WO2017045305A1 - Procédé de vérification de cordon de soudure - Google Patents

Procédé de vérification de cordon de soudure Download PDF

Info

Publication number
WO2017045305A1
WO2017045305A1 PCT/CN2015/099556 CN2015099556W WO2017045305A1 WO 2017045305 A1 WO2017045305 A1 WO 2017045305A1 CN 2015099556 W CN2015099556 W CN 2015099556W WO 2017045305 A1 WO2017045305 A1 WO 2017045305A1
Authority
WO
WIPO (PCT)
Prior art keywords
weld
verified
image
weld seam
line
Prior art date
Application number
PCT/CN2015/099556
Other languages
English (en)
Chinese (zh)
Inventor
张炯
Original Assignee
苏州中启维盛机器人科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 苏州中启维盛机器人科技有限公司 filed Critical 苏州中启维盛机器人科技有限公司
Publication of WO2017045305A1 publication Critical patent/WO2017045305A1/fr

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
    • 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
    • B23K33/00—Specially-profiled edge portions of workpieces for making soldering or welding connections; Filling the seams formed thereby
    • B23K33/004—Filling of continuous seams
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis

Definitions

  • the present invention relates to a visual recognition of a butt weld to a sheet metal structure in a robot or automated welding, and more particularly to a weld seam verification method.
  • the technical problem to be solved by the present invention is to provide a weld seam verification method which can solve any of the above problems.
  • a weld seam verification method comprising the following steps:
  • q pixels are selected on each side of the line intersecting the line of the weld to be verified to form a subset of the image, and the direction of the original image line is substantially perpendicular to the line of the weld to be verified, where q is greater than or equal to a positive integer of 1;
  • the weld position to be verified is reliable.
  • it comprises: if the maximum value and the minimum value are not located on both sides of the weld line to be verified, it is determined that the weld position to be verified is unreliable.
  • it comprises: when the distance between the obtained maximum value and the minimum value is greater than the weld line to be verified ⁇
  • the resulting weld line is adjusted based on the distance between the maximum and minimum positions.
  • the present invention adopts the above method, and can reliably verify the butt welds of the thin plates in various cases (including the welds without gaps and a certain gap) without the occurrence of verification failure.
  • FIG. 1 is an apparatus for using the weld verification method in the present application.
  • FIG. 2 is an image of an aluminum plate butt weld joint obtained by a welding robot of the prior art without active illumination.
  • FIG. 3 is an image of active welding of an aluminum plate butt weld joint obtained by a welding robot in the present application.
  • FIG. 4 is an image of a gapless butt weld inspection result of a 3 mm thick aluminum plate obtained by a welding robot in the present application.
  • FIG. 5 is an image of a gapless butt weld inspection result of a 3 mm thick steel plate obtained by a welding robot in the present application
  • FIG. 7 is an image of the detection result of the butt weld of the 3mm thick steel plate without gaps obtained by the welding robot in the present application.
  • FIG. 9 is an image of a test result of a butt weld of a 0.05 mm gap of a 3 mm thick aluminum plate obtained by a welding robot in the present application.
  • 11 is an image of a 0.5 mm butt weld joint detection result of a 3 mm thick steel plate obtained by a welding robot in the present application.
  • the present invention discloses a weld inspection method that includes the following steps:
  • the row direction of the initial image is substantially perpendicular to the weld seam, and one row is selected every n rows for processing, and m rows are selected.
  • m is a positive integer greater than 8
  • n is an integer greater than or equal to 0
  • f(i,j) represents the pixel gray value of the i-th point of the j-th row
  • p Xcos6 + Ysine , where p is the distance from the origin to the line, and ⁇ is the angle between the normal of the line and the X axis Specifically, in the present embodiment, after acquiring the thin plate image, the upper left corner of the image is defined as the origin, and the origin is defined downward as the Y direction; the origin is defined to the right by the X direction; the direction of the row is parallel to In the X direction, the direction of the column is parallel to the Y direction.
  • one line may be selected every m rows in the Y direction for sampling.
  • m may be any one of 0 to 10.
  • the origin of the coordinates is in the upper left corner of the image, the X direction is horizontal to the right, and the Y direction is vertical downward.
  • f(i,j) represents the pixel gray value of the i-th point of the jth line. If you take a line every 1 line, the last is to get 256 lines. If you take a line every 3 lines, the last line is 128 lines.
  • the gray difference values, f(i, j) and f(il, j) are the gray values at the pixel points (i, j) and (il, j), respectively.
  • the position of the k maxima pixels and the position of the k minima pixels are selected in the differential data.
  • k is a positive integer of any one of 2 to 4.
  • Hough Transform p Xcos6 + Ysine for 2k points (2k*m total) on all selected lines.
