WO2017183923A1 - 물품의 외관 검사장치 및 이를 이용한 물품의 외관 검사방법 - Google Patents
물품의 외관 검사장치 및 이를 이용한 물품의 외관 검사방법 Download PDFInfo
- Publication number
- WO2017183923A1 WO2017183923A1 PCT/KR2017/004238 KR2017004238W WO2017183923A1 WO 2017183923 A1 WO2017183923 A1 WO 2017183923A1 KR 2017004238 W KR2017004238 W KR 2017004238W WO 2017183923 A1 WO2017183923 A1 WO 2017183923A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- article
- inspection
- area
- appearance
- image data
- Prior art date
Links
Images
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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/10024—Color image
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
-
- 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
- G06T2207/30168—Image quality inspection
Definitions
- the present invention relates to a device for inspecting the appearance of the article and a method for inspecting the appearance of the article using the same, and more particularly, to a device for inspecting the appearance of the article to determine whether the appearance of the article is defective.
- various articles produced by various processing methods may have defects in terms of their functionalities, but may also have defects in appearance due to problems in the processing process or external factors.
- the problem to be solved by the present invention by obtaining the second data relative to the equation representing the distribution of the first data for the inspection area and determining whether or not the appearance of the article on the basis of this, various appearance of the article Even in the case of not having a flat planar shape, such as having a shape and a bent portion, it is possible to easily and accurately determine whether an article has a poor appearance, and a plane inspection applied when the inspection region corresponds to a plane.
- a standard can be directly applied to the second data to easily determine whether the article is defective in appearance, and the first data can be easily obtained by using data obtained for measuring a two-dimensional plane image or a three-dimensional shape.
- the above equation can be obtained simply by performing curve fitting, and drawing holes and the like formed in the article as masking areas. By removing the noise region, and correcting the equation by removing the noise area from the inspection area, and obtaining the more accurate equation, and performing the various data processing or image processing on the second data.
- the present invention provides an apparatus for inspecting the appearance of articles that can obtain more accurate inspection results.
- Another object of the present invention is to provide a method for inspecting the appearance of an article using the apparatus for inspecting the appearance of the article.
- Apparatus for inspecting the appearance of an article by using an image acquisition unit for obtaining a photographed image of at least a part of the appearance of the article and the photographed image of the article determines whether the appearance of the article is defective It includes a processing unit.
- the processor acquires first image data about an inspection area including at least a portion of the captured image, and obtains an equation for following a transition of a level of the first image data according to a position in the inspection area. Obtain second image data based on the relative level between the equation and the level of the first image data.
- the second image data may be obtained from a distance between the first image data and a value according to the equation, for each position in the inspection area.
- the processor may determine whether the article is defective in appearance based on the second image data.
- the processor may determine whether the article is defective in appearance by applying a plane inspection criterion to the second image data.
- the plane inspection criterion is an inspection criterion applied when the inspection region corresponds to a plane.
- the appearance inspection apparatus may further include at least one of a first lighting unit for providing a non-patterned lighting and a second lighting unit for providing a patterned lighting, wherein the first image data is included in the first lighting unit. It may include at least one of the brightness data obtained by the height data and the height data obtained by the second lighting unit.
- the processing unit may apply a plane inspection criterion applied to the second image data to determine whether the article is defective in appearance when the inspection region is flat. Can be determined.
- the inspection area may be a one-dimensional area or a two-dimensional area, and an equation for following the trend of the level of the first image data may be an equation for an area having the same dimension as that of the inspection area. That is, in one embodiment, the inspection area may be a one-dimensional area, and the equation for following the trend of the level of the first image data may be an equation for the one-dimensional area. In another embodiment, the inspection area may be a two-dimensional area, and the equation for following the trend of the level of the first image data may be an equation for the two-dimensional area.
- the inspection area may be a two-dimensional area
- the processing unit may obtain equations for one-dimensional areas forming the two-dimensional area, and based on the equations for the one-dimensional areas.
- the equation for the two-dimensional inspection area can be obtained.
- the processor may acquire the second image data for each position by subtracting a value according to the equation from the first image data with respect to each position in the inspection area.
- the processor may set the inspection region by setting a target region in the photographed image, obtaining a masking region to be excluded from the target region, and excluding the masking region from the target region.
- the processor may acquire an area in which the first image data is out of a predetermined range as the masking area.
- the processing unit when acquiring the equation, by first obtaining the equation for the inspection area, by extracting and removing the noise area based on the equation obtained first
- the inspection area may be modified, and the equation may be modified based on the modified inspection area.
- the processor may generate third image data by binarizing the second image data, and determine whether the article is defective in appearance by using the third image data.
- a method for inspecting an external appearance of an article is a method for inspecting an external appearance of the article using an external appearance inspecting apparatus, wherein the external appearance inspecting apparatus includes at least a part of a photographed image of at least a part of the external appearance of the article.
- Acquiring first image data relating to an inspection area acquiring an equation for following a transition of a level of the first image data according to a position in the inspection area, and the equation and the first image Acquiring second image data based on a relative level between levels of data.
- the second image data may be obtained from a distance between the first image data and a value according to the equation, for each position in the inspection area.
- the method of inspecting an appearance of the article may include: obtaining second image data based on a relative level between the equation and the level of the first image data; The method may further include determining whether the article is defective in appearance. In the determining of the appearance defect of the article based on the second image data, it is possible to determine whether the appearance of the article is defective by applying a plane inspection criterion to the second image data.
