Additional inspection device and inspection system

Abstract

An additional inspection device (12), which is added to an inspection device (11) for inspecting the appearance of a printed wiring board, is provided with: a receiving unit (31) that receives a defect image indicating a defect of the printed wiring board detected by the inspection device (11); an image processing unit (33) that performs image processing on the defect image and acquires a processed image; a re-inspection unit (34) that re-inspects, on the basis of the processed image, whether or not the defect image shows a defect; and an output unit (35) that outputs the defect image to a defect checking device (13) for a worker to check the defect image when it is determined by the re-inspection unit (34) that the defect image shows a defect.

Description

Technical Field

[0001] The present invention relates to a technology for inspecting the appearance of a printed wiring board.

[0002] [Reference to related applications]

[0003] This application claims priority from Japanese patent application JP2024-22633 filed on February 19, 2024, and incorporates the entire contents of that application into this application. Background Art

[0004] In the past, in the manufacture of printed wiring boards, inspection devices that detect defects by photographing the printed wiring boards were used. As such appearance inspection devices, there are known intermediate inspection devices called AOI (Automated Optical Inspection) and final appearance inspection devices called AVI (Automated final Visual Inspection). In these appearance inspection devices, each part divided by material and function on the printed wiring board is inspected. For example, in the inspection device of Japanese Patent Publication No. 2015-143656, defects are detected in the plating part and the screen part of the printed wiring board based on color captured images, and defects are detected in the solder resist part based on black and white images. In addition, examples are given of defects in which the material of the screen part adheres to the plating part and defects in which tiny dust particles adhere to the solder resist part.

[0005] However, visual inspection devices sometimes detect defects that are not actually defects, known as "false reports" or "pseudo-defects." Therefore, images of defects detected by the visual inspection device are sent to a defect verification device (so-called inspection device). The defect verification device displays the defect image and a main image without defects side by side. The operator observes these images to determine whether the defect image is a false report. If the defect image indicates a defect, the printed wiring board is discarded or repaired.

[0006] False reports can arise from variations in the printed wiring board manufacturing process or from lowering inspection standards to prevent missed detection of genuine defects. For example, if variations in the manufacturing process cause differences in the color or shape of the printed wiring board, or a portion thereof, these differences can be detected as defects by visual inspection equipment. Furthermore, if color differences can reliably detect solder resist peeling, foreign matter that does not affect the board's functionality can be detected as defects with a certain degree of probability. Furthermore, defects that do not affect the printed wiring board's functionality can be determined as false reports through visual reconfirmation by the operator.

[0007] When many false reports are generated, the number of images that operators must judge as true or false increases. As a result, the rate and number of human errors increase, and the possibility of falsely judging real defects as false reports increases. Summary of the Invention

[0008] An object of the present invention is to reduce the number of defect images that an operator needs to check using a defect checking device.

[0009] Embodiment 1 of the present invention is an additional inspection device, which is added to an inspection device for inspecting the appearance of a printed wiring substrate, and comprises: a receiving unit for receiving a defect image showing a defect of the printed wiring substrate detected by the inspection device; an image processing unit for performing image processing on the defect image and obtaining a processed image; a re-inspection unit for re-inspecting whether the defect image shows a defect based on the processed image; and an output unit for outputting the defect image to a defect confirmation device for an operator to confirm the defect image when the re-inspection unit determines that the defect image shows a defect.

[0010] According to the present invention, the number of defect images to be checked by an operator can be reduced using the defect checking device.

[0011] According to a second embodiment of the present invention, in the additional inspection device of the first embodiment, the defect image is an image obtained by cutting out a defect and its surrounding area from an image of the printed wiring board.

[0012] In embodiment 3 of the present invention, in the additional inspection device of embodiment 1 (it may also be embodiment 1 or 2), the receiving unit receives the defect image and inspection type from the inspection device, the inspection type is information indicating the type of inspection performed by the inspection device when the defect is detected, the image processing unit performs image processing selected according to the inspection type on the defect image, and the re-inspection unit re-inspects the defect image using a method selected according to the inspection type.

[0013] In embodiment 4 of the present invention, in the additional inspection device of embodiment 1 (or any one of embodiments 1 to 3), the defect image received by the receiving unit is determined to be at least a part of an image showing a defect based on whether the object image that becomes the inspection object in the inspection device meets the inspection criteria.

[0014] According to a fifth embodiment of the present invention, in the additional inspection device of the fourth embodiment (or any one of the first to fourth embodiments), the reinspection unit determines whether the defect image shows a defect based on whether the processed image satisfies an inspection criterion.

