Image inspection method, image management method for defective portions, and image inspection apparatus

Abstract

The present application provides an image inspection method, a defective portion image management method, and an image inspection apparatus. An image inspection method using an image inspection apparatus (e.g., an X-ray inspection apparatus) generates and displays a simulated defective portion image and a measurement value by transferring a defective portion image of a certain component to an image of another component, and can visually and numerically distinguish the difference between a good product and a defective product. The image inspection method using an image inspection apparatus of the present disclosure has a transfer step of generating a transfer image by transferring a defective portion image to a transfer destination in a good product image as part of all steps.

Description

Technical Field

[0001] This invention relates to an image inspection method and an image inspection apparatus for determining the quality of an object by inspecting an image of the object being inspected. Background Technology

[0002] Conventionally, when it is known that the required number of defective images for quality management cannot be obtained in actual manufacturing processes, simulated defective images are generated based on good product images. For example, in the case of simulating the generation of defective images required for initial learning, an automatic simulated defective image generation device capable of generating a large number of images is known. This simulated defective image generation device inputs good product images from a good product input unit as learning data for a neural network. Defective images are also input from a defective image input unit for learning. Furthermore, a defective image extraction unit extracts difference data from good product images, and in a simulated data condition setting unit, multiple sets of defect generation conditions, such as defect synthesis positions, are generated based on random values ​​from a random number generation unit. Based on these defect generation conditions, a simulated defective image generation unit synthesizes difference data from good product images, thereby generating multiple sets of simulated defective images, which are then input as learning data for a neural network (see, for example, Patent Document 1).

[0003] Furthermore, learning devices capable of performing high-precision learning on models for distinguishing good and defective products are also known. Such learning devices include an intermediate image generation unit, an intermediate image display unit, a boundary receiving unit, and a teacher image determination unit. The intermediate image generation unit generates multiple intermediate images based on a good product image representing a good product and a defective product image representing a defective product. The intermediate image display unit arranges the multiple intermediate images between the good and defective product images and displays them on a display device. The boundary receiving unit uses the boundary specified by the user for the intermediate images as the boundary between the good and defective product images. The teacher image determination unit determines the images of the good and defective product images based on the specified boundary (for example, see Patent Document 2). Moreover, image processing devices capable of easily generating good product images from defective product images are also known (for example, see Patent Document 3).

[0004] According to the content disclosed in Patent Document 1 above, a simulated defective image is generated based on data from good product images and defective images. This simulated defective image is then used as learning data for a neural network, thereby improving the accuracy of the inspection device's pass / fail determination as an automatic simulated defective image generation device. However, it cannot be said that it is possible to efficiently generate simulated defective images or their measured values ​​using previously accumulated defective images.

[0005] Furthermore, according to the content disclosed in Patent Document 2 above, when generating multiple intermediate images based on good product images and defective product images, the boundary acceptance unit uses the designation of the boundary of the intermediate images accepted by the user as the boundary between the good product images and the defective product images. However, if defective product images are not sufficiently obtained when designating the boundary of the intermediate images, the boundary of the intermediate images must be designated based solely on the good product images, which may require relying on experience for designation. This experience-based designation may cause deviations in the inspection quality of the learning device, and is therefore not preferred. In addition, it cannot be said that it is possible to efficiently generate new simulated defective images using previously accumulated defective images.

[0006] Existing technical documents

[0007] Patent documents

[0008] Patent Document 1: Japanese Patent Application Publication No. 2005-156334

[0009] Patent Document 2: Japanese Patent No. 6780769

[0010] Patent Document 3: Japanese Patent Application Publication No. 2020-008488 Summary of the Invention

[0011] The problem that the invention aims to solve

[0012] This disclosure was made in view of the aforementioned problems, and its ultimate purpose is to provide an image inspection method and an image inspection apparatus that can efficiently generate simulated images of defective parts by transferring images of previously obtained defective parts to images of good products, for use in quality management.

[0013] Methods for solving problems

[0014] The present disclosure for solving the above-mentioned problems includes an image inspection method capable of using an image inspection device to distinguish between good and defective products in an inspection object. The image inspection method includes: an image display step, displaying an image of the good product; a defective image display step, displaying images of multiple defective parts in an inspection object of the same type as the good product; a selection step, selecting a transfer destination in the image of the good product and images of one or more defective parts corresponding to the transfer destination from the images of the multiple defective parts; and a transfer step, transferring the image of one or more defective parts selected in the selection step to the transfer destination in the image of the good product to generate a transfer image.

