Substrate production system and image analysis device

Abstract

To detect a defect factor which is difficult to detect only by an appearance inspection.SOLUTION: A production system S for a substrate PX includes: a mounting line 10 including a plurality of machines 11 to 16 that perform a predetermined work on the substrate PX; and an image analysis device 20. The machines 11 to 16 include cameras 43 to 45 for photographing the substrate PX. The image analysis device 20 sequentially takes in images of the substrate PX photographed by the cameras 43 to 45 in a production process of the machines 11 to 16, extracts a difference by comparing the images of the substrate PX photographed in the production process with the images of the other substrates PX photographed in the same production process, and detects a production defect of the substrate PX based on the difference of the extracted images.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for suppressing production defects in substrates. [Background technology]

[0002] The mounting line visually inspects the manufactured boards with a camera or the like to prevent the outflow of defective boards. Patent Document 1 listed below is a document disclosing technology related to the mounting line. [Patent Document 1] Patent Publication No. 2023-100538 DISCLOSURE OF THE INVENTION [Problem to be solved by the invention]

[0003] An object of the present invention is to detect the cause of defects that are difficult to detect by visual inspection alone. [Means for solving the problem]

[0004] (1) The present invention is a circuit board production system comprising: a mounting line equipped with a plurality of machines that perform predetermined operations on the circuit board; and an image analysis device. The machines are equipped with cameras that photograph the circuit board. The image analysis device sequentially captures images of the circuit board photographed by the cameras during the machine's production process, compares the images of the circuit board photographed during the production process with images of other circuit boards photographed during the same production process, extracts differences, and detects production defects of the circuit board based on the extracted image differences. The circuit board production system described in (1) is not limited to the above configuration, and any configuration may be used.

[0005] The configuration (1) can detect slight changes or abnormalities in the production process by comparing images of a board taken during the production process with images of other boards taken during the same production process and extracting the differences. This is expected to enable the detection of defect factors that would be difficult to detect using visual inspection alone.

[0006] (2) In the board production system described in (1), the image analysis device may compare the image of the board captured from the machine with a normal image (reference image) that has been inspected as a non-defective product to extract differences. This configuration has the advantage of being able to detect image differences with high accuracy because the comparison target is a normal board that has been inspected as a non-defective product.

[0007] (3) In the circuit board production system described in (1), the image analysis device may compare images of two continuously produced circuit boards to extract differences. This configuration has the advantage of being able to detect image differences without performing test runs before producing the circuit boards (even if a normal circuit board cannot be obtained).

[0008] (4) In the circuit board production system according to any one of (1) to (3), the machine is a surface mounter that mounts components on the circuit board, and the image analysis device may extract differences by comparing an image of the circuit board near a mounting point with images of other circuit boards taken in the same production process. This configuration narrows the image analysis range (region of interest) to the vicinity of the mounting point, thereby shortening the image analysis time. The mounting point is the point where the component is mounted.

[0009] (5) In the circuit board production system according to any one of (1) to (4), when the image analysis device determines from the image difference that there is a possibility of a production defect in the circuit board, the image analysis device may notify a terminal of an operator monitoring the mounting line of the possibility of a production defect in the circuit board. This configuration leaves it up to the operator to decide whether to continue production on the mounting line, thereby making it possible to avoid stopping the mounting line due to an error in judgment by the image analysis device.

[0010] (6) In the board production system described in (5), the image analysis device may transmit to the mobile terminal, together with a notification that a production defect has occurred in the board, images of the two compared boards in a manner in which the areas where there are differences are distinguished from other areas. The mobile terminal may display the two images transmitted from the image analysis device in a manner in which the areas where there are differences are distinguished from other areas. This configuration makes it easier for an operator to recognize differences in the images displayed on the mobile terminal, allowing the operator to accurately determine board defects.

[0011] (7) In the board production system described in (5) or (6), the mobile terminal may control the stopping of work on the board when the board is determined to be defective. The determination of the production defect may be made by an operator or the mobile terminal. This configuration can prevent defective boards from being sent to a later process.