- the plane inspection criterion is an inspection criterion applied when the inspection region corresponds to a plane.
- the first image data may include at least one of brightness data and height data.
- acquiring first image data about an inspection area including at least a portion of a captured image of at least a portion of an appearance of the article may include setting a target region in the captured image, the target region.
- the method may include obtaining a masking area to be excluded from the inside and setting the inspection area by excluding the masking area from the target area.
- the obtaining of the equation for following the trend of the level of the first image data according to the position in the inspection area may include setting the inspection area as an effective area and applying the equation to the valid area. Acquiring firstly, modifying the valid region by extracting and removing the noise region based on the first obtained equation, and modifying the mathematical expression based on the modified valid region. Can be.
- a recording medium recording computer software for implementing the method for inspecting the appearance of the article may be provided.
- the appearance of the article is flat by acquiring second image data relative to the equation following the transition of the level of the first image data with respect to the inspection area and determining whether the article is defective in appearance. Even in the absence of a planar shape, it is possible to more easily and accurately determine whether or not the appearance of the article is defective, and the appearance of the article by directly applying the plane inspection criteria applied to the second image data when the inspection region is a plane. It is possible to easily determine whether or not the defect.
- the equation may be obtained simply by performing curve fitting on a quadratic or more curved equation or a curved equation.
- the first image data when the brightness data is employed as the first image data, the first image data can be easily obtained from the photographed two-dimensional plane image, and when the height data is used as the first data, the three-dimensional shape is obtained.
- the first image data can be easily obtained from the obtained data to measure.
- a more accurate inspection result may be obtained by acquiring and excluding the hole as a masking area.
- test result may be obtained by performing data processing or image processing such as binarization, various filtering, and morphology processing.
- FIG. 1 is a conceptual diagram showing the appearance inspection apparatus of the article according to an embodiment of the present invention.
- FIG. 2 is a flowchart illustrating a method of inspecting appearance of an article according to an embodiment of the present invention.
- FIG. 3 is a plan view illustrating examples of an inspection area set for an appearance inspection method of the article of FIG. 2;
- FIG. 4 is a cross-sectional view taken along the line II ′ of FIG. 3.
- FIG. 5 is a flowchart illustrating an example of a process of acquiring first data of the external appearance inspection method of the article of FIG. 2.
- FIG. 6 is a plan view illustrating an example of a process of obtaining first data of FIG. 5.
- FIG. 7 is a graph illustrating an example of a distribution of first data and an equation according to the process of obtaining the equation of FIG. 2.
- FIG. 8 is a flowchart illustrating an example of a process of obtaining a mathematical expression of an appearance inspection method of the article of FIG. 2.
- FIG. 9 is a graph illustrating a distribution of second data for each location extracted from the graph of FIG. 7 as a broken line.
- first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.
- the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component.
- the invention may be illustrated as being implemented in a suitable computing environment.
- various methods according to the present invention may be provided as a recording medium recording computer software for implementing the same.
- the recording medium typically includes a variety of computer readable media and can be provided in any available media that can be accessed by a computer.
- the recording medium may include a volatile or non-volatile medium, a removable or non-removable medium, and the like.
- the recording medium may include all media embodied by any method or technology for storing information such as computer readable instructions, data structures, program modules, or other data.
- the recording medium may be RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, DVD or other optical disk storage device, magnetic cassette, magnetic tape, magnetic disk storage device or other magnetic storage device, or And any other medium that can be accessed by a computer that can be used to store desired information, and the like.
- FIG. 1 is a conceptual diagram showing the appearance inspection apparatus of the article according to an embodiment of the present invention.
- the apparatus 100 for inspecting appearance of an article includes an image acquisition unit and a processing unit 130.
- the image acquisition unit acquires a photographed image of at least a part of the exterior of the article PD to be inspected.
- the image acquisition unit may directly acquire the photographed image by directly photographing the article PD, or may obtain the photographed image by receiving a photographed image of the article PD from the outside.
- the photographed image may include a two-dimensional plane image of at least a portion of the article PD, a pattern image of at least a portion of the article PD, and the like, and a two-dimensional shape of the article PD.
- the image may include an image including information on a three-dimensional shape and information on various attributes such as color, brightness, and saturation.
- the image acquisition unit may include an illumination unit 110 and a photographing unit 120.
- the lighting unit 110 provides light to the article PD.
- the article PD may include a mobile phone.
- the article PD may include a main body, a case, a cover, and the like of the mobile phone.
- the rear case of the mobile phone when the article PD is a rear case which covers and covers a mobile phone main body such as a circuit board from the rear, the rear case of the mobile phone generally has various protrusions, depressions, holes, openings, and the like.
- the rear case may have a rectangular parallelepiped shape including a rounded portion at least in part.
- the article PD may include a product or a workpiece of various other shapes, and may be an article having various shape features on the surface thereof, for example, protrusions, depressions, holes, openings, and the like.
- the lighting unit 110 may include at least one of the first lighting unit 112 and the second lighting unit 114.
- the first lighting unit 112 may provide non-pattern lighting.
- the non-pattern illumination may be, for example, illumination for obtaining a planar image of the two-dimensional shape of the article PD.
- the planar image may include at least one of hue, lightness or brightness, and saturation of the article PD.
- the first lighting unit 112 may include a plurality of lighting units arranged in a circle based on the article PD, which is a measurement object, to irradiate light L when viewed from a plane.
- the first lighting unit 112 may include a light source for irradiating white light or monochromatic light of a predetermined color, and may irradiate a plurality of different color lights such as red, green, and blue at different inclination angles.
