WO2016204402A1 - Component defect inspection method, and apparatus therefor - Google Patents

Component defect inspection method, and apparatus therefor Download PDF

Info

Publication number
WO2016204402A1
WO2016204402A1 PCT/KR2016/004634 KR2016004634W WO2016204402A1 WO 2016204402 A1 WO2016204402 A1 WO 2016204402A1 KR 2016004634 W KR2016004634 W KR 2016004634W WO 2016204402 A1 WO2016204402 A1 WO 2016204402A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
dimensional
defect
component
reference image
Prior art date
Application number
PCT/KR2016/004634
Other languages
French (fr)
Korean (ko)
Inventor
윤준혁
김규년
Original Assignee
주식회사 쓰리디산업영상
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 주식회사 쓰리디산업영상 filed Critical 주식회사 쓰리디산업영상
Publication of WO2016204402A1 publication Critical patent/WO2016204402A1/en

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N23/00Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
    • G01N23/02Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
    • G01N23/04Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material

Definitions

  • the present invention relates to a method and apparatus for inspecting a defect of an industrial component, and more particularly, to a method and apparatus for inspecting a defect inside a component based on a three-dimensional image and a two-dimensional transmission image of the component.
  • a conventional general method of identifying defects such as pores, cracks, and missing parts inside various industrial parts is a method of analyzing two-dimensional transmission images such as X-ray images.
  • X-ray images there is a limit in inspecting the 2D transmission image to determine whether there is a defect.
  • the technical problem to be achieved by the present invention is to improve the reliability of defect determination by comparing the two-dimensional transmission image and the defect-free reference image of the part, as well as the component that can easily generate a defect-free reference image used for defect determination.
  • an example of a component defect inspection method includes: obtaining a three-dimensional defect free image of a component; Obtaining a two-dimensional part inspection image; Generating a two-dimensional reference image from the three-dimensional defect free image based on a photographing condition including a photographing angle of the two-dimensional part inspection image; And comparing the two-dimensional component inspection image with the two-dimensional reference image to determine whether a component is defective.
  • another example of a component defect inspection method includes: obtaining a two-dimensional transmission image and a three-dimensional defect free image of a component; Determining whether there is a primary component defect based on an existence of a noise area of a predetermined size or more through image analysis of the 2D transmission image; Generating a defect-free two-dimensional reference image to contrast with all or a portion of the two-dimensional transmitted image from the three-dimensional defect free image based on photographing conditions including a photographing angle of the two-dimensional transmitted image; And comparing the 2D reference image with all or part of the 2D transmission image to determine whether there is a secondary component defect.
  • an example of a component defect inspection apparatus includes a three-dimensional image acquisition unit for acquiring a three-dimensional defect free image of a component; A two-dimensional image acquisition unit obtaining a two-dimensional transmission image of the component; A reference image generator configured to generate a two-dimensional reference image from the three-dimensional defect free image based on shooting conditions including a photographing angle of the two-dimensional transmission image; And a defect determination unit comparing the two-dimensional transmission image and the two-dimensional reference image to determine whether a component is defective.
  • the reliability of defect determination is increased by comparing the two-dimensional transmission image and the defect-free reference image.
  • a defect free reference image can be easily obtained from the 3D image to be contrasted with the 2D transmission image, it is not necessary to separately photograph and store each defect free reference image according to the photographing position or angle of the 2D transmission image.
  • FIG. 1 is a view showing an example of a method for obtaining a two-dimensional transmission image for inspection of component defects according to the present invention
  • FIG. 2 is a view showing an example of a method for obtaining a three-dimensional image for inspection of component defects according to the present invention
  • FIG. 3 is a view showing the configuration of an embodiment of a component defect inspection apparatus according to the present invention.
  • 4A to 4C are views illustrating an example of a 2D transmission image photographed for inspecting a component defect according to the present invention.
  • FIG. 5 is a view showing an example of a three-dimensional defect image photographed for inspection of component defects according to the present invention.
  • FIG. 6 is a view showing an example of a method for generating a two-dimensional reference image from a three-dimensional defect image for inspection of component defects according to the present invention
  • FIG. 7 is a diagram illustrating an example of a 2D reference image generated by the method of FIG. 6;
  • FIG. 8 is a view showing an example of an image comparison method for inspecting component defects according to the present invention.
  • FIG. 9 is a view showing another example of a method for generating a two-dimensional reference image from a three-dimensional defect image for inspection of component defects according to the present invention.
  • FIG. 10 is a diagram illustrating an example of a method of inspecting a defect for a part of a 2D transmission image by comparing with a 2D reference image
  • FIG. 11 is a flowchart illustrating an example of a method for inspecting a component defect according to the present invention
  • FIG. 12 is a flowchart illustrating an example of a method of detecting a suspected component defect region by analyzing a 2D transmission image according to the present invention
  • FIG. 13A to 13E illustrate an example of image processing results of each of the two-dimensional transmission images of FIG. 12;
  • FIG. 14 is a flowchart illustrating an example of a method of generating a 2D reference image for inspecting a component defect according to the present invention
  • 15 is a flowchart illustrating another example of a method of generating a 2D reference image for inspecting a component defect according to the present invention.
  • 16 is a flowchart illustrating another example of a component defect inspection method according to the present invention.
  • FIG. 1 is a view showing an example of a method for obtaining a two-dimensional transmission image for inspection of component defects according to the present invention.
  • the component 100 to be inspected is positioned between a radiator 110 that emits X-rays and a detector 120 that receives transmission signals such as X-rays and records them in a two-dimensional plane. .
  • the detector 120 records X-rays transmitted through the component 100 to be inspected to generate a two-dimensional transmission image (ie, an X-ray image).
  • the photographing positions or the photographing angles of the inspection target part 100 positioned between the radiator 110 and the decker 120 may be different from each other.
  • Photographing conditions such as the photographing position, the photographing angle, the brightness of the 2D transmission image, the contrast, etc. of the inspection target part 100 are used to generate a defect-free reference image for inspecting a component defect according to the present invention.
  • Photographing conditions can be identified through various conventional methods.
  • the inspection target part 100 is fixed by a fixing jig (not shown) and is located between the radiator 110 and the detector 120, and rotates the fixing jig (two or three-dimensional rotation).
  • a two-dimensional transmission image photographed at different photographing angles such as 4a to 4c may be obtained.
  • the user can directly grasp the rotation angle of the fixing jig (that is, the shooting angle) and input it through the user interface of the component defect inspection device, or the sensor can automatically measure the rotation angle of the fixing jig and provide it to the component defect inspection device. have.
  • the inspection target part 100 may have different points located between the radiator 110 and the detector 120 according to the type thereof.
  • the position of the inspection target part directly grasped by the user is input to the component defect inspection device, or various sensors measure each distance between the radiator 110, the inspection object 100, and the detector 120 to automatically inspect the component defect. It can be provided to the device.
  • the present embodiment provides an X-ray image as an example of a 2D transmission image, but is not necessarily limited thereto.
  • the 2D transmission image may be obtained through various conventional methods.
  • a 2D image may be obtained from a 3D image of the component 100 to be inspected.
  • FIG. 2 is a diagram illustrating an example of a method of obtaining a 3D image for inspection of a component defect according to the present invention.
  • the 3D image acquisition apparatus 210 captures a 3D image of the defect-free part 200.
  • the 3D image acquisition apparatus 210 acquires a 3D defect image by photographing a 3D image of the defect component 200 of the same type as the inspection target part of FIG. 1.
  • the three-dimensional defect image is used to create a reference image to determine whether there is a component defect compared to the two-dimensional transmission image of the inspection target component 100 of FIG.
  • 3D image acquisition device is a computerized tomography (CT) device or a magnetic resonance imaging (MRI) device.
  • CT computerized tomography
  • MRI magnetic resonance imaging
  • various 3D image acquisition apparatuses may be used.
  • FIG 3 is a view showing the configuration of an embodiment of a component defect inspection apparatus according to the present invention.
  • the component defect inspection apparatus 300 includes a 2D image acquisition unit 310, a 3D image acquisition unit 320, a reference image generation unit 330, and a defect determination unit 340.
  • the 2D image acquisition unit 310 receives a 2D transmission image of the inspection target part.
  • the 2D transmission image may be photographed through various conventional methods including the salping method in FIG. 1.
  • the 3D image acquisition unit 320 receives a 3D image (hereinafter, 3D flawless image) for the flawless part.
  • the 3D flawless image may be photographed through various conventional methods including the salping method in FIG. 2.
  • the reference image generator 330 generates a 2D reference image to contrast with all or part of the 2D transmission image from the 3D defect free image.
  • the reference image generator 330 generates a 2D reference image by performing a 2D simulation on the 3D flawless image. For example, when the 2D transmission image is an X-ray image and the 3D defect image is a CT image, the reference image generator 330 may perform the 3D defect image using DDR (Digitally Reconstructed Radiography) simulation. Generates a two-dimensional reference image as if it was taken with X-ray.
  • DDR Digitally Reconstructed Radiography
  • the reference image generator 330 may set a virtual X-ray photographing point to generate a 2D reference image of a desired photographing angle. For example, when there are a plurality of two-dimensional transmission images having different shooting angles for the inspection target part as shown in FIGS. 4A to 4C, the reference image generator 330 may perform the two-dimensional transmission images of FIGS. 4A to 4C. Two-dimensional reference images of different shooting angles to be contrasted with each other may be generated. A specific example of generating the reference image is illustrated in FIGS. 6 and 9.
  • the defect determiner 340 determines whether the component is defective by using the 2D transmission image and the 2D reference image. More specifically, the defect determination unit 340 may include a primary determination unit 350 that determines whether there is a primary defect through an image analysis process of the 2D transmission image, and all or part of the 2D transmission image and the 2D reference image. Comparing with the second determination unit 360 to determine whether there is a defect.
  • the defect determination method of the primary determination unit 340 will be described again with reference to FIG. 12, and the defect determination method of the secondary determination unit 370 will be described again with reference to FIGS. 8 and 10.
  • 4A to 4C are diagrams showing an example of a two-dimensional transmission image photographed for inspecting a component defect according to the present invention.
  • FIG. 5 is a diagram illustrating an example of a three-dimensional defect image photographed for component defect inspection according to the present invention.
  • the component defect inspection apparatus acquires a three-dimensional image 500 of a defective component through the method of FIG. 2.
  • Two-dimensional transmission images of the inspection target part may be photographed at different photographing angles according to inspection sites.
  • two-dimensional reference images of different defects are required for each photographing angle.
  • the component defect inspection device without the need to photograph and store the two-dimensional reference image to be contrasted with the two-dimensional transmission image in accordance with the photographing angle, the component defect inspection device easily simulates the two-dimensional reference image of the desired photographing angle through the simulation from the three-dimensional defect free image Can be generated.
