WO2012120192A2 - Système de vision de machine permettant un contrôle qualité - Google Patents

Système de vision de machine permettant un contrôle qualité Download PDF

Info

Publication number
WO2012120192A2
WO2012120192A2 PCT/FI2012/050213 FI2012050213W WO2012120192A2 WO 2012120192 A2 WO2012120192 A2 WO 2012120192A2 FI 2012050213 W FI2012050213 W FI 2012050213W WO 2012120192 A2 WO2012120192 A2 WO 2012120192A2
Authority
WO
WIPO (PCT)
Prior art keywords
cameras
images
seam
quality
inspected
Prior art date
Application number
PCT/FI2012/050213
Other languages
English (en)
Other versions
WO2012120192A3 (fr
Inventor
Kosti Kannas
Original Assignee
Oy Mapvision Ltd
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 Oy Mapvision Ltd filed Critical Oy Mapvision Ltd
Priority to EP12754525.9A priority Critical patent/EP2684033A4/fr
Publication of WO2012120192A2 publication Critical patent/WO2012120192A2/fr
Publication of WO2012120192A3 publication Critical patent/WO2012120192A3/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/001—Industrial image inspection using an image reference approach
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10141—Special mode during image acquisition
    • G06T2207/10152—Varying illumination
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20021—Dividing image into blocks, subimages or windows
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30152—Solder