[0015] In embodiment 6 of the present invention, in the additional inspection device of embodiment 1 (or any one of embodiments 1 to 5), the image processing unit calculates a threshold value based on the pixel values of the defect and the surrounding area in the defect image, and obtains the processed image by binarizing the defect image using the threshold value. The re-inspection unit determines whether the defect image shows a defect by comparing the processed image with a reference binary image.

[0016] In embodiment 7 of the present invention, in the additional inspection device of embodiment 1 (or any one of embodiments 1 to 5), the image processing unit obtains a processed image obtained by extracting a region of a certain type from the defect and the surrounding area in the defect image, and the re-inspection unit calculates the deviation degree of the deviation of the pixel values in the region of the certain type, and determines whether the defect image shows a defect by comparing the deviation degree with a specified threshold value.

[0017] In embodiment 8 of the present invention, in the additional inspection device of embodiment 1 (or any one of embodiments 1 to 5), the image processing unit obtains a processed image obtained by aligning the defect and the surrounding area in the defect image with respect to a reference area image, and the re-inspection unit determines whether the defect image shows a defect by comparing the processed image with the reference area image.

[0018] Embodiment 9 of the present invention is an inspection system that inspects the appearance of a printed wiring board, wherein the inspection system comprises: an additional inspection device described in any one of embodiments 1 to 8; and the inspection device outputs a defect image showing defects of the printed wiring board to the additional inspection device.

[0019] The above-mentioned objects and other objects, features, aspects and advantages will become more apparent from the following detailed description of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a diagram showing the configuration of an inspection system including an additional inspection device.

[0021] Figure 2 This is a diagram showing the configuration of a computer serving as an additional inspection device.

[0022] Figure 3 This is a diagram showing the operation flow of the inspection device.

[0023] Figure 4 This is a diagram showing the operation flow of the additional inspection device.

[0024] Figure 5Ais a diagram showing a defect image.

[0025] Figure 5B A diagram showing a portion of a reference binary image.

[0026] Figure 5C It is a diagram representing a portion of the target binary image.

[0027] Figure 5D It is a diagram showing the defect area.

[0028] Figure 5E : is a diagram showing a target binary image which is a processed image.

[0029] Figure 6A is a diagram showing a defect image.

[0030] Figure 6B A diagram showing a portion of a reference image.

[0031] Figure 6C A diagram showing a target processed image as a processed image.

[0032] Figure 6D A diagram showing a reference processed image.

[0033] Figure 6E It is a diagram showing the defect area.

[0034] Figure 7 It is a graph showing the distribution of pixel values.

[0035] Figure 8A is a diagram showing a defect image.

[0036] Figure 8B A diagram showing a portion of a reference image.

[0037] Figure 8C It is a diagram showing the defect area.

[0038] Figure 8D This is a diagram showing the processed image.

[0039] Description of reference numerals:

[0040] 1. Check the system

[0041] 11Inspection device

[0042] 12Additional inspection equipment

[0043] 13Defect confirmation device

[0044] 31 Receiving Department

[0045] 33 Image Processing Unit

[0046] 34 Re-inspection Department

[0047] 35 output unit

[0048] 81 defect image data

[0049] 82 Inspection Type

[0050] 811, 821, 831 defect images

[0051] 815, 836 processed images DETAILED DESCRIPTION

[0052] Figure 1 It is a diagram showing the structure of an inspection system 1 including an additional inspection device 12 of one embodiment of the present invention. The inspection system 1 inspects the appearance of a printed wiring board as an object. "Inspecting the appearance" means obtaining an image of the printed wiring board and determining whether the printed wiring board has defects based on the image. The inspection system 1 has an appearance inspection device (hereinafter referred to as "inspection device") 11 that photographs the printed wiring board and inspects its appearance, an additional inspection device 12, and a defect confirmation device 13, which are connected via a communication network 14. The inspection device 11 can be an intermediate inspection device called AOI or a final appearance inspection device called AVI. The printed wiring board is a substrate inspected by these devices, and is a printed wiring board in the manufacturing process or completed.

[0053] The inspection device 11 includes an image acquisition unit 21 and an inspection unit 22. The image acquisition unit 21 includes a camera unit and a moving mechanism. The camera unit includes a so-called line sensor in which a plurality of camera elements are arranged in a straight line. The moving mechanism moves the printed wiring board relative to the camera unit. While the printed wiring board moves relative to the camera unit, the camera unit repeatedly acquires line images, thereby acquiring a two-dimensional image of the printed wiring board. The camera unit that acquires the image of the printed wiring board can be modified in various ways. For example, the camera unit may also acquire a two-dimensional image of the printed wiring board using an image sensor in which camera elements are arranged two-dimensionally.