[0015] According to this disclosure, in the transfer step, by transferring an image of a defective part to a transfer destination in an image of a good product, a simulated image of the defective part can be efficiently generated. This simulated image of the defective part can be used to distinguish between good and defective products in the manufacturing process, thereby facilitating operations such as teaching the image inspection device to register the differences between good and defective products. The transferred image can also be provided as a criterion for distinguishing between good and defective products. Furthermore, the more images of defective parts transferred in the transfer step, the more simulated images of defective parts can be obtained.

[0016] Alternatively, in this disclosure, the image inspection method may also include: a transfer image display step, displaying the transfer image; a measurement step, measuring a predetermined feature quantity in the transfer image; and a measurement result display step, displaying the measurement result obtained in the measurement step, or the measurement result obtained in the measurement step, and the measurement result of the predetermined feature quantity in other inspected objects. Thus, the difference between good and defective products can be distinguished visually and numerically, the discrimination criteria become clearer, and therefore teaching becomes easier to implement.

[0017] Alternatively, in this disclosure, the image inspection method may be characterized in that, in the image display step and the defective image display step, the image of the good product and the images of the plurality of defective parts are respectively displayed as tomographic images in the XYZ directions. Therefore, when the defective product is three-dimensional, the risk of missing defective parts in the defective product due to blind spots can be prevented.

[0018] Alternatively, in this disclosure, the image inspection method can be configured as follows: during the transfer step, if the size of the transfer destination differs from the size of the image of the defective area, the size of the image of the defective area can be adjusted to make their respective sizes the same. Thus, transfer can be performed regardless of the size of the transfer destination.

[0019] Alternatively, in this disclosure, the image inspection method may be characterized in that, in the measurement result display step, the measurement results are displayed as a histogram. By displaying the results as a histogram, the frequency of defective products can be easily determined based on the measurement values.

[0020] Alternatively, in this disclosure, the image inspection method may be characterized in that the object to be inspected is a circuit board equipped with electronic components, the transfer destination in the image of the good product is the connection between the electronic components and the circuit board, and the boundary of the connection in the XY direction in the image of the good product is automatically determined by binarization. Thus, the transfer destination can be clearly determined.

[0021] Alternatively, in this disclosure, the image inspection method may be characterized in that the Z-direction boundary of the connecting portion is the two ends of the connecting portion, and is automatically set by capturing the positions of the two ends of the connecting portion. Therefore, even when the finished product is three-dimensional, the transfer destination can be clearly determined.

[0022] Alternatively, this disclosure could also be a method for managing images of multiple defective areas in the aforementioned image inspection method. This method involves database-izing images of the multiple defective areas obtained in past inspections, and associating each image of a defective area with the type of the inspected object, the type of defect, and a defect stage based on a numerical value representing the degree of defect. This defect stage is determined based on measurement results obtained in the previous measurement steps or on results previously determined through visual inspection. This allows for the acquisition of more detailed information related to the defective areas. More detailed information related to the defective areas can also be provided.

[0023] Additionally, this disclosure includes an image inspection apparatus capable of distinguishing between good and defective products among inspection objects. The image inspection apparatus comprises: a database storing data of images of good products; a defect database storing data of images of multiple defective parts in inspection objects of the same type as the good products; a transfer unit that transfers images of one or more defective parts corresponding to the transfer destination from the defect database to a transfer destination in the image of the good products obtained from the database, thereby generating a transfer image; and a display unit that displays the image of the good products obtained from the database and the images of the multiple defective parts obtained from the defect database.

[0024] According to this disclosure, an image of a defective part can be transferred to a transfer destination in the image of a good product using a simple device structure. This allows for the differentiation between good and defective products, and furthermore, the ability to differentiate between good and defective products facilitates teaching.

[0025] Alternatively, in this disclosure, the image inspection apparatus may also be characterized by further comprising a measuring unit that measures a predetermined feature quantity in the transferred image, and the display unit further displays the transferred image obtained in the transfer unit, the measurement result obtained in the measuring unit, and the measurement result of the predetermined feature quantity in other inspected objects. Thus, the difference between good and defective products can be distinguished visually and numerically, the discrimination criteria become clearer, and therefore teaching becomes easier to perform.