  • FIG. 6 is a diagram illustrating an example of a method of generating a 2D reference image from a 3D defect image for inspecting a component defect according to the present invention
  • FIG. 7 is a 2D reference image generated by the method of FIG. 6.
  • 1 is a diagram illustrating an example.
  • the component defect inspection apparatus performs a two-dimensional simulation on a three-dimensional defect image based on shooting conditions including a photographing angle of a two-dimensional transmission image of a component to be inspected, and a two-dimensional reference image.
  • the component defect inspection apparatus sets the virtual photographing point 600 for the two-dimensional simulation of the three-dimensional flawless image based on the photographing conditions such as the photographing distance and the photographing angle of the two-dimensional transmission image.
  • the component defect inspection apparatus transmits the virtual X-ray emitted from the virtual photographing point 600 to the virtual three-dimensional component obtained by rotating the three-dimensional defect free image according to the photographing angle of the two-dimensional transmission image.
  • a two-dimensional reference image 620 is obtained by applying a two-dimensional simulation process detected by the 610.
  • the component defect inspection apparatus may adjust the brightness and contrast of the 2D reference image based on photographing conditions such as the brightness and contrast of the 2D transmission image of the inspection target part.
  • FIG. 8 is a diagram illustrating an example of an image comparison method for inspecting component defects according to the present invention.
  • the component defect inspection apparatus acquires a 2D transmission image 810 of a component to be inspected.
  • the component defect inspection apparatus generates a flawless two-dimensional reference image 800 to be contrasted with the two-dimensional transmission image 810 by using the method of FIG.
  • the component defect inspection apparatus compares the 2D transmission image 810 and the defect free 2D reference image 800 to determine whether there is a defect. For example, the component defect inspection apparatus determines that a component defect exists if there is a difference between the 2D transmission image 810 and the 2D reference image 800.
  • the component defect inspection apparatus may determine whether the component is defective by determining the identity of the image by using a vector comparison method between the two images 800 and 810. For example, a component defect inspection apparatus generates two images as gray images, and then uses a normalized cross-correlation (NCC) method to determine whether the images match by using the correlation between the pixel values of the two gray images. It can be used, represented by the equation as follows.
  • NCC normalized cross-correlation
  • f (x, y) and t (x, y) represent pixel values of coordinates (x, y) of each image
  • n is a total number of pixels
  • f ', t' is an average of pixel values of each image.
  • the values ⁇ f and ⁇ t represent standard deviations of pixel values of two images.
  • the component defect inspection apparatus may use various conventional image comparison methods.
  • FIG. 9 is a diagram illustrating another example of a method of generating a 2D reference image from a 3D defect image for inspecting a component defect according to the present invention.
  • the component defect inspection apparatus does not generate a 2D reference image to contrast with the entire 2D transmission image, but generates a 2D reference image to contrast with a part of the 2D transmission image.
  • the component defect inspection apparatus may identify a noise region having a predetermined size or more as a defect suspect region 900 through an image analysis process on a 2D transmission image of a component to be inspected, and compare it with the defect suspect region 900. Create a dimensional reference image. The method of identifying the suspected defect site is shown in FIG.
  • the component defect inspection apparatus detects the noise region identified in the 2D transmission image, that is, the 3D region of the component corresponding to the defect suspected portion 900. Since the two-dimensional transmission image is an image obtained by photographing the inspection target part 920 located between the X-ray output point 940 and the detector 910 of the radiator, the component defect inspection apparatus may include a noise region ( The 3D subregion 930 in the inspection target part formed by connecting the corresponding region 9150 of the detector corresponding to 900 and the X-ray output point 940 is identified.
  • the component defect inspection apparatus generates two-dimensional reference image by performing two-dimensional simulation only on the three-dimensional sub-region 930 of the component among the three-dimensional defect-free images rather than performing two-dimensional simulation on the entire three-dimensional defect-free image. In addition to shortening the time required for the circuit, it is possible to perform a more accurate image comparison in the noise region where the defect is suspected.
  • FIG. 10 is a diagram illustrating an example of a method for inspecting a defect of a part of a 2D transmission image by comparing with a 2D reference image.
  • the apparatus for inspecting a component defect does not compare the entire 2D transmission image of the part with the 2D reference image, but includes a 2D component inspection image including a noise region in which a defect is suspected in the 2D transmission image. 1010 and generates a two-dimensional reference image 1000 corresponding to the two-dimensional part inspection area, and then compares the two-dimensional part inspection area 1010 and the two-dimensional reference image 1000 to determine whether there is a defect. .
  • An image comparison method for determining whether there is a defect is as described with reference to FIG. 8. Referring to the present embodiment, since different portions 1005 and 1015 exist in the two images 1000 and 1010, the component defect inspection apparatus determines that there is a component defect.
  • FIG. 11 is a flowchart illustrating an example of a component defect inspection method according to the present invention.
  • the component defect inspection apparatus receives a 3D flawless image of a component without a defect and a 2D component inspection image of a component to be inspected (S1100 and S1110).
  • the two-dimensional part inspection image may be an entirety of the two-dimensional transmission image photographing the inspection target part or an image of a partial region in which a defect is suspected.
  • the component defect inspection apparatus generates a two-dimensional reference image of a defect-free image, which may be contrasted with a two-dimensional component inspection image, based on photographing conditions of the two-dimensional transmission image (S1120). Then, the component defect inspection apparatus compares the two-dimensional component inspection image with a defect-free two-dimensional reference image, and determines whether there is a defect based on whether there is a different portion (S1130).
  • FIG. 12 is a flowchart illustrating an example of a method for detecting a suspected component defect region by analyzing a two-dimensional transmission image according to the present invention
  • FIGS. 13A to 13E are image processing results of the two-dimensional transmission image of each step of FIG. 12.
  • 1 is a diagram illustrating an example.
  • the component defect inspection apparatus obtains a two-dimensional transmission image (FIG. 13A) of the inspection target component through the same method as in FIG. 1 (S1200).
  • the component defect inspection apparatus masks the component region of the two-dimensional transmission image (FIG. 13A) to remove a background region other than the component (S1210), and corrects an image including a small hole in the remaining component region.
  • FIG. 13C is generated (S1220). Closing can be used as a way to fill small holes.
  • the component defect inspection apparatus generates a noise image (FIG. 13D) obtained by subtracting the original two-dimensional transmission image (FIG. 13A) from the corrected image (FIG. 13C) (S1230). Only small holes and noise pixels remain in the noise image (FIG. 13D) generated through the subtraction between the two images.
  • the component defect inspection apparatus removes general noise not related to a defect by applying a median filter to the noise image (FIG. 13D) (S1240).
  • the component defect inspection apparatus converts the noise image (FIG. 13D) into a binary image of 0 and 1 (S1250).
  • the user may create a binary image by setting a value such as brightness of each pixel based on a predetermined threshold value to 1 when the value of the pixel is greater than the threshold value and 0 when the value is smaller than the threshold value.
  • the component defect inspection apparatus connects small island regions separated from each other in a binary image by using a label labeling technique, and then detects a region in which a single region of a single blob is larger than a predetermined size. It is determined as 1300 of FIG. 13E (S1260).
  • One or more defect suspect regions 1300 may exist in some cases, and a sub image including the suspect defect regions may be set as a two-dimensional component inspection region to be compared with a two-dimensional reference image.
  • FIG. 14 is a flowchart illustrating an example of a method of generating a 2D reference image for component defect inspection according to the present invention.
  • the component defect inspection apparatus detects photographing conditions including a photographing position, an photographing angle, and the like, of a 2D transmission image of a component to be inspected.
  • the shooting conditions are automatically detected when the 2D transmission image is taken using the method as shown in FIG. 1 and provided to the component defect inspection apparatus, or the user may directly input the shooting conditions through the user interface of the component defect inspection apparatus. .
  • the component defect inspection apparatus sets a virtual photographing point to perform a simulation on the 3D flawless image based on the photographing conditions (S1410).
  • the component defect inspection apparatus generates a 2D reference image by performing a DDR simulation or the like based on a 3D defectless image rotated by a photographing angle of a 2D transmission image at a virtual photographing point (S1420).
  • 15 is a flowchart illustrating another example of a method of generating a 2D reference image for inspecting a component defect according to the present invention.
  • the component defect inspection apparatus does not perform a simulation process for the entire 3D defect-free image but performs a simulation process only for a specific region. To this end, the component defect inspection apparatus detects a defect suspect region of the two-dimensional transmission image through the method as shown in FIG. 12, and sets the detected defect suspect region as a two-dimensional component inspection image for defect inspection.
  • the component defect inspection apparatus detects a three-dimensional region of the component corresponding to the two-dimensional component inspection image (S1500).
  • the two-dimensional part inspection image is a three-dimensional sub-region of a part that is divided by a position 915 corresponding to the two-dimensional part inspection area of the X-ray output point 940 and the detector 910 as shown in FIG. 9. Figure out.
  • the component defect inspection apparatus generates a 2D reference image by performing a simulation process only on the 3D sub-region of the component among the 3D defect images (S1520).
  • 16 is a flowchart illustrating another example of a component defect inspection method according to the present invention.
  • the component defect inspection apparatus acquires a 2D transmission image of a component to be inspected (S1600).
  • the component defect inspection apparatus performs an image analysis process as shown in FIG. 12 on the 2D transmission image to determine whether a defect suspect region exists.
  • the component defect inspection apparatus may determine the noise region having a predetermined size or more present in the 2D transmission image as the defect suspect region 1300 as illustrated in FIG. 13E.
  • the component defect inspection apparatus If there is a suspected defect area (S1620), the component defect inspection apparatus generates a defect-free two-dimensional reference image to contrast with all or a part of the two-dimensional transmission image from the three-dimensional defect free image (S1630). Since the photographing angle of the two-dimensional transmission image is different from each other depending on which part of the part is inspected, the component defect inspection apparatus uses the photographing conditions including the photographing angle of the two-dimensional transmission image and the like. A two-dimensional reference image of the same photographing angle is generated.
  • the component defect inspection apparatus compares the whole or part of the 2D transmission image with the 2D reference image to determine whether there is a defect (S1640).
  • the invention can also be embodied as computer readable code on a computer readable recording medium.
  • the computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like.
  • the computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

Landscapes

  • Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)
  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)

Abstract

Disclosed are a component defect inspection method and an apparatus therefor. A component defect inspection apparatus acquires a three-dimensional flawless image of a component and a two-dimensional component inspection image; generates a two-dimensional reference image from the three-dimensional flawless image on the basis of photographing conditions including the photographing angle of the two-dimensional component inspection image; and determines whether the component has a defect by comparing the two-dimensional component inspection image and the two-dimensional reference image.

