WO2022003919A1 - 検査データ作成方法、検査データ作成装置および検査装置 - Google Patents

検査データ作成方法、検査データ作成装置および検査装置 Download PDF

Info

Publication number
WO2022003919A1
WO2022003919A1 PCT/JP2020/026087 JP2020026087W WO2022003919A1 WO 2022003919 A1 WO2022003919 A1 WO 2022003919A1 JP 2020026087 W JP2020026087 W JP 2020026087W WO 2022003919 A1 WO2022003919 A1 WO 2022003919A1
Authority
WO
WIPO (PCT)
Prior art keywords
component
similarity
image
parts
inspection data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2020/026087
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
孝智 水嶋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yamaha Motor Co Ltd
Original Assignee
Yamaha Motor Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yamaha Motor Co Ltd filed Critical Yamaha Motor Co Ltd
Priority to PCT/JP2020/026087 priority Critical patent/WO2022003919A1/ja
Priority to JP2022532968A priority patent/JP7448655B2/ja
Priority to TW110121908A priority patent/TWI841850B/zh
Publication of WO2022003919A1 publication Critical patent/WO2022003919A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05K—PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K13/00—Apparatus or processes specially adapted for manufacturing or adjusting assemblages of electric components
    • H05K13/08—Monitoring manufacture of assemblages
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
    • Y02P70/50—Manufacturing or production processes characterised by the final manufactured product