- a plurality of LEDs may be continuously disposed to have a ring shape.
- the second lighting unit 114 may provide pattern lighting.
- the pattern lighting may be, for example, illumination for obtaining a pattern image from which a three-dimensional shape of the article PD may be extracted.
- the second lighting unit 114 may include a plurality of pattern lighting units arranged to irradiate the grid pattern light PL inclined in different directions toward the article PD.
- the second lighting unit 114 may have a grid pattern light PL, which forms a grid pattern for acquiring three-dimensional shape information of the plurality of articles PD, perpendicular to a plane of the article PD. You can do a sloped survey based on the normal.
- the second lighting unit 114 may be disposed spaced apart from each other along the circumferential direction with respect to the article PD, or may be disposed at each vertex of the polygon around the article PD, and divides the circumference of the circumference. They may be arranged alternately spaced at equal intervals.
- the pattern lighting units of the second lighting unit 114 may be provided with M (M is a natural number of 2 or more), for example, may be provided in various numbers, such as two, four or eight.
- each pattern lighting unit of the second lighting unit 114 may irradiate the grid patterned light PL N times toward the article PD (N is a natural number of 2 or more), and the phase shifted grid.
- N is a natural number of 2 or more
- a pattern image adopting a digital light processing (DLP) method using a digital micro-mirror display (DMD) or a pattern image of a liquid crystal display device is used.
- the grid pattern can be transferred N times, and the grid pattern can be transferred using pattern images of various display methods. Alternatively, the grid pattern may be physically transferred N times using the grid transfer mechanism.
- each pattern lighting unit of the second lighting unit 114 may include a light source 114a, a grating 114b, a grating transfer mechanism 114c and a projection lens unit 114d.
- the light source 114a irradiates light toward the article PD.
- the grating 114b converts the light irradiated from the light source 114a into the grating pattern light PL.
- the grating 114b transfers N times by 2 ⁇ / N through a grating transfer mechanism 114c such as, for example, a piezo actuator (PZT) to generate a phase shifted grating patterned light PL.
- N is a natural number of 2 or more.
- the projection lens part 114d projects the grating pattern light PL generated by the grating 114b onto the article PD.
- the projection lens part 114d may be formed of, for example, a plurality of lens combinations, and focuses the grating pattern light PL formed through the grating 114b to project it onto the article PD. Accordingly, each pattern lighting unit 110 irradiates the grid pattern light PL to the article PD at every transfer while transferring the grid 114b N times.
- the photographing unit 120 acquires a photographed image by photographing at least a part of the exterior of the article PD based on the provided light.
- the photographing unit 120 may acquire a two-dimensional planar image of the article PD based on the non-pattern illumination of the first lighting unit 112, and the second lighting unit 114 may be A pattern image of the article PD may be obtained based on pattern illumination.
- the photographing unit 120 receives the light reflected from the article PD from the first lighting unit 112 to reflect the two-dimensional planar image of the article PD.
- the pattern image may be obtained by receiving the pattern light reflected from the article PD from the second lighting unit 114 and reflected by the article PD.
- the photographing unit 120 may include a camera 122 and an imaging lens 124.
- the camera 122 may employ a CCD or CMOS camera.
- the light or patterned light reflected by the article PD may be imaged by the imaging lens 124 and imaged by the camera 122.
- the photographing unit 120 may be installed in one or a plurality.
- the photographing unit 120 may be disposed on the article PD, and a plurality of photographing units may be spaced apart from each other along the circumferential direction with respect to the article PD, or may have a polygonal shape with respect to the article PD. It may be disposed at each vertex, and may be alternately spaced at equal intervals in the position to divide the circumference.
- the photographing unit 120 when a plurality of the photographing unit 120 is installed, it is possible to obtain a three-dimensional shape of the article (PD) by a known stereo method, in which case the second lighting unit 114 may be employed May be omitted.
- the processing unit 130 determines whether or not the appearance of the article is defective by using the photographed image of the article PD.
- the processor 130 obtains first data about an inspection area corresponding to at least a part of an appearance of the article PD, and calculates a formula representing a distribution of the first data according to a position in the inspection area. Obtaining second data based on the relative relationship between the equation and the first data, and determining whether the article PD is defective in appearance based on the second data.
- the appearance inspection apparatus 100 of the article may further include a stage 140 for fixing and supporting the article PD, the processing unit 130 or a separate transfer control device (not shown)
- a control operation for moving the stage 140 and setting an inspection surface of the article PD may be performed.
- the processing unit 130 or a separate transfer may be performed.
- the controller may control the transfer and rotation operations to automatically set the inspection surface on the stage 140.
- FIG. 2 is a flowchart illustrating a method of inspecting appearance of an article according to an embodiment of the present invention.
- the processing unit 130 inspects at least a portion of the appearance of the article PD.
- operation S110 first data regarding an area is acquired.
- the inspection area may correspond to the curved portion of the article PD.
- the inspection area may be set in one dimension to inspect the article PD in one direction, and may be set in two dimensions to inspect the article PD in a plan view.
- the inspection area may correspond to the inclined portion of the article PD. That is, the inspection area may include various parts other than the flat plane which is not inclined, and may include various parts forming the shape of the article PD in addition to the curved part and the inclined part described above.
- the inspection area may be set in various forms to inspect the appearance of the article PD.
- the inspection area may be set in a rectangular form or may be set in a polygonal form. In addition, it may be set in the form of a figure including a curve, or may be formed in a contour form according to the shape of the article PD.
