JP7559248B2 - Component mounting system and component mounting method - Google Patents

Description

The present invention relates to a component mounting system and a component mounting method, and in particular to a component mounting system and a component mounting method that include a component mounting device.

Conventionally, a component mounting system equipped with a component mounting device has been known. Such a component mounting system is disclosed, for example, in JP 2021-153184 A.

The above-mentioned Patent Publication No. 2021-153184 discloses a production management system (component mounting system) equipped with a component mounting machine (component mounting device).

The production management system of JP 2021-153184 A includes the component mounter, an inspection machine, and a production management computer. The component mounter includes a mounting head and a component recognition camera. The mounting head is configured to mount the component on a board by suctioning the component to a suction nozzle. The component recognition camera is configured to capture an image of the component suctioned by the suction nozzle of the mounting head to obtain a captured image. The inspection machine is configured to detect mounting defects of the component mounted on the board. The production management computer is configured to display, on a monitor, a comparison between an image of the component determined to be a mounting defect mounted on the board and a captured image that is an image of the component suctioned by the suction nozzle of the mounting head captured by the component recognition camera.

In the production management system of JP 2021-153184 A mentioned above, an operator investigates the cause of the mounting defect by using an image displayed on a monitor of the component determined to have a mounting defect mounted on a board, and an image captured by a component recognition camera of the component adsorbed to the suction nozzle of the mounting head.

JP 2021-153184 A

Here, in the production management system of JP 2021-153184 A, an operator identifies the cause of the mounting failure based on an image displayed on a monitor of the component determined to be a mounting failure mounted on a board and a captured image, which is an image of the component adsorbed to the suction nozzle of the mounting head by a component recognition camera. At this time, the operator's work in identifying the cause of the mounting failure while checking the image displayed on the monitor of the component determined to be a mounting failure mounted on a board and the captured image is cumbersome, so the operator's workload is heavy. Therefore, it is desirable to reduce the workload of the operator when identifying the cause of the mounting failure.

This invention has been made to solve the problems described above, and one object of the invention is to provide a component mounting system and a component mounting method that can reduce the workload of an operator when identifying the cause of a mounting defect.

A component mounting system according to a first aspect of the present invention comprises a component mounting device including a mounting unit that picks up a component with a nozzle and mounts it on a board, and an imaging unit that images the board or the component to obtain a recognition image, and a control device that, when a first component is determined to be mounted poorly during inspection of a board on which a component is mounted by the component mounting device, performs control to identify the operation of the component mounting device that caused the mounting failure based on a comparison between a first recognition image related to the mounting of the first component and a second recognition image related to the mounting of a second component of the same type as the first component determined to be mounted properly during inspection.

In the component mounting system according to the first aspect of the present invention, as described above, a control device is provided that performs control to identify the work of the component mounting device that caused the defect based on a comparison between the first recognition image and the second recognition image. This allows the control device to automatically identify the work of the component mounting device that caused the mounting defect, thereby reducing the workload of the worker when identifying the cause of the mounting defect. In addition, since the control device can automatically identify the work of the component mounting device that caused the mounting defect, the time required to identify the cause of the mounting defect can be shortened compared to when the worker identifies the cause of the mounting defect. In addition, by automatically identifying the work of the component mounting device that caused the mounting defect in the control device, unlike when the worker identifies the work of the component mounting device that caused the mounting defect, the judgment does not differ from worker to worker, and therefore the variability in judgment when identifying the work of the component mounting device that caused the mounting defect can be suppressed.

In the component mounting system according to the first aspect, the control device is preferably configured to correct a first component recognition image obtained by capturing an image of the first component using the imaging unit based on the amount of correction of the mounting position by the nozzle when mounting the first component, and to control the component mounting device to identify the work that caused the mounting failure based on a comparison between the first recognition image cut out based on a component dimension frame that matches the dimensions of the first component and the second recognition image captured by the imaging unit. With this configuration, the first recognition image can be aligned with the second recognition image by correcting the first component recognition image, making it easy to compare the first recognition image with the second recognition image. As a result, the control device can efficiently identify the work of the component mounting device that caused the mounting failure.

In this case, preferably, the imaging unit has a component imaging unit that images the component adsorbed to the nozzle, and the control device is configured to determine whether the amount of correction of the mounting position by the nozzle when mounting the component at the mounting position on the board is normal or not based on a comparison between a first recognition image extracted by correcting the first component recognition image captured by the component imaging unit and a second recognition image based on the image captured by the component imaging unit, and to perform control to identify the cause of the mounting failure as the component recognition work if the amount of correction of the mounting position by the nozzle is incorrect. With this configuration, when the cause of the mounting failure is the component recognition work in which the nozzle cannot correctly correct the mounting position due to incorrect recognition of the component imaged by the component imaging unit, it is possible to easily identify the cause of the mounting failure as the component recognition work.

In the component mounting system configured such that the control device performs control to identify the work of the component mounting device that caused the mounting failure based on a comparison between the first recognition image obtained by correcting the first component recognition image and the second recognition image, preferably, the imaging unit has a component imaging unit that images the component adsorbed by the nozzle, and the control device is configured to determine whether the component was adsorbed in an inverted state when the component was adsorbed by the nozzle based on a comparison between the first recognition image obtained by correcting the first component recognition image captured by the component imaging unit and the second recognition image based on the image captured by the component imaging unit, and to control to identify the cause of the mounting failure as the component supply work or the component recognition work if the component was adsorbed in an inverted state. With this configuration, if the cause of the mounting failure is a component supply work in which the component was supplied in an inverted state when the component was supplied, or a component recognition work in which an inverted component was erroneously recognized as a normal component, the cause of the mounting failure can be easily identified as the component supply work or the component recognition work.

In the component mounting system configured such that the control device performs control to identify the operation of the component mounting device that caused the mounting failure based on a comparison between the first recognition image obtained by correcting the first component recognition image and the second recognition image, preferably, the imaging unit includes a board imaging unit that images a board recognition mark for adjusting the position of the board, and the control device is configured to correct the board recognition image captured by the board imaging unit based on a correction amount based on a positional deviation amount of the board recognition mark, and to determine whether the correction amount based on the positional deviation amount of the board recognition mark is normal or not based on a comparison between the first recognition image cut out based on a mark dimension frame aligned with the board recognition mark and the second recognition image captured by the board imaging unit, and to control to identify the cause of the mounting failure as the recognition work of the board recognition mark if the correction amount of the mounting position by the nozzle is incorrect. With this configuration, if the cause of the mounting failure is the recognition work of the board recognition mark that is erroneously recognized as normal despite the board recognition mark being misaligned, the cause of the mounting failure can be easily identified as the recognition work of the board recognition mark.

