WO2022104585A1 - 一种印刷电路板远程光学检修方法及系统 - Google Patents

一种印刷电路板远程光学检修方法及系统 Download PDF

Info

Publication number
WO2022104585A1
WO2022104585A1 PCT/CN2020/129723 CN2020129723W WO2022104585A1 WO 2022104585 A1 WO2022104585 A1 WO 2022104585A1 CN 2020129723 W CN2020129723 W CN 2020129723W WO 2022104585 A1 WO2022104585 A1 WO 2022104585A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
printed circuit
defect
defect information
information
Prior art date
Application number
PCT/CN2020/129723
Other languages
English (en)
French (fr)
Inventor
柯布兰凡
胡冰峰
陈朋飞
Original Assignee
苏州康代智能科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州康代智能科技股份有限公司 filed Critical 苏州康代智能科技股份有限公司
Priority to PCT/CN2020/129723 priority Critical patent/WO2022104585A1/zh
Priority to KR1020237014079A priority patent/KR20230078729A/ko
Publication of WO2022104585A1 publication Critical patent/WO2022104585A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G01N21/956—Inspecting patterns on the surface of objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/04—Manufacturing
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8861—Determining coordinates of flaws
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/888—Marking defects
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • G01N21/956—Inspecting patterns on the surface of objects
    • G01N2021/95638—Inspecting patterns on the surface of objects for PCB's