- FIG. 3 is a plan view illustrating examples of an inspection area set for the appearance inspection method of the article of FIG. 2, and FIG. 4 is a cross-sectional view taken along the line II ′ of FIG. 3.
- the first inspection region IR1 illustrated in FIG. 3 shows an example in which the inspection region is set in one dimension, and the second inspection region IR2 has a two-dimensional inspection region. As shown in FIG. 4, the first and second inspection regions IR1 and IR2 may correspond to a rounded portion RP of the article PD.
- the article PD is a mobile phone rear case, and a portion of the side of the rear case is shown in FIGS. 3 and 4. Side surfaces of the rear case may be formed to include a curved portion (RP) as shown in FIG.
- RP curved portion
- the first data may include at least one of brightness data and height data.
- the first data may include various data obtainable from the planar image, and may include, for example, color, brightness, saturation, and the like.
- the first data may include various data related to the shape of the article PD, and may include, for example, a three-dimensional spatial coordinate, a curvature, a radius of curvature, and the like.
- the brightness data may be obtained from data obtained based on the first lighting unit 112 or from a two-dimensional planar image, and the height data may be obtained from data obtained based on the second lighting unit 114 or from a three-dimensional shape. Can be obtained.
- the first data when the brightness data is adopted as the first data, the first data can be easily obtained from the photographed two-dimensional plane image, and when the height data is used as the first data, the three-dimensional shape is obtained.
- the first data can be easily obtained from the obtained data to measure.
- the region to be excluded may be excluded in the process of obtaining the first data (S110).
- FIG. 5 is a flowchart illustrating an example of a process of acquiring first data of the appearance inspection method of the article of FIG. 2, and FIG. 6 is a plan view illustrating an example of the process of acquiring first data of FIG. 5.
- the appearance of the article PD is first obtained.
- a target area TR corresponding to at least a part of is set (S112).
- the target area TR may be an area having a two-dimensional rectangular shape.
- the masking region MR may be a portion that should be excluded because it does not correspond to the article PD.
- the masking area MR is formed to correspond to a hole for exposing the button formed in the mobile phone body to the outside, a microphone or a speaker of the mobile phone body.
- a plurality of holes and openings formed for realizing the functions of the mobile phone such as a hole, a hole for exposing the input / output connection to the mobile phone body to the outside, a hole for camera shooting, a space corresponding to the location where the battery is mounted, and the like. Can be.
- the processor 130 may acquire an area in which the first data is out of a predetermined range as the masking area.
- the brightness data when the first data includes brightness data, the brightness data may be obtained from an area outside the preset reference brightness range to obtain a masking area MR to be excluded from the target area TR.
- the masking area MR may be obtained.
- the height data when the first data includes height data, the height data may be moved from an area outside the preset reference height range to obtain a masking area MR to be excluded from the target area TR.
- the masking area MR may be obtained.
- the inspection area IR is set by excluding the masking area MR from the target area TR (S116).
- the inspection region IR is a region other than the small rectangular masking region MR in the large rectangular target region TR.
- the processor 130 obtains an equation representing a distribution of the first data according to a position in the inspection area (S120).
- the inspection region may be a one-dimensional region, for example, a first inspection region IR1 as shown in FIG. 3.
- an equation representing the distribution of the first data may be an equation for the one-dimensional area, and for example, may be at least a quadratic curve equation.
- the curve equation may be obtained by curve fitting the distribution of the first data.
- FIG. 7 is a graph illustrating an example of a distribution of first data and an equation according to the process of obtaining the equation of FIG. 2.
- a gray level of brightness data for each position obtained according to a position in the D1-direction of the first inspection area IR1 may be represented as a two-dimensional graph, and the X axis of the graph may be The position along the D-direction, the Y axis, represents the gray level at that position.
- Equation 1 the equation may be modeled as in Equation 1.
- the inspection region may be a two-dimensional region, for example, a second inspection region IR2 as shown in FIG. 3.
- the equation representing the distribution of the first data may be an equation for the two-dimensional area, and for example, may be at least a quadratic or higher surface equation.
- the curved equation may be obtained by curve fitting the distribution of the first data.
- a gray level of brightness data for each position obtained according to the positions D1- and D2- of the second inspection area IR2 may be represented by a three-dimensional graph.
- the X axis may indicate a position along the D1-direction
- the Y axis may indicate a position along the D2- direction
- the Z axis may indicate a gray level of the corresponding position.
- the gray level can be assumed to follow approximately two or more quadratic curve equations, and thus, equations representing the distribution of the first data can be obtained by curve fitting the quadratic curve equations.
- the equation may be modeled as in Equation 2.
- the above equation may be simply obtained by performing curve fitting to a quadratic or more curved equation or a curved equation as described above.
- the inspection area is a two-dimensional area
- equations for the one-dimensional areas forming the two-dimensional area are obtained, and then the two-dimensional area is based on the equations for the one-dimensional areas.
- Equation for the inspection region may be obtained. That is, the two-dimensional inspection area is divided into one-dimensional regions, the equations for the divided one-dimensional regions are obtained, and then the equations for the one-dimensional regions are arranged in a space and the two-dimensional inspection is performed. Equation for the region can be derived.
- the first data may include noise, in which case the equation is difficult to represent the exact transition of the first data. Therefore, in operation S120 of obtaining an equation representing a change of the first data according to a position in the inspection area, a more accurate equation may be obtained by removing noise.
- FIG. 8 is a flowchart illustrating an example of a process of obtaining a mathematical expression of an appearance inspection method of the article of FIG. 2.