Definitions

  • the present invention relates to an inspection data creation method, an inspection data creation device and an inspection device, and more particularly to an inspection data creation method for setting inspection data as component data, an inspection data creation device and an inspection device.
  • the above-mentioned republication 2018/105100 gazette discloses an inspection job data creation method (inspection data creation method) for setting inspection job data as component data.
  • inspection job data creation method first, the dimensions of the electronic component are extracted from the finished product image data of the finished product substrate. Then, in this inspection job data creation method, it is inquired whether or not the dimensions of the extracted electronic components match the dimensions of the electronic components in the component library in which the component data of various types of electronic components are registered in advance. .. Then, in this inspection job data creation method, inspection data as component data is set based on the fact that the dimensions of the electronic components match.
  • the electronic parts of the parts library in which the dimensions of the extracted electronic parts and the part data of various kinds of electronic parts are registered in advance are used. Since the inspection job data (inspection data) as part data is only set based only on the fact that the dimensions match, the applicable range is narrow and the inspection job data (inspection data) as part data can be easily set. There is a problem that it is difficult.
  • the present invention has been made to solve the above-mentioned problems, and one object of the present invention is to set inspection data as component data based on simply matching dimensions.
  • the present invention provides an inspection data creation method, an inspection data creation device, and an inspection device capable of easily setting inspection data as component data.
  • the inspection data creation method includes a step of acquiring a mounted board image of a board on which a component is mounted, a step of extracting a component image of a component in the mounted board image, and a component image. Based on this, it includes a step of acquiring the similarity of parts for each part and a step of setting inspection data as part data used in an inspection device for inspecting a substrate based on the similarity of parts for each part. ..
  • a step of setting inspection data as component data used in the inspection device for inspecting the above is provided. This is different from setting inspection data as part data based solely on the fact that the dimensions match, even if the dimensions do not match perfectly, based on the similarity of the parts for each part. Inspection data can be set as data. As a result, since the applicable range can be widened, it is possible to easily set the inspection data as the part data as compared with the case where the inspection data as the part data is simply set based on the matching of the dimensions. It is possible to provide a method for creating various inspection data.
  • the similarity of the parts for each part includes a numerical value
  • the part having the highest numerical value of the similarity of the parts is used as the part data.
  • steps to set as inspection data for With this configuration, the component with the highest numerical value of the similarity of the components can be set as inspection data as component data. As a result, inspection data as component data can be set more easily.
  • the step of acquiring the similarity of the parts for each part is the image processing by the machine learning model learned in advance, the shading in the image, and the height in the image. It includes image processing using at least one of them and a step of acquiring the similarity of parts for each part by using at least one of them.
  • the parts for each part can be processed by image processing using a machine learning model learned in advance, or by using image processing using at least one of the shades in the image and the height in the image.
  • the degree of similarity can be easily obtained.
  • the image processing by the machine learning model learned in advance it is possible to easily make a highly accurate judgment (acquisition of similarity) close to the judgment of a skilled person.
  • the step of acquiring the similarity of the parts for each part includes the step of acquiring the similarity of the types of parts for each part type based on the part image. ..
  • the similarity of parts for each part based on the similarity of the types of parts for each part type.
  • the step of acquiring the similarity of the parts for each part is the similarity of the types of parts in addition to the step of acquiring the similarity of the types of parts for each type of parts based on the part image.
  • it further includes a step to acquire the similarity of the shape of the part for each part shape based on the part image.
  • the step of acquiring the similarity of the parts for each part is the step of acquiring the similarity of the parts for each part type.
  • the step of including and setting the inspection data includes a step of setting the inspection data as the part data from the registered part data of the part registration information.
  • the similarity of the shape of the part is the similarity of the outer shape of the part and the main body of the part. Includes at least one of the similarity of the components and the similarity of the terminals of the component.
  • the step for acquiring the component similarity for each component is based on the component image.
  • the width of the outer shape of the part, the length of the outer shape of the part, the thickness of the outer shape of the part, the width of the main body of the part, the length of the main body of the part, the thickness of the main body of the part, the number of terminals of the part, and the part. Includes a step to obtain at least one of the terminal pitch of the component, the width of the terminal of the component, and the length of the terminal of the component.
  • the similarity of the shape of the component can be appropriately obtained based on the pitch of the terminal of the component, the width of the terminal of the component, or the length of the terminal of the component.
  • the step of acquiring the similarity of the parts for each part is machine learning learned in advance.
  • image processing with a model the similarity of the part type for each part type is acquired, and the part shape is used by image processing using at least one of the shade in the image and the height in the image. Includes a step to get the similarity of the shape of each part.
  • the image processing by the machine learning model learned in advance is used to appropriately acquire the similarity of the component types for each component type, and the shading in the image and the height in the image are included.
  • the step for acquiring the similarity of the parts for each part is the shading in the image and the image.
  • the step for acquiring at least one of the similarity of the part type for each part type and the similarity of the shape of the part for each part shape by using image processing using at least one of the medium heights. Includes a step that considers at least one of information about the color of the substrate and information about the thickness of the solder on the substrate.
  • the board and the component can be easily distinguished. As a result, it is possible to accurately obtain the degree of similarity of the types of parts for each part type or the degree of similarity of the shapes of parts for each part shape. Further, when considering the information on the thickness of the solder on the board, since the reference of the height for distinguishing the solder on the board and the terminal of the component is known, the solder on the board and the terminal of the component can be easily distinguished. As a result, it is possible to accurately obtain the degree of similarity of the types of parts for each part type or the degree of similarity of the shapes of parts for each part shape.
  • the step of extracting the component image in the mounted board image is used in the component mounting device for mounting the component on the board, and the shape of the component and the mounting of the component. It includes a step of extracting a component image in a mounted board image based on at least one of the board data including the position and the board CAD data as the image data of the board.
  • the step of setting the inspection data is when the information of the terminal of the component set as the inspection data is different from the information of the terminal of the component acquired based on the component image. , Includes a step of creating new inspection data by overwriting the terminal information of the component set as inspection data with the terminal information of the component acquired based on the component image. With this configuration, it is possible to suppress the setting of incorrect component terminal information as inspection data.