  • the parameters p and ⁇ uniquely determine a straight line, the distance from the p origin to the line, and ⁇ is the angle between the normal of the line and the X axis.
  • the line corresponding to this maximum value is our candidate weld line (the physical area corresponding to the image is approximately l l.2mm X l l.2mm, in the entire image range) It is assumed that the weld is a straight line.
  • the above weld inspection algorithm can not only reliably detect butt welds without gaps and small gaps, but also has strong detection results in the case of strong disturbances such as surface scratches and tack welds.
  • a weld seam verification method comprising the following steps:
  • the obtained weld line is adjusted by the distance between the maximum value and the minimum value. .
  • the test results are adjusted.
  • the position of the largest value here corresponds to the left edge of the gap, and the position of the minimum corresponds to the right edge of the gap.
  • the difference between the maximum and minimum position corresponds to the gap width.
  • the image of 512X512 in our experiment corresponds to a physical size of 11.2mmX12.2mm, so 0.2mm corresponds to 9 pixels.
  • a gap of 0.5 mm corresponds to 22 pixels.
  • the present application also discloses a thin-plate welding robot, which includes:
  • a robot arm having a plurality of degrees of freedom
  • a welding gun 1 the welding gun 1 is disposed on the robot arm, so that the welding gun 1 can be used for welding the thin plate as the robot arm moves;
  • a video camera 2 comprising an image sensor for acquiring the sheet and/or weld image and an electronic shutter for controlling the image sensor to acquire an image;
  • Active illumination 3 comprising a light emitting diode for providing illumination to the video camera 2, the light emitting diode operating based on the electronic shutter action.
  • the robot arm may have six or more degrees of freedom to accurately control the running trajectory of the welding torch 1.
  • the welding torch 1 is controlled by the robot arm so that the thin plate can be selectively manipulated, such as welding or stopping welding.
  • the welding torch can be a similar device that can be used for welding, such as a argon arc welding torch.
  • the video camera 2 has a high resolution and a small size, and is suitable for use by a robot.
  • the video camera 2 includes a high-precision image sensor that is controlled by an electronic shutter to acquire an image of the thin plate.
  • the active illumination 3 employs a high power light emitting diode. Since the light emitting diode can be controlled by an electrical signal, the light emitting diode can be controlled based on the electronic shutter. For example, when the electronic shutter manipulates the image sensor to acquire an image, the light emitting diode can provide light to the area of the thin plate to give the thin plate illumination with a certain brightness.
  • the active illumination used in the prior art is a conventional illumination lamp, and its gate control often has a certain degree of delay, and thus is not suitable for high precision and fast occasions.
  • the thin-plate welding robot in the present application can quickly process the thin plate, greatly increasing its accuracy and maneuverability.
  • the light emitting diode is a high power light emitting diode.
  • the light emitting diode causes the image sensor to acquire an image based on the electronic shutter
  • the light emitting diode is turned off based on the electronic shutter such that the image sensor does not acquire an image.
  • the light emitting diode adjusts the intensity of light emitted by the image sensor based on the image acquired by the image sensor
  • the video camera 2, the light emitting diode and the welding gun are arranged in a line.
  • the video camera 2 and the light emitting diode are located in front of the moving direction of the welding torch
  • the active illumination 3 employs a high power light emitting diode. Since the light emitting diode can be controlled by an electrical signal, the light emitting diode can be controlled based on the electronic shutter. For example, when an electronic shutter manipulates an image sensor to acquire an image, the LED can provide illumination to the area of the thin plate weld, The obtained sheet weld image has appropriate brightness.
  • the active illumination used in the prior art is a conventional illumination lamp, and its gate control often has a certain degree of delay, and thus is not suitable for high-precision, fast occasions.
  • the thin-plate welding robot in the present application can quickly process the thin plate, greatly increasing its accuracy and maneuverability.
  • the light emitting diode is a high power light emitting diode.
  • the light emitting diode causes the image sensor to acquire an image based on the electronic shutter
  • the light emitting diode is turned off based on the electronic shutter such that the image sensor does not acquire an image.
  • the light emitting diode adjusts the intensity of light emitted by the image sensor based on the image acquired by the image sensor
  • the video camera 2, the light emitting diode and the welding gun are arranged in a line.
  • the video camera 2 and the light emitting diode are located in front of the moving direction of the welding torch