Description

부품 결함 검사 방법 및 그 장치Component defect inspection method and device
본 발명은 산업용 부품의 결함을 검사하는 방법 및 그 장치에 관한 것으로서, 보다 상세하게는 부품의 3차원 영상과 2차원 투과 영상을 기초로 부품 내부의 결함을 검사하는 방법 및 그 장치에 관한 것이다. The present invention relates to a method and apparatus for inspecting a defect of an industrial component, and more particularly, to a method and apparatus for inspecting a defect inside a component based on a three-dimensional image and a two-dimensional transmission image of the component.
각종 산업용 부품 내부의 기공, 크랙, 부품 누락 등의 결함을 파악하는 종래의 일반적인 방법은 엑스레이(X-ray) 영상과 같은 2차원 투과 영상을 분석하는 방법이다. 그러나 2차원 투과 영상을 검사자가 일일이 확인하여 결함 여부를 파악하는 데에는 한계가 있다.A conventional general method of identifying defects such as pores, cracks, and missing parts inside various industrial parts is a method of analyzing two-dimensional transmission images such as X-ray images. However, there is a limit in inspecting the 2D transmission image to determine whether there is a defect.
본 발명이 이루고자 하는 기술적 과제는, 부품의 2차원 투과 영상과 무결점 기준 영상의 비교를 통해 결함 여부 판단의 신뢰성을 높일 뿐만 아니라, 결함 여부 판단에 사용되는 무결점 기준 영상을 용이하게 생성할 수 있는 부품 결함 검사 방법 및 그 장치를 제공하는 데 있다.The technical problem to be achieved by the present invention is to improve the reliability of defect determination by comparing the two-dimensional transmission image and the defect-free reference image of the part, as well as the component that can easily generate a defect-free reference image used for defect determination A defect inspection method and apparatus are provided.
상기의 기술적 과제를 달성하기 위한, 본 발명에 따른 부품 결함 검사 방법의 일 예는, 부품의 3차원 무결점 영상을 획득하는 단계; 2차원 부품 검사 영상을 획득하는 단계; 상기 2차원 부품 검사 영상의 촬영각도를 포함한 촬영조건을 기초로, 상기 3차원 무결점 영상으로부터 2차원 기준 영상을 생성하는 단계; 및 상기 2차원 부품 검사 영상과 상기 2차원 기준 영상을 비교하여 부품의 결함 여부를 판단하는 단계;를 포함한다.In order to achieve the above technical problem, an example of a component defect inspection method according to the present invention includes: obtaining a three-dimensional defect free image of a component; Obtaining a two-dimensional part inspection image; Generating a two-dimensional reference image from the three-dimensional defect free image based on a photographing condition including a photographing angle of the two-dimensional part inspection image; And comparing the two-dimensional component inspection image with the two-dimensional reference image to determine whether a component is defective.
상기의 기술적 과제를 달성하기 위한, 본 발명에 따른 부품 결함 검사 방법의 다른 예는, 부품의 2차원 투과 영상과 3차원 무결점 영상을 획득하는 단계; 상기 2차원 투과 영상의 이미지 분석을 통해 일정 크기 이상의 잡음 영역의 존부를 기초로 1차 부품 결함 여부를 판단하는 단계; 상기 2차원 투과 영상의 촬영 각도를 포함하는 촬영조건을 기초로, 상기 3차원 무결점 영상으로부터 상기 2차원 투과 영상의 전부 또는 일부와 대비할 무결점의 2차원 기준 영상을 생성하는 단계; 및 상기 2차원 투과 영상의 전부 또는 일부와 상기 2차원 기준 영상을 비교하여 2차 부품 결함 여부를 판단하는 단계;를 포함한다.In order to achieve the above technical problem, another example of a component defect inspection method according to the present invention includes: obtaining a two-dimensional transmission image and a three-dimensional defect free image of a component; Determining whether there is a primary component defect based on an existence of a noise area of a predetermined size or more through image analysis of the 2D transmission image; Generating a defect-free two-dimensional reference image to contrast with all or a portion of the two-dimensional transmitted image from the three-dimensional defect free image based on photographing conditions including a photographing angle of the two-dimensional transmitted image; And comparing the 2D reference image with all or part of the 2D transmission image to determine whether there is a secondary component defect.
상기의 기술적 과제를 달성하기 위한, 본 발명에 따른 부품 결함 검사 장치의 일 예는, 부품의 3차원 무결점 영상을 획득하는 3차원 영상 획득부; 부품의 2차원 투과 영상을 획득하는 2차원 영상 획득부; 상기 2차원 투과 영상의 촬영각도를 포함한 촬영조건을 기초로, 상기 3차원 무결점 영상으로부터 2차원 기준 영상을 생성하는 기준영상 생성부; 및 상기 2차원 투과 영상과 상기 2차원 기준 영상을 비교하여 부품의 결함 여부를 판단하는 결함 판단부;를 포함한다.In order to achieve the above technical problem, an example of a component defect inspection apparatus according to the present invention includes a three-dimensional image acquisition unit for acquiring a three-dimensional defect free image of a component; A two-dimensional image acquisition unit obtaining a two-dimensional transmission image of the component; A reference image generator configured to generate a two-dimensional reference image from the three-dimensional defect free image based on shooting conditions including a photographing angle of the two-dimensional transmission image; And a defect determination unit comparing the two-dimensional transmission image and the two-dimensional reference image to determine whether a component is defective.
본 발명에 따르면, 2차원 투과 영상과 무결점 기준 영상과의 비교를 통해 결함 여부 판단의 신뢰성을 높인다. 또한 2차원 투과 영상과 대비할 무결점 기준 영상을 3차원 영상으로부터 용이하게 얻을 수 있어, 2차원 투과 영상의 촬영 위치나 각도에 따른 각각의 무결점 기준 영상을 별도로 촬영하여 저장할 필요가 없다.According to the present invention, the reliability of defect determination is increased by comparing the two-dimensional transmission image and the defect-free reference image. In addition, since a defect free reference image can be easily obtained from the 3D image to be contrasted with the 2D transmission image, it is not necessary to separately photograph and store each defect free reference image according to the photographing position or angle of the 2D transmission image.
도 1은 본 발명에 따른 부품 결함 검사를 위한 2차원 투과 영상을 획득하는 방법의 일 예를 도시한 도면,1 is a view showing an example of a method for obtaining a two-dimensional transmission image for inspection of component defects according to the present invention;
도 2는 본 발명에 따른 부품 결함 검사를 위한 3차원 영상을 획득하는 방법의 일 예를 도시한 도면,2 is a view showing an example of a method for obtaining a three-dimensional image for inspection of component defects according to the present invention;
도 3은 본 발명에 따른 부품 결함 검사 장치의 일 실시 예의 구성을 도시한 도면,3 is a view showing the configuration of an embodiment of a component defect inspection apparatus according to the present invention,
도 4a 내지 도4c는 본 발명에 따른 부품 결함 검사를 위하여 촬영한 2차원 투과 영상의 일 예를 도시한 도면,4A to 4C are views illustrating an example of a 2D transmission image photographed for inspecting a component defect according to the present invention;
도 5는 본 발명에 따른 부품 결함 검사를 위하여 촬영한 3차원 무결함 영상의 일 예를 도시한 도면,5 is a view showing an example of a three-dimensional defect image photographed for inspection of component defects according to the present invention;
도 6은 본 발명에 따른 부품 결함 검사를 위하여 3차원 무결함 영상으로부터 2차원 기준 영상을 생성하는 방법의 일 예를 도시한 도면, 6 is a view showing an example of a method for generating a two-dimensional reference image from a three-dimensional defect image for inspection of component defects according to the present invention;
도 7은 도 6의 방법을 통해 생성한 2차원 기준 영상의 일 예를 도시한 도면,FIG. 7 is a diagram illustrating an example of a 2D reference image generated by the method of FIG. 6;
도 8은 본 발명에 따른 부품 결함 검사를 위한 영상 비교 방법의 일 예를 도시한 도면,8 is a view showing an example of an image comparison method for inspecting component defects according to the present invention;
도 9는 본 발명에 따른 본 발명에 따른 부품 결함 검사를 위하여 3차원 무결함 영상으로부터 2차원 기준 영상을 생성하는 방법의 다른 일 예를 도시한 도면,9 is a view showing another example of a method for generating a two-dimensional reference image from a three-dimensional defect image for inspection of component defects according to the present invention;
도 10은 2차원 투과 영상의 일부에 대해 2차원 기준 영상과 비교하여 결함을 검사하는 방법의 일 예를 도시한 도면,FIG. 10 is a diagram illustrating an example of a method of inspecting a defect for a part of a 2D transmission image by comparing with a 2D reference image; FIG.
도 11은 본 발명에 따른 부품 결함 검사 방법의 일 예를 도시한 흐름도,11 is a flowchart illustrating an example of a method for inspecting a component defect according to the present invention;
도 12는 본 발명에 따른 2차원 투과 영상을 분석하여 부품 결함 의심 영역을 검출하는 방법의 일 예를 도시한 흐름도, 12 is a flowchart illustrating an example of a method of detecting a suspected component defect region by analyzing a 2D transmission image according to the present invention;
도 13a 내지 도 13e는 도 12의 각 단계별 2차원 투과 영상의 이미지 처리 결과의 일 예를 도시한 도면,13A to 13E illustrate an example of image processing results of each of the two-dimensional transmission images of FIG. 12;
도 14는 본 발명에 따른 부품 결함 검사를 위한 2차원 기준 영상을 생성하는 방법의 일 예를 도시한 흐름도,14 is a flowchart illustrating an example of a method of generating a 2D reference image for inspecting a component defect according to the present invention;
도 15는 본 발명에 따른 부품 결함 검사를 위한 2차원 기준 영상을 생성하는 방법의 다른 일 예를 도시한 흐름도, 그리고,15 is a flowchart illustrating another example of a method of generating a 2D reference image for inspecting a component defect according to the present invention;
도 16은 본 발명에 따른 부품 결함 검사 방법의 다른 일 예를 도시한 흐름도이다.16 is a flowchart illustrating another example of a component defect inspection method according to the present invention.
이하에서, 첨부된 도면들을 참조하여 본 발명에 따른 부품 결함 검사 방법 및 그 장치에 대해 상세히 설명한다. Hereinafter, with reference to the accompanying drawings will be described in detail with respect to the component defect inspection method and apparatus according to the present invention.
도 1은 본 발명에 따른 부품 결함 검사를 위한 2차원 투과 영상을 획득하는 방법의 일 예를 도시한 도면이다.1 is a view showing an example of a method for obtaining a two-dimensional transmission image for inspection of component defects according to the present invention.
도 1을 참조하면, 검사 대상 부품(100)은 X-ray 등을 방출하는 방사장치(110)와 X-ray 등의 투과 신호를 수신하여 2차원 평면에 기록하는 디텍터(120) 사이에 위치한다. 디텍터(120)는 검사 대상 부품(100)을 투과한 X-ray 등을 기록하여 2차원 투과 영상(즉, X-ray 영상)을 생성한다. Referring to FIG. 1, the component 100 to be inspected is positioned between a radiator 110 that emits X-rays and a detector 120 that receives transmission signals such as X-rays and records them in a two-dimensional plane. . The detector 120 records X-rays transmitted through the component 100 to be inspected to generate a two-dimensional transmission image (ie, an X-ray image).
2차원 투과 영상은 방사장치(110)의 X-ray 튜브 전압이나 디텍터(120)의 노출 시간 등의 촬영조건에 따라 밝기, 컨트라스트(contrast) 등이 변화한다. 또한 도 4a 내지 도 4c와 같이, 부품의 검사 부위(400,410,420)에 따라 방사장치(110)와 디덱터(120) 사이에 위치한 검사 대상 부품(100)의 촬영 위치나 촬영 각도는 서로 다를 수 있다.In the 2D transmission image, brightness, contrast, and the like change according to imaging conditions such as the X-ray tube voltage of the radiator 110 and the exposure time of the detector 120. In addition, as shown in FIGS. 4A to 4C, depending on the inspection parts 400, 410, and 420 of the parts, the photographing positions or the photographing angles of the inspection target part 100 positioned between the radiator 110 and the decker 120 may be different from each other.