  • the inspection data creation device extracts the component image of the component in the mounted board image and the acquisition unit that acquires the mounted board image of the board on which the component is mounted, and is based on the component image.
  • the control unit is provided with a control unit that acquires the similarity of the parts for each part and sets the inspection data as the part data used in the inspection device for inspecting the substrate based on the similarity of the parts for each part.
  • the similarity of the parts for each part is acquired based on the part image, and the substrate is inspected based on the similarity of the parts for each part.
  • a control unit for setting inspection data as component data used in the inspection device is provided.
  • the inspection data as the part data can be easily set as compared with the case where the inspection data as the part data is simply set based on the fact that the dimensions match, as in the inspection data creation method according to the first aspect. It is possible to provide an inspection data creation device capable of performing.
  • the inspection device captures a board on which components are mounted and acquires a mounted board image, and extracts a component image of the component in the mounted board image and converts it into a component image. Based on this, it is provided with a control unit that acquires the similarity of parts for each part and sets inspection data as part data based on the similarity of parts for each part.
  • the similarity of the parts for each part is acquired based on the part image, and the inspection as part data is performed based on the similarity of the parts for each part.
  • a control unit for setting data is provided.
  • a data creation method, an inspection data creation device and an inspection device can be provided.
  • FIG. 1 is a diagram for explaining acquisition of similarity of parts by an inspection data creating device according to an embodiment. It is a 2nd figure for demonstrating the acquisition of the degree of similarity of a part by the inspection data creation apparatus by one Embodiment.
  • the component mounting system 100 includes an inspection data creating device 10, a database 20, and an inspection device 30.
  • the inspection data creation device 10 is a device for creating inspection data used in the inspection device 30.
  • the inspection data creating device 10 is, for example, a personal computer configured to be capable of performing various operations.
  • the inspection data creating device 10 includes a control unit 11, a storage unit 12, an input unit 13, a display unit 14, and a communication unit 15.
  • the control unit 11 is a control circuit that controls the operation of the inspection data creation device 10.
  • the control unit 11 includes a processor such as a CPU (Central Processing Unit).
  • the storage unit 12 includes a non-volatile storage medium capable of storing various types of information (data) and reading out the stored information (data).
  • the storage unit 12 stores, for example, a mounted board image 40 acquired from the inspection device 30, inspection data 50 to be transmitted to the inspection device 30, and the like.
  • the communication unit 15 is an example of the "acquisition unit" in the claims.
  • the input unit 13 includes a mouse, a keyboard, and the like, and is configured to accept input operations by the user.
  • the display unit 14 includes, for example, a liquid crystal monitor, and displays information such as inspection data 50.
  • the communication unit 15 is configured to be able to communicate with an external device.
  • the communication unit 15 is configured to be able to communicate with the database 20 and the inspection device 30.
  • the communication unit 15 is configured to acquire the mounted board image 40 from the inspection device 30.
  • the database 20 includes a component library 21 in which component data for each component C mounted on the board B (see FIG. 2) is registered.
  • the component library 21 includes information on the type of component C, information on the shape of component C, and information on inspection of component C as component data for each component C.
  • Information on the type of component C includes, for example, information on the type of component C such as chip capacitors, chip resistors, chip inductors, chip transistors, LEDs, SOPs, BGAs, QFPs, and connectors.
  • the information on the shape of the part C is, for example, the width of the outer shape of the part C, the length of the outer shape of the part C, the thickness of the outer shape of the part C, the width of the main body of the part C, the length of the main body of the part C, and the part. Thickness of the main body of C, number of terminals (leads) of component C, pitch of terminals (leads) of component C, width of terminals (leads) of component C, length of terminals (leads) of component C, etc.
  • the inspection information of the component C includes, for example, the inspection information of the component C such as the position of the inspection frame, the inspection site, and the inspection parameter.
  • the parts library 21 is an example of "parts registration information" in the claims.
  • the inspection device 30 is a device that photographs the substrate B as an inspection target and performs various inspections on the substrate B and the component C on the substrate B.
  • the inspection device 30 constitutes a part of a board manufacturing line for manufacturing a circuit board by mounting the component C on the board B.
  • solder solder paste
  • solder printing device not shown
  • the component C is mounted (mounted) on the substrate B after solder printing by a component mounting device (not shown) (mounting step), so that the terminals of the component C are arranged on the solder.
  • the substrate B on which the component C is mounted is transferred to a reflow furnace (not shown) to melt and cure (cool) the solder (reflow process), so that the terminals of the component C are connected to the wiring of the substrate B.
  • a reflow furnace not shown
  • the solder reflow process
  • the inspection device 30 may, for example, inspect the printed state of the solder on the substrate B after the solder printing process, inspect the mounted state of the component C after the mounting process, or inspect the mounted state of the component C after the reflow process. Used for. Therefore, one or more inspection devices 30 are provided in the substrate production line. As the printing state of the solder, the printing position deviation from the design printing position, the shape, volume and height (coating amount) of the solder, the presence or absence of a bridge (short circuit), and the like are inspected.
  • the type and orientation (polarity) of the component C are appropriate, whether the amount of misalignment with respect to the design mounting position of the component C is within the allowable range, and whether the solder joint state of the terminals is normal. Inspections such as whether or not are performed. In addition, as a common inspection content between each process, foreign matter such as dust and other deposits is also detected.
  • the inspection device 30 includes a substrate transfer conveyor 31 for transporting the substrate B and a head that can move above the substrate transfer conveyor 31 in the XY direction (horizontal direction) and the Z direction (vertical direction). It includes a moving mechanism 32, a measuring unit 33 held by the head moving mechanism 32, and a control device 34 for controlling the inspection device 30.
  • the substrate transfer conveyor 31 is configured so that the substrate B can be conveyed in the horizontal direction and the substrate B can be stopped and held at a predetermined inspection position. Further, the substrate transfer conveyor 31 is configured so that the substrate B for which the inspection has been completed can be transported horizontally from a predetermined inspection position and the substrate B can be carried out from the inspection device 30.
  • the head moving mechanism 32 is provided above the substrate transfer conveyor 31, and is configured by, for example, an orthogonal 3-axis (XYZ-axis) robot using a ball screw axis and a servomotor.
  • the head moving mechanism 32 includes an X-axis motor, a Y-axis motor, and a Z-axis motor for driving these X-axis, Y-axis, and Z-axis. With these X-axis motors, Y-axis motors and Z-axis motors, the head moving mechanism 32 makes the measuring unit 33 move in the XY direction (horizontal direction) and the Z direction (vertical direction) above the substrate transfer conveyor 31 (board B). It is configured so that it can be moved to.
  • the measuring unit 33 is configured to measure (photograph) two-dimensional information (two-dimensional image) and three-dimensional information (three-dimensional image).
  • the measuring unit 33 includes a photographing unit 33a and a projection unit 33b.
  • the measuring unit 33 is moved to a predetermined position above the substrate B by the head moving mechanism 32, and by using the photographing unit 33a, the projection unit 33b, or the like, the measuring unit 33 becomes the substrate B, the component C on the substrate B, or the like. It is configured to take pictures for visual inspection.