Definitions

  • the invention relates to quality control by using a machine vision system.
  • Machine vision has been used for many different applications.
  • quality control In a quality control application the quality of a manufactured item is verified by machine vision or a similar system.
  • the implementation of the system depends on the object to be inspected. For example, in timber the quality may depend on the number of knot ⁇ holes. This can be inspected by taking an image with a camera and then analyzing the image.
  • a plurality of points can be projected to the ob- ject. Then the object is imaged by using multiple cam ⁇ eras so that real coordinates of the object can be computed from the projected points and thus the cor ⁇ rect shape may be verified.
  • a seam be ⁇ tween parts that have been joined together. It is com ⁇ mon that objects are manufactured by combining a plu- rality of parts. These parts are joined together, for example, by welding, soldering, gluing, sewing or sim- ilar.
  • the method for joining or connecting depends on the materials that need to be joined. In many cases an experienced person sees from a seam if the joining fulfills the quality requirement. As the quality con- trol in seams may need to be accurate, surface scan ⁇ ners are commonly used. High accuracy requirement is very typical in objects wherein the inadequate quality of the object or part would endanger human lives or could lead into serious accidents.
  • One example of such parts is car parts.
  • the invention discloses a method for inspect ⁇ ing a portion of an object.
  • a portion of an object is first inspected by using at least one camera.
  • the imaging step a plurality of different images are taken.
  • a value from said images is computed based on at least one sample image, wherein the value represents the quality of the inspected por- tions of an object.
  • the imaging is typically done by using ordinary digital cameras and the taken images are typically common photographs; however, also spe ⁇ cial equipment may be used.
  • the term "different" here means that there is an essential difference between the images. Examples of such differences are different angle, zoom, focus, lighting, exposure and similar.
  • the inspected por ⁇ tion is a seam, such as a welding seam.
  • the seam may be divided into segments and each of the segments may be analyzed respectively.
  • the plurality of different images is taken by using a plurality of cam ⁇ eras. It is common to calibrate at least a portion of the used cameras .
  • the plurality of different images is acquired by using different lighting condi ⁇ tions when imaging.
  • the change of lighting may be com- bined with the plurality of cameras.
  • it is pos ⁇ sible to acquire a large number of images in short time with a reasonable number of cameras.
  • the invention is typically implemented as a system for inspecting a portion of an object, which system comprises at least one camera and a server for controlling said at least one camera and receiving im ⁇ ages from said at least one camera.
  • the system is con ⁇ figured to acquire a plurality of different images and compute a value representing a quality from said imag ⁇ es based on at least one sample image, wherein the value represents the quality of the inspected portions of an object.
  • the system comprises a plural ⁇ ity of cameras that are fixed to their locations by using a frame so that the object to be inspected is located inside the frame.
  • the frame further comprises a plurality of lights for providing different lighting conditions.
  • the system may be calibrated.
  • the system is configured to perform the method disclosed above.
  • the benefit of the invention is that it pro ⁇ vides a reliable method for quality controlling from images acquired by an ordinary camera.
  • the invention is particularly suitable for controlling the quality of seams.
  • the method according to the invention is able to perform the quality inspection by using small ⁇ er amount of data and thus the computing phase does not require as much computing power as prior art meth- ods . This means that more details can be inspected by machine vision and the need for human inspection is reduced. This leads into faster inspection rate and improved quality.
  • Fig. 1 is a block diagram of an example embodiment of the present invention
  • Fig. 2 is a flow chart of a method according to the present invention. DETAILED DESCRIPTION OF THE INVENTION
  • FIG 1 a block diagram according to an embodiment of the invention is shown.
  • two plates 10a and 10b are welded together by a seam 11. These seamed plates form an object to be inspect ⁇ ed.
  • the welding seam 11 in Figure 1 is the most inter ⁇ esting part of the object. If the quality of welding is inadequate the object may break into two pieces.
  • FIG 1 there is a plurality of cameras C and lights L that are attached to rails 13.
  • the rails act as a support for the cameras C and lights L.
  • the cameras C and lights L are coupled to a server 14.
  • the connection may be wireless or wired.
  • the server is configured to instruct the lights and cameras when im ⁇ aging and then receive acquired images.
  • the acquired images may be processed in the server 14; however, it is possible that the server sends the acquired images to a further server or workstation.
  • FIG 1 the welding seam 11 is divided into four segments.
  • the number of segments is deter ⁇ mined by the programmer of the system. Small details may be processed as one segment and longer seams are divided into a plurality of segments.
  • figure 2 a method according the present invention is disclosed. The method is explained with references to the embodiment of Figure 1.
  • the system is programmed, step 20.
  • the programming phase the system is taught with examples what an acceptable seam is.
  • the teaching process may include examples of an unaccepta ⁇ ble seam. This is done by using positive and negative references.
  • the programming step may be amended later if the system rejects acceptable objects.
  • the number of segments is also determined in this step. In the example of Figure 1 the number of segments is four. All of the segments are located on the welding seam 11 so that the whole seam is covered. The number and con- figuration of the used lighting conditions is also de ⁇ termined at this step. The number of different light ⁇ ing conditions is represented by N.
  • step 21 After programming, typically a plurality of objects is inspected.
  • the initial lighting is set, step 21.
  • the inspection is done by acquiring images by using a plurality of cameras C, step 22.
  • Typi ⁇ cally all cameras acquire images even if the seam 11 cannot be seen by all of them as it is beneficial to combine other quality inspections to the same measure- ments.
  • Steps 21 and 22 are repeated N times.
  • the lighting is reset dif ⁇ ferently, step 21.
  • the change of lighting provides a different view of the object to be inspected.
  • N rounds the images are processed, step 23.
  • the images are processed, step 23.
  • the maximum number of images acquired is then 6*N. However, it is not necessary to have the same value of N for all cameras.
  • the change of lighting condi ⁇ tions is a particularly good way to produce different images as some of the seam defects show better in dif- ferent lighting and thus it provides a plurality of different images from the same angle.
  • each of the images is processed so that the visible segments in each image are compared to the reference images.
  • the comparison provides a value that indicates how well the imaged segment matches with a reference segment. It must be noted that even if the method is represented here as sequential, the processing may be started in parallel as soon as the first images are acquired.
  • the re ⁇ sult may be simply an indication that the inspected object is allowable or must be rejected.
  • Other indica ⁇ tors are also possible. For example, different quali- ties of the same part may be acceptable in different applications so the result may be an indication of how much the object deviates from the sample.
  • the result may be displayed to the object as a whole or for each segment respectively.
  • the following is an example of the result computing according to the present invention.
  • the computing process the images taken from one segment are combined.
  • the combining step is repeated for each of the segments.
  • a threshold value is determined.
  • Each of the segments is compared to the predetermined threshold value. If at least one of the segments does not exceed the required threshold quality the complete seam is considered to be unac ⁇ ceptable and the object is rejected.
  • the system rejects acceptable objects. When the operator of the system notices this he can reprogram the system and teach an additional example of an acceptable ob ⁇ ject. This is a continuous process and will lead to more accurate quality control. Typically in demanding applications the automatic control is configured with very strict conditions. In these cases more accurate quality control will save money as the acceptable ob ⁇ jects are not wasted.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Biochemistry (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

La présente invention concerne un système et un procédé d'inspection de la qualité d'un objet fabriqué. Le système est conçu pour obtenir une pluralité d'images différentes d'une partie d'un objet à inspecter. La pluralité d'images différentes peut être obtenue à l'aide d'une pluralité d'appareils photo, d'une pluralité de conditions d'éclairage différentes ou d'une combinaison de ces éléments. A partir de la pluralité d'images différentes, il est possible de calculer une valeur représentant la qualité de la partie inspectée.
PCT/FI2012/050213 2011-03-10 2012-03-05 Système de vision de machine permettant un contrôle qualité WO2012120192A2 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP12754525.9A EP2684033A4 (fr) 2011-03-10 2012-03-05 Système de vision de machine permettant un contrôle qualité

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
FI20115241 2011-03-10
FI20115241A FI20115241A0 (fi) 2011-03-10 2011-03-10 Konenäköjärjestelmä laadunvalvontaan