[0054] The inspection unit 22 is implemented, for example, by a computer and / or a dedicated circuit. The inspection unit 22 detects defects from an image of the printed wiring board and acquires an image containing the defect and its vicinity as a defect image. The defect image at this stage is an image showing the defect determined to be a defect by the inspection device 11. Thus, "is a defect" means that the defect is determined to be a defect based on the determination. The defect image data, i.e., defect image data 81, is stored in the storage unit 221 of the inspection unit 22. The inspection unit 22 performs various inspections, i.e., defect detection processing, on the image of the printed wiring board.

[0055] The inspection type 82, information indicating the type of inspection performed when acquiring the defect image data 81, is stored in association with the defect image data 81 in the storage unit 221. The term "inspection type" typically refers to the type of inspection algorithm. While it generally corresponds to individual inspection algorithms, a general concept encompassing multiple inspection algorithms is considered a single inspection type. Concepts other than inspection algorithms, such as specific functional elements using machine learning or methods for inspecting materials, may also be considered a single inspection type.

[0056] The storage unit 221 may be provided outside the inspection unit 22. For example, it may be provided outside the housing of the inspection apparatus 11, and a computer or NAS (Network Attached Storage) may function as the storage unit 221.

[0057] The inspection unit 22 is usually disposed within the housing of the inspection device 11 . The inspection unit 22 may be disposed outside the housing of the inspection device 11 or may be connected to the main body of the inspection device 11 via the communication network 14 .

[0058] In the following description, defect image data is simply referred to as "defect image." Similarly, other image data is sometimes simply referred to as the image name. Furthermore, in practice, processing performed on image data is sometimes simply described as processing performed on the image. For example, binarization of defect image data is sometimes simply described as binarization of the defect image.

[0059] The additional inspection device 12 performs an additional inspection on the defect image obtained by the inspection device 11 to determine whether the defect image shows a defect or a false report of a false defect. The additional inspection device 12 includes a receiving unit 31, a storage unit 32, an image processing unit 33, a re-inspection unit 34, and an output unit 35. In the following description, the additional inspection performed by the additional inspection device 12 is referred to as "re-inspection". The additional inspection device 12 is a device added to the inspection device 11, but "addition" here means that the inspection device 11 independently has the function of outputting defect images, and the additional inspection device 12 has the function of performing additional inspections on defect images. That is, if the inspection device 11 functions as a device that performs inspections alone, the additional inspection device 12 is not physically required to be added later. For example, the additional inspection device 12 can also be set up at the same time as the inspection device 11 is set up in a specified place.

[0060] Furthermore, the additional inspection device 12 may be installed inside the housing of the inspection device 11 without being connected to the communication network 14, or may be directly connected to the inspection device 11. Conversely, the additional inspection device 12 may be installed remotely from the inspection device 11 via the communication network 14. Furthermore, the computer that implements the inspection unit 22 of the inspection device 11 may also implement the additional inspection device 12. Thus, the additional inspection device 12 may be installed in various configurations as long as it is added as a function to the inspection device 11.

[0061] The additional inspection device 12 re-inspects the defect image to determine if it is a false alarm. For example, the additional inspection device 12 executes a program on a computer to re-inspect the defect image. Only if the re-inspection determines that the defect image is not a false alarm does the additional inspection device 12 transmit the defect image to the defect confirmation device 13. At this stage, the defect image is determined to be defective by both the inspection device 11 and the additional inspection device 12.

[0062] The defect confirmation device 13 is a device for the operator to confirm the defect image when the additional inspection device 12 determines that the defect image is an image showing a defect. The defect confirmation device 13 is also a device implemented by a computer executing a program. In the defect confirmation device 13, the defect image and an image in which there is no defect corresponding to the defect image (hereinafter referred to as the "main image") are displayed on the display unit. The operator visually compares the two images to determine whether the defect image is an image showing a defect or a false report showing a false defect, and inputs the determination result to the defect confirmation device 13 via an input unit such as a mouse and a keyboard. The defect confirmation device 13 is not limited to the above-mentioned form as long as it is a device that allows the operator to confirm defects.

[0063] Defects in printed wiring board inspections include "functional defects" that affect the functionality of the printed wiring board, and "appearance defects" that do not affect the functionality but do not meet the standard appearance. Defects in the present invention can be either functional or appearance defects. However, defect images determined as false by the defect confirmation device 13 primarily depict defects that do not affect functionality. Therefore, defects in the present invention can be construed as limited to functional defects.