검사 대상 부품(100)의 촬영 위치, 촬영 각도, 2차원 투과 영상의 밝기, 컨트라스트 등과 같은 다양한 촬영조건은 이후 살펴볼 본 발명에 따른 부품 결함 검사를 위한 무결점의 기준 영상을 생성하는데 활용된다. 촬영조건은 종래의 다양한 방법을 통해 파악될 수 있다. Various photographing conditions such as the photographing position, the photographing angle, the brightness of the 2D transmission image, the contrast, etc. of the inspection target part 100 are used to generate a defect-free reference image for inspecting a component defect according to the present invention. Photographing conditions can be identified through various conventional methods.
예를 들어, 검사 대상 부품(100)은 고정지그(미도시)에 의해 고정되어 방사장치(110)와 디텍터(120) 사이에 위치하며, 고정지그를 회전(2차원 또는 3차원 회전)시켜 도 4a 내지 도 4c와 같은 서로 다른 촬영각도에서 촬영된 2차원 투과 영상을 얻을 수 있다. 이때 사용자가 고정지그의 회전각도(즉, 촬영각도)를 직접 파악하고 부품 결함 검사 장치의 사용자 인터페이스를 통해 이를 입력하거나, 센서가 고정지그의 회전각도를 자동 측정하여 부품 결함 검사 장치에 제공할 수 있다.For example, the inspection target part 100 is fixed by a fixing jig (not shown) and is located between the radiator 110 and the detector 120, and rotates the fixing jig (two or three-dimensional rotation). A two-dimensional transmission image photographed at different photographing angles such as 4a to 4c may be obtained. At this time, the user can directly grasp the rotation angle of the fixing jig (that is, the shooting angle) and input it through the user interface of the component defect inspection device, or the sensor can automatically measure the rotation angle of the fixing jig and provide it to the component defect inspection device. have.
또 다른 예로, 검사 대상 부품(100)은 그 종류에 따라 방사장치(110)와 디텍터(120) 사이에 위치한 지점이 서로 다를 수 있다. 사용자가 직접 파악한 검사 대상 부품의 위치를 부품 결함 검사 장치에 입력하거나, 각종 센서가 방사장치(110), 검사 대상 부품(100), 디텍터(120) 사이의 각 거리를 측정하여 자동으로 부품 결함 검사 장치에 제공할 수 있다. As another example, the inspection target part 100 may have different points located between the radiator 110 and the detector 120 according to the type thereof. The position of the inspection target part directly grasped by the user is input to the component defect inspection device, or various sensors measure each distance between the radiator 110, the inspection object 100, and the detector 120 to automatically inspect the component defect. It can be provided to the device.
본 실시 예는 2차원 투과 영상의 일 예로 X-ray 영상을 제시하고 있으나 반드시 이에 한정되는 것은 아니며, 종래의 다양한 방법을 통해 2차원 투과 영상을 획득할 수 있다. 또한 실시 예에 따라 검사 대상 부품(100)의 3차원 영상으로부터 2차원 영상을 획득할 수도 있다. The present embodiment provides an X-ray image as an example of a 2D transmission image, but is not necessarily limited thereto. The 2D transmission image may be obtained through various conventional methods. In addition, according to an embodiment, a 2D image may be obtained from a 3D image of the component 100 to be inspected.
도 2는 본 발명에 따른 부품 결함 검사를 위한 3차원 영상을 획득하는 방법의 일 예를 도시한 도면이다.2 is a diagram illustrating an example of a method of obtaining a 3D image for inspection of a component defect according to the present invention.
도 2를 참조하면, 3차원 영상 획득 장치(210)는 무결함 부품(200)의 3차원 영상을 촬영한다. 예를 들어, 3차원 영상 획득 장치(210)는 도 1의 검사 대상 부품과 동종의 무결함 부품(200)에 대한 3차원 영상을 촬영하여 3차원 무결함 영상을 획득한다. 3차원 무결함 영상은 도 1의 검사 대상 부품(100)에 대한 2차원 투과 영상과 비교하여 부품 결함 여부를 판단할 기준 영상을 만드는데 활용된다. Referring to FIG. 2, the 3D image acquisition apparatus 210 captures a 3D image of the defect-free part 200. For example, the 3D image acquisition apparatus 210 acquires a 3D defect image by photographing a 3D image of the defect component 200 of the same type as the inspection target part of FIG. 1. The three-dimensional defect image is used to create a reference image to determine whether there is a component defect compared to the two-dimensional transmission image of the inspection target component 100 of FIG.
3차원 영상 획득 장치의 일 예로, 컴퓨터 단층 촬영(CT, Computered Tomography) 장치나 MRI(Magnetic resonance imaging) 장치 등이 있다. 이 외에도 다양한 3차원 영상 획득 장치가 이용될 수 있다. One example of a 3D image acquisition device is a computerized tomography (CT) device or a magnetic resonance imaging (MRI) device. In addition, various 3D image acquisition apparatuses may be used.
도 3은 본 발명에 따른 부품 결함 검사 장치의 일 실시 예의 구성을 도시한 도면이다.3 is a view showing the configuration of an embodiment of a component defect inspection apparatus according to the present invention.
도 3을 참조하면, 부품 결함 검사 장치(300)는 2차원 영상 획득부(310), 3차원 영상 획득부(320), 기준 영상 생성부(330), 결함 판단부(340)를 포함한다.Referring to FIG. 3, the component defect inspection apparatus 300 includes a 2D image acquisition unit 310, a 3D image acquisition unit 320, a reference image generation unit 330, and a defect determination unit 340.
2차원 영상 획득부(310)는 검사 대상 부품의 2차원 투과 영상을 입력받는다. 2차원 투과 영상은 도 1에서 살핀 방법을 포함한 종래의 다양한 방법을 통해 촬영될 수 있다. The 2D image acquisition unit 310 receives a 2D transmission image of the inspection target part. The 2D transmission image may be photographed through various conventional methods including the salping method in FIG. 1.
3차원 영상 획득부(320)는 무결점 부품에 대한 3차원 영상(이하, 3차원 무결점 영상)을 입력받는다. 3차원 무결점 영상은 도 2에서 살핀 방법을 포함한 종래의 다양한 방법을 통해 촬영될 수 있다. The 3D image acquisition unit 320 receives a 3D image (hereinafter, 3D flawless image) for the flawless part. The 3D flawless image may be photographed through various conventional methods including the salping method in FIG. 2.
기준 영상 생성부(330)는 3차원 무결점 영상으로부터 2차원 투과 영상의 전부 또는 일부와 대비할 2차원 기준 영상을 생성한다. 기준 영상 생성부(330)는 3차원 무결점 영상에 대한 2차원 시뮬레이션을 수행하여 2차원 기준 영상을 생성한다. 예를 들어, 2차원 투과 영상이 X-ray 영상이고, 3차원 무결점 영상이 CT 영상인 경우에, 기준 영상 생성부(330)는 3차원 무결점 영상을 DDR(Digitally Reconstructed Radiography) 시뮬레이션을 이용하여 마치 X-ray로 촬영한 것과 같은 2차원 기준 영상을 생성한다. The reference image generator 330 generates a 2D reference image to contrast with all or part of the 2D transmission image from the 3D defect free image. The reference image generator 330 generates a 2D reference image by performing a 2D simulation on the 3D flawless image. For example, when the 2D transmission image is an X-ray image and the 3D defect image is a CT image, the reference image generator 330 may perform the 3D defect image using DDR (Digitally Reconstructed Radiography) simulation. Generates a two-dimensional reference image as if it was taken with X-ray.
기준 영상 생성부(330)는 가상의 X-ray 촬영 지점을 설정하여 원하는 촬영각도의 2차원 기준 영상을 생성할 수 있다. 예를 들어, 도 4a 내지 도 4c와 같이 검사 대상 부품에 대한 촬영각도가 다른 여러 장의 2차원 투과 영상이 존재하는 경우에, 기준 영상 생성부(330)는 도 4a 내지 도 4c의 2차원 투과 영상과 대비할 서로 다른 촬영각도의 2차원 기준 영상을 각각 생성할 수 있다. 기준 영상 생성의 구체적인 일 예는 도 6 및 도 9에 도시되어 있다. The reference image generator 330 may set a virtual X-ray photographing point to generate a 2D reference image of a desired photographing angle. For example, when there are a plurality of two-dimensional transmission images having different shooting angles for the inspection target part as shown in FIGS. 4A to 4C, the reference image generator 330 may perform the two-dimensional transmission images of FIGS. 4A to 4C. Two-dimensional reference images of different shooting angles to be contrasted with each other may be generated. A specific example of generating the reference image is illustrated in FIGS. 6 and 9.
결함 판단부(340)는 2차원 투과 영상과 2차원 기준 영상을 이용하여 부품의 결함 여부를 판단한다. 보다 구체적으로 결함 판단부(340)는 2차원 투과 영상의 이미지 분석과정을 통해 1차 결함 여부를 판단하는 1차 판단부(350)와, 2차원 투과 영상의 전부 또는 일부와 2차원 기준 영상을 비교하여 결함 여부를 판단하는 2차 판단부(360)를 포함한다. 1차 판단부(340)의 결함 판단 방법은 도 12에서 다시 살펴보며, 2차 판단부(370)의 결함 판단 방법은 도 8 및 도 10에서 다시 살펴본다. The defect determiner 340 determines whether the component is defective by using the 2D transmission image and the 2D reference image. More specifically, the defect determination unit 340 may include a primary determination unit 350 that determines whether there is a primary defect through an image analysis process of the 2D transmission image, and all or part of the 2D transmission image and the 2D reference image. Comparing with the second determination unit 360 to determine whether there is a defect. The defect determination method of the primary determination unit 340 will be described again with reference to FIG. 12, and the defect determination method of the secondary determination unit 370 will be described again with reference to FIGS. 8 and 10.
도 4a 내지 도4c는 본 발명에 따른 부품 결함 검사를 위하여 촬영한 2차원 투과 영상의 일 예를 도시한 도면이다. 4A to 4C are diagrams showing an example of a two-dimensional transmission image photographed for inspecting a component defect according to the present invention.
도 4를 참조하면, 동일한 부품이라고 할지라도 검사 대상 부위(400,410,420)에 따라 2차원 투과 영상의 촬영각도 등의 촬영 조건이 변경된다. 따라서 결함 판단을 위한 무결점의 2차원 기준 영상은 2차원 투과 영상의 촬영 조건에 따라 각각 생성되어야 한다. Referring to FIG. 4, even if the same components are used, imaging conditions such as the photographing angle of the 2D transmission image are changed according to the inspection target portions 400, 410, and 420. Therefore, a flawless two-dimensional reference image for defect determination must be generated according to shooting conditions of the two-dimensional transmission image.
도 5는 본 발명에 따른 부품 결함 검사를 위하여 촬영한 3차원 무결함 영상의 일 예를 도시한 도면이다.5 is a diagram illustrating an example of a three-dimensional defect image photographed for component defect inspection according to the present invention.
도 5를 참조하면, 부품 결함 검사 장치는 도 2의 방법 등을 통해 무결함 부품의 3차원 영상(500)을 획득한다. 검사 대상 부품의 2차원 투과 영상은 검사 부위에 따라 서로 다른 촬영 각도에서 여러 장 촬영될 수 있다. 이 경우 각 촬영 각도에 따른 서로 다른 무결점의 2차원 기준 영상이 필요하다. 본 실시 예에서 2차원 투과 영상과 대비할 2차원 기준 영상을 촬영각도에 따라 일일이 촬영하여 저장할 필요없이, 부품 결함 검사 장치는 3차원 무결점 영상으로부터 시뮬레이션을 통해 원하는 촬영각도의 2차원 기준 영상을 용이하게 생성할 수 있다.Referring to FIG. 5, the component defect inspection apparatus acquires a three-dimensional image 500 of a defective component through the method of FIG. 2. Two-dimensional transmission images of the inspection target part may be photographed at different photographing angles according to inspection sites. In this case, two-dimensional reference images of different defects are required for each photographing angle. In this embodiment, without the need to photograph and store the two-dimensional reference image to be contrasted with the two-dimensional transmission image in accordance with the photographing angle, the component defect inspection device easily simulates the two-dimensional reference image of the desired photographing angle through the simulation from the three-dimensional defect free image Can be generated.