In the component mounting system according to the first aspect, the control device is preferably configured to determine whether the suction state of the nozzle is normal or not based on the suction position when the component is picked up by the nozzle and the measurement value of the negative pressure of the nozzle in each case where the component mounting is determined to be defective in the board inspection and the component mounting is determined to be normal in the board inspection, and to perform control to identify the cause of the mounting failure as the suction work if the suction state of the nozzle is defective. With this configuration, when the cause of the mounting failure is the suction work in which the measurement value of the negative pressure of the nozzle is defective due to picking up a component with a nozzle with an incompatible aperture, it is possible to easily identify the cause of the mounting failure as the suction work.

In the component mounting system according to the first aspect, preferably, the imaging unit has a component imaging unit that images the component adsorbed to the nozzle, and the control device is configured to control the component mounting device to determine whether the amount of correction of the mounting position by the nozzle is appropriate based on the amount of correction of the mounting position by the nozzle based on the second component recognition image captured by the component imaging unit of the component being mounted while the component is being mounted on the board, and the second component recognition image acquired from the component mounting device. With this configuration, it is possible to predict the risk of a mounting defect by determining whether the amount of correction of the mounting position by the nozzle is appropriate based on the amount of correction of the mounting position and the component recognition image acquired while the component mounting device is mounting the component on the board, using the process of identifying the operation of the component mounting device that caused the mounting defect.

In this case, preferably, the control device is configured to determine whether the amount of correction of the mounting position by the nozzle is appropriate based on a third recognition image based on the second component recognition image captured by the component imaging unit, and a difference image between the amount of correction of the mounting position by the nozzle and the first recognition image of the first component of the same type as the component captured by the component imaging unit to obtain the second component recognition image, and to control the component mounting device to stop if the amount of correction of the mounting position by the nozzle is not appropriate. With this configuration, it is possible to predict whether a mounting defect will occur by determining whether the amount of correction of the mounting position by the nozzle is appropriate while mounting a component on a board using the process for identifying the work of the component mounting device that caused the mounting defect. As a result, if it is predicted that there is a risk of a mounting defect, the component mounting device can be stopped, thereby suppressing the occurrence of a mounting defect.

In the component mounting system according to the first aspect, the control device is preferably configured to acquire a plurality of second recognition images and to control transmission of learning results based on learning the plurality of second recognition images to the component mounting device, and the component mounting device is configured to predict whether a mounting defect will occur based on the acquired learning results and the recognition images captured by the imaging unit, and to control stopping the component mounting device if a mounting defect is predicted. With this configuration, the component mounting device can be stopped when a mounting defect is predicted based on the learning results and the recognition images, thereby suppressing the occurrence of mounting defects.

In the component mounting system according to the first aspect, the control device is preferably configured to control the display unit to associate the mounting position of the first component on the board with the work of the component mounting device that caused the defect. With this configuration, the work of the component mounting device that caused the defect corresponding to the mounting position of the first component on the board can be visually displayed on the display unit, so that the worker can easily confirm the work of the component mounting device that caused the defect corresponding to the mounting position of the first component on the board.

In this case, preferably, the control device is configured to control the display of the first recognition image, the second recognition image, and a defect information display screen including the work of the component mounting device that caused the defect on the display device based on receiving an operation to select the mounting position of the first component displayed on the display device. With this configuration, the worker can check the first recognition image, the second recognition image, and the work of the component mounting device that caused the defect by checking the defect information display screen displayed on the display device, and therefore can easily check records related to the work of the component mounting device that caused the defect.

In the component mounting system according to the first aspect, preferably, an inspection device is arranged downstream of the component mounting device and inspects the board on which the component is mounted by the component mounting device, and the control device is configured to control the component mounting device to identify the operation of the component mounting device that caused the mounting failure based on a comparison between the first recognition image and the second recognition image when a first component is determined to have a mounting failure in the inspection by the inspection device. With this configuration, a component mounting system can be realized that uses the inspection by the inspection device to identify the operation of the component mounting device that caused the mounting failure.

A component mounting method according to a second aspect of the present invention comprises the steps of mounting a component on a board by a mounting unit of a component mounting device, inspecting the board on which the component has been mounted by the mounting unit, and, if a first component is determined to be mounted poorly during inspection of the board, identifying the operation of the component mounting device that caused the mounting failure based on a comparison of a first recognition image related to the mounting of the first component with a second recognition image related to the mounting of a second component of the same type as the first component determined to be mounted properly during inspection.

In the component mounting method according to the second aspect of the present invention, as described above, when a first component is determined to be a mounting defect in the board inspection, a step is provided for identifying the operation of the component mounting device that caused the mounting defect based on a comparison between the first recognition image and the second recognition image. This makes it possible to automatically identify the operation of the component mounting device that caused the mounting defect, thereby providing a component mounting method that can reduce the workload of the worker when identifying the cause of the mounting defect.

According to the present invention, as described above, the workload of the worker when identifying the cause of mounting defects can be reduced.