Definitions

  • the invention relates to the field of circuit board detection, in particular to a method and system for remote optical repair of printed circuit boards.
  • PCB Printed Circuit Board
  • AOI Automatic optical inspection equipment
  • the present invention provides a method and system for remote optical maintenance of printed circuit boards, which can perform cooperative maintenance of circuit boards across regions through remote cooperation, rationally utilize equipment and maximize the utilization of staff resources, and save costs , and improve production efficiency, the technical scheme is as follows:
  • the present invention provides a method for remote optical maintenance of printed circuit boards, and two technical solutions are proposed for the present invention:
  • the first technical solution is as follows: a remote optical maintenance method for printed circuit boards, which is used to complete cooperative maintenance work on printed circuit boards across regions, and the maintenance method includes the following steps:
  • S101 scan the printed circuit board to be detected in the first area to obtain a scanned image, and analyze the scanned image to obtain initial defect information, where the initial defect information includes coordinate information corresponding to the defect position of the scanned image;
  • the first area remotely receives, through the relay server, the real defect information returned by the second area after excluding false point defects;
  • the second technical solution is as follows: for completing the cooperative maintenance work on the printed circuit board across regions, the maintenance method includes the following steps:
  • the second area remotely receives initial defect information obtained by analyzing the scanned image in the first area through a relay server, the initial defect information including coordinate information corresponding to the defect position of the scanned image;
  • the second area excludes false point defects from the initial defect information, and obtains real defect information
  • steps S201-S203 are performed for the second area to dock with multiple first areas.
  • the present invention provides a remote optical inspection system for printed circuit boards, which is used to complete cooperative inspection of printed circuit boards across regions.
  • AI equipment, defect repair workstation, first relay server, and virtual inspection workstation and second relay server set up in the second area, the geographic location of the first area and the second area are different, and the first relay server is different from the second area.
  • the automatic optical inspection device is used to scan the printed circuit board to be inspected to obtain a scanned image
  • the defect screening AI device is used to analyze the scanned image to obtain initial defect information
  • the virtual maintenance workstation is used to exclude false point defects from initial defect information to obtain real defect information;
  • the defect repair workstation is used to provide repair prompt information for manual inspection according to the real defect information
  • the first relay server and the second relay server are used for the first area to send initial defect information to the second area, and the second area to return real defect information to the first area.
  • the first area is further provided with a first database server, and the first database server is respectively connected to automatic optical inspection equipment, defect screening AI equipment, defect repair workstation and first transfer server.
  • the second area is further provided with a second database server, and the second database server is respectively connected to the second relay server and the virtual maintenance workstation.
  • the first area is further provided with an information proofing workstation, which is connected in communication with the first database server, and the information proofing workstation is used for proofing the initial defect information and real defect information of the printed circuit board.
  • the defect repair workstation includes spot welding equipment, a display screen, a camera device, a driving device, and a controller.
  • the driving device drives the camera device to move in sequence to be opposite to the real defect.
  • the camera device is used to enlarge and image the real defect
  • the display screen is used to display the imaging result of the camera device.
  • the first relay server and the second relay server are communicatively connected through an MES network system.
  • the first relay server is a DDE Server
  • the second relay server is a DDV Server.
  • the automatic optical inspection equipment and the defect screening AI equipment are of an integrated structure or a separate structure.
  • the number of automatic optical inspection equipment, defect repair workstations, and virtual inspection workstations is one or more, respectively.
  • the number of the first area and/or the second area is multiple, so that the multiple first area and the second area cooperate to complete the maintenance work on the printed circuit board, or the first area and the multiple second area Regional cooperation to complete the maintenance of printed circuit boards.
  • circuit boards adopts a cooperative model to separate the verification and inspection links, form a new industry, and change the inherent form of the circuit board production industry;
  • the centralized and optimal allocation of resources is conducive to improving the efficiency of production and maintenance.
  • FIG. 1 is a flowchart of a method for remote optical repair of a printed circuit board provided by an embodiment of the present invention
  • FIG. 2 is a schematic block diagram of a printed circuit board remote optical repair system provided by an embodiment of the present invention
  • FIG. 3 is a flowchart of a remote optical overhaul from a first area perspective of an overhaul system provided by an embodiment of the present invention
  • FIG. 4 is a flowchart of a remote optical overhaul from a second area perspective of the overhaul system provided by an embodiment of the present invention
  • FIG. 5 is a schematic diagram of two or more cross-region implementation scenarios of the printed circuit board remote optical repair system provided by the embodiment of the present invention.
  • a method for remote optical inspection of printed circuit boards is provided, which is used to complete cooperative inspection of printed circuit boards across regions. As shown in FIG. 1 , the inspection method includes the following steps:
  • S1 Scan the printed circuit board to be inspected in the first area (by the relevant equipment) to obtain a scanned image, and analyze the scanned image to obtain initial defect information.
  • the initial defect information includes coordinate information corresponding to the defect position of the scanned image.
  • the initial defect information includes, but is not limited to, picture 1 (corresponding to the image and/or its name) and a preliminary judgment on picture 1 as
  • the absolute coordinate information of the defect for example, can also include the defect type code (short circuit, open circuit, missing welding, etc.), each initial defect information contains the scanned image itself, and the image is re-inspected by the equipment in different places after cross-regional transmission. .
  • the scanned images obtained after the AOI equipment scans the circuit board can be classified into: good, bad, and unclear. If the scanned image is good, the corresponding circuit board does not need to be Repair, place it separately from the circuit board that needs to be repaired (to be sent to the repair station later), and filter (or delete) the scanned image, which will not be sent to cross-area for re-inspection; if the scanned image is not good , the corresponding circuit board needs to be repaired to generate initial defect information; if the scanned image is unclear, it needs to be re-scanned and inspected. If it is unclear for three consecutive times, manual inspection or direct classification as bad.
  • the first area remotely sends the initial defect information to the second area through the relay server.
  • the relay servers are set in the first area and the second area respectively, the second area and the first area are geographically different, and the relay servers in the two areas are preferably connected by the MES system for bidirectional communication.
  • the initial defect information obtained from the preliminary judgment may be mixed with false point defects, such as misjudgments caused by dust and stains, these false point defects do not need to be repaired. Compared with false point defects, it is more in line with the actual defects that need to be repaired, and does not limit 100% of the defects that need to be repaired.
  • the second area remotely sends the real defect information to the first area through the relay server.
  • the printed circuit board is repaired according to the real defect information.
  • the way of repairing the circuit board that needs to be repaired can be manual repair, semi-automatic repair, or fully automatic robot repair using AI intelligent technology.
  • one of the inventions of the present invention is to change the traditional mode of circuit board maintenance, and the maintenance work is roughly divided into preliminary judgment and initial judgment. Defects, excluding false point defects for initial defects, and repairing defects. The initial determination of initial defects and repairing of defects are completed in the first area, as shown in Figure 3:
  • S101 scan the printed circuit board to be detected in the first area to obtain a scanned image, and analyze the scanned image to obtain initial defect information, where the initial defect information includes coordinate information corresponding to the defect position of the scanned image;
  • the first area remotely receives, through the relay server, the real defect information returned by the second area after excluding false point defects;
  • the second area remotely receives initial defect information obtained by analyzing the scanned image in the first area through a relay server, the initial defect information including coordinate information corresponding to the defect position of the scanned image;
  • the second area excludes false point defects from the initial defect information, and obtains real defect information
  • the first application scenario is as follows: The geographic location difference between the second area and the first area is reflected in different locations in the same site, such as different rooms, different floors or between different buildings.
  • the first area can be implemented through a local area network or an external network.
  • the second application scenario is as follows: For a company, choose to set up the above-mentioned first area in multiple areas with lower labor costs to complete the preliminary determination of initial defects and the equipment required for repairing the defects, and concentrate them in one area ( The second area) configures the equipment required to complete the work of eliminating false point defects for initial defects, without configuring equipment required for eliminating false point defects for initial defects in each of the first areas.
  • the second area configures the equipment required to complete the work of eliminating false point defects for initial defects, without configuring equipment required for eliminating false point defects for initial defects in each of the first areas.
  • the third application scenario is as follows: This mode is also suitable for cooperative maintenance between different companies.
  • companies A, B, and C are manufacturers of printed circuit boards, while company D does not produce circuit boards, and company D can connect with them.
  • A, B, and C provide services to eliminate false point defects, making resource allocation highly concentrated.
  • the second area connects with multiple first areas to perform cooperative maintenance work.
  • the first area and the second area may be the same enterprise or different enterprises.
  • the number of the first area being multiple actually means that there are multiple sets of equipment in the first area (automatic optical inspection equipment, defect screening AI equipment, defect repair workstation, first transit server) distributed in different areas, These areas are different first areas, such as the first area 1 and the first area 2 in FIG. 5 ; similarly, the number of the second areas is multiple, which actually refers to multiple sets of equipment in the second area (virtual Maintenance workstations, second transit servers) are distributed in different areas, and these areas are different second areas (not shown).
  • the solution for remote optical inspection of circuit boards of the present invention not only saves costs, but also improves production efficiency, because VVR Stations are centrally configured in the second area, so that virtual verification inspections can be completed in batches (that is, the real defect information is obtained by re-inspecting the initial defect information). For example, it takes 100 minutes to complete the re-inspection of the scanned images of 1000 circuit boards in a local VVR Station in the original mode.
  • the remote mode of the present invention it is sent to ten large-scale VVR stations across the region. It only takes 10 minutes for the Station to complete the re-inspection, which is suitable for the batch inspection mode of large quantities.
  • a remote optical inspection system for printed circuit boards is provided, which is used to complete cooperative inspection work on printed circuit boards across regions.
  • the inspection system includes: Regional automatic optical inspection equipment (Automated Optical Inspection, referred to as AOI), defect screening AI equipment (referred to as AI, as shown in Figure 2), defect repair station (Verification Repair Station, referred to as VR Station), the first transfer server, and set up in the first Virtual Verification Repair Station (VVR Station for short) and second relay server in two areas, the first area and the second area are geographically different, and the first relay server and the second relay server are connected for bidirectional communication ;
  • AOI Regional automatic optical inspection equipment
  • AI defect screening AI equipment
  • VR Station defect repair station
  • VVR Station Virtual Verification Repair Station
  • the automatic optical inspection equipment (AOI) in the first area is used to scan the printed circuit board (PCB) to be inspected to obtain a scanned image, and the defect screening AI device is used to analyze the scanned image to obtain initial defect information;
  • the virtual maintenance workstation (VVR Station) in the second area is used to exclude false point defects from the initial defect information to obtain real defect information;
  • the defect repair workstation (VR Station) in the first area is used to provide repair prompt information for manual inspection according to the real defect information, which will be described in detail below;
  • the first relay server and the second relay server are used for the first area to send initial defect information to the second area, and the second area to return real defect information to the first area.
  • the first relay server is a dynamic data exchange mechanism (Dynamic Data Exchange, referred to as DDE) Server, so that after the first area and the second area are connected, when the data of one of them changes, it will immediately notify the other party.
  • DDE Dynamic Data Exchange
