Inspection system, inspection management device, inspection program generation method, and storage medium

Abstract

An inspection system comprising: 1+n imaging units that capture image data of a component mounting substrate, which is the object of inspection; 1+m inspection units that perform inspections corresponding to each of the 1+n image data acquired by the 1+n imaging units; an inspection suitability calculation unit that calculates an inspection suitability for each inspection item related to each component mounted on the component mounting substrate, the inspection suitability representing the suitability of the 1+m inspections of the 1+m inspection units relative to the detection of an abnormality by that inspection item; and an inspection program generation form that, based on the inspection suitability, determines whether or not the 1+m inspections are performed for each inspection item related to each component mounted on the component mounting substrate.

Description

Technical Field

[0001] This invention relates to an inspection system, an inspection management device, an inspection program generation method, and a program. Background Technology

[0002] It has long been known that there are systems in which the following are used in the manufacturing process of various substrates: measuring and inspecting images obtained by photographing the substrate; and in the case of inspection based on multiple inspection devices, the inspection items determined for each component of the substrate are shared by multiple inspection devices (Patent Document 1).

[0003] Patent Document 1 discloses a technique in which, in a substrate inspection system having multiple inspection devices, an inspection device is selected according to each component of the substrate to be inspected and according to each inspection item that needs to be performed, and this selection is reflected in the inspection program of each inspection device. This prevents the useless repetition of the same inspection items in multiple inspection devices or the generation of inspection items that are not performed by any inspection device.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2012-151250 (Patent No. 5522065) Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] However, in the technology described in Patent Document 1, the inspection device for which inspection item is performed on which component is determined based on design information such as the type of component and its arrangement on the substrate. However, in actual inspection, the allocation of inspection items may be subject to unforeseen influences based on the state of the components mounted on the substrate and the state of each inspection device. From the viewpoint of inspection accuracy and efficiency, the allocation of inspection items described above may not be optimal.

[0009] The present invention was made in view of the above-mentioned actual situation, and its purpose is to provide a technology that can improve the accuracy and efficiency of inspection in an inspection system for a component mounting substrate having multiple inspection devices.

[0010] Methods for solving problems

[0011] To achieve the above objectives, the present invention employs the following structure. That is,

[0012] An inspection system, characterized in that the inspection system has:

[0013] 1+n types of imaging units, which capture image data of the component mounting substrate that is the object of inspection;

[0014] There are 1+m inspection units, which perform inspections corresponding to each of the 1+n image data obtained by the 1+n imaging units respectively.

[0015] The inspection fit calculation unit calculates an inspection fit for each inspection item related to each component mounted on the component mounting substrate. This inspection fit represents the fit of the 1+m inspection units for each of the 1+m inspection items in relation to the detected abnormalities for that inspection item.

[0016] The inspection program generation unit generates or updates the inspection program for the component mounting substrate.

[0017] Based on the inspection suitability, the inspection program generation unit determines whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

[0018] The term "imaging unit" as used here is not limited to cameras that detect wavelengths in the visible light region, but also includes X-ray cameras that detect X-rays, light sensors used for laser scanning, and the like. Furthermore, the "inspection unit" can be, for example, an automated optical inspection (AOI) or automated X-ray inspection (AXI) device that performs inspections based on image data obtained from photographing the object being inspected. Additionally, the "inspection items" mentioned above may also include information such as the coordinates of the inspection and the extracted parameters of the object being inspected. Furthermore, in the above, n = m.

[0019] Based on this structure, in a substrate inspection system with multiple inspection units, for each inspection item determined according to each component of the substrate, a fit can be calculated based on the differences in the measurement principles of each inspection unit, and an inspection program can be set according to the fit, so that the inspection unit with the highest fit is responsible for the inspection of that item. Therefore, the reliability (accuracy) and efficiency (speed) of the inspection can be improved.

[0020] Alternatively, the inspection system may also include a sample image acquisition unit, which acquires 1+n sample images of the component mounting substrate captured by the 1+n imaging units respectively.

[0021] The inspection fit calculation unit has a first fit calculation unit, which calculates the inspection fit based on the sample image.

[0022] Based on this structure, it is possible to calculate which inspection unit is more suitable for inspection based on the actual imaging method of the inspected component, using images captured by the actual inspection equipment. For example, the shape of a component (and its surrounding solder joints) visible from the outside is generally suitable for visual inspection, but in reality, in images captured by a visible light camera, situations may occur such as secondary reflections from the solder joints, blind spots for other components, or insufficient / saturated brightness. In such cases, the suitability of visual inspection decreases (compared to general assumptions). Therefore, by considering this information to determine suitability, a more accurate suitability can be calculated. Furthermore, by setting the inspection procedure based on this, the accuracy of the inspection can be improved.

[0023] Alternatively, the first fitness calculation unit may include a learned model, which is obtained by machine learning from at least any one of the 1+m types of inspections performed in the past, using a learning dataset including inspection image data related to the component mounting substrate that is either missed or over-detected.

[0024] Furthermore, "missed detection" refers to overlooking certain aspects, while "over-detection" refers to excessive checking. With this structure, a learning model based on past performance data can be used to efficiently calculate fitness.