도 6은 본 발명에 따른 부품 결함 검사를 위하여 3차원 무결함 영상으로부터 2차원 기준 영상을 생성하는 방법의 일 예를 도시한 도면이고, 도 7은 도 6의 방법을 통해 생성한 2차원 기준 영상의 일 예를 도시한 도면이다.6 is a diagram illustrating an example of a method of generating a 2D reference image from a 3D defect image for inspecting a component defect according to the present invention, and FIG. 7 is a 2D reference image generated by the method of FIG. 6. 1 is a diagram illustrating an example.
도 6 및 도 7을 참조하면, 부품 결함 검사 장치는 검사 대상 부품의 2차원 투과 영상의 촬영 각도 등을 포함한 촬영 조건을 기초로 3차원 무결함 영상에 대한 2차원 시뮬레이션을 수행하여 2차원 기준 영상을 생성한다.6 and 7, the component defect inspection apparatus performs a two-dimensional simulation on a three-dimensional defect image based on shooting conditions including a photographing angle of a two-dimensional transmission image of a component to be inspected, and a two-dimensional reference image. Create
보다 구체적으로, 부품 결함 검사 장치는 2차원 투과 영상의 촬영거리, 촬영각도 등의 촬영조건을 기초로 3차원 무결점 영상의 2차원 시뮬레이션을 위한 가상의 촬영지점(600)을 설정한다. 그리고 부품 결함 검사 장치는 2차원 투과 영상의 촬영각도에 따라 3차원 무결점 영상을 회전시켜 얻은 가상의 3차원 부품에 대해, 가상 촬영지점(600)에서 방출된 가상 X-ray를 투과시켜 가상 디텍터(610)로 검출하는 2차원 시뮬레이션 과정을 적용하여 2차원 기준 영상(620)을 획득한다. 또한 부품 결함 검사 장치는 검사 대상 부품의 2차원 투과 영상의 밝기, 컨트라스트 등의 촬영조건을 기초로 2차원 기준 영상의 밝기, 컨트라스트 등을 조정할 수 있다. More specifically, the component defect inspection apparatus sets the virtual photographing point 600 for the two-dimensional simulation of the three-dimensional flawless image based on the photographing conditions such as the photographing distance and the photographing angle of the two-dimensional transmission image. The component defect inspection apparatus transmits the virtual X-ray emitted from the virtual photographing point 600 to the virtual three-dimensional component obtained by rotating the three-dimensional defect free image according to the photographing angle of the two-dimensional transmission image. A two-dimensional reference image 620 is obtained by applying a two-dimensional simulation process detected by the 610. In addition, the component defect inspection apparatus may adjust the brightness and contrast of the 2D reference image based on photographing conditions such as the brightness and contrast of the 2D transmission image of the inspection target part.
도 8은 본 발명에 따른 부품 결함 검사를 위한 영상 비교 방법의 일 예를 도시한 도면이다.8 is a diagram illustrating an example of an image comparison method for inspecting component defects according to the present invention.
도 8을 참조하면, 부품 결함 검사 장치는 검사 대상 부품의 2차원 투과 영상(810)을 획득한다. 또한 부품 결함 검사 장치는 2차원 투과 영상(810)과 대비할 무결점의 2차원 기준 영상(800)을 앞서 살핀 도 6의 방법 등을 이용하여 생성한다.Referring to FIG. 8, the component defect inspection apparatus acquires a 2D transmission image 810 of a component to be inspected. In addition, the component defect inspection apparatus generates a flawless two-dimensional reference image 800 to be contrasted with the two-dimensional transmission image 810 by using the method of FIG.
부품 결함 검사 장치는 2차원 투과 영상(810)과 무결점의 2차원 기준 영상(800)을 비교하여 결함 여부를 판단한다. 예를 들어, 부품 결함 검사 장치는 2차원 투과 영상(810)과 2차원 기준 영상(800)에서 차이가 나는 부위가 있다면 부품 결함으로 판단한다.The component defect inspection apparatus compares the 2D transmission image 810 and the defect free 2D reference image 800 to determine whether there is a defect. For example, the component defect inspection apparatus determines that a component defect exists if there is a difference between the 2D transmission image 810 and the 2D reference image 800.
부품 결함 검사 장치는 두 영상(800,810)의 벡터 비교 방법을 이용하여 영상의 동일성을 파악하여 부품 결함 여부를 판별할 수 있다. 예를 들어, 부품 결함 검사 장치는 두 영상을 그레이(gray) 영상으로 만든 후 두 그레이 영상의 각 화소 값의 유사도(correlation)를 이용하여 영상의 일치 여부를 파악하는 NCC(Normalized Cross-Correlation) 방법을 이용할 수 있으며, 이를 수학식으로 나타내면 다음과 같다.The component defect inspection apparatus may determine whether the component is defective by determining the identity of the image by using a vector comparison method between the two images 800 and 810. For example, a component defect inspection apparatus generates two images as gray images, and then uses a normalized cross-correlation (NCC) method to determine whether the images match by using the correlation between the pixel values of the two gray images. It can be used, represented by the equation as follows.
수학식 1
Figure PCTKR2016004634-appb-M000001
Equation 1
Figure PCTKR2016004634-appb-M000001
여기서, f(x,y), t(x,y)는 각 영상의 좌표(x,y)의 화소 값을 나타내고, n은 총 화소 수, f',t'는 각 영상의 화소 값의 평균 값, σft는 두 영상의 화소 값의 표준편차를 나타낸다. Here, f (x, y) and t (x, y) represent pixel values of coordinates (x, y) of each image, n is a total number of pixels, and f ', t' is an average of pixel values of each image. The values σ f and σ t represent standard deviations of pixel values of two images.
부품 결함 검사 장치는 이 외에도 종래의 다양한 영상 비교 방법을 이용할 수 있다.In addition, the component defect inspection apparatus may use various conventional image comparison methods.
도 9는 본 발명에 따른 본 발명에 따른 부품 결함 검사를 위하여 3차원 무결함 영상으로부터 2차원 기준 영상을 생성하는 방법의 다른 일 예를 도시한 도면이다.9 is a diagram illustrating another example of a method of generating a 2D reference image from a 3D defect image for inspecting a component defect according to the present invention.
도 9를 참조하면, 부품 결함 검사 장치는 도 6과 달리, 2차원 투과 영상의 전체와 대비할 2차원 기준 영상을 생성하는 것이 아니라, 2차원 투과 영상의 일부와 대비할 2차원 기준 영상을 생성한다.Referring to FIG. 9, unlike the case of FIG. 6, the component defect inspection apparatus does not generate a 2D reference image to contrast with the entire 2D transmission image, but generates a 2D reference image to contrast with a part of the 2D transmission image.
예를 들어, 부품 결함 검사 장치는 검사 대상 부품의 2차원 투과 영상에 대한 이미지 분석 과정을 통해 일정 크기 이상의 잡음 영역을 결함 의심 부위(900)로 파악하고, 그 결함 의심 부위(900)와 대비할 2차원 기준 영상을 생성한다. 결함 의심 부위를 파악하는 방법은 도 14에 도시되어 있다. For example, the component defect inspection apparatus may identify a noise region having a predetermined size or more as a defect suspect region 900 through an image analysis process on a 2D transmission image of a component to be inspected, and compare it with the defect suspect region 900. Create a dimensional reference image. The method of identifying the suspected defect site is shown in FIG.
구체적으로, 부품 결함 검사 장치는 2차원 투과 영상에서 파악된 잡음 영역, 즉 결함 의심 부위(900)에 대응되는 부품의 3차원 영역을 파악한다. 2차원 투과 영상은 방사장치의 X-ray 출력점(940)과 디텍터(910) 사이에 위치한 검사 대상 부품(920)을 촬영하여 얻어지는 영상이므로, 부품 결함 검사 장치는 2차원 투과 영상의 잡음 영역(900)에 대응되는 디텍터의 해당 영역(9150)과 X-ray 출력점(940)을 연결하여 형성되는 검사 대상 부품 내 3차원 서브 영역(930)을 파악한다.Specifically, the component defect inspection apparatus detects the noise region identified in the 2D transmission image, that is, the 3D region of the component corresponding to the defect suspected portion 900. Since the two-dimensional transmission image is an image obtained by photographing the inspection target part 920 located between the X-ray output point 940 and the detector 910 of the radiator, the component defect inspection apparatus may include a noise region ( The 3D subregion 930 in the inspection target part formed by connecting the corresponding region 9150 of the detector corresponding to 900 and the X-ray output point 940 is identified.
부품 결함 검사 장치는 3차원 무결점 영상 전체에 대한 2차원 시뮬레이션을 수행하는 것이 아니라, 3차원 무결점 영상 중 부품의 3차원 서브 영역(930)에 대해서만 2차원 시뮬레이션을 수행함으로써, 2차원 기준 영상의 생성의 소요시간을 단축할 수 있을 뿐만 아니라 결함이 의심되는 잡음영역에 대해 보다 정확한 영상 비교를 수행할 수 있다. The component defect inspection apparatus generates two-dimensional reference image by performing two-dimensional simulation only on the three-dimensional sub-region 930 of the component among the three-dimensional defect-free images rather than performing two-dimensional simulation on the entire three-dimensional defect-free image. In addition to shortening the time required for the circuit, it is possible to perform a more accurate image comparison in the noise region where the defect is suspected.
도 10은 2차원 투과 영상의 일부에 대해 2차원 기준 영상과 비교하여 결함을 검사하는 방법의 일 예를 도시한 도면이다.FIG. 10 is a diagram illustrating an example of a method for inspecting a defect of a part of a 2D transmission image by comparing with a 2D reference image.
도 10을 참조하면, 부품 결함 검사 장치는 부품의 2차원 투과 영상 전체에 대해 2차원 기준 영상과 비교하는 것이 아니라, 2차원 투과 영상 중 결함이 의심되는 잡음 영역을 포함하는 2차원 부품 검사 영상(1010)을 지정하고, 2차원 부품 검사 영역과 대응되는 2차원 기준 영상(1000)을 생성한 후, 2차원 부품 검사 영역(1010)과 2차원 기준 영상(1000)을 비교하여 결함 여부를 판단한다. 결함 여부 판단을 위한 영상 비교의 방법은 도 8에서 설명한 바와 같다. 본 실시 예를 살펴보면, 두 영상(1000,1010) 내 서로 다른 부분(1005,1015)가 존재하므로, 부품 결함 검사 장치는 부품 결함이 있다고 판단한다.Referring to FIG. 10, the apparatus for inspecting a component defect does not compare the entire 2D transmission image of the part with the 2D reference image, but includes a 2D component inspection image including a noise region in which a defect is suspected in the 2D transmission image. 1010 and generates a two-dimensional reference image 1000 corresponding to the two-dimensional part inspection area, and then compares the two-dimensional part inspection area 1010 and the two-dimensional reference image 1000 to determine whether there is a defect. . An image comparison method for determining whether there is a defect is as described with reference to FIG. 8. Referring to the present embodiment, since different portions 1005 and 1015 exist in the two images 1000 and 1010, the component defect inspection apparatus determines that there is a component defect.
도 11은 본 발명에 따른 부품 결함 검사 방법의 일 예를 도시한 흐름도이다.11 is a flowchart illustrating an example of a component defect inspection method according to the present invention.
도 11을 참조하면, 부품 결함 검사 장치는 결함 없는 부품을 촬영한 3차원 무결점 영상과, 검사 대상 부품을 촬영한 2차원 부품 검사 영상을 입력받는다(S1100,S1110). 여기서, 2차원 부품 검사 영상은 검사 대상 부품을 촬영한 2차원 투과 영상의 전체이거나 결함이 의심되는 일부 영역에 대한 영상일 수 있다.Referring to FIG. 11, the component defect inspection apparatus receives a 3D flawless image of a component without a defect and a 2D component inspection image of a component to be inspected (S1100 and S1110). Here, the two-dimensional part inspection image may be an entirety of the two-dimensional transmission image photographing the inspection target part or an image of a partial region in which a defect is suspected.