- the inspection area in order to obtain an equation representing a change of the first data according to a position in the inspection area (S120), first, the inspection area is set as an effective area, and the effective area is obtained.
- the equation is first obtained (S122), and then the effective area is corrected (S126) by extracting and removing a noise area based on the equation obtained first (S124), and then correcting the equation.
- the above equation may be modified based on the valid effective area (S122).
- the inspection area is set as a valid area that is a valid area for obtaining the equation, and a first equation is obtained for the valid area, and then a deviation from a preset reference value is obtained. Since the values cannot be regarded as valid values, it is determined as noise and the valid area is corrected by excluding the area corresponding to the corresponding values from the valid area.
- the noise may be spots, scratches, or the like corresponding to a defect in determining whether the article PD has a poor appearance. An area corresponding to such a defect may also be excluded from the effective area in this process.
- the equation obtained may be modified. This process may be repeated one or more times.
- the processor 130 obtains second data based on the relative relationship between the equation and the first data (S130).
- the second data may be obtained from a distance between the first data and a value according to the equation, for each position in the inspection area.
- the processor 130 subtracts a value according to the equation from the first data with respect to each position in the inspection area, thereby subtracting the second data.
- FIG. 9 is a graph illustrating a distribution of second data for each location extracted from the graph of FIG. 7 as a broken line.
- the gray level of the second data obtained by subtracting the brightness data of each position obtained according to the position of the first inspection area IR1 in the direction D1- and the value of Equation 1 at the corresponding position is 2; It can be represented as a dimensional graph, where the X axis of the graph represents a position along the D1-direction, and the Y axis represents a relative gray level of the position.
- the trend of the gray level of the second data shown in FIG. 9 is modified by following a constant value (corresponding to 0) from following the quadratic curve equation.
- the reason why the gray level trend of the first data follows the quadratic or higher curve equation instead of following the constant value is because the inspection area corresponds to the curved portion RP (see FIG. 4).
- the processor 130 may determine whether the appearance of the article PD is poor by applying a plane inspection criterion to the second data.
- the plane inspection criterion is an inspection criterion applied when the inspection region corresponds to a plane, for example, to determine a poor appearance of an article made of a flat plane, for example, foreign matter adhesion, scratches, surface stains, and the like.
- the planar inspection standard may include a predetermined inspection standard for determining whether the applicable brightness, color, height, and the like are uniformly distributed.
- the processor 130 may directly determine whether the article PD is defective in appearance as described below.
- the determination means provided separately may determine whether or not the appearance of the article PD is defective based on the second data, and the operator determines whether the appearance of the article PD is defective based on the second data. It can also be determined.
- the processing unit 130 determines whether or not the appearance of the article PD is poor based on the second data (S140).
- Poor appearance of the article PD may be caused by various causes, and may appear in various results.
- poor appearance of the article PD may include at least one of foreign material adhesion, scratches, and surface stains.
- the article PD may adhere to or adhere to foreign matters in the manufacturing process, and may form stains during anodizing, thereby resulting in appearance defects.
- the appearance defect As described above, it may be checked whether there is a region having a brightness distribution and / or a height distribution different from the surroundings in the inspection region.
- the second data is binarized to generate third data, and the generated third data. It may be determined whether or not the appearance of the article PD is poor.
- the third data may be generated by performing a binarization process of comparing the second data with a predetermined threshold and assigning 1 for a large value and 0 for a small value. Accordingly, the third data may display only extremely brightly the region having a different distribution from the surroundings, thereby easily determining whether the article PD has a poor appearance.
- the third data after generating the third data by binarizing the second data, the third data may be morphology processed. In this case, a part of the third data may be extracted and morphologically processed.
- the morphology processing includes an eroding operation (erode) in which the area is narrowed, an expansion operation (dilate) in which the area is widened, an open operation in which a detail area is removed from the area, and a closing operation in which a gap is filled in the area ( close) and the like, and by performing image processing such as changing the corresponding area wider or narrower within the inspection area, it is possible to easily extract the area having a different distribution from the surrounding area.
- a process of filtering the garb wavelet may be performed.
- the second data may be modified to further emphasize linearity by the filtering process. Therefore, when determining whether the article PD is defective in appearance by using the third data, it is possible to more easily determine the scratch defect of the article PD, so that the scratch defect is not properly exposed by the morphology process. Possible problems can be prevented.
- the Gabor wavelet filtering process can be used selectively to identify scratch defects more easily, and thus can be selectively employed. It may also determine whether or not the appearance is poor.
- more accurate inspection results may be obtained by performing data processing or image processing such as binarization, various filtering, and morphology processing.
- the images 10 to 13 are examples of images according to various data shown by the appearance inspection method of the article according to an embodiment of the present invention.
- the images show examples of side surfaces of a button hole of a mobile phone case, and are examples including a curved portion as shown in FIG. 4.
- FIG. 10 is an image corresponding to the first data acquired according to step S110 of FIG. 2, which is an actual photographed image of the article PD, and FIG. 11 illustrates the first data acquired according to step S120 of FIG. 2.
- FIG. 12 is an image corresponding to the second data acquired according to step S130 of FIG. 2, and
- FIG. 13 is a binarized image obtained to determine whether there is a defect according to step S140 of FIG. 2.
- Equation representing the distribution of the first data except for the hole HL in order to determine whether the article is defective by the appearance inspection apparatus 100 (see FIG. 1) of the article for the first spot (ST1).