  • the first area is further provided with a first database server (Database Server I in FIG. 2 ), and the second area is further provided with a second database server (Database Server II in FIG. 2 )
  • the Database Server I is respectively connected to the AOI device, the defect screening AI device, the VR Station and the first relay server;
  • the Database Server II is respectively connected to the second relay server and the VVR Station.
  • the functions of the first database server and the second database server are as follows:
  • the first database server is used to receive the scanned image output by the AOI device and send it to the defect screening AI device; after the defect screening AI device analyzes and obtains initial defect information, it sends the initial defect information to
  • the first relay server the first relay server preferably realizes the communication connection with the second relay server through the MES network system, and the MES network system can check all the output and input file information from DDE and DDV; this embodiment does not limit each time.
  • the number of scanned images corresponding to the initial defect information sent by the relay server is single or multiple, especially for the batch method, the first relay server can also package the initial defect information into a zip compressed file and then send it to the second
  • the transit server specifically, sends the image file to the folder in the DDE server, and creates a ZIP file in the DDE server, which includes all AI classification information content, and the transit MES network system automatically transfers the ZIP file to the DDV server, and after the file reaches the second relay server (DDV server), automatically delete the file from the DDE server;
  • the DDV server extracts the file from the ZIP (the object is the same as the DDE server); the operator in the second area uses the DDV system to run a virtual maintenance workstation to check the (decompressed) picture file , when the classification work from the DDV server is completed, the data after the screening and classification (that is, the real defect information obtained after the image re-examination) is automatically updated to the DDV server; after the real defect information is packaged into a new zip file, The second relay server is then sent to the first relay server, and the new zip compressed file may only include more accurate (possibly smaller capacity) defect data after filtering.
  • the MES network system outputs the ZIP file to the DDE server, then extracts the ZIP file and updates the classification information.
  • the first relay server receives the relevant information and sends it to the VR Station through Database Server I, so that the defective PCB can be repaired.
  • Described Database Server II is used for receiving the initial defect information that the second relay server obtains from the first relay server, and then forwards it to the virtual maintenance workstation (VVR Station); after the VVR Station eliminates false point defects and obtains the real defect information, The Database Server II sends the real defect information to the second relay server, which is then sent to the Database Server I by the second relay server via the first relay server.
  • VVR Station virtual maintenance workstation
  • the number of pieces of initial defect information sent by the first relay server to the second relay server (for example, picture 1, the coordinates of the initial defect position a, b, c, d, and e are counted as one piece) and the second relay server to the first relay
  • the number of pieces of real defect information returned by the server (such as picture 1, false point defect position coordinates c, e, and real defect position coordinates a, b, d are counted as one piece) are the same and correspond one-to-one.
  • VR Station can be arranged for maintenance to ensure the accuracy of maintenance data. For the case of no real defects, it is also necessary to return to the Database Server I in the first area. For example, returning NULL indicates that the initial defect information is all false point defects.
  • the first area is further provided with an information proofreading workstation, which is connected in communication with the first database server, and the information proofreading workstation is used for proofreading the initial defect information of the printed circuit board and the actual defect information. Specifically, it is used to check whether the returned real defect information corresponds to the sent initial defect information, for example, the number of pieces of information in the sent zip compressed package is inconsistent with the received zip compressed package, or the real defect information is not an initial defect If the part of the information, or the image information in the real defect information entry is inconsistent with the image information in the initial defect information entry, the result of the proofreading failure will be obtained. It is necessary to remind manual intervention to investigate the cause for correction.
  • the obtained proofreading result can be Save it locally or send it to Database Server I for saving, which can monitor the consistency of the pictures of all boards before and after they are classified to ensure the correctness and traceability of the data.
  • the defect repair workstation includes spot welding equipment, a display screen, a camera device, a driving device, and a controller.
  • the driving device drives the camera device to move in sequence to be opposite to the real defect.
  • the camera device zooms in on the real defect, and the display screen is used to display the imaging result of the camera device.
  • the automatic optical inspection device and the defect screening AI device are separate structures, and the defect screening AI device obtains the scanned image from the AOI device through the Database Server I, and the present invention does not limit the separation of the two Structure, it is obvious that the integrated structure of the automatic optical inspection equipment and the defect screening AI equipment is a simple variation of the split structure. the scope of protection of this application.
  • the number of automatic optical inspection equipment, defect repair workstations, and virtual inspection workstations is one or more, respectively.
  • a virtual maintenance workstation (VVR Station) in the second area is connected to the service work of multiple first areas, and the number of VVR Stations is preferably multiple.
  • VVR Station virtual maintenance workstation
  • its software operation efficiency is much greater than the operation (scanning or maintenance) efficiency of the physical circuit board, it is also a technical solution that can realize the service work of multiple first areas to a certain extent.
  • the AOI equipment in the first area scans the printed circuit board to be detected to obtain a scanned image. Taking the batch mode as an example, scans to obtain N scanned images (n1, n2, n3...);
  • the defect screening AI equipment in the first area analyzes the scanned images (n1, n2, n3...) to obtain initial defect information.
  • Each of the N pieces of information includes the identity information of the scanned image (such as the serial number or the image name as the image name). identification) and the location information of the initial defect (it can be simple coordinate information, or it can be in the form of marking on the image, or it can be image information plus coordinate information);
  • the first relay server After compressing the initial defect information to obtain a compressed package, the first relay server sends it to the second relay server;
  • the VVR Station excludes false point defects from the initial defect information to obtain real defect information
  • the second relay server After compressing the real defect information, the second relay server sends it to the first relay server;
  • the defect repair workstation (VR Station) in the first area starts repair work on the corresponding printed circuit board according to the real defect information.
  • the AOI equipment and defect screening AI equipment are existing equipment in the field of circuit board maintenance, refer to the Chinese invention application with publication number CN110579479A, which is incorporated into this application by full introduction, especially refer to:
  • defect list contains corresponding to the scanned image.
  • Defect coordinate information of the preliminary determined defect indicates the working principle of the defect screening AI device in this embodiment.
  • Database Server I and Database Server II in this embodiment perform similar basic functions with the database server in the prior art.
  • VVR Station the virtual maintenance workstation
  • VR Station usually exists as a sub-module in the maintenance equipment (VR Station), and the embodiment of the present invention is different from it.
  • the VVR Station is independent of the VR Station.
  • the principle of eliminating false point defects is the same as that of the prior art CN110579479A, see its record:
  • the AOI equipment can obtain the overall layout picture of the defect after scanning the PCB board, and can accurately calibrate the coordinates of the corresponding defect point in the picture. In the AOI equipment system, it also has the function of judging the type of defect, such as circuit board leakage. Welding, multi-welding and welding errors, etc.
  • a database server with data storage function which can accurately store the information input after AOI scanning, and connected to the database server is the VVR system for overhauling equipment, VVR Collect the defect information of the corresponding plate in the database server, and through its own intelligent judgment system or manual image inspection, it can accurately determine the "false point” information and "false point” coordinate information in the defect information, and then delete it through operations. After deleting the "false point” information, move to the corresponding "true point” defect coordinate position through the Video on the VVR device for manual inspection.
  • re-inspecting the initially determined defect by the exclusion method includes: extracting a local image at the defect coordinates corresponding to the initially determined defect, and judging whether the local image satisfies the short-circuit feature or the open-circuit feature, wherein, The short-circuit feature includes a straight line connecting two cables, and the open-circuit feature includes a gap in the cable. If any one of the characteristics is satisfied, the defect is determined to be a real defect, otherwise the defect is determined to be a false point. defect. "Real defects" need to be manually repaired point by point. For example, a multi-welded slit will cause a short circuit on the PCB. At this time, the slit needs to be removed manually.
  • using the feature correspondence method to re-inspect the initially determined defect includes: extracting a local image at the defect coordinates corresponding to the initially determined defect, and judging whether the local image satisfies the following conditions at the same time: non-linear, For irregular and isolated graphics, if the above characteristics are met at the same time, the defect is determined to be a false point defect.
  • the "false point defects” can be dust, stains, or fingerprints, etc., which will exist in large quantities in the PCB board, and will be judged as defect points during AOI scanning. On these large numbers of "false point defects", the embodiment of the present invention introduces a VVR system, which can greatly reduce the time spent in this aspect.
  • using the similarity matching method to recheck the initially determined defects includes: loading several preset defect template images through the database server, and the defect template images are marked as real defects or false points Defects; extract the local image at the defect coordinates corresponding to the initially determined defect, and compare it with the defect template image to find the defect template image with the highest similarity; if the defect template with the highest similarity If the image is marked as a real defect, the preliminary determined defect is determined as a real defect; if the defect template image with the highest similarity is marked as a false point defect, the preliminary determined defect is determined as a false point defect.
  • re-inspecting the initially determined defect includes: extracting a partial image at the defect coordinate corresponding to the initially determined defect, inputting it into the trained neural network model, and outputting the output according to the neural network model , determine whether the defect is a real defect or a false point defect.
  • the neural network model may adopt the deep neural network in the prior art, and the neural network can be trained by combining the back-propagation algorithm and the stochastic gradient descent method.
  • the automatic optical inspection equipment scans the printed circuit board to be inspected to obtain the scanned image
  • the defect screening AI equipment analyzes the scanned image to obtain the initial defect information
  • the virtual maintenance workstation starts from the initial defect information. Excluding false point defects from defect information, and defect repair workstations repairing circuit boards based on real defect information are all existing technologies.
  • the automatic optical inspection equipment, defect screening AI equipment, virtual repair The respective functions of each module performed by the workstation and the defect repair workstation are clear and complete, and those skilled in the art can implement the technical solutions of the present invention.
  • the VVR Station is independent from the VR Station and the separate and remote setting is new.
  • another remote optical inspection system for printed circuit boards is provided, and the inspection system includes automatic optical inspection. equipment, defect screening AI equipment, defect repair station (VR Station), first relay server, virtual repair station (VVR Station), second relay server, the first relay server and the second relay server are connected for bidirectional communication;
  • the automatic optical inspection device is used to scan the printed circuit board to be inspected to obtain a scanned image
  • the defect screening AI device is used to analyze the scanned image to obtain initial defect information, where the initial defect information includes Coordinate information of the defect location;
  • the first relay server is configured to send the initial defect information to the second relay server;
  • VVR Station The virtual maintenance workstation (VVR Station) is used to exclude false point defects from the initial defect information to obtain real defect information;
  • the second relay server is used for returning real defect information to the first relay server.
  • the defect repair station (VR Station) is used to provide repair prompt information for manual inspection according to the real defect information;
  • the number is more than the number of virtual maintenance stations (VVR Station).
  • the number of VR Stations is double the number of virtual maintenance stations (VVR Station). It is not limited to spanning or not spanning, which is free from VVR Station and VR in the prior art. Station one-to-one corresponding restrictions to further optimize the allocation of resources.
  • any reference signs placed between parentheses shall not be construed as limiting the claim.
  • the word “comprising” does not exclude the presence of other elements or steps listed in a claim statement.
  • the terms “a” or “an” are defined as one or more than one.