1 is a schematic diagram showing a component mounting system according to a first embodiment; 1 is a schematic diagram showing a configuration of a printing device of a component mounting system according to a first embodiment. FIG. 1 is a schematic diagram showing a configuration of a component mounting device of a component mounting system according to a first embodiment. FIG. 2 is a schematic diagram showing a configuration of a first inspection device of the component mounting system of the first embodiment. FIG. 2 is a schematic diagram showing a configuration of a second inspection device of the component mounting system according to the first embodiment. FIG. FIG. 2 is a schematic diagram showing a configuration of an image server of the component mounting system according to the first embodiment. 2 is a schematic diagram showing a configuration of an apparatus data server of the component mounting system according to the first embodiment; FIG. 2 is a schematic diagram showing a configuration of an inspection data server of the component mounting system according to the first embodiment; FIG. 2 is a schematic diagram showing a configuration of a control device of the component mounting system of the first embodiment. FIG. FIG. 2 is a schematic diagram showing a configuration of a client terminal of the component mounting system according to the first embodiment. 4 is a schematic diagram showing a first list created by a control device of the component mounting system of the first embodiment. FIG. 13 is a schematic diagram showing a second list created by a control device of the component mounting system of the first embodiment. FIG. 2 is a schematic diagram showing a first recognition image based on a first board recognition image of the component mounting system of the first embodiment. FIG. 13 is a schematic diagram showing a second recognition image based on a second board recognition image of the component mounting system of the first embodiment. FIG. 3 is a schematic diagram showing a first recognized image based on a first component recognized image of the component mounting system of the first embodiment. FIG. 10 is a schematic diagram showing a second recognition image based on a second component recognition image of the component mounting system of the first embodiment. FIG. 1 is a schematic diagram showing an inverted first recognition image based on a first component recognition image of the component mounting system of the first embodiment. FIG. 10 is a schematic diagram showing a non-inverted second recognition image based on the second component recognition image of the component mounting system of the first embodiment. FIG. 4 is a schematic diagram showing a defect information display screen by a control device of the component mounting system of the first embodiment. FIG. 13 is a schematic diagram showing a third list created by the control device of the component mounting system of the first embodiment. FIG. 13 is a schematic diagram showing a fourth list created by the control device of the component mounting system of the first embodiment. FIG. 4 is a Pareto diagram created by the control device of the component mounting system of the first embodiment. FIG. 4 is a flowchart showing a component mounting method performed by a control device of the component mounting system of the first embodiment. 10 is a flowchart showing an FI mark recognition image determination of the component mounting method performed by the control device of the component mounting system of the first embodiment. 5 is a flowchart showing a component recognition image determination of a component mounting method performed by a control device of the component mounting system of the first embodiment. 10 is a flowchart showing a component inversion recognition image determination of the component mounting method performed by the control device of the component mounting system of the first embodiment. 5 is a flowchart showing a pickup position deviation amount determination of the component mounting method performed by the control device of the component mounting system of the first embodiment. FIG. 13 is a schematic diagram showing a component mounting system according to a second embodiment. 13 is a schematic diagram showing a first recognition image based on a first component recognition image of the component mounting system of the second embodiment. FIG. 13 is a schematic diagram showing a second recognition image based on a second component recognition image of the component mounting system of the second embodiment. FIG. 10 is a flowchart showing a mounting defect prediction method executed by a control device of a component mounting system according to a second embodiment. FIG. 13 is a schematic diagram showing a component mounting system according to a third embodiment. 13 is a flowchart showing a component recognition image determination of a component mounting method performed by a control device of a component mounting system according to a third embodiment. 13 is a flowchart showing a component recognition image determination of a component mounting method performed by a control device of a component mounting system according to a first modified example of the first embodiment. 13 is a flowchart showing a component recognition image determination of a component mounting method performed by a control device of a component mounting system according to a second modified example of the first embodiment.

Below, an embodiment of the present invention is described with reference to the drawings.

[First embodiment]
The configuration of a component mounting system 100 according to a first embodiment of the present invention will be described with reference to FIGS.

(Component mounting system configuration)
As shown in FIG. 1, a component mounting system 100 is a system that mounts components E (electronic components) such as ICs, transistors, capacitors, and resistors on a printed circuit board Pb or the like to manufacture a board P on which the components E are mounted.

The component mounting system 100 includes a printing device 1, a component mounting device 2, a first inspection device 3, a reflow furnace 4, a second inspection device 5, an image server 6, an apparatus data server 7, an inspection data server 8, a control device 9, and a client terminal 10.

(Printing device)
1 and 2, the printing device 1 is a device that performs a printing operation of screen-printing solder on a printed circuit board Pb. The printing device 1 includes a screen mask 11, a squeegee 12, a mask imaging unit 13, a print inspection imaging unit 14, a communication unit 15, a CPU (Central Processing Unit) 16, and a storage unit 17.

The squeegee 12 is configured to spread the solder placed on the screen mask 11 pressed against the printed circuit board Pb. This causes the solder of a predetermined pattern to be printed (transferred) onto the printed circuit board Pb.

The mask imaging unit 13 is configured to capture an image of the screen mask 11. In the printing device 1, the screen mask 11 is inspected based on the captured image of the screen mask 11. This makes it possible to detect, for example, clogging of the screen mask 11.

The print inspection imaging unit 14 is configured to capture an image of the printed circuit board Pb after the printing operation on which the solder has been printed. The printing device 1 inspects the solder on the printed circuit board Pb based on the captured image of the printed circuit board Pb after the printing operation. This makes it possible to inspect, for example, the solder bleeding, fading, and amount of misalignment on the printed circuit board Pb. The printing device 1 also determines whether the solder on the printed circuit board Pb is good or bad based on the inspection results of the solder on the printed circuit board Pb.

The communication unit 15 is configured to communicate with other devices other than the printing device 1 (the component mounting device 2, the first inspection device 3, the reflow furnace 4, and the second inspection device 5).

The memory unit 17 stores the captured image captured by the mask imaging unit 13 and the captured image captured by the print inspection imaging unit 14 as image data.

(Component mounting device)
1 and 3, the component mounting device 2 is a device that performs a mounting operation of mounting a component E on a printed circuit board Pb on which solder is printed. The component mounting device 2 includes a mounting unit 21, a component imaging unit 22, a board imaging unit 23, a tape feeder 24, a communication unit 25, a CPU 26, and a storage unit 27. Each of the component imaging unit 22 and the board imaging unit 23 is an example of an "imaging unit" in the claims.

The mounting unit 21 includes a mounting head 21a and a negative pressure measuring unit 21b. The mounting head 21a is configured to adsorb a component E with a nozzle N and mount the component E on the printed circuit board Pb. The nozzle N that adsorbs the component E when mounting the component E on the printed circuit board Pb is removably attached to the mounting head 21a. The negative pressure measuring unit 21b is configured to measure the negative pressure of the mounting head 21a. The negative pressure measuring unit 21b is a sensor that measures the negative pressure of the mounting head 21a.

The component imaging unit 22 is configured to image the component E adsorbed to the nozzle N of the mounting head 21a from below. As a result, a component recognition image Gc (see Figures 15 and 16) of the component E is acquired by the component imaging unit 22. The component mounting device 2 is configured to inspect the component E adsorbed to the nozzle N of the mounting head 21a based on the component recognition image Gc acquired by the component imaging unit 22. As a result, it is possible to inspect, for example, the adsorption state of the component E as viewed from below, the amount of positional deviation of the component E from the center position of the nozzle N, and the size (depth and width) of the component E. In addition, when the component mounting device 2 acquires the amount of positional deviation of the component E from the center position of the nozzle N, it acquires a correction amount for correcting the mounting position by the nozzle N when mounting the component E.

The board imaging unit 23 is configured to image the FI mark (fiducial mark) Fi1 (Fi2) (see Figures 13 and 14) attached to the upper surface of the printed circuit board Pb prior to mounting the component E on the printed circuit board Pb. As a result, a board recognition image Gs (see Figures 13 and 14) is acquired by imaging the FI mark Fi1 (Fi2) of the printed circuit board Pb by the board imaging unit 23. The FI mark Fi1 (Fi2) is a mark for adjusting the position of the board. The component mounting device 2 is configured to inspect the position of the printed circuit board Pb based on the board recognition image Gs imaged by the board imaging unit 23. This makes it possible to adjust, for example, the positional deviation of the printed circuit board Pb in the board transport direction. The FI mark Fi1 and the FI mark Fi2 are examples of the "board recognition mark" in the claims.