[0026] Alternatively, in this disclosure, the image inspection apparatus may be characterized in that, in the display unit, the image of the good product and the images of the plurality of defective parts are respectively displayed as tomographic images in the XY, YZ, and XZ directions. This prevents the risk of missing defective parts in a three-dimensional defective product due to blind spots. Since the display unit can display the image of the good product and the images of the plurality of defective parts as tomographic images in all directions—XY, YZ, and XZ—it is easy to grasp the overall image of the good product and the defective parts.

[0027] Alternatively, in this disclosure, the image inspection device can be configured such that, in the transfer unit, if the size of the transfer destination differs from the size of the image of the defective area, the size of the image of the defective area can be adjusted to make their respective sizes the same. Thus, transfer can be performed regardless of the size of the transfer destination.

[0028] Alternatively, in this disclosure, the image inspection apparatus may be characterized in that the measurement results are displayed as a histogram on the display unit. By displaying the results as a histogram, the frequency of defective products can be easily determined based on the measurement values. Furthermore, if the transfer image and the histogram can be displayed simultaneously on the display unit, it is easier to visually and numerically distinguish the differences between good and defective products.

[0029] Alternatively, in this disclosure, the image inspection apparatus may be characterized in that the object to be inspected is a circuit board equipped with electronic components, the transfer destination in the image of the good product is the connection between the electronic components and the circuit board, and the XY-direction boundary of the connection in the image of the good product is automatically determined by binarization. Thus, the transfer destination can be clearly determined.

[0030] Alternatively, in this disclosure, the image inspection device may be characterized in that the Z-direction boundary of the connecting portion is the two ends of the connecting portion, and is automatically set by capturing the positions of the two ends of the connecting portion. Therefore, even when the finished product is three-dimensional, the transfer destination can be clearly determined.

[0031] Alternatively, in this disclosure, the image inspection apparatus may be characterized in that the defect database contains images of the plurality of defective areas obtained in past inspections. Each image of the plurality of defective areas is associated with the type of the object being inspected, the type of defect, and a defect stage based on a numerical value representing the degree of defect. This defect stage is determined based on measurement results previously obtained in the measurement unit or based on results previously determined through visual inspection. This allows for the acquisition of more detailed information related to the defective areas.

[0032] Furthermore, the methods used to solve the above problems can be combined and used in combination as much as possible.

[0033] Invention Effects

[0034] According to this disclosure, by transferring an image of a defective part of a component onto an image of a good product, simulated images of defective parts can be generated efficiently for quality management. As a result, simulated images and measured values ​​of defective parts can be generated and displayed, allowing for visual and numerical differentiation between good and defective products. Furthermore, sufficient images of defective parts are ensured, facilitating teaching. Attached Figure Description

[0035] [ Figure 1 ] Figure 1 This is a functional block diagram illustrating an example of an X-ray inspection apparatus according to an embodiment.

[0036] [ Figure 2 ] Figure 2 This is an explanatory diagram illustrating an example of a method for managing images of multiple defective areas in an image inspection method using an X-ray inspection apparatus according to an embodiment.

[0037] [ Figure 3 ] Figure 3 Figure (A) is the first flow diagram illustrating an example of the content displayed by the display unit in an image inspection method using the X-ray inspection apparatus of the embodiment. Figure 3 Figure (B) is a flowchart illustrating an example of the content displayed by the display unit in an image inspection method using the X-ray inspection apparatus of the embodiment.

[0038] [ Figure 4 ] Figure 4 Figure (A) is the third flowchart illustrating an example of the content displayed by the display unit in an image inspection method using the X-ray inspection apparatus of the embodiment. Figure 4 Figure (B) is the fourth figure illustrating an example of the content displayed by the display unit in an image inspection method using the X-ray inspection apparatus of the embodiment.

[0039] [ Figure 5 ] Figure 5 Figure 5 is a flowchart illustrating an example of the content displayed by the display unit in an image inspection method using the X-ray inspection apparatus of the embodiment.

[0040] [ Figure 6 ] Figure 6 This is a flowchart illustrating the steps of an image inspection method using an X-ray inspection apparatus according to an embodiment.

[0041] Label Explanation

[0042] 1: X-ray inspection device; 10: Imaging unit; 11: Worktable; 12: Imaging condition storage unit; 13: X-ray generator; 14: X-ray detector; 20: Calculation unit; 21: 3D data generation unit; 22: Database; 23: Defect database; 24: Transfer unit; 25: Measurement unit; 30: Display unit; 40: Chip; 50: Pin; 51: Defective pin; 52: Transfer pin; 60: Measurement result. Detailed Implementation

[0043] [Application Example]

[0044] The following description uses some accompanying drawings to illustrate a general overview of applications of this disclosure. This disclosure can be applied to... Figure 1 The X-ray inspection apparatus 1 shown is as described. Furthermore, by using the X-ray inspection apparatus 1, this disclosure can be applied to… Figure 6 The process is as shown in the flowchart.