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Signal Processing (AREA)
  • Manufacturing & Machinery (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

一种印刷电路板远程光学检修方法及系统，远程光学检修方法包括：S1、在第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析得到初始缺陷信息；S2、第一区域将所述初始缺陷信息通过中转服务器远程发送至第二区域；S3、在第二区域对所述初始缺陷信息排除假点缺陷，得到真实缺陷信息；S4、第二区域将所述真实缺陷信息通过中转服务器远程发送至第一区域；S5、在第一区域，根据真实缺陷信息对所述印刷电路板进行检修。跨区域完成对印刷电路板的合作检修工作的解决方案，突破了印刷电路板的传统检修模式，实现资源的高度配置，节约成本，提高生产效率。

Description

一种印刷电路板远程光学检修方法及系统 技术领域
本发明涉及电路板检测领域，尤其涉及一种印刷电路板远程光学检修方法及系统。
背景技术
现如今在高度发展的电子工业时代，印刷电路板(Printed Circuit Board，简称PCB)已成为计算机、电子通信等产品上必不可缺的一样重要部件之一。PCB电路板在制作完成之后，需要经过一道检测流程，行业内普遍采用自动光学检测设备(Automated Optical Inspection，简称AOI)，AOI能够检测PCB上的缺陷，然后根据AOI检测到的缺陷进行检修。
电路板在AOI(自动光学检测)设备处扫描完后，通过软件分类后得到不同的缺陷种类，之后通过人工判断这些被分类的缺陷种类是否正确，确认好是真正的缺陷后，最后通过人工对缺陷处进行检修。现有技术中，用户是在同一个工厂实现这些流程。
现有技术中尚未出现远程光学检修电路板的模式。
发明内容
为了解决现有技术的问题，本发明提供了一种印刷电路板远程光学检修方法及系统，通过远程协作跨区域对电路板进行协作检修，合理利用设备及工作人员的资源最大化利用，节省成本，且提高生产效率，所述技术方案如下：
一方面，本发明提供了一种印刷电路板远程光学检修方法，对其本发明提出两种技术方案：
第一种技术方案如下：一种印刷电路板远程光学检修方法，用于跨区域完成对印刷电路板的合作检修工作，所述检修方法包括以下步骤：
S101、在第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析得到初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的 缺陷位置的坐标信息；
S102、将所述初始缺陷信息通过中转服务器远程发送至第二区域；
S103、响应于第二区域对所述初始缺陷信息排除假点缺陷，第一区域通过中转服务器远程接收第二区域返回的排除假点缺陷后的真实缺陷信息；
S104、根据真实缺陷信息，在第一区域对所述印刷电路板进行检修
第二种技术方案如下：用于跨区域完成对印刷电路板的合作检修工作，所述检修方法包括以下步骤：
S201、响应于第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析，第二区域通过中转服务器远程接收第一区域分析扫描图像得到的初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息；
S202、第二区域对所述初始缺陷信息排除假点缺陷，得到真实缺陷信息；
S203、将所述真实缺陷信息通过中转服务器远程发送至第一区域，所述真实缺陷信息用于为第一区域处印刷电路板的检修工作提供检修提示。
进一步地，所述第二区域对接多个第一区域执行步骤S201-S203。
另一方面，本发明提供了一种印刷电路板远程光学检修系统，用于跨区域完成对印刷电路板的合作检修工作，所述检修系统包括设置在第一区域的自动光学检测设备、缺陷筛选AI设备、缺陷检修工作站、第一中转服务器，以及设置在第二区域的虚拟检修工作站、第二中转服务器，所述第一区域与第二区域的地理位置不同，所述第一中转服务器与第二中转服务器双向通讯连接；
所述自动光学检测设备用于对待检测的印刷电路板进行扫描得到扫描图像，所述缺陷筛选AI设备用于对扫描图像分析得到初始缺陷信息；
所述虚拟检修工作站用于从初始缺陷信息中排除假点缺陷，得到真实缺陷信息；
所述缺陷检修工作站用于根据真实缺陷信息，为人工检修提供检修提示信息；
所述第一中转服务器和第二中转服务器用于第一区域向第二区域发送初始缺陷信息，及第二区域向第一区域返回真实缺陷信息。
进一步地，所述第一区域还设有第一数据库服务器，所述第一数据库服务器分别连接自动光学检测设备、缺陷筛选AI设备、缺陷检修工作站和第一中转 服务器。
进一步地，所述第二区域还设有第二数据库服务器，所述第二数据库服务器分别连接第二中转服务器和虚拟检修工作站。
优选地，所述第一区域还设有信息校对工作站，其与所述第一数据库服务器通讯连接，所述信息校对工作站用于校对印刷电路板的初始缺陷信息和真实缺陷信息。
具体地，所述缺陷检修工作站包括点焊设备、显示屏、摄像装置、驱动装置和控制器，在所述控制器的控制下，所述驱动装置驱动所述摄像装置依次移动至与真实缺陷相对的位置，所述摄像装置用于对真实缺陷放大成像，所述显示屏用于显示所述摄像装置的成像结果。
优选地，所述第一中转服务器与第二中转服务器通过MES网络系统通讯连接。
可选地，所述第一中转服务器为DDE Server，所述第二中转服务器为DDV Server。
可选地，所述自动光学检测设备与缺陷筛选AI设备为集成结构或分体结构。
可选地，所述自动光学检测设备、缺陷检修工作站、虚拟检修工作站的数量分别为一个或多个。
可选地，所述第一区域和/或第二区域的数量为多个，使得多个第一区域与第二区域合作完成对印刷电路板的检修工作，或者第一区域与多个第二区域合作完成对印刷电路板的检修工作。
本发明具有如下有益效果：
a.突破检修电路板的本地地理限制，打破传统电路板本地检修模式；
b.实现设备资源和人工资源的双重优化配置和最大化利用；
c.电路板的生产采用合作模式将核检环节分离，形成新的产业，改变电路板生产行业固有形态；
d.资源的集中优化配置，有利于提高生产和检修效率。
附图说明
被视为本发明的主题在说明书的结论部分中被特别指出并清楚地主张权利。然而，当结合附图一起参阅时，通过参考以下详细描述可以最佳地理解本 发明的组织、操作方法，以及主题、特征和优点，其中：
图1是本发明实施例提供的印刷电路板远程光学检修方法的流程图；
图2是本发明实施例提供的印刷电路板远程光学检修系统的示意框图；
图3是本发明实施例提供的检修系统在第一区域视角下远程光学检修流程图；
图4是本发明实施例提供的检修系统在第二区域视角下远程光学检修流程图；
图5是本发明实施例提供的印刷电路板远程光学检修系统的两个以上跨区实施场景示意图。
具体实施方式
在以下详细描述中，阐述了许多具体细节以便提供对本发明的透彻理解。然而，本领域技术人员将理解，可以在没有这些具体细节的情况下实践本发明。在其他情况下，没有详细描述众所周知的方法，过程和组件，以免模糊本发明。
被视为本发明的主题在说明书的结论部分中被特别指出并清楚地主张权利。然而，当结合附图一起参阅时，通过参考以下详细描述可以最佳地理解本发明的组织、操作方法，以及主题、特征和优点。
应当理解，为了说明的简单和清楚，图中所示的元件不一定按比例绘制。例如，为了清楚起见，一些元件的尺寸可能相对于其他元件被放大。
由于本发明的说明性实施例在很大程度上可使用本领域技术人员熟知的电子元件和电路来实施，如上文所述，在认为必要的范围之外，不会对细节作更大的解释，以便理解和体会本发明的基本概念，以免混淆或分散本发明的教导。
在本发明的一个实施例中，提供了一种印刷电路板远程光学检修方法，用于跨区域完成对印刷电路板的合作检修工作，如图1所述，所述检修方法包括以下步骤：
S1、在第一区域(由相关的设备)对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析得到初始缺陷信息。
具体地，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息，比如所述初始缺陷信息包括但不限于图片1(对应图像和/或其名称)以及图片1上初步判定为缺陷的绝对坐标信息，比如还可以包括缺陷类型代码(短 路、断路、漏焊等)，每一条初始缺陷信息中都包含扫描图像本身，在跨区传送后由异地的设备对其进行图像复检。
在本发明的一个优选实施例中，可以对AOI设备扫描电路板后得到的扫描图像进行分类，主要分为：良好、不好、不清楚，若扫描图像是良好，对应的电路板则不需要检修，将其与需要检修的(后续要送到检修站的)电路板分开放置，并将该扫描图像过滤(或删除)，后续不会发送至跨区进行复检；若扫描图像是不好，则对应的电路板需要检修，生成初始缺陷信息；若扫描图像是不清楚，则需要对其重新扫描检查，若连续三次都是不清楚，则人工核查或者直接归为不好。
需要说明的是，以上扫描、图像分析分别由第一区域的相关设备完成，下述其他操作，包括第二区域的操作同理。具体的相关设备在系统实施例中详述。
S2、第一区域将所述初始缺陷信息通过中转服务器远程发送至第二区域。
本实施例中采用在第一区域和第二区域分别设置中转服务器，第二区域与第一区域地理位置不同，两地的中转服务器优选采用MES系统双向通讯连接。
S3、在第二区域(由相关的设备)对所述初始缺陷信息排除假点缺陷，得到真实缺陷信息。
