding pool surface configuration based on binocular vision according to claim 4 is obtained and sensor-based system, it is characterized in that: the broad band pass filter wavelength coverage is 580nm-720nm, and peak transmittance is 25%, and neutral light damping plate transmitance is 10%.
6. obtain and sensor-based system according to claim 4 or 5 described welding pool surface configurations based on binocular vision, it is characterized in that: coated optics reflective mirror 1 mounting center position is (40,120; 90); Coated optics reflective mirror 1 surface normal is (126 °, 90 °, 108 °); Coated optics reflective mirror 2 installation sites are (40 ,-120,90), and coated optics reflective mirror 2 surface normals are (126 °, 90 °, 108 °); Coated optics reflective mirror 3 installation sites are (40,10,90), and coated optics reflective mirror 3 surface normal positions are (90 °, 45 °, 45 °); Coated optics reflective mirror 4 installation sites are (40 ,-10,90), and coated optics reflective mirror 4 surface normal positions are (90 °, 135 °, 45 °).
7. the surface elevation computing method based on binocular vision is characterized in that: comprise the steps: 1) crater image is carried out the image enhancement processing based on wavelet analysis; 2) image is carried out rim detection; 3) same point on two width of cloth images is calculated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010105662062A CN102478390A (en) | 2010-11-30 | 2010-11-30 | Welding pool surface shape acquiring and sensing system on basis of binocular vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010105662062A CN102478390A (en) | 2010-11-30 | 2010-11-30 | Welding pool surface shape acquiring and sensing system on basis of binocular vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102478390A true CN102478390A (en) | 2012-05-30 |
Family
ID=46091104
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010105662062A Withdrawn CN102478390A (en) | 2010-11-30 | 2010-11-30 | Welding pool surface shape acquiring and sensing system on basis of binocular vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102478390A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104325234A (en) * | 2014-10-16 | 2015-02-04 | 上海交通大学 | Visual sensor for acquiring images of double-sided weld pool during plasma welding |
CN105290576A (en) * | 2015-09-29 | 2016-02-03 | 湘潭大学 | Real-time detecting method and device for surface topography of swing arc MAG welding molten pool |
CN107962278A (en) * | 2016-10-19 | 2018-04-27 | 南京理工大学 | A kind of arc welding weld crater image harvester based on saturating infrared fileter |
CN108620752A (en) * | 2017-03-17 | 2018-10-09 | 丰田自动车株式会社 | Measuring apparatus and laser welding apparatus |
CN111805062A (en) * | 2019-04-12 | 2020-10-23 | 住友重机械工业株式会社 | Welding observation device and welding system |
CN114283272A (en) * | 2021-12-15 | 2022-04-05 | 上海电力大学 | Molten pool image acquisition method based on spatial filtering technology |
-
2010
- 2010-11-30 CN CN2010105662062A patent/CN102478390A/en not_active Withdrawn
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104325234A (en) * | 2014-10-16 | 2015-02-04 | 上海交通大学 | Visual sensor for acquiring images of double-sided weld pool during plasma welding |
CN104325234B (en) * | 2014-10-16 | 2015-11-25 | 上海交通大学 | Plasma welds the vision sensor of two-sided Molten pool image gathering |
CN105290576A (en) * | 2015-09-29 | 2016-02-03 | 湘潭大学 | Real-time detecting method and device for surface topography of swing arc MAG welding molten pool |
CN107962278A (en) * | 2016-10-19 | 2018-04-27 | 南京理工大学 | A kind of arc welding weld crater image harvester based on saturating infrared fileter |
CN108620752A (en) * | 2017-03-17 | 2018-10-09 | 丰田自动车株式会社 | Measuring apparatus and laser welding apparatus |
CN108620752B (en) * | 2017-03-17 | 2019-11-26 | 丰田自动车株式会社 | Measuring device and laser welding apparatus |
CN111805062A (en) * | 2019-04-12 | 2020-10-23 | 住友重机械工业株式会社 | Welding observation device and welding system |
CN114283272A (en) * | 2021-12-15 | 2022-04-05 | 上海电力大学 | Molten pool image acquisition method based on spatial filtering technology |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102478390A (en) | Welding pool surface shape acquiring and sensing system on basis of binocular vision | |
CN101486124B (en) | Multi-structured light binocular composite vision weld joint tracking method and device | |
CN108088390B (en) | Optical losses three-dimensional coordinate acquisition methods based on double eye line structure light in a kind of welding detection | |
CN102814574B (en) | Narrow gap welding monitoring and welding line deviation detecting method based on infrared vision sensing | |
CN104588838A (en) | Automatic tracking control and online correction system with welding gun and control method thereof | |
CN101334264B (en) | Laser welding narrow butt-jointed seam measurement method and device | |
CN105157603A (en) | Line laser sensor and method for calculating three-dimensional coordinate data of line laser sensor | |
CN106392267A (en) | Line laser real-time welding seam tracking method for six-freedom-degree welding robot | |
CN102455171B (en) | Method for detecting geometric shape of back of tailor-welding weld and implementing device thereof | |
CN110246175A (en) | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera | |
CN105414710B (en) | Active and passive visual welding pool composited sensing device and sensing method realized through same | |
CN112191993B (en) | Argon arc welding seam tracking system and method | |
CN103134599A (en) | Method and system for real-time monitoring of molten bath state in direct molding process of laser metal | |
CN113634954B (en) | Multi-sensor fusion welding seam track tracking method and system | |
CN102284769A (en) | System and method for initial welding position identification of robot based on monocular vision sensing | |
CN110793572A (en) | Selective laser melting process monitoring system based on multi-sensor fusion | |
CN102155920B (en) | Welding seam measuring method based on micro field depth | |
CN103398669A (en) | Multi-axis linkage visual inspection method and equipment used for measuring free-form surface | |
CN110385505A (en) | Large-sized structural parts welding automatic tracking method and system based on passive light vision | |
CN111761266A (en) | Welding robot and welding method | |
Zhao et al. | Extraction of three-dimensional parameters for weld pool surface in pulsed GTAW with wire filler | |
CN112935541A (en) | Laser tracking large-scale curved plate fillet welding system and method | |
Xia et al. | Real-time seam tracking during narrow gap GMAW process based on the wide dynamic vision sensing method | |
Zhang et al. | Automatic gap tracking during high power laser welding based on particle filtering method and BP neural network | |
CN102528811A (en) | Mechanical arm positioning and obstacle avoiding system in Tokamak cavity |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C04 | Withdrawal of patent application after publication (patent law 2001) | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20120530 |