[0045] Figure 1 This is a functional block diagram illustrating an example of the X-ray inspection apparatus 1 disclosed herein that can be applied. For example... Figure 1 As shown, the X-ray inspection apparatus 1 is configured to generally include an imaging unit 10, a processing unit 20, and a display unit 30. The imaging unit 10 is configured to include a worktable 11, an imaging condition storage unit 12, an X-ray generator 13, and an X-ray detector 14. The object to be inspected by the X-ray inspection apparatus 1 (e.g., a circuit board with soldered electronic components) is placed on the worktable 11, and an image is taken by a camera (not shown) based on imaging conditions (e.g., imaging distance, brightness) read from the imaging condition storage unit 12. In addition, the object to be inspected placed on the worktable 11 is analyzed for its three-dimensional structure by X-rays generated from the X-ray generator 13. The X-ray detector 14 detects the intensity of the X-rays that irradiate and pass through the object to be inspected from the X-ray generator 13.

[0046] The computing unit 20 is configured to include a 3D data generation unit 21, a database 22, a defect database 23, a transfer unit 24, and a measurement unit 25. The 3D data generation unit 21 generates 3D data based on an image of the object to be inspected captured by a camera. Furthermore, 3D data refers to a tomographic image viewed from the X, Y, and Z directions. Here, due to reasons such as the low frequency of defects in the manufacturing process of the object to be inspected, it is often difficult to obtain 3D data of defective products from the object to be inspected. Therefore, the 3D data of the object to be inspected is mostly generated as good product data. In addition, the generated 3D data is stored in the database 22. Furthermore, for example, if the object to be inspected is a planar structure, or if only surface data of the object to be inspected is desired, the computing unit 20 may not have a 3D data generation unit 21, but instead have a structure for generating 2D data. The defect database 23 stores 3D data of multiple defective parts from components of the same type as the object to be inspected. Furthermore, the data stored in the defect database 23 may also be 2D data.

[0047] Display unit 30 displays the three-dimensional data generated by three-dimensional data generation unit 21 (hereinafter, to clearly distinguish it from the three-dimensional data of defective parts, this three-dimensional data is referred to as "image of good product") and the three-dimensional data of defective parts stored in defect database 23 (hereinafter referred to as "image of defective parts"). The defect database 23 pre-databases images of defective parts obtained in past inspections. In display unit 30, by selecting a portion of the image of the good product and the image of the defective part, transfer unit 24 uses a portion of the image of the good product as the transfer destination and transfers the image of the defective part. This updates the image of the good product, generating a transfer image containing simulated defective parts. Here, transfer refers to combining general image processing methods (e.g., binarization-based methods) to perform image clipping processing in a way that does not become unnatural (the same applies to the transfer described in the following embodiments). Furthermore, the transfer destination refers to the area where the image clipping processing is performed (the same applies to the transfer destination described in the following embodiments). Display unit 30 displays the transfer image, and measurement unit 25 measures the displayed transfer image. When the measurement is completed, the display unit 30 displays the measurement results (illustration omitted). Based on the measurement results, the frequency of defects occurring in the inspected object can be confirmed.

[0048] Furthermore, details regarding the computational processing performed by the arithmetic unit 20 and the content displayed by the display unit 30 will be provided in the following embodiments. Figure 3 (A) Figure 3 (B) Figure 4 (A) Figure 4 (B) Figure 5 Please provide an explanation.

[0049] Figure 6 This is a flowchart illustrating the steps of using the image inspection method of the X-ray inspection apparatus 1 to which the present disclosure is applied. The process is only described in summary in this application example; details are provided in the following embodiments.

[0050] In the image inspection method of the X-ray inspection apparatus 1 using this application example, as described above, firstly, the image of the good product generated by the three-dimensional data generation unit 21 is displayed on the display unit 30, and then the image of the defective parts obtained from the defect database 23 is also displayed on the display unit 30. Next, an arbitrary point is selected as the transfer destination from the image of the good product displayed on the display unit 30, and furthermore, one or more images of defective parts are selected from the multiple images of defective parts displayed on the display unit 30. When the transfer destination is determined, the image of the selected defective part is transferred to that transfer destination. Thus, the image of the good product is processed to generate a transfer image containing simulated defective parts. Next, the transfer image is displayed on the display unit 30, and a predetermined characteristic quantity is measured in the measurement unit 25. Finally, the measurement result is displayed on the display unit 30.