- a curved equation can be obtained and the resulting image can be represented as shown in FIG. 11.
- the image corresponding to the curved equation shown in FIG. 11 is subtracted from the image corresponding to the first data shown in FIG. 10 to obtain an image according to the second data as shown in FIG. 12.
- the first spot ST1 may appear as the second spot ST2, and the entire gradation due to the bent portion may be removed.
- the second data is binarized and other image processing is selectively performed to obtain a final image for determining whether the appearance is defective as shown in FIG. 13.
- the second spot ST2 may appear to be clearly visible as the third spot ST3, and the entire gradation due to the bent portion may be almost removed.
- the external appearance inspection apparatus of the article and the external appearance inspection method of the article by obtaining the second data relative to the equation representing the distribution of the first data for the inspection area by determining whether the appearance of the article defective
- the appearance of the article does not have a flat plane shape
- Direct application can easily determine whether the article is defective in appearance.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Description
Claims (15)
- 물품의 외관의 적어도 일부에 대한 촬영 이미지를 획득하는 이미지 획득부; 및상기 물품의 촬영 이미지를 이용하여 상기 물품의 외관 불량 여부를 판정하는 처리부를 포함하고,상기 처리부는,상기 촬영 이미지의 적어도 일부를 포함하는 검사영역에 관한 제1 영상데이터를 획득하고, 상기 검사영역 내의 위치에 따른 상기 제1 영상데이터의 레벨의 추이를 추종하는 수학식을 획득하며, 상기 수학식과 상기 제1 영상데이터의 레벨 사이의 상대적 레벨에 기초하여 제2 영상데이터를 획득하는 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 제2 영상데이터는, 상기 검사영역 내의 각 위치에 대하여, 상기 제1 영상데이터와 상기 수학식에 따른 값 사이의 거리로부터 획득되는 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 처리부는, 상기 제2 영상데이터를 기초로 상기 물품의 외관 불량 여부를 판정하는 것을 특징으로 하는 물품의 외관 검사장치.
- 제3항에 있어서,상기 처리부는, 평면검사기준을 상기 제2 영상데이터에 적용하여 상기 물품의 외관 불량 여부를 판정하고,상기 평면검사기준은 상기 검사영역이 평면에 대응하는 경우 적용되는 검사기준인 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,비패턴조명을 제공하는 제1 조명부; 및패턴조명을 제공하는 제2 조명부 중 적어도 하나를 더 포함하고,상기 제1 영상데이터는 상기 제1 조명부에 의해 획득되는 밝기 데이터 및 상기 제2 조명부에 의해 획득되는 높이 데이터 중 적어도 하나를 포함하는 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 검사영역은 1차원 영역 또는 2차원 영역이고,상기 제1 영상데이터의 레벨의 추이를 추종하는 수학식은 상기 검사영역의 차원과 동일한 차원의 영역에 대한 수학식인 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 검사영역은 2차원 영역이고,상기 처리부는, 상기 2차원 검사영역을 형성하는 1차원 영역들에 대한 수학식들을 획득하고, 상기 1차원 영역들에 대한 수학식들을 기초로 상기 2차원 검사영역에 대한 수학식을 획득하는 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 처리부는,상기 촬영 이미지에서 대상영역을 설정하고, 상기 대상영역 내에서 제외될 마스킹(masking) 영역을 획득하여, 상기 대상영역에서 상기 마스킹 영역을 제외시킴에 의해 상기 검사영역을 설정하는 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 처리부는, 상기 수학식을 획득할 때, 상기 검사영역을 유효영역으로 설정하고 상기 유효영역에 대하여 상기 수학식을 일차적으로 획득하고,일차적으로 획득된 상기 수학식을 기초로 노이즈 영역을 추출하여 제거함에 의해 상기 유효영역을 수정하며,상기 수정된 유효영역을 기초로 상기 수학식을 수정하는 것을 특징으로 하는 물품의 외관 검사장치.
- 제1항에 있어서,상기 처리부는,상기 제2 영상데이터를 이진화 처리하여 제3 영상데이터를 생성하고,상기 제3 영상데이터를 이용하여 상기 물품의 외관 불량 여부를 판정하는 것을 특징으로 하는 물품의 외관 검사장치.
- 물품의 외관 검사장치를 이용한 물품의 외관 검사방법으로서,상기 외관 검사장치가,물품의 외관의 적어도 일부에 대한 촬영 이미지의 적어도 일부를 포함하는 검사영역에 관한 제1 영상데이터를 획득하는 단계;상기 검사영역 내의 위치에 따른 상기 제1 영상데이터의 레벨의 추이를 추종하는 수학식을 획득하는 단계; 및상기 수학식과 상기 제1 영상데이터의 레벨 사이의 상대적 레벨에 기초하여 제2 영상데이터를 획득하는 단계;를 포함하는 물품의 외관 검사방법.
- 제11항에 있어서, 상기 수학식과 상기 제1 영상데이터의 레벨 사이의 상대적 레벨에 기초하여 제2 영상데이터를 획득하는 단계 이후에,상기 제2 영상데이터를 기초로 상기 물품의 외관 불량 여부를 판정하는 단계를 더 포함하는 것을 특징으로 하는 물품의 외관 검사방법.
- 제12항에 있어서,상기 제2 영상데이터를 기초로 상기 물품의 외관 불량 여부를 판정하는 단계에서, 평면검사기준을 상기 제2 영상데이터에 적용하여 상기 물품의 외관 불량 여부를 판정하고,상기 평면검사기준은 상기 검사영역이 평면에 대응하는 경우 적용되는 검사기준인 것을 특징으로 하는 물품의 외관 검사방법.