[0012] (8) In the circuit board production system according to any one of (1) to (7), the image analysis device may sequentially capture images of components mounted on the circuit board in addition to images of the circuit board taken by the camera during the machine's production process. The image of the component taken during the production process may be compared with images of other components taken during the same production process to extract differences, and production defects of the circuit board may be detected based on the extracted image differences. This configuration can detect changes in the component mounting process, etc., and therefore more easily detect defective circuit boards. [Effects of the Invention]

[0013] According to the present invention, it is possible to detect factors of defects that are difficult to detect by visual inspection alone. [Brief explanation of the drawings]

[0014] [Figure 1] Production system configuration diagram [Figure 2] Plan view of surface mounter [Figure 3] Diagram showing the head unit support structure [Figure 4] Block diagram showing the electrical configuration of the surface mounter [Figure 5] Examples of images taken at each production process [Figure 6] Comparison of component bottom images and mounting point images [Figure 7] Diagram showing foreign matter that gets mixed in between parts and pads [Figure 8] Flowchart of time-lapse image analysis processing (primary judgment processing of difference images) [Figure 9A] Diagram showing how images are compared [Figure 9B] Diagram showing how images are compared [Figure 10] Luminance Histogram [Figure 11] Image correction diagram [Figure 12] Diagram showing anomaly detection processing in a production system [Figure 13] Mobile device display image [Figure 14] A diagram showing the current position on the mounting line of the board where the abnormality was detected BEST MODE FOR CARRYING OUT THE INVENTION

[0015] <Embodiment 1> 1. Description of the production system 1 is a system configuration diagram of a production system S. The production system S includes a mounting line 10 and an image analysis device 20.

[0016] The mounting line 10 is a line where components E are mounted on boards PX. The mounting line 10 is equipped with a printing machine 11, an SPI 12, a surface mounter 13, an inspection machine 14, a reflow device 15, and an inspection machine 16. These machines 11 to 16 are connected in series via a conveyor, and the boards PX are sent to each machine in order, while performing a predetermined operation (printing ⇒ inspection after printing ⇒ component mounting ⇒ inspection after component mounting ⇒ reflow ⇒ inspection after reflow). Note that machine is a collective term for each device that makes up the mounting line 10.

[0017] Hereinafter, the transport direction of the substrate PX is referred to as the X direction, the direction perpendicular to this is referred to as the Y direction, and the vertical direction is referred to as the Z direction.

[0018] 2. Explanation of surface mounter The surface mounter 13 is a device that mounts components E onto the printed board PX. The number of surface mounters 13 may be one or more, and in this embodiment, a line configuration is provided in which three surface mounters 13A to 13C are arranged in series.

[0019] As shown in FIGS. 2 and 3, the surface mounter 13 includes a base 31, a transport conveyor 32, a head unit 33, and a drive device .

[0020] The transfer conveyor 32 is disposed in the center of the base 31. The transfer conveyor 32 is provided with a pair of transfer belts 35, and transfers the substrate PX in the X direction.

[0021] A large number of feeders F for supplying components E are arranged side by side on the base 31 so as to surround a work position in the center of the base. The components E are electronic components such as resistors and capacitors.

[0022] At the work position in the center of the base, a mounting process is performed by the head unit 33 to mount the components E supplied by the feeder F onto the substrate PX.

[0023] The driving device 34 is a device that moves the head unit 33 in a planar direction (X and Y directions) on the base 31. The driving device 34 is made up of a Y-axis ball screw 36, a Y-axis motor 37, an X-axis ball screw 38, and an X-axis motor 39.

[0024] By driving the Y-axis motor 37, the X-beam 40 and the head unit 33 can be moved in the Y direction along the guide rails 41 (Y-axis servo mechanism).

[0025] By driving the X-axis motor 39, the head unit 33 can be moved in the X direction relative to the X-beam 40 (X-axis servo mechanism).

[0026] 3, the head unit 33 includes a plurality of mounting heads 42. The mounting heads 42 are configured to be able to move up and down independently of the head unit 33 by a linear motion mechanism (for example, a screw mechanism) using a Z-axis motor 48 as a drive source.

[0027] A negative pressure is supplied to the mounting head 42 from a negative pressure means (not shown), which generates a suction force at the tip of the head. The mounting head 42 can hold the component E by the negative pressure.