부품 결함 검사 장치는 2차원 투과 영상의 촬영조건을 기초로 2차원 부품 검사 영상과 대비될 수 있는 무결점의 2차원 기준 영상을 3차원 무결점 영상으로부터 생성한다(S1120). 그리고 부품 결함 검사 장치는 2차원 부품 검사 영상과 무결점의 2차원 기준 영상을 비교하여, 상이한 부분이 있는지 여부를 기초로 결함 여부를 판단한다(S1130).The component defect inspection apparatus generates a two-dimensional reference image of a defect-free image, which may be contrasted with a two-dimensional component inspection image, based on photographing conditions of the two-dimensional transmission image (S1120). Then, the component defect inspection apparatus compares the two-dimensional component inspection image with a defect-free two-dimensional reference image, and determines whether there is a defect based on whether there is a different portion (S1130).
도 12는 본 발명에 따른 2차원 투과 영상을 분석하여 부품 결함 의심 영역을 검출하는 방법의 일 예를 도시한 흐름도이고, 도 13a 내지 도 13e는 도 12의 각 단계별 2차원 투과 영상의 이미지 처리 결과의 일 예를 도시한 도면이다.12 is a flowchart illustrating an example of a method for detecting a suspected component defect region by analyzing a two-dimensional transmission image according to the present invention, and FIGS. 13A to 13E are image processing results of the two-dimensional transmission image of each step of FIG. 12. 1 is a diagram illustrating an example.
도 12 및 도 13을 함께 참조하면, 부품 결함 검사 장치는 도 1과 같은 방법을 통해 검사 대상 부품의 2차원 투과 영상(도 13a)을 얻는다(S1200). 부품 결함 검사 장치는 2차원 투과 영상(도 13a) 중 부품 영역을 마스킹(도 13b)하여 부품 이외의 배경 영역을 제거하고(S1210), 남아 있는 부품 영역 내 작은 구멍(hole) 등을 메운 보정 영상(도 13c)을 생성한다(S1220). 작은 구멍 등을 매우기 위한 방법으로 닫기(closing) 연산을 이용할 수 있다.12 and 13 together, the component defect inspection apparatus obtains a two-dimensional transmission image (FIG. 13A) of the inspection target component through the same method as in FIG. 1 (S1200). The component defect inspection apparatus masks the component region of the two-dimensional transmission image (FIG. 13A) to remove a background region other than the component (S1210), and corrects an image including a small hole in the remaining component region. FIG. 13C is generated (S1220). Closing can be used as a way to fill small holes.
부품 결함 검사 장치는 보정 영상(도 13c)에서 원 2차원 투과 영상(도 13a)을 뺀 노이즈 영상(도 13d)을 생성한다(S1230). 두 영상 사이의 차감을 통해 생성한 노이즈 영상(도 13d)에는 작은 구멍(hole)들과 노이즈 픽셀들만 남게 된다. 부품 결함 검사 장치는 노이즈 영상(도 13d)에서 중간값(median) 필터를 적용하여 결함과 관련없는 일반 노이즈를 제거한다(S1240).The component defect inspection apparatus generates a noise image (FIG. 13D) obtained by subtracting the original two-dimensional transmission image (FIG. 13A) from the corrected image (FIG. 13C) (S1230). Only small holes and noise pixels remain in the noise image (FIG. 13D) generated through the subtraction between the two images. The component defect inspection apparatus removes general noise not related to a defect by applying a median filter to the noise image (FIG. 13D) (S1240).
그리고 부품 결함 검사 장치는 노이즈 영상(도 13d)을 0과 1의 이진 영상(binary)으로 변환한다(S1250). 예를 들어, 사용자가 기 설정된 임계값을 기준으로 각 픽셀의 명도 등의 값이 임계값보다 크면 1, 작으면 0으로 설정하여 이진 영상을 만들 수 있다. The component defect inspection apparatus converts the noise image (FIG. 13D) into a binary image of 0 and 1 (S1250). For example, the user may create a binary image by setting a value such as brightness of each pixel based on a predetermined threshold value to 1 when the value of the pixel is greater than the threshold value and 0 when the value is smaller than the threshold value.
부품 결함 검사 장치는 이진 영상에서 서로 떨어져 있는 작은 섬(island) 영역들을 레이블링(blob labeling) 기법 등을 통해 서로 연결한 후, 단일 영역(blob)의 크기가 기 설정된 일정 크기 이상인 부위를 결함 의심 영역(도 13e의 1300)으로 판단한다(S1260). 결함 의심 영역(1300)은 경우에 따라 하나 이상 존재할 수 있으며, 이러한 결함 의심 영역을 포함하는 서브 영상을 2차원 기준 영상과 비교되는 2차원 부품 검사 영역으로 설정할 수 있다. The component defect inspection apparatus connects small island regions separated from each other in a binary image by using a label labeling technique, and then detects a region in which a single region of a single blob is larger than a predetermined size. It is determined as 1300 of FIG. 13E (S1260). One or more defect suspect regions 1300 may exist in some cases, and a sub image including the suspect defect regions may be set as a two-dimensional component inspection region to be compared with a two-dimensional reference image.
도 14는 본 발명에 따른 부품 결함 검사를 위한 2차원 기준 영상을 생성하는 방법의 일 예를 도시한 흐름도이다.14 is a flowchart illustrating an example of a method of generating a 2D reference image for component defect inspection according to the present invention.
도 14를 참조하면, 부품 결함 검사 장치는 검사 대상 부품의 2차원 투과 영상의 촬영위치, 촬영각도 등을 포함한 촬영조건을 파악한다. 촬영조건은 도 1과 같은 방법을 이용하여 2차원 투과 영상을 촬영될 때 자동으로 파악되어 부품 결함 검사 장치에 제공되거나, 사용자가 부품 결함 검사 장치의 사용자 인터페이스를 통해 촬영조건을 직접 입력할 수 있다. Referring to FIG. 14, the component defect inspection apparatus detects photographing conditions including a photographing position, an photographing angle, and the like, of a 2D transmission image of a component to be inspected. The shooting conditions are automatically detected when the 2D transmission image is taken using the method as shown in FIG. 1 and provided to the component defect inspection apparatus, or the user may directly input the shooting conditions through the user interface of the component defect inspection apparatus. .
부품 결함 검사 장치는 촬영 조건을 기초로 3차원 무결점 영상에 대한 시뮬레이션을 수행할 가상의 촬영지점을 설정한다(S1410). 그리고 부품 결함 검사 장치는 도 6과 같이, 가상의 촬영지점에서 2차원 투과 영상의 촬영각도만큼 회전된 3차원 무결점 영상을 기초로 DDR 시뮬레이션 등을 수행하여 2차원 기준 영상을 생성한다(S1420).The component defect inspection apparatus sets a virtual photographing point to perform a simulation on the 3D flawless image based on the photographing conditions (S1410). In operation S1420, the component defect inspection apparatus generates a 2D reference image by performing a DDR simulation or the like based on a 3D defectless image rotated by a photographing angle of a 2D transmission image at a virtual photographing point (S1420).
도 15는 본 발명에 따른 부품 결함 검사를 위한 2차원 기준 영상을 생성하는 방법의 다른 일 예를 도시한 흐름도이다.15 is a flowchart illustrating another example of a method of generating a 2D reference image for inspecting a component defect according to the present invention.
도 15를 참조하면, 부품 결함 검사 장치는 3차원 무결점 영상 전체에 대한 시뮬레이션 과정을 수행하는 것이 아니라 특정 영역에 대해서만 시뮬레이션 과정을 수행한다. 이를 위해, 부품 결함 검사 장치는 도 12와 같은 방법을 통해 2차원 투과 영상 중 결함 의심 영역을 검출하고, 검출한 결함 의심 영역을 결함 조사를 위한 2차원 부품 검사 영상으로 설정한다. Referring to FIG. 15, the component defect inspection apparatus does not perform a simulation process for the entire 3D defect-free image but performs a simulation process only for a specific region. To this end, the component defect inspection apparatus detects a defect suspect region of the two-dimensional transmission image through the method as shown in FIG. 12, and sets the detected defect suspect region as a two-dimensional component inspection image for defect inspection.
부품 결함 검사 장치는 2차원 부품 검사 영상에 해당하는 부품의 3차원 영역을 파악한다(S1500). 예를 들어, 2차원 부품 검사 영상은 도 9와 같이 X-ray 출력점(940)과 디텍터(910)의 2차원 부품 검사 영역에 대응하는 위치(915)를 이어 구분되는 부품의 3차원 서브 영역을 파악한다. The component defect inspection apparatus detects a three-dimensional region of the component corresponding to the two-dimensional component inspection image (S1500). For example, the two-dimensional part inspection image is a three-dimensional sub-region of a part that is divided by a position 915 corresponding to the two-dimensional part inspection area of the X-ray output point 940 and the detector 910 as shown in FIG. 9. Figure out.
그리고 부품 결함 검사 장치는 3차원 무결함 영상 중 부품의 3차원 서브 영역에 대해서만 시뮬레이션 과정을 수행하여 2차원 기준 영상을 생성한다(S1520).The component defect inspection apparatus generates a 2D reference image by performing a simulation process only on the 3D sub-region of the component among the 3D defect images (S1520).
도 16은 본 발명에 따른 부품 결함 검사 방법의 다른 일 예를 도시한 흐름도이다.16 is a flowchart illustrating another example of a component defect inspection method according to the present invention.
도 16을 참조하면, 부품 결함 검사 장치는 검사 대상 부품의 2차원 투과 영상을 획득한다(S1600). 부품 결함 검사 장치는 2차원 투과 영상에 대해 도 12와 같은 이미지 분석 과정을 수행하여 결함 의심 영역이 존재하는지 파악한다(S1610). 예를 들어, 부품 결함 검사 장치는 도 13e와 같이 2차원 투과 영상에 존재하는 일정 크기 이상의 잡음 영역을 결함 의심 영역(1300)으로 판단할 수 있다.Referring to FIG. 16, the component defect inspection apparatus acquires a 2D transmission image of a component to be inspected (S1600). The component defect inspection apparatus performs an image analysis process as shown in FIG. 12 on the 2D transmission image to determine whether a defect suspect region exists. For example, the component defect inspection apparatus may determine the noise region having a predetermined size or more present in the 2D transmission image as the defect suspect region 1300 as illustrated in FIG. 13E.
결함 의심 영역이 존재하는 경우에(S1620), 부품 결함 검사 장치는 3차원 무결점 영상으로부터 2차원 투과 영상의 전체 또는 그 일부와 대비할 무결점의 2차원 기준 영상을 생성한다(S1630). 부품의 어느 부위의 결함을 검사하는지에 따라 2차원 투과 영상의 촬영 각도 등은 서로 달라지므로, 부품 결함 검사 장치는 2차원 투과 영상의 촬영 각도 등을 포함하는 촬영 조건을 이용하여 2차원 투과 영상과 동일한 촬영 각도 등으로 이루어진 2차원 기준 영상을 생성한다. If there is a suspected defect area (S1620), the component defect inspection apparatus generates a defect-free two-dimensional reference image to contrast with all or a part of the two-dimensional transmission image from the three-dimensional defect free image (S1630). Since the photographing angle of the two-dimensional transmission image is different from each other depending on which part of the part is inspected, the component defect inspection apparatus uses the photographing conditions including the photographing angle of the two-dimensional transmission image and the like. A two-dimensional reference image of the same photographing angle is generated.
그리고 부품 결함 검사 장치는 2차원 투과 영상의 전체 또는 일부와 2차원 기준 영상을 비교하여 결함 여부를 판단한다(S1640).The component defect inspection apparatus compares the whole or part of the 2D transmission image with the 2D reference image to determine whether there is a defect (S1640).
본 발명은 또한 컴퓨터로 읽을 수 있는 기록매체에 컴퓨터가 읽을 수 있는 코드로서 구현하는 것이 가능하다. 컴퓨터가 읽을 수 있는 기록매체는 컴퓨터 시스템에 의하여 읽혀질 수 있는 데이터가 저장되는 모든 종류의 기록장치를 포함한다. 컴퓨터가 읽을 수 있는 기록매체의 예로는 ROM, RAM, CD-ROM, 자기 테이프, 플로피디스크, 광데이터 저장장치 등이 있다. 또한 컴퓨터가 읽을 수 있는 기록매체는 네트워크로 연결된 컴퓨터 시스템에 분산되어 분산방식으로 컴퓨터가 읽을 수 있는 코드가 저장되고 실행될 수 있다.The invention can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.
이제까지 본 발명에 대하여 그 바람직한 실시예들을 중심으로 살펴보았다. 본 발명이 속하는 기술 분야에서 통상의 지식을 가진 자는 본 발명이 본 발명의 본질적인 특성에서 벗어나지 않는 범위에서 변형된 형태로 구현될 수 있음을 이해할 수 있을 것이다. 그러므로 개시된 실시예들은 한정적인 관점이 아니라 설명적인 관점에서 고려되어야 한다. 본 발명의 범위는 전술한 설명이 아니라 특허청구범위에 나타나 있으며, 그와 동등한 범위 내에 있는 모든 차이점은 본 발명에 포함된 것으로 해석되어야 할 것이다.So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