[0025] Alternatively, the inspection system may also include an inspection history acquisition unit that acquires past inspection history information, including missed detections and / or over-detections, related to components of the same type as those mounted on the component mounting substrate.

[0026] The inspection suitability calculation unit has a second suitability calculation unit, which calculates the inspection suitability based on the inspection history information.

[0027] The term "same type of component" here is not limited to components with the same component number, but also includes other components with similar shapes, uses, etc. Based on this structure, by considering the inspection results of components that are prone to missed or over-inspection (with many past instances) for each inspection unit, a more accurate fit can be calculated.

[0028] Alternatively, the inspection system may also include a design information acquisition unit that acquires design information related to the component mounting substrate.

[0029] The fitness calculation unit has an initial value calculation unit that calculates an initial value for the fitness calculation based on the design information.

[0030] Regarding the inspection items for each component mounted on the substrate, the general understanding has been that by referring to design information such as the component configuration and dimensions on the substrate, it is possible to infer which inspection unit is suitable for inspection. Therefore, based on this understanding, for each inspection item of each component mounted on the substrate that is the object of inspection, the suitability for the first / second inspection can be calculated and set as the initial value. This makes it easier to calculate the suitability, and based on the actual inspection situation, the initial value can be updated with a more accurate suitability calculation, thereby improving the accuracy of the inspection.

[0031] Alternatively, the inspection fit can be calculated separately for each of the 1+m inspections, and the inspection fit calculation unit calculates the inspection fit for all 1+m inspections according to each inspection item related to each component.

[0032] Specifically, for example, the suitability of each of the 1+m types of checks can be calculated using a 10-level scale from 1 to 10. This allows for a determination of how to allocate each check item based on the calculated and compared suitability scores, making the check procedure more accurately reflect the suitability of the checks. However, the suitability of checks is not limited to the representation described above. For example, the suitability of other checks relative to one check can be expressed as a ratio, or the total suitability of all checks can be assigned values ​​such that it is always 100. Furthermore, the suitability of all checks can be ranked.

[0033] Alternatively, the inspection program generation unit may, based on the inspection suitability, determine whether to perform the 1+m types of inspections for each inspection item related to the components mounted on the component mounting substrate in the following manner: perform at least one of the 1+m types of inspections for each inspection item related to the components, and perform any one of the 1+m types of inspections for each inspection item for which the inspection suitability of each of the 1+m types of inspections does not reach the specified benchmark.

[0034] To achieve efficient inspection, it is preferable to avoid performing repetitive inspections across multiple inspection units. However, it is necessary to perform all components thoroughly and accurately. In this regard, with the structure described above, coverage can be ensured by repeatedly performing all inspections, even if no single inspection can guarantee the required accuracy.

[0035] Alternatively, for inspection items where the difference in suitability of each of the 1+m inspections is within a specified range, the inspection program generation unit may determine to implement any one of the 1+m inspections in a manner that minimizes the production line cycle time related to the inspection of the component mounting substrate.

[0036] If the suitability of the first and second checks is a value that ensures the accuracy of the checks without any issues, then regarding which inspection unit should perform the inspection for that item, it is preferable to generate the inspection procedure in a way that minimizes the cycle time of the entire inspection process (i.e., in a way that improves efficiency). Specifically, inspections can be assigned based on the following criteria: by assigning the inspection item to a unit that does not increase the number of shooting fields; by assigning the inspection to a unit that does not cause bottleneck problems by referring to historical information on the time required for the inspection.

[0037] Alternatively, the 1+n imaging units may include a first imaging unit as a visible light camera and a second imaging unit as an X-ray camera, and the 1+m inspections may include a first inspection based on first image data obtained by the first imaging unit and a second inspection based on first image data obtained by the second imaging unit. This combination of inspection units is suitable for inspecting component mounting substrates.

[0038] Furthermore, the present invention can also be understood as an inspection management device, which is a device for managing inspections using an inspection system. This inspection system has: 1+n types of imaging units that capture images of the component mounting substrate, which is the object of inspection, to obtain image data; and 1+m types of inspection units that perform inspections corresponding to each of the 1+n types of image data acquired by the 1+n types of imaging units. The inspection management device is characterized by having:

[0039] The inspection fit calculation unit calculates an inspection fit for each inspection item related to each component mounted on the component mounting substrate. This inspection fit represents the fit of the 1+m inspection units for each of the 1+m inspection items in relation to the detected abnormalities for that inspection item.

[0040] Inspection program generation unit, which generates or updates the inspection program for the component mounting substrate;

[0041] Based on the inspection suitability, the inspection program generation unit determines whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

[0042] Furthermore, this invention can also be understood as an inspection program generation method, which is an inspection program generation method in an inspection system having: 1+n types of imaging units that capture images of the component mounting substrate, which is the object of inspection, to obtain image data; and 1+m types of inspection units that perform inspections corresponding to each of the 1+n types of image data obtained by the 1+n types of imaging units respectively. The inspection program generation method is characterized by having:

[0043] The fit calculation step involves calculating the fit for each inspection item related to each component mounted on the component mounting substrate, where the fit represents the fit of the 1+m inspection units for each of the 1+m inspection items to detect anomalies for that inspection item; and

[0044] The inspection decision step involves determining, based on the inspection suitability, whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

[0045] Alternatively, the inspection procedure generation method may also include: a sample image acquisition step, acquiring 1+n sample images of the component mounting substrate captured by the 1+n imaging units respectively.