- 제11항에 있어서,상기 물품의 외관의 적어도 일부에 대한 촬영 이미지의 적어도 일부를 포함하는 검사영역에 관한 제1 영상데이터를 획득하는 단계는,상기 촬영 이미지에서 대상영역을 설정하는 단계;상기 대상영역 내에서 제외될 마스킹 영역을 획득하는 단계; 및상기 대상영역에서 상기 마스킹 영역을 제외시켜 상기 검사영역을 설정하는 단계를 포함하는 것을 특징으로 하는 물품의 외관 검사방법.
- 제11항에 있어서,상기 검사영역 내의 위치에 따른 상기 제1 영상데이터의 레벨의 추이를 추종하는 수학식을 획득하는 단계는,상기 검사영역을 유효영역으로 설정하고 상기 유효영역에 대하여 상기 수학식을 일차적으로 획득하는 단계;일차적으로 획득된 상기 수학식을 기초로 노이즈 영역을 추출하여 제거함에 의해 상기 유효영역을 수정하는 단계; 및상기 수정된 유효영역을 기초로 상기 수학식을 수정하는 단계;를 포함하는 것을 특징으로 하는 물품의 외관 검사방법.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/094,970 US11100629B2 (en) | 2016-04-20 | 2017-04-20 | Appearance inspecting apparatus for article and appearance inspecting method for article using the same |
EP17786191.1A EP3447725B1 (en) | 2016-04-20 | 2017-04-20 | Apparatus for inspecting appearance of article and method for inspecting appearance of article using same |
JP2018554687A JP2019514012A (ja) | 2016-04-20 | 2017-04-20 | 品物の外観検査装置及びそれを用いた品物の外観検査方法 |
CN201780024295.8A CN109074647B (zh) | 2016-04-20 | 2017-04-20 | 物品的外观检查装置及利用其的物品的外观检查方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2016-0048299 | 2016-04-20 | ||
KR1020160048299A KR101802812B1 (ko) | 2016-04-20 | 2016-04-20 | 물품의 외관 검사장치 및 이를 이용한 물품의 외관 검사방법 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017183923A1 true WO2017183923A1 (ko) | 2017-10-26 |
Family
ID=60117075
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2017/004238 WO2017183923A1 (ko) | 2016-04-20 | 2017-04-20 | 물품의 외관 검사장치 및 이를 이용한 물품의 외관 검사방법 |
Country Status (6)
Country | Link |
---|---|
US (1) | US11100629B2 (ko) |
EP (1) | EP3447725B1 (ko) |
JP (1) | JP2019514012A (ko) |
KR (1) | KR101802812B1 (ko) |
CN (1) | CN109074647B (ko) |
WO (1) | WO2017183923A1 (ko) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3664023A1 (en) * | 2018-12-04 | 2020-06-10 | General Electric Company | System and method for work piece inspection |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102012125B1 (ko) * | 2018-03-19 | 2019-10-21 | 주식회사 삼영정보 | 물품조사시스템 |
KR102180648B1 (ko) * | 2019-08-05 | 2020-11-19 | 주식회사 휴비츠 | 3차원 단층촬영 검사 장치 및 방법 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005292081A (ja) * | 2004-04-05 | 2005-10-20 | Mitsubishi Electric Corp | 円板形状部品の欠け検査装置及び欠け検査方法 |
KR20090052369A (ko) * | 2007-05-25 | 2009-05-25 | 도요타지도샤가부시키가이샤 | 형상 평가 방법, 형상 평가 장치 및 3차원 검사 장치 |
JP2010256053A (ja) * | 2009-04-22 | 2010-11-11 | Visco Technologies Corp | 形状欠損検査装置、形状モデリング装置および形状欠損検査プログラム |
KR20110105512A (ko) * | 2010-03-19 | 2011-09-27 | 주식회사 고영테크놀러지 | 검사방법 |
JP2015148447A (ja) * | 2014-02-04 | 2015-08-20 | 東レエンジニアリング株式会社 | 自動外観検査装置 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5091963A (en) | 1988-05-02 | 1992-02-25 | The Standard Oil Company | Method and apparatus for inspecting surfaces for contrast variations |
KR100558943B1 (ko) * | 2004-01-14 | 2006-03-10 | 삼성전자주식회사 | 평판 디스플레이 검사방법 |
JP5217221B2 (ja) * | 2007-04-16 | 2013-06-19 | 新日鐵住金株式会社 | 溶接部の表面欠陥形状検出方法及びコンピュータプログラム |
JP4894628B2 (ja) | 2007-05-28 | 2012-03-14 | パナソニック電工株式会社 | 外観検査方法および外観検査装置 |
CN102084213B (zh) | 2008-06-04 | 2013-09-25 | 株式会社神户制钢所 | 轮胎形状检查方法、轮胎形状检查装置 |
JP5347418B2 (ja) * | 2008-10-10 | 2013-11-20 | 新日鐵住金株式会社 | 表面欠陥検査システム、方法及びプログラム |
JP5726045B2 (ja) * | 2011-11-07 | 2015-05-27 | 株式会社神戸製鋼所 | タイヤ形状検査方法、及びタイヤ形状検査装置 |
WO2014070156A1 (en) | 2012-10-31 | 2014-05-08 | Compagnie Generale Des Etablissements Michelin | System and method for analyzing tire tread parameters |
JP6506914B2 (ja) | 2013-07-16 | 2019-04-24 | 株式会社キーエンス | 三次元画像処理装置、三次元画像処理方法及び三次元画像処理プログラム並びにコンピュータで読み取り可能な記録媒体及び記録した機器 |