由于初步判定得到的初始缺陷信息中可能夹杂着假点缺陷，比如灰尘、污点引起的误判，这部分假点缺陷是不需要检修的，对其排除后得到真实缺陷信息，而所谓真实缺陷信息是相对于假点缺陷更符合实际需要检修的缺陷，而不限定100％需要检修的缺陷。
S4、第二区域将所述真实缺陷信息通过中转服务器远程发送至第一区域。
S5、在第一区域，根据真实缺陷信息对所述印刷电路板进行检修。
对需要检修的电路板实施修补的方式可以是人工检修，可以是半自动化检修，也可以是利用AI智能技术实现机械手全自动化检修。
相比于将缺陷的初步判定和排除假点缺陷设置在同一区域的现有技术而言，本发明的发明点之一在于改变了电路板检修的传统模式，将检修工作大致分为初步判定初始缺陷、对初始缺陷排除假点缺陷、对缺陷进行检修，将其中的初步判定初始缺陷和对缺陷进行检修放在第一区域完成，具体如图3所示：
S101、在第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析得到初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的 缺陷位置的坐标信息；
S102、将所述初始缺陷信息通过中转服务器远程发送至第二区域；
S103、响应于第二区域对所述初始缺陷信息排除假点缺陷，第一区域通过中转服务器远程接收第二区域返回的排除假点缺陷后的真实缺陷信息；
S104、根据真实缺陷信息，在第一区域对所述印刷电路板进行检修。
而将对初始缺陷排除假点缺陷的工作放在第二区域完成，具体如图4所示：
S201、响应于第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析，第二区域通过中转服务器远程接收第一区域分析扫描图像得到的初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息；
S202、第二区域对所述初始缺陷信息排除假点缺陷，得到真实缺陷信息；
S203、将所述真实缺陷信息通过中转服务器远程发送至第一区域，所述真实缺陷信息用于为第一区域处印刷电路板的检修工作提供检修提示。
这种全新的模式实现了资源的高度优化配置，如图5所示：
第一种应用场景如下：第二区域与第一区域地理位置不同体现在同一场区的不同地点，比如不同房间、不同楼层或者不同楼宇之间，可以通过局域网，也可以通过外网实现第一区域的第一中转服务器与第二区域的第二中转服务器的通讯连接；
甚至跨区域的第一区域和第二区域在地理位置上不容易实现建立局域网，比如跨区、跨市、跨省或者跨国，对应有两种应用场景：
第二种应用场景如下：对于一家公司而言，选择在人工成本等较低的多个区域设置上述第一区域完成初步判定初始缺陷和对缺陷进行检修所需的设备，并集中在一个区域(第二区域)配置完成对初始缺陷排除假点缺陷工作所需的设备，而无需在各个第一区域配置对初始缺陷排除假点缺陷所需的设备。选择第二区域优选与各个第一区域建立MES网络成本、采购排除假点缺陷设备的成本等为考虑因素。
第三种应用场景如下：这种模式还适用于不同的企业之间进行合作检修，比如，A、B、C企业是印刷电路板的生产厂家，而D企业不生产电路板，D企业可以对接A、B、C提供排除假点缺陷的服务，使得资源配置高度集中。
即第二区域对接多个第一区域执行合作检修工作的步骤，第一区域和第二 区域可以是同一家企业，也可以是不同的企业。
本申请主张保护“第二区域对接多个第一区域”，但是不限定第二区域的数量为多个或一个，比如A企业将部分的服务委托给D企业，部分的服务委托给E企业，这种情况下，就不排除第二区域为多个。具体地，所述第一区域的数量为多个实际是指有多套第一区域的设备(自动光学检测设备、缺陷筛选AI设备、缺陷检修工作站、第一中转服务器)分布在不同的区域，这些区域为不同的第一区域，如图5中的第一区域①、第一区域②；同理，所述第二区域的数量为多个实际是指有多套第二区域的设备(虚拟检修工作站、第二中转服务器)分布在不同的区域，这些区域为不同的第二区域(未图示)。
本发明的远程光学检修电路板的解决方案不仅节约成本，而且提高生产效率，因为在第二区域集中配置VVR Station，使得批量完成虚拟验证检修(即对初始缺陷信息进行复检得到真实缺陷信息)的工作效率提升了，比如原来的模式对1000片电路板的扫描图像在本地一台VVR Station完成复检需要100分钟，利用本发明的远程模式，发送至跨区的大规模的十台，VVR Station完成复检只需要10分钟，适合于大批量的批次检修模式。
在本发明的一个实施例中，提供了一种印刷电路板远程光学检修系统，用于跨区域完成对印刷电路板的合作检修工作，如图2所示，所述检修系统包括设置在第一区域的自动光学检测设备(Automated Optical Inspection，简称AOI)、缺陷筛选AI设备(简称AI，如图2)、缺陷检修工作站(Verification Repair Station，简称VR Station)、第一中转服务器，以及设置在第二区域的虚拟检修工作站(Virtual Verification Repair Station，简称VVR Station)、第二中转服务器，所述第一区域与第二区域的地理位置不同，所述第一中转服务器与第二中转服务器双向通讯连接；
所述第一区域的自动光学检测设备(AOI)用于对待检测的印刷电路板(PCB)进行扫描得到扫描图像，所述缺陷筛选AI设备用于对扫描图像分析得到初始缺陷信息；
所述第二区域的虚拟检修工作站(VVR Station)用于从初始缺陷信息中排除假点缺陷，得到真实缺陷信息；
所述第一区域的缺陷检修工作站(VR Station)用于根据真实缺陷信息，为人工检修提供检修提示信息，具体在下文中详述；
所述第一中转服务器和第二中转服务器用于第一区域向第二区域发送初始缺陷信息，及第二区域向第一区域返回真实缺陷信息。可选地，所述第一中转服务器为动态数据交换机制(Dynamic Data Exchange，简称DDE)Server，使得第一区域与第二区域建立起连接关系后，当其中一方的数据发生变化后就会马上通知另一方。
具体如图2所示，所述第一区域还设有第一数据库服务器(图2中的Database Server Ⅰ)，所述第二区域还设有第二数据库服务器(图2中的Database Server Ⅱ)，所述Database Server Ⅰ分别连接AOI设备、缺陷筛选AI设备、VR Station和第一中转服务器；所述Database Server Ⅱ分别连接第二中转服务器和VVR Station。所述第一数据库服务器和第二数据库服务器的作用如下：
所述第一数据库服务器用于接收所述AOI设备输出的扫描图像，并将其发送至缺陷筛选AI设备；待所述缺陷筛选AI设备分析得到初始缺陷信息后，将所述初始缺陷信息发送至第一中转服务器，所述第一中转服务器优选通过MES网络系统实现与第二中转服务器通讯连接，MES网络系统可以检查所有从DDE和DDV的输出、输入的文件信息；本实施例不限定每次通过中转服务器发送与初始缺陷信息对应的扫描图像的数量为单个或多个，特别地对于批量的方式，第一中转服务器还可以将所述初始缺陷信息打包成zip压缩文件后再发送至第二中转服务器，具体地，将图像文件发送到DDE服务器里面的文件夹，并且，在DDE服务器中创建一个ZIP文件，该文件包括了所有的AI分类信息内容，中转MES网络系统自动将ZIP文件传输到DDV服务器，并且在文件到达第二中转服务器(DDV服务器)之后，自动删除来自DDE服务器的文件；
当该ZIP文件到达DDV服务器之后，DDV服务器从ZIP中提取文件(该物件与DDE服务器相同)；位于第二区域的操作者使用DDV系统运行虚拟检修工作站对(解压缩后的)图片文件进行检查，当完成从DDV服务器的分类工作后自动将筛选分类后的资料(即对图像复检后得到的真实缺陷信息)更新到DDV服务器；将所述真实缺陷信息打包成新的zip压缩文件后，第二中转服务器再发送至第一中转服务器，该新的zip压缩文件可以只包括过滤之后更准确(可能更小容量)的缺陷数据。
MES网络系统输出ZIP文件到DDE服务器，然后提取该ZIP文件并更新分类信息，第一中转服务器接收到相关信息后通过Database Server Ⅰ发送给VR  Station，即可对有缺陷的PCB进行修复。
所述Database Server Ⅱ用于接收第二中转服务器从第一中转服务器获取的初始缺陷信息，然后转发给虚拟检修工作站(VVR Station)；在所述VVR Station排除假点缺陷，得到真实缺陷信息后，所述Database Server Ⅱ将所述真实缺陷信息发送至第二中转服务器，再由第二中转服务器经由第一中转服务器发送至Database Server Ⅰ。
显然，第一中转服务器向第二中转服务器发送的初始缺陷信息的条数(比如图片1、初始缺陷位置坐标a、b、c、d、e算作一条)与第二中转服务器向第一中转服务器返回的真实缺陷信息的条数(比如图片1、假点缺陷位置坐标c、e，真实缺陷位置坐标a、b、d算作一条)是相同且一一对应的。通过校正后可以安排VR Station进行检修，确保检修数据的准确性。对于真实缺陷无的情况，同样需要返回至第一区域的Database Server Ⅰ，比如返回NULL表示初始缺陷信息全部为假点缺陷。在本发明的一个具体实施例中，所述第一区域还设有信息校对工作站，其与所述第一数据库服务器通讯连接，所述信息校对工作站用于校对印刷电路板的初始缺陷信息和真实缺陷信息。具体地，用于校对返回来的真实缺陷信息与发出去的初始缺陷信息是否对应，比如发出去的zip压缩包与收到的zip压缩包内的信息条数不一致，或者真实缺陷信息不是初始缺陷信息中的部分，又或者真实缺陷信息条目中的图像信息与初始缺陷信息条目中的图像信息不一致，就会得到校对失败的结果，需要提醒人工介入排查原因以进行纠正，所得到的校对结果可以保存在本地或者发送至Database Server Ⅰ进行保存，可以监控所有电路板在被分类之前和之后的图片的一致性，确保数据的正确性和可追溯性。