A plurality of tape feeders 24 are arranged in the feeder arrangement section of the component mounting device 2. The tape feeders 24 are configured to supply components E to be mounted on the printed circuit board Pb. The tape feeder 24 holds a reel (not shown) wound with a component supply tape that holds a plurality of components E at a predetermined interval.

The communication unit 25 is configured to communicate with other devices (printing device 1, first inspection device 3, reflow furnace 4, and second inspection device 5) other than the component mounting device 2. In the component mounting device 2, the communication unit 25 transmits the feeder number, which is the ID (Identification) of the tape feeder 24, to the device data server 7. In the component mounting device 2, the communication unit 25 transmits the nozzle number, which is the ID of the nozzle N, to the device data server 7. In the component mounting device 2, the communication unit 25 transmits the head number, which is the ID of the mounting head 21a, to the device data server 7. In the component mounting device 2, the communication unit 25 transmits the first camera number, which is the ID of the component imaging unit 22, to the device data server 7. In the component mounting device 2, the communication unit 25 transmits the second camera number, which is the ID of the board imaging unit 23, to the device data server 7.

The memory unit 27 stores the captured images captured by the component imaging unit 22 and the captured images captured by the board imaging unit 23 as image data.

(First inspection device)
1 and 4, the first inspection device 3 is a device that inspects components E on a board P on which the components E are mounted by the component mounting device 2. The first inspection device 3 is disposed downstream in the board transport direction from the component mounting device 2. The first inspection device 3 includes an inspection imaging unit 31, a communication unit 32, a CPU 33, and a storage unit 34.

The inspection imaging unit 31 is configured to image the component E at the mounting position on the printed circuit board Pb. The first inspection device 3 is configured to inspect the component E on the printed circuit board Pb based on the captured image of the component E on the printed circuit board Pb after the mounting work. This makes it possible to inspect, for example, a lack of solder on the printed circuit board Pb, the amount of misalignment of the component E, and floating of the component E. Furthermore, the first inspection device 3 determines whether the mounting of the component E on the printed circuit board Pb is normal or defective based on the inspection results of the component E on the printed circuit board Pb.

The communication unit 32 is configured to communicate with other devices other than the first inspection device 3 (the printing device 1, the component mounting device 2, the reflow furnace 4, the second inspection device 5 and the inspection data server 8).

The memory unit 34 stores the captured image captured by the inspection imaging unit 31 as image data.

(reflow oven)
1, the reflow furnace 4 performs a reflow operation of joining the components E to the printed circuit board Pb by melting and solidifying solder printed on the printed circuit board Pb as a board manufacturing operation for manufacturing a board P on which components E are mounted. Although not shown, the reflow furnace 4 includes a communication unit capable of communicating with other devices (the printing device 1, the component mounting device 2, the first inspection device 3, and the second inspection device 5) other than the reflow furnace 4.

(Second Inspection Device)
1 and 5, the second inspection device 5 is a device that inspects components E on a printed circuit board Pb after a reflow operation. The second inspection device 5 is disposed downstream in the board transport direction from the component mounting device 2. The second inspection device 5 includes an inspection imaging unit 51, a communication unit 52, a CPU 53, and a storage unit 54.

The inspection imaging unit 51 is configured to image the component E at the mounting position on the printed circuit board Pb. The second inspection device 5 is configured to inspect the component E on the printed circuit board Pb based on the captured image of the component E on the printed circuit board Pb after the reflow operation. This makes it possible to inspect, for example, a lack of solder on the printed circuit board Pb, the amount of misalignment of the component E, and floating of the component E. Furthermore, the second inspection device 5 determines whether the mounting of the component E on the printed circuit board Pb is normal or defective based on the inspection results of the component E on the printed circuit board Pb.

The communication unit 52 is configured to communicate with other devices other than the second inspection device 5 (the printing device 1, the component mounting device 2, the first inspection device 3, the reflow furnace 4 and the inspection data server 8).

The memory unit 54 stores the captured image captured by the inspection imaging unit 51 as image data.

(Image Server)
As shown in FIG. 1 and FIG. 6, the image server 6 is a server that manages the component recognition image Gc of the component E captured by the component imaging unit 22 and the board recognition image Gs of the FI mark Fi1 (Fi2) captured by the board imaging unit 23. The image server 6 includes a communication unit 61, a CPU 62, a storage unit 63, and the like. The communication unit 61 is configured to communicate with the component mounting device 2. The storage unit 63 stores the component recognition image Gc and the board recognition image Gs. The image server 6 receives the component recognition image Gc from the component mounting device 2 every time the component recognition image Gc is captured by the component imaging unit 22 in the component mounting device 2. The image server 6 receives the board recognition image Gs from the component mounting device 2 every time the board recognition image Gs is captured by the board imaging unit 23 in the component mounting device 2.

(Device Data Server)
1 and 7, the device data server 7 is a server that manages the suction position deviation amount Vp relative to the center position of the nozzle N of the component E based on the component recognition image Gc captured by the component imaging unit 22, the negative pressure measurement value Pr of the mounting head 21a measured by the negative pressure measurement unit 21b, and the like. The device data server 7 includes a communication unit 71, a CPU 72, a storage unit 73, and the like. The communication unit 71 is configured to communicate with the component mounting device 2. The storage unit 73 stores the suction position deviation amount Vp and the negative pressure measurement value Pr. In addition to the suction position deviation amount Vp and the negative pressure measurement value Pr, the storage unit 73 stores a feeder number, a board identification number, a line number, a nozzle number, a feeder number, a component number, and the like.

(Test data server)
As shown in FIG. 1 and FIG. 8, the inspection data server 8 is a server that manages a first judgment result R1 in the first inspection device 3 as to whether the mounting of the component E mounted at the mounting position on the printed circuit board Pb is normal or defective, and a second judgment result R2 in the second inspection device 5 as to whether the mounting of the component E mounted at the mounting position on the printed circuit board Pb is normal or defective. The inspection data server 8 includes a communication unit 81, a CPU 82, a storage unit 83, and the like. The communication unit 71 is configured to communicate with the component mounting device 2. The storage unit 73 stores the first judgment result R1 and the second judgment result R2. The storage unit 73 stores a first inspection image associated with the first judgment result R1 and a second inspection image associated with the second judgment result R2. The inspection data server 8 is configured to acquire the first judgment result R1, the first inspection image, the second judgment result R2, and the second inspection image every time one board P is manufactured.

(Control device)
As shown in Fig. 1 and Fig. 7, the control device 9 is configured to perform control to identify the cause of the mounting defect included in each of the first judgment result R1 and the second judgment result R2 based on the information stored in the image server 6, the information stored in the device data server 7, and the information stored in the inspection data server 8. The control device 9 includes a display unit 91, a communication unit 92, a CPU 93, and a storage unit 94. The display unit 91 is configured with a liquid crystal screen. The communication unit 92 is configured to communicate with each of the image server 6, the device data server 7, and the inspection data server. The storage unit 94 stores a defect cause identification program Pg that identifies the cause of the mounting defect. The defect cause identification program Pg will be described in detail later.