[0028] Furthermore, the surface mounter 13 is equipped with multiple inspection cameras to check the status of the production process (board loading, component pickup, component mounting, etc.). Specifically, it is equipped with a component recognition camera 43, a side view camera 44, and a board recognition camera 45.

[0029] The component recognition camera 43 is fixed on the base 31 with its imaging surface facing upward, and captures an image of the bottom surface of the component E picked up by the mounting head .

[0030] Side view camera 44 is installed on X-beam 40 and captures a side image of component E picked up by mounting head 42. From the images taken by component recognition camera 43 and side view camera 44, the state of suction of component E to mounting head 42 can be confirmed.

[0031] The board recognition camera 45 is fixed to the head unit 33 with its imaging surface facing downward. The board recognition camera 45 can capture a planar image of the board brought into the work position. The image from the board recognition camera 45 can be used to check the state of the board PX before and after component mounting.

[0032] 4, the surface mounter 13 includes a control device 50, an X-axis motor 39, a Y-axis motor 37, a Z-axis motor 48, a component recognition camera 43, a side-view camera 44, and a board recognition camera 45. The control device 50 includes a CPU 51 and a memory 52.

[0033] The control device 50 controls the X-axis motor 39, the Y-axis motor 37, and the Z-axis motor 48 in accordance with a mounting program recorded in the memory 52, and executes the process of mounting the component E on the board PX.

[0034] 3. Image analysis equipment 1, the image analysis device 20 includes a server computer 21 and a database 22. The server computer 21 is connected to each of the machines 11 to 16 on the mounting line 10 via a LAN. The server computer 21 is also capable of wireless communication with a mobile terminal 23 of an operator who monitors the mounting line 10. The mobile terminal 23 is a smartphone, a laptop computer, or the like.

[0035] The server computer 21 sequentially captures images G taken in each production process from each machine 11 to 16 on the mounting line 10 and stores all the images in a database 22.

[0036] FIG. 5 shows an example of an image G captured in each production process of the mounting line 10. Image G of No. 1 is an image of the mounting point (board image) after reflow. Image G of No. 2 is an image of the mounting point (board image) before reflow. Image G of No. 3 is a mounting point image (board image) after the component is mounted. Image G of No. 4 is a mounting point image (board image) before the component is mounted.

[0037] Image No. 5, G, is an image of the bottom of the component (image of the component after picking up). Image No. 6, G, is a side image of the part (image of the part after picking up). Image G of No. 7 is an image of the feeder before parts are picked up. Image No. 8, G, is an image of the feeder during part pick-up. Image No. 9, G, is an image of the feeder after parts have been picked up. Image G of No. 10 is the mounting point image (board image) after printing.

[0038] When these images G taken in the same production process are compared, there is a high probability that they are identical, and there is almost no change between the boards PX. Until now, the purpose of storing images G taken in the production process in the database 22 was to check if the inspection failed, and there was no idea of ​​detecting changes between images.

[0039] However, for example, as shown in FIG. 6, by comparing the image of the bottom surface of the component with other images taken in the same process, it may be possible to detect a foreign object 80 on the bottom surface of the component, and by comparing the image of the mounting point with other image G taken in the same process, it may be possible to detect a foreign object 80 on the mounting point.

[0040] This may enable detection of defects that are difficult to detect by visual inspection alone, such as a foreign substance 80 that gets mixed in between a pad T and a component E during the production process of a substrate PX, as shown in FIG.

[0041] 4. Time-lapse differential analysis using an image analyzer 8 is a flowchart of the temporal difference analysis process (primary determination process of the difference image) by the image analyzer. When the server computer 21 imports the image G taken during the production process from each machine 11 to 16 into the database 22, the server computer 21 compares the imported image G with other images G taken during the same production process (S10, S20).

[0042] For example, when capturing image G No. 4 in Fig. 5, it is compared with image G No. 4 that has already been acquired, and when capturing image G No. 5, it is compared with image G No. 5 that has already been acquired. Then, two difference images ΔG are generated by method (A) or (B), and the difference amount is calculated.

[0043] Method (A) As shown in Figure 9A, image G of the substrate PX is compared with a normal image G to generate a difference image ΔG from the normal image G. For example, the normal image G can be data obtained by test printing before production begins and confirming that the product is good. This method has the advantage of high accuracy in detecting the difference.