[0051] As described above, according to the image inspection method of the X-ray inspection apparatus 1 using this application example, even if no defects occur during the manufacturing process of the object being inspected and data on the defective parts of the object cannot be obtained from the three-dimensional data generation unit 21, a transfer image containing simulated defective parts can be generated by the transfer unit 24. Furthermore, by measuring the transfer image in the measurement unit 25, the difference between good and defective products can be visually and numerically determined. This transfer image can also be provided as a criterion for determining the difference between good and defective products.

[0052] [Example]

[0053] Hereinafter, the image inspection method using the X-ray inspection apparatus 1 according to the embodiments of the present disclosure will be described in more detail with the help of the accompanying drawings (including the drawings temporarily described in the above application examples). Furthermore, the X-ray inspection apparatus of the present invention is not limited to the following structure. In addition, X-ray inspection apparatus 1 is illustrated as an example of an image inspection apparatus in the embodiments, but it is not limited thereto, and other types of apparatus may also be used.

[0054] <Device Structure>

[0055] Here, return Figure 1 The X-ray inspection apparatus 1 of this embodiment has the same structure as the X-ray inspection apparatus 1 described in the application examples; therefore, detailed descriptions of the contents already described in the application examples are omitted. Furthermore, in this specification, the same reference numerals are used to describe the same constituent elements.

[0056] The 3D data generation unit 21 can retrieve images of good products previously stored in the database 22. Therefore, even if no object to be inspected is placed on the worktable 11, the transfer unit 24 can transfer images of defective areas onto images of good products on the object to be inspected. Thus, the 3D data generation unit 21 can generate transfer images as a function independent of the imaging function in the X-ray inspection apparatus 1. Furthermore, for example, as... Figure 2 As shown, in the defect database 23, images of multiple defective parts obtained in past inspections are pre-databaseed. Each image of a defective part is associated with the type of the object being inspected, the type of defect, and the defect stage based on a numerical value representing the degree of defect. Furthermore, these four items—image of the defective part, type of component, type of defect, and numerical defect stage—are grouped together and assigned an ID. The display unit 30 can also display information for these four items. Additionally, filtering and sorting functions allow retrieval of one of these four items. In the case of an object being inspected, such as a circuit board with soldered electronic components, types of defects include, for example, non-wetting and voids (in...). Figure 2 The image shown is an example of a defective pin 51 at a solder joint (details will be explained below). The numerical defect stage is determined based on past measurement results obtained by the measurement unit 25 or on past visual assessments. Detailed information related to the image of the defective area can also be provided to manage these four items as a set. Figure 2 The management method shown is equivalent to the image management method for defective parts in this disclosure.

[0057] <Image inspection methods using X-ray inspection equipment>

[0058] The following is based on Figure 3 (A) Figure 3 (B) Figure 4 (A) Figure 4 (B) Figure 5 The flow of an example of the content displayed on the display unit 30 in the image inspection method using the X-ray inspection apparatus 1 of this embodiment will be described. The content displayed on the display unit 30 is based on the calculation processing performed by the arithmetic unit 20. Furthermore, as an example, the object of inspection is assumed to be an electronic component on a circuit board, namely, a plurality of pins 50 in a chip 40.

[0059] Figure 3(A) is an example of the content displayed on the display unit 30 when the 3D data generation unit 21 generates an image of a good chip 40. The display unit 30 displays tomographic images of a chip 40 viewed from the XY direction (top view), the YZ direction (side view), and the XZ direction (front or rear view). To clearly indicate the viewing direction, labels such as "XY," "YZ," and "XZ" can be added to each image. Furthermore, in the case of a good chip, the pins 50 are circular with high roundness when viewed from the XY direction, and elliptical when viewed from the YZ and XZ directions. Additionally, the display unit 30 displays buttons such as "Transfer Setting," "Transfer Execution," and "Measurement," indicating commands to the calculation unit 20. When the user presses a button, the calculation unit 20 begins processing.