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Mechanical Engineering (AREA)
  • Optics & Photonics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

L'invention concerne un procédé permettant de vérifier les résultats de mesure d'un cordon de soudure de raccordement comprenant les étapes suivantes: l'obtention d'une image de cordon de soudure et d'une ligne droite de cordon de soudure devant être vérifié dans l'image de cordon de soudure; la sélection, à partir des deux bords des points où chaque rangée croise la ligne droite de cordon de soudure devant être vérifié, d'un nombre q de pixels pour constituer un sous-ensemble d'image; la direction des rangées de l'image d'origine étant généralement perpendiculaire au cordon de soudure devant être vérifié, q étant un nombre entier positif supérieur ou égal à 1; l'addition de toutes les rangées dudit sous-ensemble d'image pour obtenir un réseau, et la réalisation d'un différentiel de premier ordre sur le réseau et l'obtention des positions de la valeur maximale et de la valeur minimale; si la valeur maximale et la valeur minimale sont situées des deux côtés de la ligne droite de cordon de soudure devant être vérifié, la détermination de la fiabilité de la position du cordon de soudure devant être vérifié. Au moyen du procédé décrit, il est possible de vérifier de manière fiable le cordon de soudure de raccordement d'une feuille comprenant un cordon de soudure sans interstice et un cordon de soudure présentant un interstice sans aucune défaillance lors de l'identification.
PCT/CN2015/099556 2015-09-15 2015-12-29 Procédé de vérification de cordon de soudure WO2017045305A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510582953.8A CN106514064A (zh) 2015-09-15 2015-09-15 焊缝验证方法
CN201510582953.8 2015-09-15

Publications (1)

Publication Number Publication Date
WO2017045305A1 true WO2017045305A1 (fr) 2017-03-23

Family

ID=58288138

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/099556 WO2017045305A1 (fr) 2015-09-15 2015-12-29 Procédé de vérification de cordon de soudure

Country Status (2)

Country Link
CN (1) CN106514064A (fr)
WO (1) WO2017045305A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106925924A (zh) * 2017-05-17 2017-07-07 山东双源印铁制罐有限公司 一种焊缝外涂补保护加强器