Claims (12)

  1. 부품의 3차원 무결점 영상을 획득하는 단계;Obtaining a three-dimensional defect free image of the component;
    2차원 부품 검사 영상을 획득하는 단계;Obtaining a two-dimensional part inspection image;
    상기 2차원 부품 검사 영상의 촬영각도를 포함한 촬영조건을 기초로, 상기 3차원 무결점 영상으로부터 2차원 기준 영상을 생성하는 단계; 및Generating a two-dimensional reference image from the three-dimensional defect free image based on a photographing condition including a photographing angle of the two-dimensional part inspection image; And
    상기 2차원 부품 검사 영상과 상기 2차원 기준 영상을 비교하여 부품의 결함 여부를 판단하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And comparing the two-dimensional component inspection image with the two-dimensional reference image to determine whether a component is defective.
  2. 제 1항에 있어서, 상기 2차원 부품 검사 영상을 획득하는 단계는,The method of claim 1, wherein the obtaining of the 2D component inspection image comprises:
    부품의 2차원 투과 영상을 획득하는 단계;Obtaining a two-dimensional transmission image of the part;
    상기 2차원 투과 영상의 작은 구멍들을 메운 보정 영상을 생성하는 단계;Generating a corrected image filling small holes of the 2D transmission image;
    상기 보정 영상에서 상기 2차원 투과 영상을 뺀 노이즈 영상을 생성하는 단계;Generating a noise image obtained by subtracting the 2D transmission image from the corrected image;
    상기 노이즈 영상을 이진 영상으로 변환하는 단계; 및Converting the noise image into a binary image; And
    상기 이진 영상에서 기 설정된 크기 이상의 잡음영역을 2차원 부품 검사 영상으로 지정하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And specifying a noise area of a predetermined size or more in the binary image as a 2D component inspection image.
  3. 제 1항에 있어서, 상기 2차원 기준 영상을 생성하는 단계는,The method of claim 1, wherein the generating of the 2D reference image comprises:
    상기 2차원 부품 검사 영상의 촬영위치 및 촬영각도를 포함하는 촬영조건을 기초로, 상기 3차원 무결점 영상의 2차원 시뮬레이션을 위한 가상의 촬영지점을 설정하는 단계; 및Setting a virtual photographing point for a 2D simulation of the 3D flawless image based on a photographing condition including a photographing position and a photographing angle of the 2D part inspection image; And
    상기 가상의 촬영지점을 기초로 상기 3차원 무결점 영상으로부터 상기 2차원 부품 검사 영상에 대응되는 무결점의 2차원 기준 영상을 획득하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And acquiring a two-dimensional reference image of a zero defect corresponding to the two-dimensional component inspection image from the three-dimensional defect free image based on the virtual photographing point.
  4. 제 1항에 있어서, 상기 2차원 기준 영상을 생성하는 단계는, The method of claim 1, wherein the generating of the 2D reference image comprises:
    상기 2차원 부품 검사 영상에 해당하는 3차원 부품 영역을 파악하는 단계; 및Identifying a 3D part region corresponding to the 2D part inspection image; And
    상기 3차원 무결점 영상 중 상기 3차원 부품 영역에 해당하는 무결점 서브 영역을 파악하는 단계; 및Identifying a defect-free subregion corresponding to the three-dimensional component region of the three-dimensional defect free image; And
    상기 무결점 서브 영역에 대한 2차원 시뮬레이션을 통해, 상기 2차원 부품 검사 영상과 대비되는 무결점의 2차원 기준 영상을 생성하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And generating a two-dimensional reference image of a defect free from the two-dimensional component inspection image by performing a two-dimensional simulation on the defect-free subregion.
  5. 제 1항에 있어서, The method of claim 1,
    상기 3차원 무결점 영상은 컴퓨터 단층촬영 영상으로 구성되고,The three-dimensional defect free image is composed of a computed tomography image,
    상기 2차원 부품 검사 영상은 엑스레이 영상으로 구성되는 것을 특징으로 하는 부품 결함 검사 방법.The two-dimensional component inspection image is a component defect inspection method, characterized in that consisting of the x-ray image.
  6. 제 1항에 있어서, 상기 결함 여부를 판단하는 단계는,The method of claim 1, wherein the determining of the defect comprises:
    상기 2차원 부품 검사 영상과 상기 2차원 기준 영상의 벡터 차를 이용하여 유사도를 판단하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And determining similarity using the vector difference between the two-dimensional part inspection image and the two-dimensional reference image.
  7. 부품의 2차원 투과 영상과 3차원 무결점 영상을 획득하는 단계;Obtaining a two-dimensional transmission image and a three-dimensional defect free image of the component;
    상기 2차원 투과 영상의 이미지 분석을 통해 일정 크기 이상의 잡음 영역의 존부를 기초로 1차 부품 결함 여부를 판단하는 단계;Determining whether there is a primary component defect based on an existence of a noise area of a predetermined size or more through image analysis of the 2D transmission image;
    상기 2차원 투과 영상의 촬영 각도를 포함하는 촬영조건을 기초로, 상기 3차원 무결점 영상으로부터 상기 2차원 투과 영상의 전부 또는 일부와 대비할 무결점의 2차원 기준 영상을 생성하는 단계; 및Generating a defect-free two-dimensional reference image to contrast with all or a portion of the two-dimensional transmitted image from the three-dimensional defect free image based on photographing conditions including a photographing angle of the two-dimensional transmitted image; And
    상기 2차원 투과 영상의 전부 또는 일부와 상기 2차원 기준 영상을 비교하여 2차 부품 결함 여부를 판단하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And comparing the 2D reference image with all or part of the 2D transmission image to determine whether there is a secondary component defect.
  8. 제 7항에 있어서, 상기 2차원 기준 영상을 생성하는 단계는,The method of claim 7, wherein the generating of the 2D reference image comprises:
    상기 2차원 투과 영상의 촬영위치 및 촬영각도를 포함하는 촬영조건을 기초로, 상기 3차원 무결점 영상의 2차원 시뮬레이션을 위한 가상의 촬영지점을 설정하는 단계; 및Setting a virtual photographing point for two-dimensional simulation of the three-dimensional defect free image based on photographing conditions including a photographing position and an photographing angle of the two-dimensional transmission image; And
    상기 가상의 촬영지점을 기초로 상기 3차원 무결점 영상으로부터 상기 2차원 부품 검사 영상에 대응되는 무결점의 2차원 기준 영상을 획득하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And acquiring a two-dimensional reference image of a zero defect corresponding to the two-dimensional component inspection image from the three-dimensional defect free image based on the virtual photographing point.
  9. 제 7항에 있어서, 상기 2차원 기준 영상을 생성하는 단계는, The method of claim 7, wherein the generating of the 2D reference image comprises:
    상기 2차원 투과 영상 중 상기 잡음 영역에 해당하는 3차원 부품 영역을 파악하는 단계; 및Identifying a 3D component region corresponding to the noise region of the 2D transmission image; And
    상기 3차원 무결점 영상 중 상기 3차원 부품 영역에 해당하는 3차원 서브 영역을 파악하는 단계; 및Identifying a three-dimensional sub-region corresponding to the three-dimensional part region among the three-dimensional defect free images; And
    상기 3차원 서브 영역에 대한 2차원 시뮬레이션을 통해, 상기 2차원 부품 검사 영상과 대비되는 무결점의 2차원 기준 영상을 생성하는 단계;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And generating a defect-free two-dimensional reference image contrasted with the two-dimensional component inspection image by performing two-dimensional simulation on the three-dimensional sub-region.
  10. 부품의 3차원 무결점 영상을 획득하는 3차원 영상 획득부;A three-dimensional image acquisition unit for acquiring a three-dimensional defect free image of the component;
    부품의 2차원 투과 영상을 획득하는 2차원 영상 획득부;A two-dimensional image acquisition unit obtaining a two-dimensional transmission image of the component;
    상기 2차원 투과 영상의 촬영각도를 포함한 촬영조건을 기초로, 상기 3차원 무결점 영상으로부터 2차원 기준 영상을 생성하는 기준영상 생성부; 및A reference image generator configured to generate a two-dimensional reference image from the three-dimensional defect free image based on shooting conditions including a photographing angle of the two-dimensional transmission image; And
    상기 2차원 투과 영상과 상기 2차원 기준 영상을 비교하여 부품의 결함 여부를 판단하는 결함 판단부;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And a defect determination unit which compares the two-dimensional transmission image with the two-dimensional reference image to determine whether a component is defective.
  11. 제 10항에 있어서, 상기 결함 판단부는, The method of claim 10, wherein the defect determination unit,
    상기 2차원 투과 영상의 이미지 분석을 통해 일정 크기 이상의 잡음 영역의 존부를 기초로 1차 부품 결함 여부를 판단하는 1차 판단부; 및A primary determination unit determining whether a primary component is defective based on an existence of a noise area of a predetermined size or more through image analysis of the 2D transmission image; And
    상기 2차원 투과 영상의 전부 또는 일부와 상기 2차원 기준 영상을 비교하여 2차 부품 결함 여부를 판단하는 2차 판단부;를 포함하는 것을 특징으로 하는 부품 결함 검사 방법.And a second determination unit which compares all or a part of the 2D transmission image with the 2D reference image to determine whether there is a secondary component defect.
  12. 제 1항 내지 제 9항 중 어느 한 항에 기재된 방법을 수행하기 위한 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록매체.A computer-readable recording medium having recorded thereon a program for performing the method according to any one of claims 1 to 9.
PCT/KR2016/004634 2015-06-17 2016-05-03 Component defect inspection method, and apparatus therefor WO2016204402A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2015-0086184 2015-06-17
KR1020150086184A KR101717768B1 (en) 2015-06-17 2015-06-17 Component inspecting method and apparatus