[0060] Figure 3 (B) is an example of what the display unit 30 displays when images of multiple defective parts (hereinafter referred to as "defective pins 51") are obtained from the defect database 23. The images of the multiple defective pins 51 are displayed in a list format by the user pressing the "Transfer Setting" button. Regarding the defective pins 51, images viewed from the XY, YZ, and XZ directions are also displayed as a set of tomographic images. Furthermore, the defective pins 51 may be labeled with markers such as "XY", "YZ", and "XZ". In this embodiment, a defective pin 51 refers to a pin whose shape, for example, is disrupted (e.g., a perfect circle or an ellipse).

[0061] Figure 4 (A) is an example of what the display unit 30 displays when selecting a portion of the image of the chip 40 as the transfer destination and an image of the defective pin 51 as the transfer source. When determining the transfer destination and transfer source, from... Figure 3 Select any point from the image of chip 40 displayed in (A), from Figure 3 Select one image of a faulty pin 51 from the multiple images of faulty pins 51 shown in (B). The selection locations are, for example, as shown in... Figure 4As shown in (A), black dots are used to indicate the location. Furthermore, when any point is selected from the image of chip 40, the transfer destination is determined based on that point. Specifically, regarding the pins 50 of the transfer destination, the boundaries in the XY directions are automatically determined through binarization, or, based on manually set regions, the boundaries within that region are automatically determined through binarization. The boundaries in the Z direction are the two ends of pin 50, which are automatically set by capturing the positions of the two ends of pin 50. By automatically setting the Z-direction boundaries, the features of chip 40 and pins 50 are captured to perform image processing on the image of chip 40. Thus, the portion of the image of chip 40 that becomes the transfer destination and the image of the defective pins 51 that become the transfer source corresponding to the transfer destination are determined respectively. Furthermore, in the image of chip 40, a region can be selected that includes images of multiple pins 50, and images of multiple defective pins 51 can be selected corresponding to each of these multiple pins 50. Here, the transfer destination corresponds to the pins 50, which are the connection points between electronic components and the circuit board.

[0062] Figure 4 (B) is an example of what the display unit 30 displays when transferring the image of the faulty pin 51 to a portion of the image on the chip 40. The transfer unit 24 will... (The sentence is incomplete and requires more context to translate accurately.) Figure 4 The image of the defective pin 51 selected in (A) is transferred to a portion of the image of the chip 40. Through this transfer, pin 50 in the chip 40 is replaced with the defective pin 51 (hereinafter, the pin 50 replaced by this transfer will be referred to as "transfer pin 52"), generating a transfer image. During transfer, the images of the transfer destination and the transfer source often differ in size; therefore, adjustments can be made by enlarging or reducing the size of the image of the defective pin 51 that becomes the transfer source to make them the same size. Furthermore, in Figure 4 In (A), for example, when a portion of the image of chip 40 is selected as viewed from the XY direction, this region is determined not only in the XY direction, but also in the YZ and XZ directions, therefore in Figure 4 In (B), when generating the transfer image, the transfer pin 52 is displayed in all directions, including the XY, YZ, and XZ directions (i.e., if the transfer pin 52 is displayed when viewed from the XY direction, the transfer pin 52 viewed from the YZ and XZ directions corresponding to that transfer pin 52 is also displayed). The transfer pin 52 is, for example, as shown in... Figure 4 As shown in (B), it is displayed surrounded by a box. Additionally, as mentioned above, it can also be displayed in... Figure 4In (A), a portion of the image of chip 40 is selected at multiple locations, and images of multiple defective pins 51 are selected corresponding to these multiple locations. Therefore, in this case, multiple defective pins 51 can be transferred to obtain more images of transfer pins 52 (i.e., images of defective locations).

[0063] Figure 5 This is an example of what the display unit 30 displays when measuring a transferred image. The transferred image has pins 50 and transfer pins 52. By having the user press the "Measure" button, the measurement unit 25 measures specified characteristic quantities in all pins 50 and transfer pins 52 on the transferred image, as well as in all pins 50 and transfer pins 52 on other chips 40 that are not part of the transferred image. When the measurement is performed, the display unit 30 displays the measurement result 60. The measurement result 60 is, for example, as shown below. Figure 5 As shown, this is displayed using a histogram with the measured values ​​on the horizontal axis and frequency on the vertical axis. In this case, based on the measured values ​​of specified characteristic quantities in the newly generated transfer image and images of other chips 40, a histogram of the measured values ​​of the specified characteristic quantities of the object being measured in the process can be displayed. It can also display a histogram of the measured values ​​of all pins 50 and transfer pins 52 in the chip 40 involved in the transfer image displayed at that moment. Furthermore, the specified characteristic quantities refer, for example, to the roundness and area of ​​the surfaces of pins 50 and transfer pins 52. Pins 50 and transfer pins 52 are roughly classified as good or defective based on specified thresholds in these measured values.