[0046] The fitness calculation step includes: a first fitness calculation step, which calculates the fitness of the examination based on the sample image.

[0047] Alternatively, the inspection procedure generation method may also include an inspection history acquisition step, which acquires past inspection history information, including missed detections and / or over-detections, related to components of the same type as those mounted on the component mounting substrate.

[0048] The inspection fit calculation step includes: a second fit calculation step, which calculates the inspection fit based on the inspection history information.

[0049] Alternatively, the inspection procedure generation method may also include: a design information acquisition step, which acquires design information related to the component mounting substrate.

[0050] The check fit calculation step includes: an initial value calculation step, which calculates the initial value of the check fit based on the design information.

[0051] Furthermore, the present invention can also be understood as a program for causing a computer to perform the above-described method, and a computer-readable recording medium that non-temporarily records such a program.

[0052] Furthermore, the above-mentioned structures and processes can be combined with each other to constitute the present invention, provided that they are not technically contradictory.

[0053] Invention Effects

[0054] According to the present invention, a technique can be provided that improves the accuracy and efficiency of inspection in an inspection system for component mounting substrates having multiple inspection devices. Attached Figure Description

[0055] Figure 1 This is a schematic diagram illustrating the general structure of an inspection system for an application example.

[0056] Figure 2 This is a block diagram illustrating the general structure of the inspection system according to an embodiment.

[0057] Figure 3 This is a flowchart illustrating the process of generating an inspection program in an inspection system according to an embodiment. Detailed Implementation

[0058] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, unless otherwise specified, the dimensions, materials, shapes, and relative arrangements of the structural elements described in the following examples are not intended to limit the scope of the present invention.

[0059] <Application Example>

[0060] (Structure of an application example)

[0061] This invention can be used, for example, as an inspection management device for generating inspection procedures for a substrate inspection system. Figure 1 This is a schematic diagram illustrating the general structure of the substrate inspection system applying the present invention. Figure 1 As shown, the substrate inspection system 9 of this application example is configured to include multiple inspection devices 91 and 92 equipped on a production line (not shown) for mounting component substrates, an inspection management device 93 for managing the content and results of the inspection, and communication lines such as a LAN (Local Area Network) connecting them to each other. Furthermore, this application example is described using a system including two types of imaging units and an inspection device having each imaging unit, but the number of imaging units and inspection devices is not limited to this.

[0062] Inspection devices 91 and 92 are respectively devices that inspect the component mounting substrate O, which is the object of inspection, based on image data obtained by capturing images of the component mounting substrate being transported from the production line by a conveyor roller (not shown) using an imaging unit. Figure 1As shown, each inspection device 91 and 92 has a structure including imaging units 911 and 921, image data acquisition units 912 and 922, and inspection processing units 913 and 923. Additionally, the white arrows in the figure indicate the direction of the transport component mounting substrate O.

[0063] Here, the imaging unit 911 of the inspection device 91 and the imaging unit 921 of the inspection device 92 are different types of imaging units. For example, the imaging unit 911 can be a visible light camera, and the imaging unit 921 can be an X-ray camera. Moreover, in each inspection device, the component mounting substrate O is inspected by applying a prescribed inspection procedure in the inspection processing unit to determine whether the image data obtained by the imaging unit and the image data acquisition unit is qualified or not.

[0064] The inspection management device 93 can be configured, for example, by a general-purpose computer, and includes functional units such as an inspection suitability calculation unit 931, an inspection program generation unit 932, and a storage unit 933. Furthermore, although not shown, it includes various input units such as a mouse and keyboard, and output units such as a display.

[0065] The inspection suitability calculation unit 931 calculates the inspection suitability for each inspection item related to each component mounted on the component mounting substrate O. This inspection suitability indicates the suitability of the inspection by the inspection device 91 and the inspection by the inspection device 92 for detecting abnormalities through that inspection item. The specific calculation method is described below.

[0066] Furthermore, the inspection program generation unit 932 generates an inspection program for the inspection processes performed by the inspection devices 91 and 92. The generated inspection program includes flag information indicating whether to perform inspections related to each component mounted on the component mounting substrate O in the inspection devices 91 and 92. That is, if the flag is "on," the inspection of the target inspection item is performed in the inspection device; if the flag is "off," it is not performed. Additionally, the "generation" of the program here includes not only generating a program from the beginning but also updating existing programs.

[0067] The storage unit 933 is composed of storage units such as RAM and HDD, and stores various design information related to the component mounting substrate O (the mounted components, the configuration relationship of the components, etc.), information related to the components (component type, component number, batch number, component image, etc.), inspection procedures (inspection items, inspection standards, etc.), past inspection image data, past inspection result information, and other information.