具体地，所述缺陷检修工作站包括点焊设备、显示屏、摄像装置、驱动装置和控制器，在所述控制器的控制下，所述驱动装置驱动所述摄像装置依次移动至与真实缺陷相对的位置，所述摄像装置对真实缺陷放大成像，所述显示屏用于显示所述摄像装置的成像结果。
在图2对应的实施例中，所述自动光学检测设备与缺陷筛选AI设备为分体结构，缺陷筛选AI设备通过Database Server Ⅰ从AOI设备获取扫描图像，本发明并不限定两者的分体结构，显然易见地，所述自动光学检测设备与缺陷筛选AI设备为集成结构是分体结构的简单变型，数据通过内部传输，集成结构直接 向Database Server Ⅰ输出初始缺陷信息的技术方案同样落入本申请的保护范围。
可选地，所述自动光学检测设备、缺陷检修工作站、虚拟检修工作站的数量分别为一个或多个。尤其是从资源配置的角度，一个第二区域的虚拟检修工作站(VVR Station)对接多个第一区域的服务工作，VVR Station的数量优选为多个。但即使是单个VVR Station，由于其软件运行效率远远大于对实物电路板的操作(扫描或检修)效率，因此在一定程度上也是可以实现对接多个第一区域的服务工作的技术方案。
本实施例的印刷电路板远程光学检修系统的工作过程如下：
第一区域的AOI设备对待检测的印刷电路板进行扫描得到扫描图像，以批量方式为例，扫描得到N张扫描图像(n1、n2、n3……)；
再由第一区域的缺陷筛选AI设备对扫描图像(n1、n2、n3……)分析得到初始缺陷信息，N条信息中的每一条都包括扫描图像的身份信息(比如编号或以图像名称作为识别)以及初始缺陷的位置信息(可以是单纯的坐标信息，也可以是标注在图像上的形式，也可以是图像信息加坐标信息)；
压缩初始缺陷信息得到压缩包后，第一中转服务器将其发送至第二中转服务器；
从所述第二中转服务器提取所述初始缺陷信息，并分配至第二区域的虚拟检修工作站(VVR Station)；
所述VVR Station从所述初始缺陷信息中排除假点缺陷，得到真实缺陷信息；
将真实缺陷信息压缩后，第二中转服务器将其发送至第一中转服务器；
第一区域的缺陷检修工作站(VR Station)提取所述真实缺陷信息后，按照真实缺陷信息对相应的印刷电路板启动检修工作。
所述AOI设备和缺陷筛选AI设备为电路板检修领域的现有设备，参见公开号为CN110579479A的中国发明申请，通过全文引入的方式并入本申请，尤其参见：
“扫描得到扫描图像，并将其与通过数据库服务器加载的对应标准图像作比较，将比较得到的差异点作为初步判定的缺陷，构建缺陷列表，所述缺陷列表中包含对应于所述扫描图像的初步判定的缺陷的缺陷坐标信息”即表明本实施例中缺陷筛选AI设备的工作原理。
本实施例中的Database Server Ⅰ、Database Server Ⅱ与该现有技术中的数据 库服务器执行类似的基本功能。
现有技术中虚拟检修工作站(VVR Station)通常是作为检修设备(VR Station)中的子模块而存在，本发明实施例与之不同，在本实施例中，VVR Station是独立于VR Station的单独的设备，但是其排除假点缺陷的原理是与现有技术CN110579479A相同的，参见其记载：
“AOI设备其在扫描PCB板后，可以得到缺陷的整体布局图片，并能在图片中准确的标定对应缺陷点的坐标，在AOI设备系统中，还具有判定缺陷类型的功能，例如线路板漏焊、多焊和焊接错误等。与AOI连接的是带有数据储存功能的数据库服务器，该数据库服务器可以准确地存储AOI扫描后输入的信息，与数据库服务器连接的是检修设备的VVR系统，VVR采集数据库服务器中对应板材的缺陷信息，通过自身的智能判定系统或者人工图片验视，可以准确地判断出缺陷信息中“假点”信息和“假点”坐标信息，然后通过操作可以删除判断出来的“假点”信息，在删除“假点”后，通过VVR设备上的Video移动到对应“真点”缺陷坐标位置处进行人工检修。
作为第一种可选技术方案，利用排除法对初步判定的缺陷进行复检包括：提取初步判定的缺陷对应的缺陷坐标处的局部图像，判断该局部图像是否满足短路特征或者断路特征，其中，所述短路特征包括具有连接着两根排线的直线，所述断路特征包括在排线上存在缺口，若满足任意一个特征，则判定所述缺陷为真实缺陷，否则判定所述缺陷为假点缺陷。“真实缺陷”是需要人工逐个点检修的，如多焊接的窄缝，会导致PCB短路，这时就需要人工将该窄缝去除。
作为第二种可选技术方案，利用特征对应法对初步判定的缺陷进行复检包括：提取初步判定的缺陷对应的缺陷坐标处的局部图像，判断该局部图像是否同时满足以下条件：非直线、不规则且孤立存在的图形，若同时满足以上特征，则判定所述缺陷为假点缺陷。所述“假点缺陷”可以是灰尘，污点，或者指纹等，在PCB板材中会大量存在，在AOI扫描时候均会判定为缺陷点，若不智能排除，在后续检修时候，将花费大量人工在这些大量的“假点缺陷”上面，本发明实施例引入VVR系统，可以大大减少该方面的时间花费。
作为第三种可选技术方案，利用相似度匹配法对初步判定的缺陷进行复检包括：通过数据库服务器加载预设的若干个缺陷模板图像，所述缺陷模板图像被标定为真实缺陷或假点缺陷；提取初步判定的缺陷对应的缺陷坐标处的局部 图像，并将其与所述缺陷模板图像进行相似度比较，找到与之相似度最高的缺陷模板图像；若所述相似度最高的缺陷模板图像被标定为真实缺陷，则判定该初步判定的缺陷为真实缺陷；若所述相似度最高的缺陷模板图像被标定为假点缺陷，则判定该初步判定的缺陷为假点缺陷。
作为第四种可选技术方案，对初步判定的缺陷进行复检包括：提取初步判定的缺陷对应的缺陷坐标处的局部图像，将其输入完成训练的神经网络模型，根据所述神经网络模型输出的结果，判定所述缺陷为真实缺陷还是假点缺陷。其中，所述神经网络模型可采用现有技术中的深度神经网络，结合反向传播算法及随机梯度下降法对该神经网络进行训练。”
由上述公开号为CN110579479A的现有技术公开的以上内容可知，自动光学检测设备对待检测的印刷电路板进行扫描得到扫描图像、缺陷筛选AI设备对扫描图像分析得到初始缺陷信息、虚拟检修工作站从初始缺陷信息中排除假点缺陷、缺陷检修工作站根据真实缺陷信息对电路板进行检修均为现有技术，通过引用CN110579479A现有技术的全文内容，所述自动光学检测设备、缺陷筛选AI设备、虚拟检修工作站、缺陷检修工作站执行各模块各自的功能是清楚且完整的，本领域技术人员以此能够实现本发明的技术方案。
本发明实施例中VVR Station相对于VR Station分体独立且异地设置是新的，在本发明的一个实施例中，提供另一种印刷电路板远程光学检修系统，所述检修系统包括自动光学检测设备、缺陷筛选AI设备、缺陷检修工作站(VR Station)、第一中转服务器，虚拟检修工作站(VVR Station)、第二中转服务器，所述第一中转服务器与第二中转服务器双向通讯连接；
所述自动光学检测设备用于对待检测的印刷电路板进行扫描得到扫描图像，所述缺陷筛选AI设备用于对扫描图像分析得到初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息；
所述第一中转服务器用于将初始缺陷信息发送给第二中转服务器；
所述虚拟检修工作站(VVR Station)用于从初始缺陷信息中排除假点缺陷，得到真实缺陷信息；
所述第二中转服务器用于向第一中转服务器返回真实缺陷信息。
所述缺陷检修工作站(VR Station)用于根据真实缺陷信息，为人工检修提供检修提示信息；本实施例中将VVR Station相对于VR Station分体独立设置， 所述缺陷检修工作站(VR Station)的数量多于虚拟检修工作站(VVR Station)的数量，比如VR Station的数量成倍于虚拟检修工作站(VVR Station)的数量，不限定跨区或者不跨区，摆脱了现有技术中VVR Station与VR Station一一对应的限制，实现资源的进一步优化配置。
在此基础上进一步提出跨区域完成对印刷电路板的合作检修工作的解决方案，即将上述方案全文引入至此实施例，如图5所示，创造性地改变了电路板检测检修的传统模式，实现资源的高度配置，节约成本，提高生产效率。
此外，本领域技术人员将意识到，上述操作之间的界限仅为示例性的。多个操作可以合并为单个操作，单个操作可以分布于额外操作中，且可在至少部分重叠的时间下执行操作。此外，可选实施例可包括特定操作的多个举例说明，并且操作顺序可在各种其他实施例中变化。
然而，其他修改、变化及替代也是可能的。因此，应在示例性意义上而非限制性意义上看待说明书及附图。
在权利要求声明中，置于圆括号之间的任何参考符号不应被视为限制请求项。词语“包括”并不排除那些列在权利要求声明中的其他元件或步骤的存在。此外，本文所使用的术语“一”或“一个”，被定义为一个或多于一个。而且，引言短语例如权利要求声明中的“至少一个”及“一个或多个”的使用不应该解释为暗示不定冠词“一”或“一个”引入另一个权利要求要素将包含这种引入的权利要求的任何特定权利要求限制于仅包含一个这样的要素的发明，即使同一权利要求包括引言短语“一个或多个”或“至少一个”和不定冠词，如“一个”或“一个”。使用定冠词也是如此。除非另有说明，否则诸如“第一”和“第二”之类的术语用于任意区分这些术语所描述的元素。因此，这些术语不一定旨在表示这些元素的时间或其他优先级。在彼此不同的权利要求中叙述某些措施的仅有事实并不表示这些措施的组合不能加以利用。
虽然本文已经说明和描述了本发明的某些特征，但是本领域普通技术人员现在将想到许多修改、替换、改变和等同物。因此，应该理解，所附权利要求旨在覆盖落入本发明的真正精神内的所有这些修改和变化。