[0064] <Flowchart>

[0065] Here, return to Figure 6 The following explanation uses... Figure 6 The steps of the image inspection method using the X-ray inspection apparatus 1 of this embodiment will be described in detail. In this flowchart, firstly, the three-dimensional data generation unit 21 generates images of the chip 40 when viewed from the XY, YZ, and XZ directions, and the display unit 30 displays these images of the chip 40 as tomographic images (step S101). Here, step S101 corresponds to the image display step in this disclosure and the step in this embodiment. Figure 3 (A). Furthermore, the display unit 30 also displays images of multiple defective pins 51 obtained from the defect database 23 when viewed from the XY, YZ, and XZ directions alongside the image of the chip 40 (step S102). Here, step S102 corresponds to the defective image display step in this disclosure and the step in this embodiment. Figure 3(B). Next, an arbitrary point is selected from the image of the chip 40 displayed on the display unit 30, and corresponding to this arbitrary point, one or more images of defective pins 51 are selected from the images of a plurality of defective pins 51. When an arbitrary point is selected from the image of the chip 40, the boundaries in the XY direction and the Z direction are determined based on this arbitrary point, thereby determining the transfer destination (step S103). Here, step S103 is equivalent to the selection step in this disclosure and the step in this embodiment. Figure 4 (A). If a transfer destination is determined, the image of the selected defective pin 51 is transferred to that transfer destination. Thus, the image of chip 40 is processed to generate a transfer image containing the transfer pin 52 as a simulated defective area (step S104). Here, step S104 corresponds to the transfer step in this disclosure and the step in this embodiment. Figure 4 (B). The display unit 30 displays the transfer image (step S105), and the measurement unit 25 measures, for example, the roundness and area of ​​the surfaces of all pins 50 and transfer pins 52 on the transfer image and other chips 40 that are not transfer images (step S106). Here, step S105 corresponds to the transfer image display step in this disclosure and the step in this embodiment. Figure 5 Furthermore, step S106 corresponds to the measurement step in this disclosure and the step in this embodiment. Figure 5 The measurement result 60 in step S106 is displayed on the display unit 30, for example, as a histogram (step S107). Here, step S107 corresponds to the measurement result display step in this disclosure and the step in this embodiment. Figure 5 .

[0066] <Postscript 1>

[0067] An image inspection method is provided, which uses an image inspection device (1) to distinguish between good products (50) and defective products (51, 52) in an object (40) to be inspected, the image inspection method having:

[0068] Image display step (step S101): Display an image of the good product;

[0069] The defective image display step (step S102) displays images of multiple defective parts (51) in an inspection object of the same type as the good product;

[0070] In the selection step (step S103), the transfer destination in the image of the good product and one or more defective parts corresponding to the transfer destination are selected from the images of the plurality of defective parts; and

[0071] In the transfer step (step S104), the image of one or more of the defective parts selected in the selection step is transferred to the transfer destination in the image of the good product to generate a transfer image.

[0072] <Appendix 2>

[0073] An image inspection device (1) is capable of distinguishing between good products (50) and defective products (51, 52) in an object (40) to be inspected. The image inspection device has the following features:

[0074] Database (22), which stores data of images of the good products;

[0075] The defect database (23) stores data of images of multiple defective parts (51) in the same type of inspection object as the good product;

[0076] The transfer unit (24) transfers images of one or more defective parts corresponding to the transfer destination from the images of the plurality of defective parts obtained from the defect database to the transfer destination in the image of the good product obtained from the database, thereby generating a transfer image; and

[0077] The display unit (30) displays an image of the good product obtained from the database and images of the plurality of defective parts obtained from the defective database.

Claims

1. An image inspection method, which uses an image inspection device to distinguish between good and defective products in an inspection object, the image inspection method comprising: The image display step displays a screen that simultaneously shows a good product image display area and a defective product image display area. The good product image display area displays an image of the good product, and the defective product image display area displays images of multiple defective parts of an inspection object of the same type as the good product. In the selection step, select the transfer destination in the image of the good product and one or more defective parts in the images of the multiple defective parts that correspond to the transfer destination; as well as In the transfer step, the image of one or more defective areas selected in the selection step is transferred to the transfer destination in the image of the good product, thereby generating a transfer image. The defective image display area displays images of the plurality of defective parts that are filtered or arranged based on the type of the inspected object, the type of defect, or the degree of defect.