[0068] (Check fit calculation)

[0069] Next, in this application example, the fit calculation performed by the fit calculation unit 931 will be explained. Based on various design information related to the component mounting substrate O stored in the storage unit 933, the fit calculation unit 931 sets an initial value for the fit (i.e., the degree to which abnormalities can be appropriately detected) of the inspection performed in each inspection device 91, 92, according to each component mounted on the component mounting substrate O and each inspection item for each component. For example, inspection items such as "different serial numbers," which involve component colors or characters printed on components, cannot be inspected using images captured by an X-ray camera; therefore, the fit of the X-ray inspection device is 0. On the other hand, bottom-mounted components (BGAs) of the substrate cannot be inspected using visible light images obtained by photographing the upper surface of the substrate; therefore, the fit of the appearance inspection device is 0.

[0070] The fit calculation unit 931 also performs a process to update the set initial values ​​in conjunction with the actual inspection environment. For example, it can acquire images of qualified product samples (hereinafter referred to as sample images) pre-captured by the inspection devices 91 and 92, and correct the fit based on the sample images in a way that reflects the actual situation. For example, if the acquired sample image has insufficient brightness / saturation, the inspection fit of the inspection device that captured the sample image is reduced and corrected.

[0071] The inspection fit calculation unit 931 calculates the inspection fit for each inspection item related to each component mounted on the component mounting substrate O. The inspection fit indicates the suitability of the inspection by the inspection device 91 and the inspection by the inspection device 92 for detecting abnormalities through the inspection item.

[0072] Furthermore, the inspection program generation unit 932 generates an inspection program based on the inspection suitability calculated in this way. Here, the inspection suitability is used to ensure coverage in a way that necessary and sufficient inspection is performed on all components mounted on the component mounting substrate O, and to determine the distribution of inspection items in a way that avoids repeated inspections in the inspection apparatus 91 and inspection apparatus 92 as much as possible.

[0073] In addition, the inspection program generated by the inspection program generation unit 932 is sent to each inspection device 91, 92, and the inspection of the component mounting substrate O begins according to the inspection program generated based on the inspection suitability.

[0074] According to the inspection management system 9 described above, inspections can be carried out by multiple inspection devices with different imaging systems according to the suitability of each inspection item, thus achieving both improved inspection accuracy and efficiency.

[0075] <Implementation Method>

[0076] Hereinafter, taking Figure 2 and Figure 3 as an example, a system for inspecting a substrate by an appearance inspection device and an X-ray inspection device will be used to further describe the embodiments of the present invention in detail.

[0077] (System Structure)

[0078] Figure 2 FIG. is a schematic block diagram showing the structure of the substrate inspection system 1 of the present embodiment. The substrate inspection system 1 of the present embodiment is generally configured to include an appearance inspection device 10, an X-ray inspection device 20, a data server 30, and an inspection management device 40, which are connected in a communicable manner through a communication unit not shown.

[0079] The appearance inspection device 10 is, for example, a device that inspects the appearance of a component mounting substrate by an inspection method that combines a so-called phase shift method and a color highlight method. Regarding the inspection method that combines the phase shift method and the color highlight method, since it is already a well-known technique, a detailed description thereof will be omitted. However, through such inspection, at the pad portion of the substrate, the shape of the electrode visible from the appearance and the inclination degree of the solder leg can be detected with high precision. In addition, the phase shift method is one of the methods for restoring the three-dimensional shape of an object surface by analyzing the deformation of a pattern when the pattern light is projected onto the object surface. Further, the color highlight method is a method in which light of multiple colors (wavelengths) is irradiated onto the substrate at different incident angles, and the substrate is photographed in a state where a color feature corresponding to the normal direction appears on the solder surface (the color of the light source located in the specular reflection direction when observed from the camera), thereby capturing the three-dimensional shape of the solder surface as two-dimensional tone information.

[0080] The appearance inspection device 10 generally has functional units such as an appearance image capturing unit 110, an appearance measuring unit 120, and an appearance inspection unit 130, as well as a projector, a light source, a workbench for holding the substrate (all not shown). The appearance image capturing unit 110 captures the substrate in a state where light is irradiated from a projector and a light source not shown, and outputs an image for appearance inspection. The appearance measuring unit 120 measures the appearance shape of the substrate (mounted components) based on the image for appearance inspection. The appearance inspection unit 130 compares the measured appearance shape with an inspection reference to perform an appearance inspection of the substrate (mounted components), that is, a determination of pass or fail. In addition, hereinafter, even when it is only set as "inspection of the substrate", it includes inspection of the components mounted on the substrate.

[0081] In addition, each of the above information of the image for appearance inspection, the measured value of the appearance shape, and the appearance inspection result is sent from the appearance inspection device 10 to the data server 30 and stored in the data server 30.

[0082] The X-ray inspection device 20 is, for example, a device that measures the three-dimensional shape of a substrate by means of CT (Computed Tomography), tomography, etc., and determines whether the substrate is qualified or not based on the three-dimensional shape.

[0083] The X-ray inspection apparatus 20 generally includes functional units such as an X-ray image capturing unit 210, an X-ray measuring unit 220, and an X-ray inspection unit 230, as well as an X-ray source and a worktable for holding the substrate (neither shown). The X-ray image capturing unit 210 captures X-rays (not shown) that have irradiated and passed through the substrate, outputting a tomographic image (hereinafter referred to as an X-ray image) of the substrate. The X-ray measuring unit 220 measures the three-dimensional shape of the substrate based on multiple X-ray images. The X-ray inspection unit 230 performs a three-dimensional shape inspection of the substrate, i.e., determines whether it is qualified or not, by comparing the measured three-dimensional shape with an inspection standard.