Claims (10)

  1. 一种印刷电路板远程光学检修方法，其特征在于，用于跨区域完成对印刷电路板的合作检修工作，所述检修方法包括以下步骤：
    S101、在第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析得到初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息；
    S102、将所述初始缺陷信息通过中转服务器远程发送至第二区域；
    S103、响应于第二区域对所述初始缺陷信息排除假点缺陷，第一区域通过中转服务器远程接收第二区域返回的排除假点缺陷后的真实缺陷信息；
    S104、根据真实缺陷信息，在第一区域对所述印刷电路板进行检修。
  2. 一种印刷电路板远程光学检修方法，其特征在于，用于跨区域完成对印刷电路板的合作检修工作，所述检修方法包括以下步骤：
    S201、响应于第一区域对待检测的印刷电路板进行扫描得到扫描图像，并对扫描图像分析，第二区域通过中转服务器远程接收第一区域分析扫描图像得到的初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息；
    S202、第二区域对所述初始缺陷信息排除假点缺陷，得到真实缺陷信息；
    S203、将所述真实缺陷信息通过中转服务器远程发送至第一区域，所述真实缺陷信息用于为第一区域处印刷电路板的检修工作提供检修提示。
  3. 根据权利要求2所述的印刷电路板远程光学检修方法，其特征在于，所述第二区域对接多个第一区域执行步骤S201-S203。
  4. 一种印刷电路板远程光学检修系统，其特征在于，用于跨区域完成对印刷电路板的合作检修工作，所述检修系统包括设置在第一区域的自动光学检测设备、缺陷筛选AI设备、缺陷检修工作站、第一中转服务器，以及设置在第二区域的虚拟检修工作站、第二中转服务器，所述第一区域与第二区域的地理位 置不同，所述第一中转服务器与第二中转服务器双向通讯连接；
    所述自动光学检测设备用于对待检测的印刷电路板进行扫描得到扫描图像，所述缺陷筛选AI设备用于对扫描图像分析得到初始缺陷信息，所述初始缺陷信息包括对应于所述扫描图像的缺陷位置的坐标信息；
    所述虚拟检修工作站用于从初始缺陷信息中排除假点缺陷，得到真实缺陷信息；
    所述缺陷检修工作站用于根据真实缺陷信息，为人工检修提供检修提示信息；
    所述第一中转服务器和第二中转服务器用于第一区域向第二区域发送初始缺陷信息，及第二区域向第一区域返回真实缺陷信息。
  5. 根据权利要求4所述的印刷电路板远程光学检修系统，其特征在于，所述第一区域还设有第一数据库服务器，所述第一数据库服务器分别连接自动光学检测设备、缺陷筛选AI设备、缺陷检修工作站和第一中转服务器；
    所述第二区域还设有第二数据库服务器，所述第二数据库服务器分别连接第二中转服务器和虚拟检修工作站。
  6. 根据权利要求5所述的印刷电路板远程光学检修系统，其特征在于，所述第一区域还设有信息校对工作站，其与所述第一数据库服务器通讯连接，所述信息校对工作站用于校对印刷电路板的初始缺陷信息和真实缺陷信息。
  7. 根据权利要求4所述的印刷电路板远程光学检修系统，其特征在于，所述缺陷检修工作站包括点焊设备、显示屏、摄像装置、驱动装置和控制器，在所述控制器的控制下，所述驱动装置驱动所述摄像装置依次移动至与真实缺陷相对的位置，所述摄像装置用于对真实缺陷放大成像，所述显示屏用于显示所述摄像装置的成像结果。
  8. 根据权利要求4所述的印刷电路板远程光学检修系统，其特征在于，所述第一中转服务器与第二中转服务器通过MES网络系统通讯连接，所述第一中 转服务器为DDE Server。
  9. 根据权利要求4所述的印刷电路板远程光学检修系统，其特征在于，所述自动光学检测设备、缺陷检修工作站、虚拟检修工作站的数量分别为一个或多个。
  10. 根据权利要求4所述的印刷电路板远程光学检修系统，其特征在于，所述第一区域和/或第二区域的数量为多个，使得多个第一区域与第二区域合作完成对印刷电路板的检修工作，或者第一区域与多个第二区域合作完成对印刷电路板的检修工作。
PCT/CN2020/129723 2020-11-18 2020-11-18 一种印刷电路板远程光学检修方法及系统 WO2022104585A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2020/129723 WO2022104585A1 (zh) 2020-11-18 2020-11-18 一种印刷电路板远程光学检修方法及系统
KR1020237014079A KR20230078729A (ko) 2020-11-18 2020-11-18 인쇄 회로 기판 원격 광학 유지보수 방법 및 시스템