[0064] Figure 2This diagram shows the configuration of a computer serving as the additional inspection device 12. The additional inspection device 12 has a typical computer system configuration, including a CPU 301, a GPU 302, a ROM 303, a RAM 304, a fixed disk 305, a display 306, an input unit 307, a reader 308, a communication unit 309, and a bus 30. The CPU 301 performs various computations. The GPU 302 performs various computations related to image processing. The ROM 303 stores basic programs. The RAM 304 and the fixed disk 305 store various information. The display 306 displays various information, including images. The input unit 307 includes a keyboard 307a and a mouse 307b for receiving input from the operator. The reader 308 reads information from a computer-readable storage medium 9, such as an optical disc, a magnetic disc, a magneto-optical disc, or a memory card. The communication unit 309 transmits and receives signals with other components of the additional inspection device 12 and external devices. The bus 30 is a signal circuit that connects the CPU 301 , the GPU 302 , the ROM 303 , the RAM 304 , the fixed disk 305 , the display 306 , the input unit 307 , the reader 308 , and the communication unit 309 .

[0065] In the additional inspection device 12, the program 91 is read from the storage medium 9 via the reading device 308 and stored in the fixed disk 305. The program 91 may also be stored in the fixed disk 305 via a communication network. The CPU 301 and the GPU 302 perform calculations according to the program 91 while using the RAM 304 and the fixed disk 305. The CPU 301 and the GPU 302 function as a calculation unit in the additional inspection device 12. In addition to the CPU 301 and the GPU 302, other structures that function as a calculation unit may also be used.

[0066] In the additional inspection device 12, the computer performs calculations and the like according to the program 91 to realize Figure 1 The functional structure shown in FIG. Specifically, the CPU 301, GPU 302, ROM 303, RAM 304, fixed disk 305, communication unit 309, and their peripheral components implement the receiving unit 31, storage unit 32, image processing unit 33, review unit 34, and output unit 35. All or part of these functions may be implemented by dedicated circuits, or by separate programs. Alternatively, these functions may be implemented by multiple computers.

[0067] The receiving unit 31 receives the defect image (more precisely, the defect image data 81) from the inspection device 11 and stores it in the storage unit 32. For example, the receiving unit 31 is a communication unit 309 such as an interface for communicating with the communication network 14 and a functional structure for controlling the communication unit 309. The defect image is stored in the storage unit 32. Figure 2The receiving unit 31 receives the defect image data and the inspection type from the inspection device 11. The inspection type is information indicating the type of inspection when the defect is detected. The inspection type is also stored in the fixed disk 305.

[0068] The image processing unit 33 performs image processing on the defect image and obtains a processed image. The re-inspection unit 34 determines whether the defect image is a false report based on the processed image and the inspection type. If the re-inspection determines that the defect image shows a defect, the defect image data and inspection type are transmitted to the defect confirmation device 13 via the output unit 35 and the communication network 14. If the re-inspection determines that the defect image does not show a defect, the defect image data and inspection type are not transmitted to the defect confirmation device 13. The output unit 35 is a functional structure that includes the communication unit 309, such as an interface for communicating with the communication network 14, and controls the communication unit 309.

[0069] Next, refer to Figure 3 and Figure 4 A specific example of the operation of the inspection device 11 and the additional inspection device 12 will be described.

[0070] First, the printed wiring board is moved into the inspection device 11, and the image acquisition unit 21 acquires a two-dimensional image of the entire printed wiring board (hereinafter referred to as the "substrate image") (step S11). The substrate image can be a color image or a monochrome image, or both images can be acquired. Next, the inspection unit 22 performs a process for detecting defects on the printed wiring board based on the substrate image (step S12). The inspection unit 22 performs various processes on the substrate image according to the type of defect to be detected. In the inspection device 11 of a typical example, the defects in the substrate image (more precisely, the areas showing defects) are determined based on whether each area of the substrate image that becomes the inspection unit meets the inspection benchmark.

[0071] Figure 5A This figure shows an example of a defect image, which is an image showing a defect detected by the inspection unit 22. In defect image 811, the area indicated by reference numeral 813 within the circular bright area 812 is slightly darker, indicating that area 813 is detected as a defect. Such defects are detected due to a change in the overall color of the printed wiring board or a difference in the color of a portion of the printed wiring board.