2. The image inspection method according to claim 1, characterized in that, The image inspection method also has the following characteristics: The transfer image display step displays the transfer image; The measurement step involves measuring a specified feature quantity in the transferred image; as well as The measurement result display step displays the measurement results obtained in the measurement step, or displays the measurement results obtained in the measurement step and the measurement results of the specified characteristic quantities in other inspected objects.

3. The image inspection method according to claim 1 or 2, characterized in that, In the image display step and the defective image display step, the image of the good product and the images of the multiple defective parts are displayed as tomographic images in the XYZ directions, respectively.

4. The image inspection method according to claim 1 or 2, characterized in that, In the transfer step, if the size of the transfer destination is different from the size of the image of the defective area, the size of the image of the defective area can be adjusted so that their respective sizes are the same.

5. The image inspection method according to claim 2, characterized in that, In the measurement result display step, the measurement results are displayed as a histogram.

6. The image inspection method according to claim 1 or 2, characterized in that, The object to be inspected is a circuit board containing electronic components. The transfer destination in the image of the good product is the connection between the electronic component and the circuit board. The XY direction boundaries of the connecting portion in the image of the good product are automatically determined through binarization.

7. The image inspection method according to claim 6, characterized in that, The Z-direction boundary of the connecting part is the two ends of the connecting part, which is automatically set by capturing the positions of the two ends of the connecting part.

8. A method for managing images of defective areas, which is a method for managing images of multiple defective areas in the image inspection method of claim 2, characterized in that, Images of the aforementioned defective areas obtained during past inspections were databased. Each of the images of the plurality of defective areas is associated with the type of the object being inspected, the type of defect, and a defect stage based on a numerical value representing the degree of defect. The defect stage based on the numerical value is determined based on measurement results previously measured in the measurement steps or based on results previously determined by visual inspection.

9. An image inspection apparatus capable of distinguishing between good and defective products among inspected objects, said image inspection apparatus comprising: A database that stores data on images of the good products; A defect database, which stores data of images of multiple defective parts in inspection objects of the same type as the good products; The transfer unit transfers images of one or more defective parts corresponding to the transfer destination from the images of the plurality of defective parts obtained from the defective database to the transfer destination of the good product image obtained from the database, thereby generating a transfer image; as well as The display unit simultaneously shows a good product image display area and a defective product image display area. The good product image display area shows images of the good products obtained from the database, and the defective product image display area shows images of products previously identified as defective, obtained from the defective product database. The defective image display area displays images of the plurality of defective parts that are filtered or arranged based on the type of the inspected object, the type of defect, or the degree of defect.

10. The image inspection apparatus according to claim 9, characterized in that, The image inspection device also includes a measuring unit for measuring specified feature quantities in the transferred image. The display unit also displays the transfer image obtained in the transfer unit, the measurement results obtained in the measurement unit, and the measurement results of the specified characteristic quantities in other objects under inspection.

11. The image inspection apparatus according to claim 9 or 10, characterized in that, In the display unit, the images of the good products and the images of the plurality of defective parts are displayed as tomographic images in the XYZ directions.

12. The image inspection apparatus according to claim 9 or 10, characterized in that, In the transfer unit, if the size of the transfer destination is different from the size of the image of the defective area, the size of the image of the defective area can be adjusted so that their respective sizes are the same.

13. The image inspection apparatus according to claim 10, characterized in that, The measurement results are displayed in a histogram in the display unit.

14. The image inspection apparatus according to claim 9 or 10, characterized in that, The object to be inspected is a circuit board containing electronic components. The transfer destination in the image of the good product is the connection between the electronic component and the circuit board. The XY direction boundaries of the connecting portion in the image of the good product are automatically determined through binarization.

15. The image inspection apparatus according to claim 14, characterized in that, The Z-direction boundary of the connecting part is the two ends of the connecting part, which is automatically set by capturing the positions of the two ends of the connecting part.

16. The image inspection apparatus according to claim 10, characterized in that, The defect database contains images of the various defective areas obtained from past inspections. Each of the images of the plurality of defective areas is associated with the type of the object being inspected, the type of defect, and a defect stage based on a numerical value representing the degree of defect. The defect stage based on the numerical value is determined based on measurement results previously measured in the measurement unit or results previously determined by visual inspection.

Images