  • the photographing unit 33a is configured to photograph the substrate B irradiated with the striped pattern light by the projection unit 33b.
  • the photographing unit 33a has an image pickup element such as a CCD image sensor and a CMOS image sensor.
  • the photographing unit 33a is configured so that the substrate B can be photographed in a substantially rectangular photographing area.
  • the optical axis 33c is arranged in the direction perpendicular to the reference plane in the horizontal direction. That is, the photographing unit 33a is configured to acquire a two-dimensional image of the upper surface of the substrate B from a position substantially vertically above.
  • the photographing unit 33a obtains a two-dimensional image under the illumination light of the projection unit 33b.
  • a plurality of projection units 33b are provided. Further, each of the plurality of projection units 33b is configured to project a measurement pattern to be photographed by the photographing unit 33a from a direction inclined with respect to the optical axis 33c direction of the photographing unit 33a. That is, the measurement unit 33 is configured to project measurement patterns from a plurality of directions to measure three-dimensional information.
  • a plurality (for example, four) of the projection units 33b are arranged so as to surround the periphery of the photographing unit 33a when viewed from above. Further, the plurality of projection units 33b are arranged at positions equidistant from the center of photography (photographing unit 33a) at intervals of substantially equal angles (approximately 90 degrees).
  • the projection unit 33b is configured to project a grid-like light-dark pattern (striped pattern light) having a sinusoidal light intensity distribution at equal intervals as a measurement pattern. Further, the projection unit 33b is configured to shift the position (phase) of the light / dark pattern for projection.
  • the control device 34 is configured to control each part of the inspection device 30.
  • the control device 34 includes a control unit 34a, a storage unit 34b, an image processing unit 34c, an imaging control unit 34d, a projection control unit 34e, and a motor control unit 34f.
  • the control unit 34a includes a processor such as a CPU that executes a logical calculation, a ROM (Read Only Memory) that stores a program that controls the CPU, and a RAM (Random Access Memory) that temporarily stores various data during the operation of the device. ) And so on.
  • the control unit 34a passes through the image processing unit 34c, the photographing control unit 34d, the projection control unit 34e, and the motor control unit 34f according to the program stored in the ROM and the software (program) stored in the storage unit 34b. It is configured to control each part of the inspection device 30. Then, the control unit 34a controls the measurement unit 33 to perform various visual inspections on the substrate B.
  • the storage unit 34b includes a non-volatile storage medium capable of storing various information (data) and reading the stored information (data).
  • the storage unit 34b stores, for example, captured image data (mounted board image 40) captured by the photographing unit 33a, inspection data 50 acquired from the inspection data creating device 10, and the like. Based on the inspection data 50, the inspection apparatus 30 performs various inspections on the substrate B and the component C on the substrate B.
  • the image processing unit 34c performs image processing on the captured image (photographed signal) captured by the photographing unit 33a, and recognizes (image recognition) the substrate B, the component C on the substrate B, and the solder joint portion (solder). It is configured to generate image data suitable for.
  • the shooting control unit 34d is configured to read the shooting signal from the shooting unit 33a at a predetermined timing based on the control signal output from the control unit 34a and output the read shooting signal to the image processing unit 34c. Has been done.
  • the projection control unit 34e is configured to control the projection by the projection unit 33b based on the control signal output from the control unit 34a.
  • the motor control unit 34f drives each servomotor of the inspection device 30 (X-axis motor, Y-axis motor and Z-axis motor of the head moving mechanism 32, and the substrate transfer conveyor 31 based on the control signal output from the control unit 34a. It is configured to control the drive of a motor (not shown) for this purpose. Further, the motor control unit 34f is configured to acquire the positions of the measurement unit 33, the substrate B, and the like based on a signal from an encoder (not shown) of each servomotor.
  • step S101 the mounted board image (completed board image) 40 of the board B on which the component C is mounted is acquired. Specifically, first, the mounted board image 40 is acquired by photographing the board B on which the component C is mounted in the inspection device 30. Then, in step S101, the mounted board image 40 acquired by the inspection device 30 is acquired by the inspection data creating device 10 via the communication unit 15.
  • the mounted board image 40 includes a three-dimensional image in which the component C has information on the three-dimensional shape of the mounted board B and a two-dimensional image in which the component C has information on the two-dimensional shape of the mounted board B. Includes.
  • step S102 the component image 41 in the mounted board image 40 is extracted.
  • the component image 41 in the mounted board image 40 is extracted based on at least one of the board data and the board CAD data.
  • the board data is used in a component mounting device (not shown) for mounting the component C on the substrate B, and is data including the shape of the component C and the mounting position of the component C.
  • the board CAD data is image data of the board B including the mounting position of the component C.
  • the component image in the mounted board image 40 is used by using at least one of the information on the mounting position of the component C included in the board data and the information on the mounting position of the component C included in the board CAD data. 41 is extracted.
  • the component image 41 includes a three-dimensional image having information on the three-dimensional shape of the component C and a two-dimensional image having information on the two-dimensional shape of the component C.
  • step S103 the similarity degree 60 of the component C for each component C is acquired based on the component image 41. Specifically, in step S103, the processes of steps S201 to S203 of FIG. 4 are performed.
  • step S201 the similarity degree 61 of the type of the part C for each part type is acquired based on the part image 41.
  • step S201 as shown in FIG. 6, the similarity degree 61 of the type of the part C for each part type is acquired by using the image processing by the machine learning model (discriminator) 70 learned in advance.
  • a part image 41 as a two-dimensional image and a three-dimensional image is input to the machine learning model 70, and the similarity degree 61 of the type of the part C for each part type is acquired as an output result.
  • the similarity degree 61 of the type of the part C for each part type is acquired as an output result.
  • step S201 the similarity degree 61 of the type of the part C for each part type is acquired as the degree of certainty (probability) for each part type.
  • the similarity degree 61 of the type of the component C for each component type includes a numerical value.
  • the machine learning model 70 can be created, for example, by training as shown in FIG. 7. That is, as shown in FIG. 7, first, the field image 71 is prepared.
  • the field-of-view image 71 is not particularly limited as long as it is an image including the component C, but for example, an image acquired by the inspection device 30 can be used.
  • the visual field image 71 includes a two-dimensional image and a three-dimensional image. Then, by cutting out (extracting) the region of the component C from the visual field image 71, the component image 72 in the visual field image 71 is acquired. At this time, the component image 72 as the two-dimensional image and the three-dimensional image is acquired from the field image 71 as the two-dimensional image and the three-dimensional image. In the process of cutting out the region of the component C, the size of the clipped image differs depending on the type and size of the component C.
  • the part image 73 of the determined size is acquired as a learning image.
  • the component image 73 as the two-dimensional image and the three-dimensional image is acquired from the component image 72 as the two-dimensional image and the three-dimensional image.
  • the type of the component C is labeled on the component image 73.
  • the trained machine learning model 70 is created by performing machine learning based on the two-dimensional image and the part image 73 as the three-dimensional image labeled with the type of the component C.