[0072] The operation of the inspection unit 22 upon detecting a defect image 811 is as follows. First, an image of the substrate image's inspection target region (hereinafter referred to as the "target image") and an image of the target region of the main image (hereinafter referred to as the "reference image") are prepared. The inspection target region is wider than the final defect image 811. Next, a reference binary image is obtained by binarizing the reference image using a predetermined threshold. Figure 5B It only represents the reference binary image with Figure 5A In the figure of the corresponding area, the outline is indicated by a dotted line to express an actually large image (the same applies hereinafter).

[0073] Next, the object image is also binarized using the same threshold value to obtain an object binary image. Figure 5C It only represents the binary image of the object Figure 5A The inspection unit 22 obtains a differential image between the object binary image and the reference binary image, such as Figure 5D As shown in FIG. 8 , the difference area is detected as the defect area 814. Then, the inspection unit 22 cuts out the defect area 814 and the surrounding area in the target image, and obtains Figure 5A As mentioned above, in 5A to 5D In the example shown, defects are detected using the aforementioned "comparison check" (algorithm).

[0074] In practice, when multiple types of inspections are performed on each area of the substrate image and more than one defect is detected, Figure 1 As shown, the defect image data 81 and the inspection type 82 are associated and stored in the storage unit 221. The inspection type 82 is information indicating the type of inspection performed by the inspection device 11 when the defect image is detected. 5A to 5D In the case of the example, the inspection type is information indicating “comparison inspection.” Then, when the inspection of one substrate image is completed, the combination of the defect image data 81 and the inspection type 82 is output to the additional inspection device 12 (step S13).

[0075] The combination of defect image data and inspection type is received by the receiving unit 31 of the additional inspection device 12 and stored in the storage unit 32 (step S21). Next, the image processing unit 33 performs processing corresponding to the inspection type on the defect image (step S22). For example, 5A to 5D In the example shown, the image processing unit 33 performs the following operations: Figure 5A A threshold value for binarization is determined for the defect image 811, and the threshold value is used to binarize the defect image 811. For example, the K-means method is used to determine the threshold value.

[0076] In the threshold determination based on the K-means method, first, the pixels in the image are divided into two appropriate categories, (1) the average value of each category is calculated, (2) the pixels are reallocated to the category with the pixel value closest to the average value, (3) (1) and (2) are repeated until no more redistribution is done, and (4) the pixel value between the two categories is determined as the threshold. Figure 5E The figure is an example of the processed image after the binarization, that is, the target binary image. Since the threshold is determined based on the defect image 811, the target binary image is suppressed. Figure 5A The dark area 813 is affected.

[0077] Meanwhile, the additional inspection device 12 pre-stores a main image of the printed wiring board, extracts the area corresponding to the defect image 811 from the main image, and performs binarization in the same manner as the inspection device 11 to obtain a reference binary image. Then, as a re-inspection, a difference image is obtained between the target binary image and the reference binary image. Since no difference area exists in the difference image, the defect image 811 is determined to be a false alarm (step S23). Thus, the re-inspection unit 34 re-inspects the defect image based on the processed image of the defect image by the image processing unit 33 to determine whether the defect image has a result.

[0078] Furthermore, the threshold used by the image processing unit 33 to binarize the defect image does not need to use the pixel values of the entire defect image. Instead, it can be determined based on the pixel values of the defect in the defect image (more precisely, the region showing the defect, the same applies hereinafter) and the surrounding region (including the case where the region is the defect image itself). Alternatively, the pixel values of only a portion of the defect image may be used. The image processing unit 33 uses a threshold different from the threshold used in the inspection device 11 for binarization, and the re-inspection unit 34 compares the processed image with the reference binary image to determine whether the defect image shows a defect. This allows for appropriate re-inspection even when the substrate image is acquired in a color different from that of the main image due to slight differences in manufacturing conditions, imaging conditions, etc.

[0079] If the defect image is determined to be a false alarm, the additional inspection device 12 completes processing of the defect image (step S24). On the other hand, if the defect image is also determined to be defective during the re-inspection, the output unit 35 outputs the defect image (data) and the corresponding inspection type to the defect confirmation device 13 (step S25). In the defect confirmation device 13, as described above, the operator visually confirms whether the defect image indicates a defect.

[0080] By providing the additional inspection device 12 in the inspection system 1, the number of defect images that the operator must verify is reduced by the defect confirmation device 13. As a result, the probability of human error due to operator fatigue can be reduced, and it is possible to appropriately prevent printed wiring boards with genuine defects from being used as products.