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/129723 WO2022104585A1 (zh) 2020-11-18 2020-11-18 一种印刷电路板远程光学检修方法及系统

Publications (1)

Publication Number Publication Date
WO2022104585A1 true WO2022104585A1 (zh) 2022-05-27

Family

ID=81708166

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/129723 WO2022104585A1 (zh) 2020-11-18 2020-11-18 一种印刷电路板远程光学检修方法及系统

Country Status (2)

Country Link
KR (1) KR20230078729A (zh)
WO (1) WO2022104585A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117409261A (zh) * 2023-12-14 2024-01-16 成都数之联科技股份有限公司 一种基于分类模型的元件角度分类方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05324406A (ja) * 1992-05-21 1993-12-07 Nec Field Service Ltd 遠隔保守制御装置
CN110006903A (zh) * 2018-01-05 2019-07-12 皓琪科技股份有限公司 印刷电路板复检系统、标示方法与复检方法
CN110473170A (zh) * 2019-07-10 2019-11-19 苏州卓融新能源科技有限公司 一种适用于pcb板真假缺陷判定的人工智能检测方法
CN110579479A (zh) * 2019-08-09 2019-12-17 康代影像科技(苏州)有限公司 一种基于假点缺陷检测的pcb检修系统及检修方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05324406A (ja) * 1992-05-21 1993-12-07 Nec Field Service Ltd 遠隔保守制御装置
CN110006903A (zh) * 2018-01-05 2019-07-12 皓琪科技股份有限公司 印刷电路板复检系统、标示方法与复检方法
CN110473170A (zh) * 2019-07-10 2019-11-19 苏州卓融新能源科技有限公司 一种适用于pcb板真假缺陷判定的人工智能检测方法
CN110579479A (zh) * 2019-08-09 2019-12-17 康代影像科技(苏州)有限公司 一种基于假点缺陷检测的pcb检修系统及检修方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117409261A (zh) * 2023-12-14 2024-01-16 成都数之联科技股份有限公司 一种基于分类模型的元件角度分类方法及系统
CN117409261B (zh) * 2023-12-14 2024-02-20 成都数之联科技股份有限公司 一种基于分类模型的元件角度分类方法及系统

Also Published As

Publication number Publication date
KR20230078729A (ko) 2023-06-02

Similar Documents

Publication Publication Date Title
TWI757825B (zh) 基於假點缺陷檢測之pcb檢修系統及檢修方法
CN110659662B (zh) 利用人工智能的瑕疵检测系统及其方法
CN106096851A (zh) 一种基于bim技术的钢结构工程可视化验收方法
CN110021005A (zh) 电路板瑕疵筛选方法及其装置与计算机可读取记录介质
CN111105167A (zh) 一种基于bim技术的钢结构工程可视化验收方法
CN111208140A (zh) 一种缺陷复检系统及方法
CN110186375A (zh) 智能化高铁白车身焊接装配特征检测装置及检测方法
WO2022104585A1 (zh) 一种印刷电路板远程光学检修方法及系统
CN105184519A (zh) 标准化安全作业痕迹化管理系统
JP2021119341A (ja) 生産ラインスマート監視システム及び監視方法
CN214201239U (zh) 一种印刷电路板远程光学检修系统
TWI780580B (zh) 圖像複檢方法、電腦裝置及儲存介質
WO2015119450A1 (ko) 공정지연 및 시험과 검사의 불량을 최소화하고 신뢰성이 향상된 모바일 토탈 품질관리 매니지먼트 시스템 및, 이의 방법
TWI769688B (zh) 一種印刷電路板遠程光學檢修方法及系統
CN117589770A (zh) Pcb贴片板检测方法、装置、设备及介质
CN110531718A (zh) 一种石化工业生产线中设备的巡检方法及巡检系统
KR20230056006A (ko) 3차원 모델을 통한 인공지능 기반 안전진단 장치 및 그 방법
CN115829191A (zh) 生成检验计划的方法、设备及存储介质
CN215749764U (zh) 一种板件封边质量自动检验系统
CN113378810B (zh) 一种螺丝锁附检测方法、装置、设备和可读存储介质
CN107894425A (zh) 一种基于二维自动光学检测机的缺陷检测辅助系统
CN115379197A (zh) 用于在生产线上检测摄像头产品的镜头表面的方法和装置
JP2003140735A (ja) ポンプ場点検システム
CN216870414U (zh) 一种用于检测整板元件外观缺陷的装置
Zhang et al. Machine Vision On-line Detection System: Applications and Standardization Requirements

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: 20961876

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20237014079

Country of ref document: KR

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: 20961876

Country of ref document: EP

Kind code of ref document: A1