[0084] In addition, the aforementioned X-ray images, three-dimensional shape data, and X-ray examination results are sent from the X-ray examination device 20 to the data server 30 and stored in the data server 30.

[0085] The inspection management device 40 can be, for example, a general-purpose computer. That is, although not shown, it has a storage unit including a processor such as a CPU or DSP, a main storage unit such as a read-only memory (ROM) or random access memory (RAM) and an auxiliary storage unit such as an EPROM, a hard disk drive (HDD), or a removable medium, an input unit such as a keyboard and a mouse, and an output unit such as a liquid crystal display. Furthermore, the inspection management device 40 can be composed of a single computer or multiple computers working together.

[0086] The auxiliary storage unit stores the operating system (OS), various programs, information related to the object being inspected, and various inspection standards. The programs stored therein are loaded into the working area of ​​the main storage unit and executed. The execution of these programs controls various structural units, thereby enabling the functional units that perform the specified purposes, as described later. Alternatively, some or all of the functional units can be implemented using hardware circuits such as ASICs or FPGAs.

[0087] Next, the functional units of the inspection management device 40 will be described. The inspection management device 40 has functional units including an inspection suitability calculation unit 410, a design information acquisition unit 420, a sample image acquisition unit 430, a historical information acquisition unit 440, and an inspection program generation unit 450.

[0088] As described below, the inspection fit calculation unit 410 calculates the inspection fit based on the information obtained by the design information acquisition unit 420, the sample image acquisition unit 430, and the historical information acquisition unit 440. The inspection fit represents the degree of suitability of the appearance inspection device 10 and the X-ray inspection device 20 for each inspection item related to the components mounted on the substrate, indicating the degree to which any abnormalities are detected by that inspection item. In this embodiment, the appearance inspection fit representing the suitability of the appearance inspection device 10 and the X-ray inspection fit representing the suitability of the X-ray inspection device 20 are calculated as values ​​from 0 to 10.

[0089] In addition, in detail, the fitness calculation unit 410 is configured to also have a functional unit including an initial value calculation unit 411, an image information reflection unit 412, and a historical information reflection unit 413.

[0090] The design information acquisition unit 420 obtains design information about the substrate, including the shape, size, and arrangement of components (and pads) mounted on the substrate to be inspected, from the data server 30. Furthermore, the sample image acquisition unit 430 obtains sample image data from the data server 30 obtained by taking images of qualified samples of the substrate to be inspected by the visual inspection device 10 and the X-ray inspection device 20, respectively. Additionally, the historical information acquisition unit 440 obtains information from the data server 30 regarding past inspection history, including inspection results of missed and / or over-inspections, related to components of the same type as those mounted on the substrate. Furthermore, the term "same type of component" here is not limited to components with the same component number, but also includes other components with similar shapes, uses, etc.

[0091] The inspection program generation unit 450 generates an inspection program for the inspection process performed by the visual inspection device 10 and the X-ray inspection device 20. The generation of the inspection program will be described in detail later. Furthermore, the term "generation" here includes not only generating a program from the outset, but also updating an existing program.

[0092] The initial value calculation unit 411 calculates an initial value for the inspection suitability based on the design information obtained by the design information acquisition unit 420. Specifically, for example, inspection items such as "different product numbers" that require identification of component colors or characters printed on components cannot be inspected using X-ray inspection; therefore, the X-ray inspection suitability is set to 0, and the visual inspection suitability is set to 10. On the other hand, for inspection items related to components mounted on the bottom surface of the substrate, components covered by shielding members, etc., where visual inspection cannot be applied, the visual inspection suitability is set to 0, and the X-ray inspection suitability is set to 10. Furthermore, for inspections related to solder shape, such as front solder pads, it is sufficient to set the visual inspection suitability to be higher than the X-ray inspection suitability (e.g., visual inspection suitability = 7, X-ray inspection suitability = 4, etc.). However, even for inspection items related to solder shape, depending on the positional relationship between the target component and adjacent components, components that are more likely to become blind spots for visual inspection or be affected by secondary reflections from solder pads of adjacent components are set to have a low visual inspection suitability.

[0093] The image information reflecting unit 412 uses the data acquired by the sample image acquisition unit 430 to calculate a corrected inspection fit, which is obtained by correcting the initial value of the inspection fit calculated by the initial value calculation unit 411. Specifically, for example, if a component with secondary reflection or a blind spot is identified in the sample image of the appearance inspection, or if there is insufficient brightness / saturation in the image, a corrected inspection fit is calculated by reducing the appearance inspection fit related to that component. Furthermore, regarding X-ray inspection fit, for example, if there is significant noise caused by a component on the back of the substrate in the sample image of the X-ray inspection, a corrected inspection fit is also calculated by reducing the X-ray inspection fit. In addition, such processing based on sample images can be performed based on brightness and noise levels during image processing of the sample image, or the sample image can be input into a learned model obtained by learning from past inspection performance to obtain the value. In this embodiment, the image information reflecting unit 412 corresponds to the first fit calculation unit.