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005014027A (ja) * 2003-06-24 2005-01-20 Enshu Ltd 溶接部のイメージ処理方法、溶接管理システム、溶接機のためのフィードバックシステム、突合わせ線検出システム
US20050041852A1 (en) * 2001-11-15 2005-02-24 Joachim Schwarz Method and device for evaluation of jointing regions on workpieces
CN101526480A (zh) * 2009-04-27 2009-09-09 清华大学 基于视觉的薄板对接焊缝实时检测方法
CN101559512A (zh) * 2009-05-21 2009-10-21 山东大学 基于激光测距的平板对接焊缝焊接轨迹检测与控制方法
CN104636760A (zh) * 2015-03-11 2015-05-20 西安科技大学 一种焊缝的定位方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1872184B1 (fr) * 2005-04-04 2011-05-25 Fisher-Rosemount Systems, Inc. Methodes de depouillement statistique utilisees a des fins de detection de situations anormales
CN102175700B (zh) * 2011-01-20 2012-07-25 山东大学 数字x射线图像焊缝分割和缺陷检测方法
CN103759648B (zh) * 2014-01-28 2016-05-25 华南理工大学 一种基于激光双目视觉的复杂角焊缝位置检测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050041852A1 (en) * 2001-11-15 2005-02-24 Joachim Schwarz Method and device for evaluation of jointing regions on workpieces
JP2005014027A (ja) * 2003-06-24 2005-01-20 Enshu Ltd 溶接部のイメージ処理方法、溶接管理システム、溶接機のためのフィードバックシステム、突合わせ線検出システム
CN101526480A (zh) * 2009-04-27 2009-09-09 清华大学 基于视觉的薄板对接焊缝实时检测方法
CN101559512A (zh) * 2009-05-21 2009-10-21 山东大学 基于激光测距的平板对接焊缝焊接轨迹检测与控制方法
CN104636760A (zh) * 2015-03-11 2015-05-20 西安科技大学 一种焊缝的定位方法

Also Published As

Publication number Publication date
CN106514064A (zh) 2017-03-22

Similar Documents

Publication Publication Date Title
Shao et al. A novel weld seam detection method for space weld seam of narrow butt joint in laser welding
CN110064819B (zh) 基于结构光的柱面纵向焊缝特征区域提取、焊缝跟踪方法及系统
Dinham et al. Autonomous weld seam identification and localisation using eye-in-hand stereo vision for robotic arc welding
Liu et al. Precise initial weld position identification of a fillet weld seam using laser vision technology
US20160377422A1 (en) Device and method for detecting flatness of surface of display panel
US8244040B2 (en) Object position recognition system, object positioning system, and system and method for adjoining objects
JP6296206B2 (ja) 形状測定装置及び形状測定方法
CN108907526A (zh) 一种具有高鲁棒性的焊缝图像特征识别方法
CN108344693B (zh) 一种面向自动焊接的薄板焊缝错边量视觉测量方法
JP2011089826A (ja) ねじ穴または穴の内部表面欠陥検査装置
CN110596139A (zh) 一种屏缺陷检测方法和系统
WO2017045303A1 (fr) Procédé de test de cordon de soudure
WO2017045306A1 (fr) Robot de soudage de tôle
US20170309035A1 (en) Measurement apparatus, measurement method, and article manufacturing method and system
JP2014149182A (ja) ワークとの相関位置決め方法
KR20170142379A (ko) 영상처리를 이용한 용접로봇의 용접선 인식장치
WO2017045305A1 (fr) Procédé de vérification de cordon de soudure
CN114581368A (zh) 一种基于双目视觉的棒材焊牌方法及装置
US9636824B2 (en) Transfer system
WO2021022956A1 (fr) Procédé de correction de position de module d'affichage et écran d'affichage
CN113579601A (zh) 一种焊道的定位方法、装置、焊接机器人和存储介质
JP2009058459A (ja) 形状測定装置
US10054515B2 (en) Focusing state measuring apparatus
JP2004042118A (ja) 溶接位置自動倣い制御装置及び自動倣い溶接方法
JPH0545117A (ja) 光学式３次元位置計測方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15903992

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 20/08/2018)

122 Ep: pct application non-entry in european phase

Ref document number: 15903992

Country of ref document: EP

Kind code of ref document: A1