  • the machine learning model 70 can be created (constructed) by using a convolutional neural network. Further, the machine learning model 70 is learned by using various types of component images 73 such as chip capacitors, chip resistors, chip inductors, chip transistors, LEDs, SOPs, BGAs, QFPs, and connectors.
  • step S202 the similarity 62 of the shape of the component C is acquired based on the component image 41. That is, in addition to step S201 for acquiring the similarity 61 of the type of component C for each component type based on the component image 41, the component image is separate from step S201 for acquiring the similarity 61 for the type of component C. Based on 41, step S202 is performed to acquire the similarity 62 of the shape of the part C for each part shape.
  • step S202 the shape of the component C is acquired based on the component image 41. Specifically, based on the part image 41, the width W1 of the outer shape of the part C, the length L1 of the outer shape of the part C, the thickness T1 of the outer shape of the part C, the width W2 of the main body of the part C, and the part.
  • the outer shape of the component C means the entire component C including both the main body and the terminal of the component C. Further, the main body of the component C means a chip portion that does not include the terminal of the component C. Further, the terminal of the component C means an electrode (lead) portion that does not include the main body of the component C.
  • the width W1 of the outer shape of the part C, the length L1 of the outer shape of the part C, and the thickness T1 of the outer shape of the part C are obtained.
  • the width W3 and the length L3 of the terminal of the component C are acquired.
  • the image processing using the shading in the image and the height in the image is not particularly limited, but binarization processing, edge extraction processing, contour tracking processing, and the like can be used.
  • the main body and terminals of the component C can be identified by using the difference in the shade (luminance value) between the main body and the terminal of the component C and the substrate B.
  • the width W1 of the outer shape of the part C, the length L1 of the outer shape of the part C, the width W2 of the main body of the part C, and the length L2 of the main body of the part C can be acquired.
  • the main body and terminals of the component C can be identified by utilizing the difference in height between the main body and terminals of the component C and the substrate B.
  • the width W1 of the outer shape of the part C the length L1 of the outer shape of the part C, the thickness T1 of the outer shape of the part C, and the width W2 of the main body of the part C.
  • the length L2 of the main body of the component C, the thickness T2 of the main body of the component C, the number of terminals of the component C, the pitch P3 of the terminals of the component C, the width W3 of the terminals of the component C, and the terminals of the component C The length L3 and can be obtained.
  • the component image 41 as a two-dimensional image and a three-dimensional image may be used in combination.
  • the component image 41 as a two-dimensional image is masked to hide unnecessary parts (parts other than the component C) in two dimensions.
  • the shape of the component C can be obtained from the component image 41 as an image. As a result, the shape of the component C can be acquired with higher accuracy.
  • the information on the color of the substrate B and the information on the thickness of the solder on the substrate B are taken into consideration. Will be done.
  • the information regarding the color of the substrate B includes the color of the surface of the substrate B, the color of the silk print of the substrate B, and the like.
  • the information regarding the thickness of the solder of the substrate B includes the thickness of the solder of the substrate B or the thickness of the screen mask used for printing the solder.
  • the information regarding the color of the substrate B is used as auxiliary information for distinguishing the component C from the substrate B at the time of image processing using shades in the image, for example.
  • the information regarding the thickness of the solder of the substrate B is used as auxiliary information for distinguishing the terminal of the component C from the solder of the substrate B at the time of image processing using the height in the image.
  • step S202 the similarity 62 of the shape of the part C for each part shape is acquired by using the image processing using the shading in the image and the height in the image. Further, in step S202, when the similarity 62 of the shape of the component C for each component shape is acquired by using the image processing using the shading in the image and the height in the image, the information regarding the color of the substrate B, And information about the thickness of the solder on the substrate B is taken into account.
  • step S202 the similarity 62 of the shape of the part C for each part shape is acquired based on the shape of the part C.
  • the similarity 62 of the shape of the component C for each component shape is acquired for the component library 21 in which the types and shapes of the plurality of components C are registered in advance.
  • the similarity 62 in the shape of the component C includes the similarity in the outer shape of the component C, the similarity in the main body of the component C, and the similarity in the terminals of the component C.
  • the similarity 62 of the shapes of the parts C is the similarity of the width W1 of the outer shape of the part C, the similarity of the length L1 of the outer shape of the part C, and the similarity of the thickness T1 of the outer shape of the part C.
  • the degree of similarity 62 in the shapes of these parts C can be obtained as the degree of deviation between the measurement part (part C in the part image 41) and the registered part in the part library 21 by the following equation (1).
  • the similarity 62 of the shape of the part C for each part shape includes a numerical value.
  • S s 1.0-
  • X2 Value of shape of registered component.
  • the outer shape length L1 of the measurement part is 6.40 (X1)
  • the outer shape length L1 of the registered part is obtained.
  • the similarity (S s ) of the external length L1 of the component C is 1.0-
  • / 6.40 0.9843.
  • W2 of the main body of the component C in the example shown in FIG.
  • the width W2 of the main body of the measurement component is 4.37 (X1)
  • the width W2 of the main body of the registered component is Since it is 4.40 (X2)
  • the similarity (S s ) of the width W2 of the main body of the component C is 1.0-
  • /4.37 0.9931.
  • the similarity (S s ) of the number of terminals of the component C is 1.0-
  • / 6 0.8333.
  • step S203 the component type and the component shape are based on the similarity 61 of the type of the component C for each component type and the similarity 62 of the shape of the component C for each component shape.
  • the similarity degree 60 of the part C for each part C is acquired.
  • the types and shapes of the plurality of parts C are different based on the similarity 61 of the types of the parts C for each part type and the similarity 62 of the shapes of the parts C for each part shape.
  • the similarity 60 of the component C for each component C is acquired for the component library 21 registered in advance.
  • the similarity 60 of the part C is acquired for all the registered parts of the part library 21, so that the similarity 60 of the part C for each part C whose type and shape are both specified is 60. Is obtained.
  • the similarity 60 of the component C can be obtained as a comprehensive evaluation index including the evaluation index (similarity) of both the component type and the component shape by the following formula (2).
  • the similarity 60 of the component C for each component C includes a numerical value.
  • S p, S t of the equation (2), S sow, S sol, S sot, S sbw, S sbl, S sbt, S srn, S srp, S srw, S srl is between 0 and 1 or less The value.
  • the weighting coefficients a, b, c, and d can be obtained in advance by an experiment or the like. Further, when the maximum degree of similarity 60 parts C (maximum value of S p) and 100, the sum of weighting coefficients a, b, c, d becomes 100.
  • the similarity 60 of the component C is obtained by weighting and adding the similarity 61 of the type of the component C and the similarity 62 of the shape of the component C, respectively. Will be done.
  • the similarity 60 of the component C includes a group of the similarity 61 of the type of the component C, a group of the similarity of the outer shape of the component C, a group of the similarity of the main body of the component C, and the group of the component C. It is obtained by weighting and adding to the group of terminal similarity.
  • the component is used by using the image processing by the machine learning model 70 learned in advance and the image processing using at least one of the shading in the image and the height in the image.
  • the similarity 60 of the component C for each C is acquired.