[0094] The historical information reflection unit 413 uses information from past inspection history acquired by the historical information acquisition unit 440 to calculate a revised inspection fit, which is obtained by further revising the initial value of the inspection fit or the revised inspection fit. Specifically, for example, if over-inspection of parts of the same type as the inspection object frequently occurred in past appearance inspections, a revised inspection fit, obtained by reducing the appearance inspection fit, can be calculated. Furthermore, if there is no specific historical information that should be reflected, the historical information reflection unit 413 does not need to calculate the revised inspection fit. In this embodiment, the historical information reflection unit 413 corresponds to the second fit calculation unit.

[0095] (Check the process generated)

[0096] Next, refer to Figure 3 The processing flow for generating inspection programs in the inspection management system 1 of this embodiment will be explained. Figure 3 This is a flowchart illustrating the process. For example... Figure 3 As shown, firstly, the components to be inspected, the inspection items for those components, and the inspection criteria for determining whether the inspection items are qualified or not are registered, and an initial program is generated (S101). This process can be performed by the user through manual input or by the inspection program generation unit 450 of the inspection management device 40.

[0097] Next, the visual inspection device 10 and the X-ray inspection device 20 are used to photograph qualified samples of the substrate, and the sample image data is stored in the data server 30 (S102).

[0098] Next, the inspection management device 40 obtains the design information of the substrate to be inspected from the data server 30 via the design information acquisition unit 420 (S103). Alternatively, if the design information of the substrate to be inspected is not pre-stored in the data server 30, the design information can be registered at this stage. Next, the inspection management device 40 calculates an initial value of the inspection fit based on the design information obtained in step S103 via the initial value calculation unit 411 of the inspection fit calculation unit 410 (S104). The calculation of the initial value of the inspection fit is as described above, therefore, the explanation here is omitted.

[0099] Next, the inspection management device 40 acquires the sample image of the substrate to be inspected, captured in step S102, from the data server 30 via the sample image acquisition unit 430 (S105). Then, the inspection management device 40 calculates a corrected inspection fit based on the sample image information acquired in step S105 via the image information reflection unit 412 of the inspection fit calculation unit 410 (S106). The calculation of the corrected inspection fit by the image information reflection unit 412 is as described above, therefore, its explanation here is omitted.

[0100] Next, the inspection management device 40 obtains past inspection history information, including inspection results of missed detections and / or over-detections, related to components of the same type as those mounted on the substrate, from the data server 30 via the historical information acquisition unit 440 (S107). Following this, the inspection management device 40 calculates a corrected inspection fit based on the historical information obtained in step S107 via the historical information reflection unit 413 of the inspection fit calculation unit 410 (S108). The calculation of the corrected inspection fit by the historical information reflection unit 413 is as described above, therefore, the explanation here is omitted.

[0101] Next, the inspection management device 40 updates the initial program using the inspection suitability calculated through steps S104 to S108 via the inspection program generation unit 450. More specifically, firstly, for each inspection item related to each component mounted on the substrate, a decision is made on whether to perform inspections in the visual inspection device 10 and the X-ray inspection device 20 (S109). Specifically, coverage is ensured by performing necessary and sufficient inspections on all components mounted on the substrate, and the decision to enable / disable inspections in each inspection device is made in a way that avoids repeating the same inspection items in the visual inspection device 10 and the X-ray inspection device 20 as much as possible. As a result, inspection efficiency can be improved, and over-inspection can be suppressed by performing inspections with lower inspection suitability.

[0102] Furthermore, the decision to enable / disable inspection items can be made in accordance with policies such as those described below. For example, "select inspection items with high suitability for each possible non-conformity category and enable them," or "for non-conformity categories with only low suitability in any inspection device (e.g., visual inspection suitability = 5, X-ray inspection suitability = 4, etc.), intentionally enable the inspection in any inspection device to prevent omissions." Additionally, flexible settings can be implemented, such as "for the inspection of the front solder joint, only perform inspection using the X-ray inspection device 20 if a non-conformity is found in the visual inspection device 10."

[0103] Next, the inspection management device 40 checks whether the inspection procedure, for which the inspection start / stop was determined in step S109, can be updated to further shorten the production line cycle time (S110). Specifically, if the inspection suitability of either the visual inspection device 10 or the X-ray inspection device 20 meets the prescribed benchmarks, and the inspection accuracy of the object inspection items is ensured in either case (e.g., visual inspection suitability = 8, X-ray inspection suitability = 8, etc.), the inspection start / stop is updated in a way that minimizes the overall production line cycle time of the inspection process (i.e., in a way that improves efficiency). For example, the inspection procedure can be updated based on benchmarks such as "performing inspections using inspection devices that do not increase the number of fields of view even when performing the corresponding inspection items" and "predicting the time required for similar inspection items by referring to past historical information, and performing inspections in devices that will not cause bottleneck problems."

[0104] The inspection management device 40 saves the inspection program generated (updated) in this way to the data server 30 (or each inspection device (S111)) and ends a series of inspection program generation processes. Then, the appearance inspection device 10 and the X-ray inspection device 20 perform the inspection of the substrate according to the inspection program.