[0044] Method (B) As shown in FIG. 9B, images G of two substrates PX that are continuously produced are compared to generate a differential image ΔG. For example, the first image and the second image are compared to generate a differential image ΔG. Next, the second image and the third image are compared to generate a differential image ΔG. Next, the third image and the fourth image are compared to generate a differential image ΔG. This method has the advantage that the differential image ΔG can be obtained without having to perform a trial run before starting production.

[0045] Furthermore, due to the production process, component misalignment and lighting conditions may cause differences in the captured images G. The server computer 21 may perform image correction processing before comparing the images to eliminate or reduce image differences.

[0046] The image correction process can be exemplified by processes (A) to (C) as shown in Fig. 10. The image correction process may involve all of (A) to (C), or only one of them. Alternatively, two of (A) to (C) may be performed. The image correction process is optional and may be omitted.

[0047] (A) Tone correction (brightness and color correction) (B) Part shape recognition, part angle correction, and part center position correction (C) Emphasis correction (In the image, areas other than the area of ​​interest are painted black, etc., to highlight the area of ​​interest. The area of ​​interest is the part to be analyzed or the vicinity of the mounting point.)

[0048] FIG. 11 is a histogram of an image, with the vertical axis representing frequency (occurrence frequency) and the horizontal axis representing gradation (brightness value). In this embodiment, the difference amount of the differential image ΔG is obtained from the histogram. Specifically, the frequency difference of each gradation is calculated, and the sum of these is used as the difference amount. In addition to the histogram, a QL value may also be used. The QL value is a quantized value of the brightness value that represents the intensity of light.

[0049] When the server computer 21 detects the difference amount of the differential image ΔG, it makes a primary judgment as to whether the substrate PX is a production defect. Specifically, if the difference amount of the differential image ΔG is less than a predetermined value, it is judged to be normal, and if the difference amount of the differential image ΔG is equal to or greater than the predetermined value, it is judged to be a possible production defect (S30).

[0050] If the server computer 21 determines as a result of the primary judgment that there is a possibility of a production defect (S30: YES), it notifies the operator's mobile terminal 23 of the occurrence of an abnormality (that there is a possibility of a production defect), as shown in FIG. 12, and also outputs information about the substrate PX in which the abnormality was detected (S40).

[0051] <Information about the PX board where the abnormality was detected> (1) Tracking information for the board PX (2) Abnormality detection image of the board PX

[0052] The tracking information of the substrate PX is the manufacturing number etc. The abnormality detection image of the substrate PX is the two compared images G and the difference image ΔG thereof. The abnormality detection image may be only the two compared images G.

[0053] When the mobile terminal 23 receives a notification of the occurrence of an abnormality, the operator makes a secondary judgment as to whether or not there is a production defect based on the abnormality detection image of the board PX transmitted from the server computer 21.

[0054] To facilitate the secondary judgment by the operator, the server computer 21 may transmit the two compared images G in a manner that distinguishes the areas where differences exist from the other areas. In this example, as shown in Fig. 13, the server computer 21 transmits an image G in which the areas where differences exist are surrounded by a frame 75, and displays the image G on the mobile terminal 23.

[0055] If the operator determines from the image G displayed on the mobile terminal 23 that a production defect has occurred, he or she uses the mobile terminal 23 to instruct the server computer 25 that controls the operation of the mounting line 10 to stop the machine, as shown in Figure 12.

[0056] When the server computer 25 receives the instruction to stop the machine, it takes the following measures depending on the working state of the board PX in which the abnormality was detected.

[0057] If the machine 13A that detected the foreign object is in operation (case (1) in FIG. 14), the operation of the machine 13A or the operation of the entire mounting line is stopped.

[0058] If the substrate PX has been carried out to the downstream machines 13B to 16 and is being worked on (case (2) in FIG. 14), the operation of the downstream machines 13B to 16 working on the substrate PX is stopped, or the operation of the entire mounting line is stopped.

[0059] If the substrate PX has leaked to a later process (case (3) in FIG. 14), the relevant parties are notified and measures such as recovering the substrate PX are taken.