  • step S104 inspection data 50 as component data used in the inspection apparatus 30 for inspecting the substrate B is set based on the similarity 60 of the component C for each component C. Will be done. Specifically, in step S104, the component C having the highest numerical value of the similarity 60 of the component C is set as the inspection data 50 as the component data from the registered component data of the component library 21. In the example shown in FIG. 10, the component (registered component) C of "SOP14pin-p1.27-L6.5 x W4.4" is set as the inspection data 50 as the component data.
  • step S104 when the terminal information of the component C set as the inspection data 50 (information of the terminal of the component C in the registered component data) is different from the information of the terminal of the component C acquired based on the component image 41, New inspection data is created by overwriting the terminal information of the component C set as the inspection data 50 with the terminal information of the component C acquired based on the component image 41.
  • the information on the terminals of the component C includes the number of terminals of the component C, the pitch P3 of the terminals of the component C, the width W3 of the terminals of the component C, and the length L3 of the terminals of the component C.
  • step S104 the terminal information of the component C set as the inspection data 50 and the terminal information of the component C acquired based on the component image 41 are compared, and the two pieces of information set the threshold value based on the comparison result. If it is determined that the difference is significantly larger than the above, the information on the terminal of the component C set as the inspection data 50 is overwritten by the information on the terminal of the component C acquired based on the component image 41.
  • steps S103 and S104 is performed on all the components C (all component images 41) in the mounted board image 40, so that the component C inspects the mounted substrate B.
  • the automatic setting of the data 50 is completed.
  • step S104 even if the similarity 60 of the component C is small, the component C having the highest numerical value of the similarity 60 of the component C is set as the inspection data 50 as the component data. .. That is, regardless of the magnitude of the numerical value of the similarity 60 of the component C, the component C having the highest numerical value of the similarity 60 of the component C is set as the inspection data 50 as the component data.
  • the substrate B is provided with the step of acquiring the similarity 60 of the component C for each component C based on the component image 41 and the similarity 60 of the component C for each component C.
  • a step of setting inspection data 50 as component data used in the inspection device 30 to be inspected is provided.
  • inspection data 50 can be set as component data.
  • the inspection data 50 as the part data can be easily set as compared with the case where the inspection data 50 as the part data is simply set based on the matching of the dimensions. Can be done.
  • the similarity 60 of the component C for each component C includes a numerical value.
  • the step of setting the inspection data 50 includes a step of setting the component C having the highest numerical value of the similarity 60 of the component C as the inspection data 50 as the component data.
  • the step of acquiring the similarity 60 of the component C for each component C is the image processing by the machine learning model 70 learned in advance, the shading in the image, and the height in the image. It includes a step of acquiring the similarity 60 of the component C for each component C by using the image processing using the above.
  • the similarity 60 of the component C for each component C can be obtained by using the image processing by the machine learning model 70 learned in advance or the image processing using the shading in the image and the height in the image. It can be easily obtained.
  • the image processing by the machine learning model 70 learned in advance it is possible to easily make a highly accurate judgment (acquisition of similarity) close to the judgment of a skilled person.
  • the step of acquiring the similarity 60 of the component C for each component C acquires the similarity 61 of the type of component C for each component type based on the component image 41. Including steps. Thereby, the similarity degree 60 of the part C for each part C can be acquired based on the similarity degree 61 of the type of the part C for each part type. As a result, the inspection data 50 as the component data can be set while identifying the component C having different types based on the similarity 60 of the component C in consideration of the type of the component C.
  • the step of acquiring the similarity 60 of the component C for each component C acquires the similarity 61 of the type of component C for each component type based on the component image 41.
  • a step of acquiring the similarity 62 of the shape of the component C for each component shape is further included based on the component image 41.
  • the inspection data 50 as the part data can be set while identifying the parts C having different types and shapes based on the similarity 60 of the parts C in consideration of the type and shape of the parts C. Further, since both the type and shape of the part C are considered, unlike the case where only the shape of the part C is considered, the inspection data 50 as the part data while identifying the parts C having the same shape but different types. Can be set.
  • the step of acquiring the similarity 60 of the component C for each component C is the similarity 61 of the type of component C for each component type and the shape of the component C for each component shape.
  • a step of acquiring the similarity 60 of the component C for each component C with respect to the component library 21 in which the types and shapes of the plurality of components C are registered in advance based on the similarity 62 is included.
  • the step of setting the inspection data 50 includes a step of setting the inspection data 50 as the component data from the registered component data of the component library 21. Thereby, the inspection data 50 as the component data can be easily set by using the registered component data of the component library 21.
  • the similarity 62 in the shape of the component C includes the similarity in the outer shape of the component C, the similarity in the main body of the component C, and the similarity in the terminals of the component C. ..
  • the similarity 62 of the shape of the component C can be appropriately acquired based on the similarity of the outer shape of the component C, the similarity of the main body of the component C, or the similarity of the terminals of the component C.
  • the step of acquiring the similarity 60 of the component C for each component C is the width W1 of the outer shape of the component C and the length of the outer shape of the component C based on the component image 41.
  • L1 the outer thickness T1 of the component C
  • the width W2 of the main body of the component C the width W2 of the main body of the component C
  • the length L2 of the main body of the component C the thickness T2 of the main body of the component C
  • the number of terminals of the component C and the component. It includes a step of acquiring the pitch P3 of the terminal of C, the width W3 of the terminal of the component C, and the length L3 of the terminal of the component C.
  • Appropriate degree of similarity 62 in the shape of component C based on the thickness T2, the number of terminals of component C, the pitch P3 of the terminals of component C, the width W3 of the terminals of component C, or the length L3 of the terminals of component C. Can be obtained in.
  • the step of acquiring the similarity 60 of the component C for each component C is the step of acquiring the similarity 60 of the component C for each component type by using the image processing by the machine learning model 70 learned in advance. It includes a step of acquiring the similarity degree 61 of the type and acquiring the similarity degree 62 of the shape of the part C for each part shape by using the image processing using the shading in the image and the height in the image.
  • image processing by the machine learning model 70 learned in advance the shading in the image and the height in the image are used while appropriately acquiring the similarity 61 of the type of the component C for each component type.
  • the similarity 62 of the shape of the part C for each part shape can be appropriately acquired.
  • the step of acquiring the similarity 60 of the component C for each component C is for each component shape by using image processing using the shading in the image and the height in the image.
  • the step of considering the information about the color of the substrate B and the information about the thickness of the solder of the substrate B is included.