[0105] According to the inspection management system of this embodiment as described above, in an inspection system for a component mounting substrate equipped with a visual inspection device and an X-ray inspection device, it is possible to calculate the inspection suitability for each inspection item related to each component mounted on the substrate, and based on this, to generate an inspection program that ensures inspection coverage and performs efficient inspections. Therefore, inspection accuracy can be ensured and inspection efficiency can be improved.

[0106] <Other>

[0107] The examples described above are merely illustrative of the invention, and the invention is not limited to the specific embodiments described. The invention can be modified and combined in various ways within its conceptual scope. For example, in the systems described above, two inspection devices are configured to correspond to the two types of imaging units, but the invention can also be applied to systems that include inspection devices with other imaging units. Furthermore, conversely, a structure can be constructed where multiple imaging units and corresponding inspection units are provided in a single inspection device. Moreover, while the examples described above are of systems including inspection devices, the invention can also be understood as a management device for inspection systems including the inspection devices described above.

[0108] Furthermore, in the inspection procedure generation process described above, the order of steps S105 to S108 can be changed. Also, any one or all of steps S106, S108, and S110 can be omitted.

[0109] Furthermore, in the above embodiment, an inspection management device 40 for generating inspection procedures is configured to be separate from the appearance inspection device 10 and the X-ray inspection device 20. However, it is also possible not to provide an inspection management device 40 separately, but to provide each functional unit of the inspection management device 40 in either the appearance inspection device 10 or the X-ray inspection device 20 to perform the above-mentioned steps.

[0110] Furthermore, in the above embodiment, the appearance inspection device 10 was described as an inspection device that combines phase shift mode and color highlight mode, but it may also be an appearance inspection device that performs inspection only by phase shift mode or only by color highlight mode.

[0111] Furthermore, the present invention is not limited to the combination of a visual inspection device and an X-ray inspection device, but can also be applied to the combination of a laser scanning measurement device and an X-ray inspection device.

[0112] <Postscript 1>

[0113] An inspection system (1) is characterized in that the inspection system (1) comprises: 1+n types of imaging units (110, 210) which capture images of the component mounting substrate, which is the object of inspection, and obtain image data; and 1+m types of inspection units (10, 20) which perform inspections corresponding to each of the 1+n types of image data obtained by the 1+n types of imaging units respectively.

[0114] The inspection fit calculation unit (410) calculates an inspection fit for each inspection item related to each component mounted on the component mounting substrate, wherein the inspection fit represents the fit of the 1+m inspection units for each of the 1+m inspections to detect an abnormality for that inspection item; and

[0115] The inspection program generation unit (450) generates or updates the inspection program for the component mounting substrate.

[0116] Based on the inspection suitability, the inspection program generation unit determines whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

[0117] <Appendix 2>

[0118] An inspection management device (93) manages inspections using an inspection system (9), which includes: 1+n imaging units (911, 921) that capture image data of a component mounting substrate (O) to be inspected; and 1+m inspection units (91, 92) that perform inspections corresponding to each of the 1+n image data acquired by the 1+n imaging units, characterized in that the inspection management device (93) has:

[0119] The inspection fit calculation unit (931) calculates an inspection fit for each inspection item related to each component mounted on the component mounting substrate. This inspection fit represents the fit of the 1+m inspection units for each of the 1+m inspection items in relation to the detection of abnormalities for that inspection item.

[0120] The inspection program generation unit (932) generates or updates the inspection program for the component mounting substrate.

[0121] Based on the inspection suitability, the inspection program generation unit determines whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

[0122] <Appendix 3>

[0123] An inspection program generation method is provided, which is an inspection program generation method in an inspection system. The inspection system includes: 1+n types of imaging units that capture images of the mounting substrate of the component being inspected, obtaining image data; and 1+m types of inspection units that perform inspections corresponding to each of the 1+n types of image data obtained by the 1+n types of imaging units. The method is characterized by:

[0124] The fit calculation steps (S104, S106, S108) calculate the fit for each inspection item related to each component mounted on the component mounting substrate, wherein the fit represents the fit of the 1+m inspection units for each of the 1+m inspections to detect an abnormality for that inspection item; and

[0125] The inspection decision step (S109) is implemented, which, based on the inspection suitability, determines whether to implement the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

[0126] Label Explanation

[0127] 1, 9: Substrate inspection system; 10: Appearance inspection device; 110: Appearance image capturing unit; 120: Appearance measurement unit; 130: Appearance inspection unit; 20: X-ray inspection device; 210: X-ray image capturing unit; 220: X-ray measurement unit; 230: X-ray inspection unit; 30: Data server; 40, 93: Inspection management device; 410, 931: Inspection fit calculation unit; 450, 932: Inspection program generation unit; 91, 92: Inspection device; 911, 921: Imaging unit; O: Component mounting substrate.