[0060] 4.Effects According to the present invention, by comparing an image G of a substrate PX taken during a production process with an image G of a substrate PX taken during the same production process and extracting the differences, it is possible to detect slight changes or abnormalities in the production process. Therefore, it is expected that it will be possible to detect factors of defects that would be difficult to detect by visual inspection alone.

[0061] <Other embodiments> The present invention is not limited to the embodiments described above and illustrated in the drawings, and the following embodiments, for example, are also included within the technical scope of the present invention.

[0062] (1) In the above embodiment, images G of the substrates PX photographed in the same production process are compared to detect production defects. Instead of comparing images G of the substrates PX, images G of the components E mounted on the substrates PX may also be compared to detect production defects. Furthermore, images G of the feeders F that supply the components E may also be compared to detect production defects.

[0063] (2) In the above embodiment, the image G of the substrate PX photographed by the surface mounter 13 is compared with images G of other substrates PX photographed in the same production process to detect production defects. The image G of the substrate PX photographed by the surface mounter 13 is not limited to images G, but images G photographed by other machines on the mounting line 10, such as the printing machine 11 and inspection machines 12, 14, and 16, may be compared with other images G photographed in the same production process to detect production defects.

[0064] (3) In the above embodiment, if the image analysis device 20 determines in its primary judgment that there is a possibility of a production defect, the image G of the substrate PX is sent to the operator's mobile terminal 23. Then, the image G is displayed on the mobile terminal 23, and the operator makes a secondary judgment as to whether there is a production defect. The secondary judgment may be made by a program on the mobile terminal 23. It may also be made by AI or the like. If the mobile terminal 23 determines that the substrate PX is a production defect, it may perform control to stop work on the substrate PX.

[0065] (4) In the above embodiment, the images G taken by each of the machines 11 to 16 on the mounting line 10 are analyzed by the image analyzer 20. However, the images G may be analyzed by each of the machines 11 to 16 on the mounting line 10. [Explanation of symbols]

[0066] S Production System 10 Mounting Line 13 Surface Mounting Machine 20 Image analysis equipment 21 Server Computer 22 Databases 23 Mobile devices 43 Parts Recognition Camera 44 Side view camera 45 Circuit Board Recognition Camera

Claims

1. A substrate production system, a mounting line equipped with a plurality of machines that perform predetermined operations on the board; an image analysis device, The machine includes a camera for photographing the substrate; The image analysis device Sequentially capturing images of the substrate taken by the camera during the production process of the machine; A board production system that compares an image of the board taken during a production process with images of other boards taken during the same production process to extract differences, and detects production defects of the board based on the extracted image differences.

2. The substrate production system according to claim 1, The image analysis device compares the image of the board taken from the machine with a normal image that has been inspected for non-defective products, thereby extracting differences.

3. The substrate production system according to claim 1, The image analysis device is a board production system that compares images of two boards that are continuously produced and extracts differences.

4. The substrate production system according to claim 1, the machine is a surface mount machine that mounts components on the board, The image analysis device extracts differences by comparing the vicinity of the mounting point in the image of the board with images of other boards taken in the same production process.

5. The substrate production system according to claim 1, When the image analysis device determines from the image difference that there is a possibility of a production defect in the board, it notifies a terminal of an operator monitoring the mounting line that there is a possibility of a production defect in the board.

6. The substrate production system according to claim 5, the image analysis device notifies the mobile terminal that a production defect has occurred in the board, and also transmits images of the two compared boards in a manner that distinguishes areas where differences have occurred from other areas; The mobile terminal displays the two images transmitted from the image analysis device in a manner that distinguishes areas where differences occur from other areas.

7. The substrate production system according to claim 5 or 6, The mobile terminal controls the production system to stop work on the board when the board is determined to be defective.

8. The substrate production system according to claim 1, The image analysis device In addition to the images of the board taken by the camera during the production process of the machine, images of components to be mounted on the board are sequentially captured; A board production system that compares an image of the component taken during a production process with images of other components taken during the same production process to extract differences, and detects production defects of the board based on the extracted image differences.

9. An image analysis device that constitutes a part of a mounting line and sequentially captures and analyzes images taken during a production process from a machine that performs a predetermined operation on a board, An image analysis device that compares an image of the substrate taken during a production process with images of other substrates taken during the same production process to extract differences, and detects production defects of the substrate based on the extracted image differences.

Images