  • the similarity 62 of the shape of the part C for each part shape can be accurately obtained. Further, when considering the information on the thickness of the solder of the board B, the reference of the height for distinguishing the solder of the board B and the terminal of the component C can be known, so that the solder of the board B and the terminal of the component C can be easily distinguished. can do. As a result, the similarity 62 of the shape of the part C for each part shape can be accurately obtained.
  • the step of extracting the component image 41 in the mounted substrate image 40 is used in the component mounting device for mounting the component C on the substrate B, and the shape and component of the component C.
  • the step includes extracting the component image 41 in the mounted board image 40 based on at least one of the board data including the mounting position of C and the board CAD data as the image data of the board B.
  • the position of the component C in the mounted board image 40 can be easily specified based on the board data or the board CAD data.
  • the component image 41 in the mounted board image 40 can be easily extracted as compared with the case where the board data or the board CAD data is not used.
  • the terminal information of the component C set as the inspection data 50 is the information of the terminal of the component C acquired based on the component image 41. If they are different, the step of creating new inspection data 50 is included by overwriting the terminal information of the component C set as the inspection data 50 with the terminal information of the component C acquired based on the component image 41. As a result, it is possible to suppress setting the wrong terminal information of the component C as the inspection data 50.
  • the inspection data creating apparatus shows an example of creating the inspection data of the present invention, but the present invention is not limited to this.
  • the inspection device may create the inspection data of the present invention.
  • the control unit 34a of the inspection device 30 of the above embodiment functions in the same manner as the control unit 11 of the inspection data creation device 10 of the above embodiment.
  • inspection data may be manually set based on the similarity of parts for each part.
  • the similarity of the parts for each part may be presented to the operator, and the inspection data may be manually set by the operator.
  • the component having the highest component similarity value is set as inspection data as component data regardless of the magnitude of the component similarity value. Not limited to this.
  • the operator may be notified that the numerical value of the similarity of parts is equal to or less than the threshold value.
  • the similarity of the parts for each part is acquired by using the image processing by the machine learning model and the image processing using the shading in the image and the height in the image.
  • the present invention is not limited to this.
  • the similarity of the parts for each part is acquired by using only one of the image processing by the machine learning model and the image processing using the shading in the image and the height in the image. May be good.
  • image processing by a machine learning model may acquire both the similarity of component types and the similarity of component shapes.
  • both the similarity of the types of parts and the similarity of the shapes of parts may be obtained by image processing using the shading in the image and the height in the image.
  • the arrangement state of the terminal of the component is acquired by image processing using the shading in the image and the height in the image, and the component is based on the arrangement state of the terminal of the component and the rule of the terminal shape of the component.
  • Kind similarity may be acquired.
  • a rule of the terminal shape for example, in the case of SOP, it can be a rule that gull-wing-shaped terminals exist on two opposite sides of the main body of the component. Further, for example, in the case of QFP, it can be a rule that gull-wing-shaped terminals exist on four sides of the main body of the component.
  • image processing is performed using the shading in the image and the height in the image
  • the present invention is not limited to this.
  • image processing may be performed using only one of the shading in the image and the height in the image.
  • the mounted board image and the component image include a two-dimensional image and a three-dimensional image
  • the present invention is not limited to this.
  • the mounted board image and the component image may include only one of the two-dimensional image and the three-dimensional image.
  • the similarity of parts for each part is acquired by acquiring the similarity of parts for all the registered parts of the parts library (part registration information).
  • the present invention is not limited to this.
  • the similarity of parts for each part may be acquired by acquiring the similarity of parts for some registered parts of the parts registration information.
  • the similarity in the shape of the component includes the similarity in the outer shape of the component, the similarity in the main body of the component, and the similarity in the terminals of the component.
  • the invention is not limited to this.
  • the similarity of the shape of the component includes only one or two of the similarity of the outer shape of the component, the similarity of the main body of the component, and the similarity of the terminals of the component. May be good.
  • the width of the outer shape of the part, the length of the outer shape of the part, the thickness of the outer shape of the part, the width of the main body of the part, the length of the main body of the part, and the length of the main body of the part are based on the part image.
  • An example is shown in which the thickness of the main body, the number of terminals of the component, the pitch of the terminal of the component, the width of the terminal of the component, and the length of the terminal of the component are obtained, but the present invention is limited to this. No.
  • the width of the outer shape of the part based on the component image, the width of the outer shape of the part, the length of the outer shape of the part, the thickness of the outer shape of the part, the width of the main body of the part, the length of the main body of the part, and the thickness of the main body of the part. It is not necessary to acquire all of the number of terminals of the component, the pitch of the terminals of the component, the width of the terminals of the component, and the length of the terminals of the component.
  • the similarity of the parts type for each part type is acquired by using the image processing by the machine learning model, and the image processing using the shading in the image and the height in the image is used.
  • the present invention is not limited to this, although an example is shown in which the similarity of the shape of a part is acquired for each part shape.
  • the similarity of the component types for each component type is acquired by using image processing using at least one of the shading in the image and the height in the image, and the image by the machine learning model is used. By using the process, the similarity of the shape of the part for each part shape may be acquired.
  • the present invention when the similarity of the shape of the component for each component shape is acquired by using the image processing using the shading in the image and the height in the image, the information regarding the color of the substrate and the information regarding the color of the substrate are obtained.
  • the present invention is not limited to this.
  • information on the color of the substrate is obtained when the similarity of the shape of the component for each component shape is acquired by using image processing using at least one of the shade in the image and the height in the image. , And only one of the information about the thickness of the solder on the substrate may be considered.
  • the processing operation of the control unit has been described using a flow-driven flowchart in which the processing operations are sequentially performed along the processing flow, but the present invention is not limited to this.
  • the processing operation of the control unit may be performed by event-driven type (event-driven type) processing in which processing is executed in event units. In this case, it may be completely event-driven, or it may be a combination of event-driven and flow-driven.
  • Inspection data creation device 11 10 Inspection data creation device 11, 34a Control unit 15 Communication unit 21 Parts library (parts registration information) 30 Inspection device 33a Imaging unit 40 Mounted board image 41 Part image 50 Inspection data 60 Part similarity 61 Part type similarity 62 Part shape similarity 70 Machine learning model B Board C Part L1 Part external length L2 Length of the main body of the part L3 Length of the terminal of the part P3 Pitch of the terminal of the part T1 Thickness of the outer shape of the part T2 Thickness of the main body of the part W1 Width of the outer shape of the part W2 Width of the main body of the part W3 width