-
2016
- 2016-04-20 KR KR1020160048299A patent/KR101802812B1/ko active IP Right Grant
-
2017
- 2017-04-20 CN CN201780024295.8A patent/CN109074647B/zh active Active
- 2017-04-20 WO PCT/KR2017/004238 patent/WO2017183923A1/ko active Application Filing
- 2017-04-20 US US16/094,970 patent/US11100629B2/en active Active
- 2017-04-20 JP JP2018554687A patent/JP2019514012A/ja active Pending
- 2017-04-20 EP EP17786191.1A patent/EP3447725B1/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005292081A (ja) * | 2004-04-05 | 2005-10-20 | Mitsubishi Electric Corp | 円板形状部品の欠け検査装置及び欠け検査方法 |
KR20090052369A (ko) * | 2007-05-25 | 2009-05-25 | 도요타지도샤가부시키가이샤 | 형상 평가 방법, 형상 평가 장치 및 3차원 검사 장치 |
JP2010256053A (ja) * | 2009-04-22 | 2010-11-11 | Visco Technologies Corp | 形状欠損検査装置、形状モデリング装置および形状欠損検査プログラム |
KR20110105512A (ko) * | 2010-03-19 | 2011-09-27 | 주식회사 고영테크놀러지 | 검사방법 |
JP2015148447A (ja) * | 2014-02-04 | 2015-08-20 | 東レエンジニアリング株式会社 | 自動外観検査装置 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3447725A4 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3664023A1 (en) * | 2018-12-04 | 2020-06-10 | General Electric Company | System and method for work piece inspection |
CN111272763A (zh) * | 2018-12-04 | 2020-06-12 | 通用电气公司 | 用于工件检查的系统和方法 |
US10755401B2 (en) | 2018-12-04 | 2020-08-25 | General Electric Company | System and method for work piece inspection |
CN111272763B (zh) * | 2018-12-04 | 2023-04-07 | 通用电气公司 | 用于工件检查的系统和方法 |
Also Published As
Publication number | Publication date |
---|---|
EP3447725A4 (en) | 2019-05-08 |
KR101802812B1 (ko) | 2017-11-29 |
CN109074647A (zh) | 2018-12-21 |
KR20170119952A (ko) | 2017-10-30 |
CN109074647B (zh) | 2022-03-15 |
US11100629B2 (en) | 2021-08-24 |
JP2019514012A (ja) | 2019-05-30 |
US20190147575A1 (en) | 2019-05-16 |
EP3447725A1 (en) | 2019-02-27 |
EP3447725B1 (en) | 2020-05-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2014163375A1 (ko) | 기판의 이물질 검사방법 | |
WO2012050378A2 (ko) | 기판 검사방법 | |
KR101735403B1 (ko) | 검사 방법, 템플릿 기판 및 포커스 오프셋 방법 | |
WO2017183923A1 (ko) | 물품의 외관 검사장치 및 이를 이용한 물품의 외관 검사방법 | |
WO2019164381A1 (ko) | 부품의 실장 상태를 검사하기 위한 방법, 인쇄 회로 기판 검사 장치 및 컴퓨터 판독 가능한 기록매체 | |
TW201606260A (zh) | 膜厚測定裝置、膜厚測定方法及非暫時性電腦記錄媒體 | |
WO2018044111A2 (ko) | 색수차를 이용한 3차원 스캐너 및 스캐닝 방법 | |
TW201405120A (zh) | 外觀檢查裝置及外觀檢查方法 | |
JP7157580B2 (ja) | 基板検査方法及び基板検査装置 | |
TWI498543B (zh) | 自動晶圓光學檢測裝置及檢測晶圓表面均勻性的方法 | |
US11074684B2 (en) | Inspection apparatus and inspection method | |
WO2023033627A1 (ko) | 돔 조명 및 동축 조명을 포함하는 광학 시스템 및 광학 시스템의 설계 방법 | |
KR100710703B1 (ko) | 반도체 리드프레임 도금 선폭 측정 검사장치 및 그 방법 | |
WO2020263054A1 (ko) | 대상체의 3차원 형상을 결정하기 위한 장치 및 방법 | |
JPH02128284A (ja) | 欠陥検査における画像処理方法 | |
TW201929116A (zh) | 工件處理裝置、工件輸送系統 | |
JP3865156B2 (ja) | 画像比較装置およびこれを用いたウエハ検査装置とウエハ検査システム | |
KR20150049799A (ko) | 기판의 표면 검사 방법 및 이를 수행하기 위한 장치 | |
WO2024063471A1 (ko) | 제품의 외관 검사 장치 및 방법 | |
TW201834102A (zh) | 外觀檢查裝置 | |
WO2016093597A9 (ko) | 기판 상에 형성된 부품의 터미널 검사방법 및 기판 검사장치 | |
WO2021101142A1 (ko) | 비접촉식 위치정보 획득장치 및 이를 이용한 위치정보 획득방법 | |
JP2001236493A (ja) | 外観検査装置 | |
TW202109022A (zh) | 晶圓外觀檢查裝置及方法 | |
JP2002015303A (ja) | パターンの欠陥検出方法と装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
ENP | Entry into the national phase |
Ref document number: 2018554687 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2017786191 Country of ref document: EP |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17786191 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017786191 Country of ref document: EP Effective date: 20181120 |