[0081] Figure 6A FIG. 1 is a diagram showing another example of a defect image. Figure 6A The defect image 821 includes a solder resist region 822, a substantially circular plating region 823, and an outer region 824 outside the printed wiring board. In the defect image 821, a slightly brighter region 825 exists in the solder resist region 822, and the inspection device 11 detects the region 825 as a defect.

[0082] The operation of the inspection unit 22 when acquiring defect image 821 is as follows. First, a target image showing the inspection target area of the substrate image and a reference image showing the area of the main image corresponding to the target image are prepared. As described above, the inspection target area is wider than the final defect image 821. Figure 6B It only indicates the Figure 6A The corresponding area diagram.

[0083] Next, in the target image and the reference image, the design information of the printed wiring board is referred to, and the area outside the solder resist is masked, and the image is obtained as Figure 6C The processed image of the object image shown and the object processed image as Figure 6D Then, by binarizing the difference image between the target processed image and the reference processed image, a reference processed image is obtained. Figure 6E The image shown in FIG. 826 shows the defect area. Further, the inspection unit 22 cuts out the defect area 826 and the surrounding area in the target image and obtains Figure 6A The defect image 821. Figures 6A to 6E In the example shown, defects are detected by the above-mentioned "unevenness inspection" (algorithm).

[0084] When the combination of defect image data and inspection type based on "uneven inspection" is output from the inspection device 11 and received by the additional inspection device 12, the image processing unit 33 of the additional inspection device 12 first performs the same processing as the inspection device 11 on the defect image and the corresponding main image area, and obtains the same Figure 6E The same is true for the defective area 826. Alternatively, the defective area 826 may be stored in advance in the storage unit 221 of the inspection device 11 and sent to the additional inspection device 12 together with the defective image data.

[0085] Next, the reinspection unit 34 acquires the pixel values of the defect region 826 in the defect image and obtains the distribution of the pixel values. Figure 7 It is a graph showing the distribution. The re-inspection unit 34 obtains the standard deviation, which is a value representing the deviation of the pixel values, from the distribution of the pixel values. Then, the standard deviation is compared with a preset threshold value. When the standard deviation is greater than the threshold value, that is, when the deviation of the pixel values in the defective area is large and the color is unstable, it is determined that the defect image is an image showing an uneven defect. In this way, the re-inspection unit 34 re-inspects whether the defect image shows a defect based on the processed image processed by the image processing unit 33. The additional inspection device 12 refers to the image after determining that the defect image is an image showing a defect or a false report. Figure 4 As described.

[0086] Figures 6A to 6E and Figure 7 The processing shown is not limited to defect detection in the solder resist area. Similar processing can also be performed on other specific types of areas, such as copper areas, plating areas, and screen printing areas. Furthermore, the entire defect image need not be the target of area extraction. Generally speaking, the image processing unit 33 obtains a processed image in which a specific type of area is extracted from the defect in the defect image and its surrounding area (including when the area is the defect image itself).

[0087] The re-inspection unit 34 may also determine a value other than the standard deviation as a degree of deviation indicating the deviation of pixel values in the above-specified type of region. The re-inspection unit 34 determines whether the defect image shows a defect by comparing the degree of deviation with a predetermined threshold value.

[0088] Figure 8A 2 is a diagram showing another example of a defective image. Figure 8A The defect image 831 has a solder resist area 832 , a linear copper area 833 , and a screen printing area 834 where screen printing is performed.

[0089] The operation of the inspection unit 22 when acquiring defect image 831 is as follows. First, a target image showing the inspection target area of the substrate image and a reference image showing the area of the main image corresponding to the target image are prepared. As described above, the inspection target area is wider than the final defect image 831. Figure 8B It only indicates the Figure 8A In the inspection device 11, as Figure 8A For the corresponding area, use Figure 8B The area shown, Figure 8A The position of the pattern in Figure 8B The positional shift of the pattern occurs due to deformation and warping of the printed wiring board.

[0090] Next, the inspection unit 22 obtains a difference image between the target image and the reference image, and obtains the difference image by binarizing the difference image. Figure 8C The image shown shows the defect area 835. The difference image can also be obtained by other methods. 5A to 5D Similarly, the inspection unit 22 may obtain a difference image of the binary images after binarizing the target image and the reference image. Figure 8A The defect image 831. However, in Figure 8C In the case of , the defect area 835 appears in a large area, so the defect image is appropriately cut out from the target image with a predetermined size. Figures 8A to 8C In the example shown, defects are detected using the aforementioned "comparison check" (algorithm).