(Client terminal)
1 and 10, the client terminal 10 is a terminal such as a stationary personal computer operated by an operator. The client terminal 10 includes a display unit 101, a communication unit 102, a CPU 103, and a storage unit 104. The display unit 101 is configured with a liquid crystal screen. The display unit 101 is configured to display information transmitted from the control device 9 on the screen. The communication unit 102 is configured to communicate with the control device 9.

(Program to identify the cause of defects)
11 and 12, the control device 9 of the first embodiment is configured to perform control to identify the operation of the component mounting device 2 that caused the mounting failure, based on a comparison between a first recognition image Gr1 related to the mounting of the first component E1 and a second recognition image Gr2 related to the mounting of a second component E2 of the same type as the first component E1 that was determined to be normally mounted in the inspection, when a first component E1 is determined to be defective in the inspection of a board on which the component E is mounted by the component mounting device 2. That is, the control device 9 is configured to perform control to identify the operation of the component mounting device 2 that caused the mounting failure, based on a comparison between the first recognition image Gr1 and the second recognition image Gr2, when a first component E1 is determined to be defective in the inspections of the first inspection device 3 and the second inspection device 5.

Figure 11 shows an example of the results of identifying the operation of the component mounting device 2 that caused the mounting failure. That is, the cause of the mounting failure of component E mounted at mounting position number 2 is identified as erroneous recognition of the FI mark Fi1. The cause of the mounting failure of component E mounted at mounting position number 13 is identified as erroneous component recognition. The cause of the mounting failure of component E mounted at mounting position number 23 is identified as component inversion. The cause of the mounting failure of component E mounted at mounting position number 24 is unknown and has not been identified. The cause of the mounting failure of component E mounted at mounting position number 30 is identified as the suction posture.

Figure 12 shows an example of the results of a determination made to identify the operation of component mounting device 2 that caused the mounting failure and related numerical values. The above determination is made to identify the cause of the mounting failure at each of mounting position numbers 2, 13, 23, 24, and 30.

With reference to Figures 11 and 12, the identification of the cause of the mounting failure of component E at mounting position number 2 is described below as an example.

<FI mark recognition image judgment>
First, an FI mark recognition image judgment is performed. That is, the control device 9 is configured to perform control to judge whether or not the correction amount based on the positional deviation amount of the FI mark Fi1 is normal based on a comparison between a first recognition image Gr1 (see FIG. 13) obtained by converting the board recognition image Gs based on the correction amount based on the positional deviation amount of the FI mark Fi1 and the mark size frame Fs aligned with the FI mark Fi1, and a second recognition image Gr2 (see FIG. 14). When the correction amount based on the positional deviation amount of the FI mark Fi1 is defective, the control device 9 is configured to perform control to identify the cause of the mounting defect as the recognition work of the FI mark Fi1.

Specifically, the control device 9 is configured to obtain the mounting position of the first component E1, whose mounting of the component E is judged to be defective, from the inspection data server 8, and to control the acquisition of the first board recognition image Gs1 associated with the mounting position from the image server 6. The control device 9 is configured to control the movement (correction) of the first board recognition image Gs1 obtained by capturing the FI mark Fi1 by the board imaging unit 23 based on the correction amount based on the positional deviation amount of the FI mark Fi1. Here, the correction amount is stored in the device data server 7. The control device 9 is configured to control the cutting out of the portion of the mark dimension frame Fs of the moved first board recognition image Gs1 based on the mark dimension frame Fs that matches the dimensions of the FI mark Fi1. This cut-out image is the converted first recognition image Gr1 (see FIG. 13).

The control device 9 is configured to obtain the mounting position of the second component E2, which is determined to be normally mounted and is of the same variety as the first component E1, from the inspection data server 8, and to control the acquisition of the second board recognition image Gs2 associated with the mounting position from the image server 6. The control device 9 is configured to control the movement (correction) of the second board recognition image Gs2, which is an image of the FI mark Fi2 captured by the board imaging unit 23, based on the correction amount based on the positional deviation amount of the FI mark Fi2. Here, the correction amount is stored in the device data server 7. The control device 9 is configured to control the cutting out of the part of the mark dimension frame Fs of the moved second board recognition image Gs2, based on the mark dimension frame Fs that matches the dimensions of the FI mark Fi2. This cut-out image is the converted second recognition image Gr2. Here, multiple second recognition images Gr2 are acquired.

The control device 9 is configured to control the acquisition of FI mark recognition image learning data by performing machine learning to determine that the second recognition image Gr2 is an image in which the component E is correctly mounted, based on a plurality of second recognition images Gr2, which are images obtained during normal mounting. The control device 9 is configured to control the acquisition of a numerical value indicating the similarity between the FI mark recognition image learning data and the first recognition image Gr1, based on pixel value information (histogram) included in the FI mark recognition image learning data and pixel value information (histogram) of the first recognition image Gr1. The control device 9 is configured to control the determination of whether the numerical value indicating the similarity between the FI mark recognition image learning data and the first recognition image Gr1 is within a first similarity range.

Here, for mounting position number 2, the numerical value indicating the acquired similarity is outside the first similarity range, so the control device 9 determines that the amount of correction of the mounting position by the nozzle N of component E based on the recognition of the FI mark Fi1 is defective. As a result, the control device 9 determines that the recognition of the FI mark Fi1 was incorrect, and that the cause of the mounting defect was the incorrect amount of correction of the mounting position by the nozzle N of component E. Furthermore, as a measure to resolve this mounting defect, the control device 9 indicates confirmation of board recognition parameters including the amount of movement of the board imaging unit 23 when the FI mark Fi1 is imaged by the board imaging unit 23.

<Parts recognition image judgment>
Next, as shown in Fig. 11 and Fig. 12, a component recognition image judgment is performed. That is, the control device 9 is configured to perform control to judge whether or not the amount of correction of the mounting position by the nozzle N when mounting the component E on the mounting position on the board is normal based on a comparison between a first recognition image Gr1 (see Fig. 15) obtained by converting the first component recognition image Gc1 based on the amount of correction of the mounting position by the nozzle N when mounting the first component E1 and a component size frame Fc that matches the size of the first component E1, and a second recognition image Gr2 obtained by converting the second component recognition image Gc2 based on the amount of correction of the mounting position by the nozzle N when mounting the second component E2 and the component size frame Fc. In addition, the control device 9 is configured to perform control to identify whether the cause of the mounting defect is the component recognition work when the amount of correction of the mounting position by the nozzle N is defective.