Landscapes

  • Engineering & Computer Science (AREA)
  • Operations Research (AREA)
  • Manufacturing & Machinery (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Supply And Installment Of Electrical Components (AREA)
PCT/JP2020/026087 2020-07-02 2020-07-02 検査データ作成方法、検査データ作成装置および検査装置 Ceased WO2022003919A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
PCT/JP2020/026087 WO2022003919A1 (ja) 2020-07-02 2020-07-02 検査データ作成方法、検査データ作成装置および検査装置
JP2022532968A JP7448655B2 (ja) 2020-07-02 2020-07-02 検査データ作成方法、検査データ作成装置および検査装置
TW110121908A TWI841850B (zh) 2020-07-02 2021-06-16 檢查資料製作方法、檢查資料製作裝置及檢查裝置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/026087 WO2022003919A1 (ja) 2020-07-02 2020-07-02 検査データ作成方法、検査データ作成装置および検査装置

Publications (1)

Publication Number Publication Date
WO2022003919A1 true WO2022003919A1 (ja) 2022-01-06

Family

ID=79315842

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/026087 Ceased WO2022003919A1 (ja) 2020-07-02 2020-07-02 検査データ作成方法、検査データ作成装置および検査装置

Country Status (3)

Country Link
JP (1) JP7448655B2 (https=)
TW (1) TWI841850B (https=)
WO (1) WO2022003919A1 (https=)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2023171412A1 (https=) * 2022-03-08 2023-09-14
US20250086779A1 (en) * 2023-09-13 2025-03-13 Kabushiki Kaisha Toshiba Information processing apparatus and information processing method

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004151057A (ja) * 2002-11-01 2004-05-27 Omron Corp 部品コード変換テーブルの登録方法、変換テーブル登録装置、部品コード変換テーブルの登録用プログラムおよび記憶媒体
JP2004340832A (ja) * 2003-05-16 2004-12-02 Matsushita Electric Ind Co Ltd 回路基板の外観検査方法及び回路基板の外観検査装置
JP2006250609A (ja) * 2005-03-09 2006-09-21 Matsushita Electric Ind Co Ltd 回路実装基板の外観検査方法
JP2008032525A (ja) * 2006-07-28 2008-02-14 Omron Corp 基板外観検査用の検査基準データの設定方法およびこの方法を用いた基板外観検査装置
WO2015115274A1 (ja) * 2014-01-31 2015-08-06 オムロン株式会社 物体判別装置、画像センサ、物体判別方法
WO2018105100A1 (ja) * 2016-12-09 2018-06-14 株式会社Ｆｕｊｉ 装着ジョブデータの作成方法および作成装置
JP2019100917A (ja) * 2017-12-05 2019-06-24 パナソニックＩｐマネジメント株式会社 検査プログラム生成システム、検査プログラムの生成方法、及び検査プログラムの生成用プログラム

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5278634A (en) * 1991-02-22 1994-01-11 Cyberoptics Corporation High precision component alignment sensor system

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004151057A (ja) * 2002-11-01 2004-05-27 Omron Corp 部品コード変換テーブルの登録方法、変換テーブル登録装置、部品コード変換テーブルの登録用プログラムおよび記憶媒体
JP2004340832A (ja) * 2003-05-16 2004-12-02 Matsushita Electric Ind Co Ltd 回路基板の外観検査方法及び回路基板の外観検査装置
JP2006250609A (ja) * 2005-03-09 2006-09-21 Matsushita Electric Ind Co Ltd 回路実装基板の外観検査方法
JP2008032525A (ja) * 2006-07-28 2008-02-14 Omron Corp 基板外観検査用の検査基準データの設定方法およびこの方法を用いた基板外観検査装置
WO2015115274A1 (ja) * 2014-01-31 2015-08-06 オムロン株式会社 物体判別装置、画像センサ、物体判別方法
WO2018105100A1 (ja) * 2016-12-09 2018-06-14 株式会社Ｆｕｊｉ 装着ジョブデータの作成方法および作成装置
JP2019100917A (ja) * 2017-12-05 2019-06-24 パナソニックＩｐマネジメント株式会社 検査プログラム生成システム、検査プログラムの生成方法、及び検査プログラムの生成用プログラム

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2023171412A1 (https=) * 2022-03-08 2023-09-14
WO2023171412A1 (ja) * 2022-03-08 2023-09-14 三菱電機株式会社 設計支援装置、および学習装置
JP7638438B2 (ja) 2022-03-08 2025-03-03 三菱電機株式会社 設計支援装置、学習装置、および学習済モデル
US20250086779A1 (en) * 2023-09-13 2025-03-13 Kabushiki Kaisha Toshiba Information processing apparatus and information processing method

Also Published As

Publication number Publication date
TWI841850B (zh) 2024-05-11
JPWO2022003919A1 (https=) 2022-01-06
TW202220542A (zh) 2022-05-16
JP7448655B2 (ja) 2024-03-12

Similar Documents

Publication Publication Date Title
US10991088B2 (en) Defect inspection system and method using artificial intelligence
JP6922168B2 (ja) 表面実装ラインの品質管理システム及びその制御方法
JP3092809B2 (ja) 検査方法、並びに検査プログラムデータの自動作成機能を有する検査装置
CN1299556C (zh) 具有元件布局检查功能的抓取式设备
JP7122524B2 (ja) 検査プログラム生成システム、検査プログラムの生成方法、及び検査プログラムの生成用プログラム
US20120327215A1 (en) High speed optical sensor inspection system
JP4493421B2 (ja) プリント回路基板検査装置とプリント回路基板組み立て検査ラインシステムおよびプログラム
CN114022436A (zh) 一种印制电路板的检测方法、装置及检测设备
CN104837302B (zh) 品质管理系统以及内部检查装置
JP2008185514A (ja) 基板外観検査装置
JP7448655B2 (ja) 検査データ作成方法、検査データ作成装置および検査装置
CN116758160B (zh) 基于正交视觉系统的光学元件装配过程位姿检测方法及装配方法
EP3104169B1 (en) Quality management system
TWI244359B (en) Automatic optical detecting system for defect components on printed circuit board
JP4333349B2 (ja) 実装外観検査方法及び実装外観検査装置
KR20040096277A (ko) 메탈 마스크 검사 장치 및 그의 검사 방법
JP3123275B2 (ja) 電子部品の欠品検査における検査データの作成方法
JP5778685B2 (ja) ボールグリッドアレイデバイスの位置合わせ及び検査のためのシステム及び方法
JP2004301620A (ja) 自動部品マウント装置の検査方法
TWI449119B (zh) Circuit board placement method
KR101495323B1 (ko) 벤트홀을 이용한 플립칩 얼라인먼트 검사 장치 및 방법
JP2576147B2 (ja) 実装部品の欠品検査装置
JP2012252003A (ja) 高速光学センサ検査システム
TWI773035B (zh) 影像辨識系統及方法
JP2884581B2 (ja) 実装済プリント基板の検査装置における基準データーの教示方法

Legal Events

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

Ref document number: 20942432

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022532968

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20942432

Country of ref document: EP

Kind code of ref document: A1