Publications (1)

Publication Number Publication Date
WO2016204402A1 true WO2016204402A1 (en) 2016-12-22

Family

ID=57545778

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2016/004634 WO2016204402A1 (en) 2015-06-17 2016-05-03 Component defect inspection method, and apparatus therefor

Country Status (2)

Country Link
KR (1) KR101717768B1 (en)
WO (1) WO2016204402A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020002704A1 (en) * 2018-06-29 2020-01-02 Universiteit Antwerpen Item inspection by dynamic selection of projection angle
CN114460093A (en) * 2022-01-29 2022-05-10 新拓三维技术(深圳)有限公司 Method and system for detecting defects of aircraft engine

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102300562B1 (en) 2018-01-25 2021-09-13 삼성전자주식회사 Detecting method, detecting apparatus and computer readable storage medium
KR102075872B1 (en) * 2019-08-09 2020-02-11 레이디소프트 주식회사 Method for non-destructive inspection based on image and apparatus thereof
KR102249836B1 (en) * 2019-08-26 2021-05-10 레이디소프트 주식회사 Method for non-destructive inspection based on image and Computer-readable storage medium
WO2021029625A1 (en) * 2019-08-09 2021-02-18 레이디소프트 주식회사 Transmission image-based non-destructive inspecting method, method of providing non-destructive inspection function, and device therefor
KR102415928B1 (en) * 2019-08-26 2022-07-05 레이디소프트 주식회사 Method for non-destructive inspection based on transmission image
KR102616867B1 (en) * 2021-04-20 2024-01-03 레이디소프트 주식회사 Method for non-destructive inspection

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020146120A1 (en) * 2001-04-05 2002-10-10 Hugh Anglin Inspecting print quality using digital watermarks
JP2003510568A (en) * 1999-09-20 2003-03-18 松下電器産業株式会社 LCD inspection method and LCD inspection apparatus by pattern comparison
KR100942379B1 (en) * 2009-06-30 2010-02-12 한국전력공사 Crack detection apparatus and method of box culvert for electric power transmission using three-dimension laser scanner
KR20110120686A (en) * 2010-04-29 2011-11-04 주식회사 쓰리디산업영상 Method and system for inspecting internal defect using reference information
KR101094412B1 (en) * 2011-05-23 2011-12-15 주식회사 쓰리디산업영상 Defect detection method using 3d images in ball joint region of bga chip

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101293532B1 (en) * 2011-05-12 2013-08-07 주식회사 쎄크 Computer tomography inspecting method of semiconductor chip
KR20150056148A (en) 2013-11-15 2015-05-26 (주)자비스 Apparatus Having I Improved Efficiency in X-Ray Inspection and Method For the Same

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003510568A (en) * 1999-09-20 2003-03-18 松下電器産業株式会社 LCD inspection method and LCD inspection apparatus by pattern comparison
US20020146120A1 (en) * 2001-04-05 2002-10-10 Hugh Anglin Inspecting print quality using digital watermarks
KR100942379B1 (en) * 2009-06-30 2010-02-12 한국전력공사 Crack detection apparatus and method of box culvert for electric power transmission using three-dimension laser scanner
KR20110120686A (en) * 2010-04-29 2011-11-04 주식회사 쓰리디산업영상 Method and system for inspecting internal defect using reference information
KR101094412B1 (en) * 2011-05-23 2011-12-15 주식회사 쓰리디산업영상 Defect detection method using 3d images in ball joint region of bga chip

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020002704A1 (en) * 2018-06-29 2020-01-02 Universiteit Antwerpen Item inspection by dynamic selection of projection angle
US11953451B2 (en) 2018-06-29 2024-04-09 Universiteit Antwerpen Item inspection by dynamic selection of projection angle
CN114460093A (en) * 2022-01-29 2022-05-10 新拓三维技术(深圳)有限公司 Method and system for detecting defects of aircraft engine
CN114460093B (en) * 2022-01-29 2024-05-28 新拓三维技术(深圳)有限公司 Aeroengine defect detection method and system

Also Published As

Publication number Publication date
KR101717768B1 (en) 2017-03-27
KR20160149106A (en) 2016-12-27

Similar Documents

Publication Publication Date Title
WO2016204402A1 (en) Component defect inspection method, and apparatus therefor
WO2019107614A1 (en) Machine vision-based quality inspection method and system utilizing deep learning in manufacturing process
WO2019132589A1 (en) Image processing device and method for detecting multiple objects
WO2021066253A1 (en) Artificial intelligence (ai)-based object inspection system and method
WO2019172728A1 (en) Method for measuring foot size and shape by using image processing
WO2015005577A1 (en) Camera pose estimation apparatus and method
WO2019164074A1 (en) Fine dust analysis method, and apparatus for performing same
WO2011037344A2 (en) Multi-energy x-ray system, image processing apparatus for discriminating multi-energy x-ray material, and image processing method for material discrimination of multi-energy x-ray system
WO2016043397A1 (en) Glass defect detection method and apparatus using hyperspectral imaging technique
CN110390661A (en) It checks management system, check managing device and checks management method
WO2014035103A1 (en) Apparatus and method for monitoring object from captured image
WO2019132588A1 (en) Image analysis device and method based on image feature and context
US11715195B2 (en) Machine learning-based circuit board inspection
CN105934757A (en) Method and apparatus for detecting incorrect associations between keypoints of first image and keypoints of second image
WO2018101746A2 (en) Apparatus and method for reconstructing road surface blocked area
WO2015099249A1 (en) Apparatus and method for determining lesion similarity of medical image
JP2018004272A (en) Pattern inspection device and pattern inspection method
CN105787429A (en) Method and apparatus for inspecting an object employing machine vision
Londe et al. Automatic PCB defects detection and classification using MATLAB
WO2019124731A1 (en) Multiplexing signal processing apparatus and method using clustering and deep learning
WO2019098421A1 (en) Object reconstruction device using motion information and object reconstruction method using same
WO2023113274A1 (en) Ai-based product surface inspection device and method
WO2021130978A1 (en) Motion analysis system and motion analysis program
WO2013077508A1 (en) Device and method for depth map generation and device and method using same for 3d image conversion
WO2017111515A1 (en) Apparatus and method for inspecting steel plate

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: 16811815

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16811815

Country of ref document: EP

Kind code of ref document: A1