Specifically, the control device 9 is configured to obtain the mounting position of the first component E1, whose mounting of the component E is determined to be defective, from the inspection data server 8, and to control the acquisition of the first component recognition image Gc1 associated with the mounting position from the image server 6. The control device 9 is configured to control the movement (correction) of the first component recognition image Gc1 obtained by capturing the image of the first component E1 by the component imaging unit 22, based on the amount of correction of the mounting position by the nozzle N when mounting the first component E1. Here, the amount of correction is stored in the device data server 7. The control device 9 is configured to control the cutting out of the part of the moved first component recognition image Gc1 that corresponds to the part dimension frame Fc based on the part dimension frame Fc that matches the dimensions of the first component E1. This cut-out image is the converted first recognition image Gr1 (see FIG. 15).

The control device 9 is configured to control the acquisition of the mounting position of the second component E2, which is determined to be normal for the same type as the first component E1, from the inspection data server 8, and the acquisition of the board recognition image Gs associated with the mounting position from the image server 6. The control device 9 is configured to control the movement (correction) of the second component recognition image Gc2, which is an image of the second component E2 captured by the component imaging unit 22, based on the amount of correction of the mounting position by the nozzle N when mounting the second component E2. Here, the correction amount is stored in the device data server 7. The control device 9 is configured to control the cutting out of the part of the component dimension frame Fc of the moved second component recognition image Gc2 based on the part dimension frame Fc. This cut-out image is the converted second recognition image Gr2 (see FIG. 16). Here, a plurality of second recognition images Gr2 are acquired.

The control device 9 is configured to control the acquisition of component recognition image learning data by performing machine learning to determine that the second recognition image Gr2 is an image in which the component E is normally mounted, based on a plurality of second recognition images Gr2, which are images of normal mounting. The control device 9 is configured to control the acquisition of a numerical value indicating the similarity between the component recognition image learning data and the first recognition image Gr1, based on pixel value information (histogram) included in the component recognition image learning data and pixel value information (histogram) of the first recognition image Gr1. The control device 9 is configured to control the determination of whether the numerical value indicating the similarity between the component recognition image learning data and the first recognition image Gr1 is within a second similarity range.

Here, at mounting position number 2, the numerical value indicating the acquired similarity is within the second similarity range. As a result, at mounting position number 2, the control device 9 determines that the component recognition image judgment is normal.

<Parts inversion recognition image judgment>
Furthermore, a determination is made as to whether or not the component E is inverted as a component inversion recognition image determination. That is, the control device 9 is configured to perform control to determine whether or not the component E was picked up in an inverted state when picked up by the nozzle N, based on a comparison between the first recognition image Gr1 obtained by converting the first component recognition image Gc1 and the second recognition image Gr2 obtained by converting the second component recognition image Gc2. Furthermore, the control device 9 is configured to perform control to determine that the cause of the mounting defect is the component supply operation or the component recognition operation, when the component E was picked up in an inverted state.

Specifically, the control device 9 is configured to obtain the mounting position of the first component E1, whose mounting of the component E is determined to be defective, from the inspection data server 8, and to control the acquisition of the first component recognition image Gc1 associated with the mounting position from the image server 6. The control device 9 is configured to control the movement (correction) of the first component recognition image Gc1 obtained by the component imaging unit 22 capturing the first component E1 based on the amount of correction of the mounting position by the nozzle N when mounting the first component E1. Here, the amount of correction is stored in the device data server 7. The control device 9 is configured to control the cutting out of the part of the moved first component recognition image Gc1 that corresponds to the part dimension frame Fc based on the part dimension frame Fc that matches the dimensions of the first component E1. This cut-out image is the converted first recognition image Gr1 (see FIG. 17).

The control device 9 is configured to control the acquisition of the mounting position of the second component E2, which is determined to be normal for the same type as the first component E1, from the inspection data server 8, and the acquisition of the board recognition image Gs associated with the mounting position from the image server 6. The control device 9 is configured to control the movement (correction) of the second component recognition image Gc2, which is an image of the second component E2 captured by the component imaging unit 22, based on the amount of correction of the mounting position by the nozzle N when mounting the second component E2. Here, the correction amount is stored in the device data server 7. The control device 9 is configured to control the cutting out of the part of the component dimension frame Fc of the moved second component recognition image Gc2 based on the part dimension frame Fc. This cut-out image is the converted second recognition image Gr2 (see FIG. 18). Here, a plurality of second recognition images Gr2 are acquired.

The control device 9 is configured to control the acquisition of component inverted recognition image learning data by performing machine learning to determine that the second recognition image Gr2 is an image in which the component E is normally mounted, based on a plurality of second recognition images Gr2, which are images at the time of normal mounting. The control device 9 is configured to control the acquisition of a numerical value indicating the similarity between the component inverted recognition image learning data and the first recognition image Gr1, based on pixel value information (histogram) included in the component inverted recognition image learning data and pixel value information (histogram) of the first recognition image Gr1. The control device 9 is configured to control the determination of whether the numerical value indicating the similarity between the component inverted recognition image learning data and the first recognition image Gr1 is within a third similarity range.

Here, at mounting position number 2, the numerical value indicating the acquired similarity is within the third similarity range. As a result, the control device 9 determines that the component inversion recognition image judgment at mounting position number 2 is normal.

<Determining the amount of deviation in the pickup position>
Next, as shown in Fig. 11 and Fig. 12, a suction position deviation amount is determined. That is, the control device 9 is configured to perform control to determine whether or not the suction state of the nozzle N is normal based on the suction position when the nozzle N picks up the component E and the measurement value of the negative pressure of the nozzle N, in each case where the mounting of the component E is determined to be defective in the inspection of the board and where the mounting of the component E is determined to be normal in the inspection of the board. Furthermore, the control device 9 is configured to perform control to identify the cause of the mounting failure as the suction work, in the case where the suction state of the nozzle N is defective.

Specifically, the control device 9 is configured to obtain the mounting position of the first component E1, whose mounting of the component E has been determined to be defective, from the inspection data server 8, and to control the acquisition of the suction position deviation amount Vp and negative pressure measurement value Pr associated with the mounting position from the device data server 7.

The control device 9 is configured to control the acquisition of the mounting position of the second component E2, which is determined to be normally mounted and is of the same variety as the first component E1, from the inspection data server 8, and the acquisition of the pickup position deviation amount Vp and the negative pressure measurement value Pr associated with the mounting position from the device data server 7. The control device 9 is configured to control the acquisition of the average value and standard deviation of the pickup position deviation amount Vp during normal mounting, based on the pickup position deviation amount Vp during normal mounting. The control device 9 is configured to control the acquisition of the average value and standard deviation of the negative pressure measurement value Pr during normal mounting, based on the negative pressure measurement value Pr during normal mounting.