Publications (2)

Publication Number Publication Date
WO2012120192A2 true WO2012120192A2 (fr) 2012-09-13
WO2012120192A3 WO2012120192A3 (fr) 2012-11-08

Family

ID=43806452

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/FI2012/050213 WO2012120192A2 (fr) 2011-03-10 2012-03-05 Système de vision de machine permettant un contrôle qualité

Country Status (3)

Country Link
EP (1) EP2684033A4 (fr)
FI (1) FI20115241A0 (fr)
WO (1) WO2012120192A2 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3040782A1 (fr) * 2015-09-08 2017-03-10 Eurostat Group Dispositif et procede de controle d’une piece thermoformee

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4863268A (en) * 1984-02-14 1989-09-05 Diffracto Ltd. Diffractosight improvements
JPS61293657A (ja) * 1985-06-21 1986-12-24 Matsushita Electric Works Ltd 半田付け外観検査方法
PT836093E (pt) * 1996-10-10 2004-06-30 Elpatronic Ag Processo e dispositivo para a verificacao optica de costuras de soldadura
US6204469B1 (en) * 1999-03-04 2001-03-20 Honda Giken Kogyo Kabushiki Kaisha Laser welding system
US7075565B1 (en) * 2000-06-14 2006-07-11 Landrex Technologies Co., Ltd. Optical inspection system
FI20041414A0 (fi) * 2004-11-03 2004-11-03 Valtion Teknillinen Laserhitsausmenetelmä
WO2006128317A1 (fr) * 2005-06-03 2006-12-07 Elpatronic Ag Procede d'eclairage et systeme d'eclairage
WO2009094489A1 (fr) * 2008-01-23 2009-07-30 Cyberoptics Corporation Système d'inspection optique à haute vitesse avec imagerie par multiples éclairages
WO2009122393A2 (fr) * 2008-03-31 2009-10-08 Brightview Systems Ltd. Procédé et système d'amélioration de rendement de production de cellules photovoltaïques
US20100326962A1 (en) * 2009-06-24 2010-12-30 General Electric Company Welding control system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of EP2684033A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3040782A1 (fr) * 2015-09-08 2017-03-10 Eurostat Group Dispositif et procede de controle d’une piece thermoformee

Also Published As

Publication number Publication date
EP2684033A4 (fr) 2014-10-01
EP2684033A2 (fr) 2014-01-15
FI20115241A0 (fi) 2011-03-10
WO2012120192A3 (fr) 2012-11-08

Similar Documents

Publication Publication Date Title
CN106657983B (zh) 全景摄像机的参数测试方法和装置
JP5109633B2 (ja) 測定方法及び検査方法並びに測定装置及び検査装置
CN116879308A (zh) 一种工业机器视觉系统图像处理方法
CN110207951B (zh) 一种基于视觉的飞机线缆支架装配状态检测方法
TWI444613B (zh) 攝像檢查裝置及攝像檢查方法
Janóczki et al. Automatic optical inspection of soldering
KR101630596B1 (ko) 차량하부 촬영장치 및 이를 운용하는 차량하부 촬영방법
EP4202424A1 (fr) Procédé et système d'inspection de soudures
US20240171838A1 (en) System and method for positioning of a visual production line inspection appliance
JP4834373B2 (ja) Ｘ線検査装置、ｘ線検査方法およびｘ線検査プログラム
US20130230144A1 (en) System and method for automated x-ray inspection
JP2022126587A (ja) 複雑な曲面を有するワークの自動視覚検査装置
JP2018185177A (ja) 画像検査装置、生産システム、画像検査方法、プログラム及び記憶媒体
CN106018415A (zh) 基于显微视觉的微小零件质量检测系统
WO2012120192A2 (fr) Système de vision de machine permettant un contrôle qualité
CN116256366A (zh) 一种芯片缺陷的检测方法、检测系统及存储介质
JP2005283267A (ja) 貫通穴計測装置及び方法並びに貫通穴計測用プログラム
US20080152211A1 (en) Rotating prism component inspection system
JP2000009880A (ja) 燃料集合体検査装置及び検査方法
KR101442666B1 (ko) 복수 행의 조명부재를 포함하는 비전검사장치
JPH1073419A (ja) エンジン外付け部品の誤欠品検査装置
KR20210058329A (ko) 다면 비전 검사 알고리즘 및 이를 이용한 시스템
JP2006177760A (ja) Ｘ線検査装置、ｘ線検査方法およびｘ線検査プログラム
JP7440975B2 (ja) 撮像装置と識別方法
US20220330420A1 (en) Method of verifying fault of inspection unit, inspection apparatus and inspection system

Legal Events

Date Code Title Description
NENP Non-entry into the national phase in:

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2012754525

Country of ref document: EP