[0091] Next, Figure 8A The operation of the additional inspection device 12 corresponding to the defect image is described. Figure 8A The defect is a defect caused by "position shift". The defect caused by "position shift" is detected, for example, when a detailed inspection is carried out on an area where a solder resist is formed on a fine line pattern. In addition, the defect caused by "position shift" also becomes a cause of detection of various types of defects, so the action of the additional inspection device 12 described below can also be performed on other types of inspections. That is, the re-inspection of the defects detected by the inspection device 11 is not limited to one type of re-inspection, and multiple types of re-inspections can also be performed. In this case, depending on the inspection type, if a "false report" is determined in one of the multiple re-inspections, the re-inspection unit 34 may conclude that it is a "false report", or it may conclude that it is a "false report" only when it is determined as a "false report" in all the re-inspections.

[0092] When re-inspecting defects caused by "positional misalignment," the image processing unit 33 first performs image processing for alignment of the defect image and the corresponding main image area. For example, the defect image is offset relative to the main image, and the total difference in pixel values between the defect image and the main image is repeatedly calculated. The position of the defect image where the total difference in pixel values is minimized is determined. Figure 8D 1 and 2 are diagrams showing a processed image 836 on which the image processing unit 33 has performed image processing for alignment. Figure 8B The pixel position of the upper left vertex corresponds to Figure 8D The pixel position of the upper left vertex.

[0093] The reinspection unit 34 obtains a difference image of the overlapping portion between the processed image 836 and the main image in this state and binarizes the difference image. If a difference region exists in the difference image, the defect image 831 is determined to have a defect. If no difference region exists, the defect image 831 is determined to be a false alarm.

[0094] Furthermore, alignment processing (e.g., image processing that changes the coordinate values of each pixel) between the defect image and the corresponding region of the main image (hereinafter referred to as the "reference region image") can also be performed by processing only the reference region image without processing the defect image. Because the relationship between the defect image and the reference region image in alignment is relative, processing the reference region image is essentially processing the defect image. Processing the defect image in the image processing unit 33 is defined as including processing the reference region image.

[0095] In this case, the process of comparing the processed image with the reference area image by the re-inspection unit 34 to determine whether the defect image shows a defect includes not only the case of using the processed image obtained by directly processing the defect image, but also the case of re-inspecting the defect image by indirectly processing the reference area image. In addition, the entire defect image does not need to be used for alignment. Generally speaking, the processed image after alignment is the defect in the defect image and the surrounding area (including the case where the area is the defect image itself) aligned with the reference area image.

[0096] Figure 5A 、 Figure 6A and Figure 8A The defect image is an image obtained by cutting out a defect and its surrounding area detected from a substrate image showing a printed wiring board. As a result, the processing range during re-inspection can be easily limited to the range of the defect image. In addition, "one defect" means that the number of defects of concern is one, including the case where a plurality of defective areas are regarded as one defect. The form of the defect image can be any other form as long as it is an image showing a defect detected by the inspection device 11. For example, the defect image can also be a combination of a substrate image and coordinates indicating the position of the defect. Alternatively, the object image in the above description, that is, the image of the area in the substrate image used for inspection in the inspection device 11, can be used as the defect image.

[0097] In addition, the defect image is an image showing a defect detected by the inspection device 11, but the defect image does not need to be an image showing only one defect. Since multiple defects are close to each other, multiple defects can also be included in one defect image, that is, including one defect of concern and other defects.

[0098] In the above description, the receiving unit 31 of the additional inspection device 12 receives the defect image and inspection type from the inspection device 11 (step S21). The image processing unit 33 performs image processing selected based on the inspection type on the defect image to obtain a processed image (step S22). Furthermore, the re-inspection unit 34 re-inspects the defect image using the method selected based on the inspection type (step S23). By receiving the inspection type from the inspection device 11, the additional inspection device 12 can quickly perform appropriate re-inspections. However, the re-inspection content can also be selected based on the defect image, regardless of the inspection type.

[0099] For example, the type of defect may be determined from the defect image, and the type of image processing and re-inspection may be selected based on the type of defect determined. Alternatively, the type of image processing and re-inspection may be selected from the defect image based on the functional components of the printed wiring board shown in the defect image (e.g., pads, wiring, through-holes, etc.) and the material of the printed wiring board (e.g., solder resist, screen printing, copper, plating, etc.). In this way, the inspection type may not be sent from the inspection device 11 to the additional inspection device 12. In addition, instead of the inspection type, the type of defect, the functional components of the printed wiring board shown in the defect image, the material of the printed wiring board shown in the defect image, etc. may be sent from the inspection device 11 to the additional inspection device 12 together with the defect image.