The control device 9 is configured to perform control based on the average value and standard deviation of the adsorption position deviation amount Vp during normal mounting to determine whether the adsorption position deviation amount Vp during defective mounting is between a value obtained by subtracting three times the standard deviation of the adsorption position deviation amount Vp during normal mounting from the average value of the adsorption position deviation amount Vp during normal mounting and a value obtained by adding three times the standard deviation of the adsorption position deviation amount Vp during normal mounting to the average value of the adsorption position deviation amount Vp during normal mounting (a first predetermined range).

The control device 9 is configured to perform control based on the average value and standard deviation of the negative pressure measurement value Pr during normal installation to determine whether the negative pressure measurement value Pr during poor installation is between the value obtained by subtracting three times the standard deviation of the negative pressure measurement value Pr during normal installation from the average value of the negative pressure measurement value Pr during normal installation and the value obtained by adding three times the standard deviation of the negative pressure measurement value Pr during normal installation to the average value of the negative pressure measurement value Pr during normal installation (second predetermined range).

Here, at mounting position number 2, the suction position deviation amount Vp during poor mounting is within a first specified range, and the negative pressure measurement value Pr during poor mounting is within a second specified range, so the suction position deviation amount judgment is determined to be normal.

<Defect information display screen>
11, 12 and 19, the control device 9 displays on the display unit 101 a display showing the mounting position of the first component E1 on the board in association with the operation of the component mounting device 2 that caused the mounting defect. The present invention is configured to control the display of the

Specifically, as shown in Fig. 11, the control device 9 is configured to control the display unit 101 to create a first list L1 that associates the mounting position of the first component E1 on the board with the cause of the defect in the board manufacturing work based on the mounting position of the first component E1 on the board and the work of the component mounting device 2 that caused the mounting defect. Also, as shown in Fig. 12, the control device 9 is configured to control the display unit 101 to create a second list L2 that associates the mounting position of the first component E1 on the board with the judgment result based on the judgment result performed to identify the mounting position of the first component E1 on the board and the cause of the mounting defect.

As shown in FIG. 19, the control device 9 is configured to control the display of the first recognition image Gr1, the second recognition image Gr2, and a defect information display screen Gf including the operation of the component mounting device 2 that caused the defect, on the display unit 101 based on receiving an operation to select the mounting position of the first component E1 in the first list L1 or the second list L2 displayed on the display unit 101.

The defect information display screen Gf shown in FIG. 19 shows, as an example, a case in which a mounting failure of component E occurs due to a misalignment amount at the time of pickup of component E in the pickup position misalignment amount judgment. As an example, the defect information display screen Gf in FIG. 19 shows that the cause is pickup position misalignment amount Vp. As an example, the defect information display screen Gf in FIG. 19 shows that the inspection result of the misalignment amount of the mounting position of the first inspection device 3 is bad. As an example, the defect information display screen Gf in FIG. 19 shows the judgment result of the component recognition image judgment and the judgment result of the pickup position misalignment amount judgment. 19 shows, as an example, a first component recognition image Gc1 after the first component E1 is adsorbed by the nozzle N, a first inspection image Gh1 of the first component E1 captured by the first inspection device 3, a second inspection image Gh2 of the first component E1 captured by the second inspection device 5, a second component recognition image Gc2 after the second component E2 is adsorbed by the nozzle N, and a first board recognition image Gs1 of the FI mark Fi1 captured by the board imaging unit 23. The defect information display screen Gf in FIG. 19 shows, as an example, related part information including the number of the mounting head 21a and the number of the nozzle N.

As shown in Fig. 20, the control device 9 is configured to control the display unit 101 to display a third list L3 that lists the results of the control device 9 automatically identifying the cause of the mounting defect at the mounting position of the board of the first component E1 and the results of the worker manually identifying the cause of the mounting defect at the mounting position of the board of the first component E1. This allows the worker an opportunity to improve the algorithm for automatically identifying the cause of the mounting defect at the mounting position of the board of the first component E1 in the control device 9 so that the automatically identified results and the manually identified results match for items where there is a mismatch. Note that the mismatched items are hatched in Fig. 20.

As shown in FIG. 21, the control device 9 is configured to control the display of a fourth list L4 of related part information displayed on the defect information display screen Gf shown in FIG. 19 on the display unit 101.

22, the control device 9 is configured to control the display unit 101 to display a Pareto diagram in which the causes of mounting defects at the mounting position of the first component E1 on the board are sorted in descending order of the number of cases, based on the summary information that summarizes the number of cases by cause of mounting defects at the mounting position of the first component E1 on the board. This makes it easier for the worker to determine the priority of the work to eliminate the causes of mounting defects at the mounting position of the first component E1 on the board.

In addition, the control device 9 is configured to control the display of the first list L1, the second list L2, the third list L3, the fourth list L4, and the defect information display screen Gf on the display unit 101 in order to visualize the causes of mounting defects at the mounting positions of the first component E1 on the board. The control device 9 is configured to control the display of the number of occurrences of mounting defects at the mounting positions of the first component E1 on the board for each of the production model names and mounting position numbers displayed in the fourth list L4 on the display unit 101.

The control device 9 is configured to control the display unit 101 to display the number of mounting defects occurring at the mounting position of the first component E1 on the board for each line number displayed in the fourth list L4. The control device 9 is configured to control the display unit 101 to display the number of mounting defects occurring at the mounting position of the first component E1 on the board for each nozzle number displayed in the fourth list L4. The control device 9 is configured to control the display unit 101 to display the number of mounting defects occurring at the mounting position of the first component E1 on the board for each component number displayed in the fourth list L4.

The control device 9 is configured to control the display unit 101 to display the cross-tabulation results, for example, of multiplying the component number and line number displayed in the fourth list L4 to tally up the number of mounting defects occurring at the mounting positions of the first component E1 on the board. The control device 9 is configured to control the display unit 101 to display the determination rate, which is the number of cases identified as having an unknown cause divided by the total number of mounting defects occurring.

(Component mounting method)
23 to 27, a component mounting method performed in the control device 9 will be described below. The component mounting method is one example of the "component mounting method" set forth in the claims.

As shown in FIG. 23, in step S1, an FI mark image recognition judgment is performed. The FI mark image recognition judgment will be described in detail later. In step S2, a component recognition image judgment is performed. The component recognition image judgment will be described in detail later. In step S3, a component inverted recognition image judgment is performed. The component inverted recognition image judgment will be described in detail later. In step S4, an adsorption position deviation amount judgment is performed, and then the component mounting method ends. The adsorption position deviation amount judgment will be described in detail later.

<FI mark image recognition judgment>
With reference to FIG. 24, the FI mark image recognition determination performed by the control device 9 will be described.