Claims

1. An inspection system, characterized in that, The inspection system has the following features: 1+n types of imaging units, which capture image data of the component mounting substrate that is the object of inspection; The sample image acquisition unit acquires 1+n sample images of qualified product samples that serve as component mounting substrates, captured by the 1+n imaging units respectively. There are 1+m inspection units, which perform inspections corresponding to each of the 1+n image data obtained by the 1+n imaging units respectively. The inspection fit calculation unit calculates the inspection fit for each inspection item related to each component mounted on the component mounting base plate. The inspection fit represents the fit of the 1+m inspections of the 1+m inspection units for each inspection item that detects an abnormality. as well as The inspection program generation unit generates or updates the inspection program for the component mounting substrate. The inspection fit calculation unit has a first fit calculation unit, which calculates the inspection fit based on the sample image and by combining it with the actual inspection environment. Based on the inspection suitability, the inspection program generation unit determines whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

2. The inspection system according to claim 1, characterized in that, The first fitness calculation unit includes a learned model, which is obtained by machine learning from at least any one of the 1+m types of inspections performed in the past, using a learning dataset including inspection image data related to the component mounting substrate that is either missed or over-detected.

3. The inspection system according to claim 1 or 2, characterized in that, The inspection system also includes an inspection history acquisition unit, which acquires past inspection history information, including missed detection and / or over-detection results, related to components of the same type as those mounted on the component mounting substrate. The inspection suitability calculation unit has a second suitability calculation unit, which calculates the inspection suitability based on the inspection history information.

4. The inspection system according to claim 1 or 2, characterized in that, The inspection system also includes a design information acquisition unit, which acquires design information related to the component mounting substrate. The fitness calculation unit has an initial value calculation unit that calculates an initial value for the fitness calculation based on the design information.

5. The inspection system according to claim 1 or 2, characterized in that, The fitness of the inspection is calculated separately for each of the 1+m inspections. The inspection suitability calculation unit calculates the inspection suitability of all 1+m inspections for each inspection item related to each component.

6. The inspection system according to claim 1 or 2, characterized in that, The inspection program generation unit determines, based on the inspection suitability, whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate in the following manner: performing at least one of the 1+m types of inspections for each inspection item related to each component, and performing any one of the 1+m types of inspections for each inspection item for which the inspection suitability of each of the 1+m types of inspections does not reach the specified benchmark.

7. The inspection system according to claim 1 or 2, characterized in that, For inspection items for which the differences in suitability of each of the 1+m types of inspections are within a specified range, the inspection program generation unit determines to implement any one of the 1+m types of inspections in a manner that minimizes the production line cycle time related to the inspection of the component mounting substrate.

8. The inspection system according to claim 1 or 2, characterized in that, The 1+n imaging units include a first imaging unit as a visible light camera and a second imaging unit as an X-ray camera. The 1+m types of checks include a first check based on first image data obtained through the first shooting unit and a second check based on second image data obtained through the second shooting unit.

9. An inspection management device for managing inspections using an inspection system, the inspection system comprising: 1+n types of imaging units that capture image data of a component mounting substrate, which is the object of inspection; and 1+m types of inspection units that perform inspections corresponding to each of the 1+n types of image data acquired by the 1+n types of imaging units, characterized in that... The inspection management device has: The sample image acquisition unit acquires 1+n sample images of qualified product samples that serve as component mounting substrates, captured by the 1+n imaging units respectively. The inspection fit calculation unit calculates the inspection fit for each inspection item related to each component mounted on the component mounting base plate. The inspection fit represents the fit of the 1+m inspections of the 1+m inspection units for each inspection item that detects an abnormality. as well as The inspection program generation unit generates or updates the inspection program for the component mounting substrate. The examination fit calculation unit calculates the examination fit based on the sample image and by combining it with the actual examination environment. Based on the inspection suitability, the inspection program generation unit determines whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate.

10. A method for generating an inspection program, which is an inspection program generation method in an inspection system, the inspection system comprising: 1+n types of imaging units that capture images of a component mounting substrate, which is the object of inspection, to obtain image data; and 1+m types of inspection units that perform inspections corresponding to each of the 1+n types of image data acquired by the 1+n types of imaging units, characterized in that... The inspection program generation method has the following characteristics: The sample image acquisition step involves acquiring 1+n sample images of qualified product samples that serve as component mounting substrates, captured by the 1+n imaging units respectively. The inspection fit calculation step involves calculating the inspection fit for each inspection item related to each component mounted on the component mounting base. The inspection fit represents the fit of the 1+m inspection units for each of the 1+m inspection items for detecting abnormalities through that inspection item. as well as The inspection decision-making step involves determining, based on the inspection suitability, whether to perform the 1+m types of inspections for each inspection item related to each component mounted on the component mounting substrate. The inspection fit calculation step includes: a first fit calculation step, which calculates the inspection fit based on the sample image by combining it with the actual inspection environment.

11. The inspection program generation method according to claim 10, characterized in that, The inspection procedure generation method further includes an inspection history acquisition step, which acquires past inspection history information, including missed detection and / or over-detection results, related to components of the same type as those mounted on the component mounting substrate. The inspection fit calculation step includes: a second fit calculation step, which calculates the inspection fit based on the inspection history information.

12. The inspection procedure generation method according to claim 10 or 11, characterized in that, The inspection procedure generation method further includes: a design information acquisition step, which acquires design information related to the component mounting substrate. The check fit calculation step includes: an initial value calculation step, which calculates the initial value of the check fit based on the design information.

13. A computer-readable storage medium that non-temporarily stores a program, which, when executed by a computer, implements the steps of the inspection program generation method according to any one of claims 10 to 12.

Images