[0100] The inspection performed in the inspection device 11 may be an inspection using machine learning, but is preferably performed based on whether the object image that is the inspection object satisfies the inspection criteria. That is, the inspection device 11 performs a so-called "rule-based" inspection. "Based on whether the inspection criteria are met" typically means that the presence or absence of defects is determined by comparing the values derived from the object image (the area of the differential area, the number of pixels of a certain color, etc.) with the specified values. Moreover, at least a portion of the object image that is determined to show a defect is used as a defect image. Through the above-mentioned processing, a large amount of processing is achieved quickly in the inspection device 11.

[0101] The re-inspection of the re-inspection unit 34 of the additional inspection device 12 may also be an inspection based on machine learning, but preferably, it is determined whether the defect image shows a defect based on whether the processed image meets the inspection benchmark. In this way, re-inspection can be performed stably. From the perspective of "re-inspection", it is preferred that, in addition to image processing, the inspection processing performed on the object image in the inspection device 11 and the inspection processing performed on the processed image in the additional inspection device 12 are the same type of inspection (inspection type). In this case, the inspection processing in the inspection device 11 and the inspection processing in the additional inspection device 12 also become rule-based inspections. Of course, as shown in reference Figures 6A to 6E and Figure 7As illustrated, the inspection process in the inspection device 11 and the inspection process in the additional inspection device 12 may be greatly different.

[0102] The above-mentioned examples of inspections by the inspection device 11 and the examples of re-inspections by the additional inspection device 12 are merely a few examples. In reality, various inspections are performed on various parts of the substrate image. For example, various inspections are performed for pattern warping, pattern shorts, pattern disconnections, through-hole anomalies, screen printing anomalies, solder resist peeling from copper, solder resist peeling from the base material, foreign matter on copper, foreign matter on solder resist, foreign matter on the pattern, solder resist unevenness, copper pattern anomalies, and plating area anomalies.

[0103] The configurations in the above-described embodiment and various modifications may be appropriately combined as long as they do not contradict each other.

[0104] While the invention has been described and illustrated in detail, the above description is illustrative and not restrictive, and therefore, various modifications and embodiments are possible without departing from the scope of the invention.

Claims

1. An additional inspection device, which is added to an inspection device for inspecting the appearance of a printed wiring board, wherein: have: a receiving unit that receives a defect image showing a defect of the printed wiring board detected by the inspection device; An image processing unit, performing image processing on the defect image and obtaining a processed image; a re-inspection unit that re-inspects whether the defect image shows a defect based on the processed image; and The output unit outputs the defect image to a defect checking device for a worker to check the defect image, when the reinspection unit determines that the defect image shows a defect.

2. The additional inspection device according to claim 1, wherein: The defect image is an image obtained by cutting out a defect and its surrounding area from an image of the printed wiring board.

3. The additional inspection device according to claim 1, wherein: The receiving unit receives the defect image and an inspection type from the inspection device, the inspection type being information indicating a type of inspection performed by the inspection device when the defect is detected. The image processing unit performs image processing selected according to the inspection type on the defect image. The reinspection section reinspects the defect image using a method selected according to the inspection type.

4. The additional inspection device according to claim 1, wherein: The defect image received by the receiving unit is at least a portion of an image determined to show a defect based on whether the target image serving as an inspection target in the inspection device satisfies an inspection criterion.

5. The additional inspection device according to claim 4, wherein: The reinspection unit determines whether the defect image shows a defect based on whether the processed image satisfies an inspection criterion.

6. The additional inspection device according to claim 1, wherein: The image processing unit obtains a threshold value based on pixel values of the defect and its surrounding area in the defect image, and obtains the processed image by binarizing the defect image using the threshold value. The reinspection unit determines whether the defect image shows a defect by comparing the processed image with a reference binary image.

7. The additional inspection device according to claim 1, wherein: The image processing unit obtains a processed image obtained by extracting a region of a specific type from the defect and its surrounding region in the defect image. The reinspection unit obtains a degree of deviation indicating a deviation of pixel values in the region of the specified type, and determines whether the defect image shows a defect by comparing the degree of deviation with a predetermined threshold value.

8. The additional inspection device according to claim 1, wherein: The image processing unit obtains a processed image obtained by aligning the defect and its surrounding area in the defect image with a reference area image. The reinspection unit determines whether the defect image shows a defect by comparing the processed image with the reference area image.

9. An inspection system for inspecting the appearance of a printed wiring board, wherein: have: The additional inspection device according to any one of claims 1 to 8; as well as The inspection device outputs a defect image showing a defect of the printed wiring board to the additional inspection device.

Images