As shown in FIG. 24, in step S11, the mounting position of the first component E1, which is determined to be defective when mounted at the mounting position on the printed circuit board Pb, and a first board recognition image Gs1 associated with the mounting position of the first component E1 are acquired. In step S12, a first recognition image Gr1 is acquired by performing the above-mentioned conversion on the first board recognition image Gs1 based on the first board recognition image Gs1. In step S13, the mounting position of the second component E2, which is determined to be normal when mounted at the mounting position on the printed circuit board Pb, and a second board recognition image Gs2 associated with the mounting position of the second component E2 are acquired. In step S14, a second recognition image Gr2 is acquired by performing the above-mentioned conversion on the second board recognition image Gs2 based on the second board recognition image Gs2.

In step S15, the second recognition image Gr2 is learned by machine learning to obtain FI mark recognition image learning data. In step S16, the FI mark recognition image learning data is compared with the first recognition image Gr1 to obtain a similarity value. In step S17, it is determined whether the similarity value is within the first similarity range. If the similarity value is within the first similarity range, the process proceeds to step S18, where it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the FI mark Fi1 is normal, and the FI mark recognition image judgment is terminated. If the similarity value is outside the first similarity range, the process proceeds to step S19, where it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the FI mark Fi1 is defective, and the FI mark recognition image judgment is terminated.

<Parts recognition image judgment>
Referring to FIG. 25, the component recognition image determination performed by the control device 9 will be described.

As shown in FIG. 25, in step S21, the mounting position of the first component E1, which is determined to be defective when mounted at the mounting position on the printed circuit board Pb, and the first component recognition image Gc1 associated with the mounting position of the first component E1 are obtained. In step S22, the first recognition image Gr1 is obtained by performing the above conversion on the first component recognition image Gc1 based on the first component recognition image Gc1. In step S23, the mounting position of the second component E2, which is determined to be normal when mounted at the mounting position on the printed circuit board Pb, and the second component recognition image Gc2 associated with the mounting position of the second component E2 are obtained. In step S24, the second recognition image Gr2 is obtained by performing the above conversion on the second component recognition image Gc2 based on the second component recognition image Gc2.

In step S25, the second recognition image Gr2 is learned by machine learning to obtain component recognition image learning data. In step S26, the component recognition image learning data is compared with the first recognition image Gr1 to obtain a similarity value. In step S27, it is determined whether the similarity value is within the second similarity range. If the similarity value is within the second similarity range, the process proceeds to step S28, where it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the first component E1 is normal, and the component recognition image judgment is terminated. If the similarity value is outside the second similarity range, the process proceeds to step S29, where it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the first component E1 is defective, and the component recognition image judgment is terminated.

<Parts inversion recognition image judgment>
Referring to FIG. 26, the component inversion recognition image determination performed by the control device 9 will be described.

As shown in FIG. 26, in step S31, the mounting position of the first component E1, which is determined to be defective when mounted at the mounting position on the printed circuit board Pb, and the first component recognition image Gc1 associated with the mounting position of the first component E1 are acquired. In step S32, the first recognition image Gr1 is acquired by performing the above conversion on the first component recognition image Gc1 based on the first component recognition image Gc1. In step S33, the mounting position of the second component E2, which is determined to be normal when mounted at the mounting position on the printed circuit board Pb, and the second component recognition image Gc2 associated with the mounting position of the second component E2 are acquired. In step S34, the second recognition image Gr2 is acquired by performing the above conversion on the second component recognition image Gc2 based on the second component recognition image Gc2.

In step S35, the second recognition image Gr2 is learned by machine learning to obtain part inverted recognition image learning data. In step S36, the part inverted recognition image learning data is compared with the first recognition image Gr1 to obtain a similarity value. In step S37, it is determined whether the similarity value is within the third similarity range. If the similarity value is within the third similarity range, the process proceeds to step S38, where it is determined that the suction posture of the first part E1 picked up by the nozzle N is normal, and then the part inverted recognition image judgment is terminated. If the similarity value is outside the third similarity range, the process proceeds to step S39, where it is determined that the suction posture of the first part E1 picked up by the nozzle N is defective because it was inverted, and then the part inverted recognition image judgment is terminated.

<Determining the amount of deviation in the pickup position>
The adsorption position deviation amount determination performed by the control device 9 will be described with reference to FIG.

As shown in FIG. 27, in step S41, the mounting position, negative pressure measurement value Pr, and suction position deviation amount Vp of a first component E1, which is determined to be defectively mounted at the mounting position on the printed circuit board Pb, are acquired. In step S42, the mounting position, negative pressure measurement value Pr, and suction position deviation amount Vp of a second component E2, which is determined to be normally mounted at the mounting position on the printed circuit board Pb, are acquired. In step S43, the average value and standard deviation of the negative pressure measurement value Pr of the second component E2 are acquired, and the average value and standard deviation of the suction position deviation amount Vp of the second component E2 are acquired.

In step S44, it is determined whether the suction position deviation amount Vp of the first component E1 is within the first predetermined range and the negative pressure measurement value Pr of the first component E1 is within the second predetermined range. The first predetermined range is a range between a value obtained by subtracting three times the standard deviation of the suction position deviation amount Vp during normal mounting from the average value of the suction position deviation amount Vp during normal mounting and a value obtained by adding three times the standard deviation of the suction position deviation amount Vp during normal mounting to the average value of the suction position deviation amount Vp during normal mounting. The second predetermined range is a range between a value obtained by subtracting three times the standard deviation of the negative pressure measurement value Pr during normal mounting from the average value of the negative pressure measurement value Pr during normal mounting and a value obtained by adding three times the standard deviation of the negative pressure measurement value Pr during normal mounting to the average value of the negative pressure measurement value Pr during normal mounting.

In step S45, it is determined whether the suction position deviation amount Vp of the first part E1 is within a first predetermined range and the negative pressure measurement value Pr of the first part E1 is within a second predetermined range. If the suction position deviation amount Vp of the first part E1 is within the first predetermined range and the negative pressure measurement value Pr of the first part E1 is within the second predetermined range, the process proceeds to step S46, where it is determined that the suction posture of the first part E1 picked up by the nozzle N is normal, and the suction position deviation amount determination is terminated. If the suction position deviation amount Vp of the first part E1 is outside the first predetermined range and/or the negative pressure measurement value Pr of the first part E1 is outside the second predetermined range, the process proceeds to step S47, where it is determined that the suction posture of the first part E1 picked up by the nozzle N is defective, and the suction position deviation amount determination is terminated.

(Effects of the First Embodiment)
In the first embodiment, the following effects can be obtained.

In the first embodiment, as described above, the component mounting system 100 includes a control device 9 that performs control to identify the work of the component mounting device 2 that caused the defect based on a comparison between the first recognition image Gr1 and the second recognition image Gr2. As a result, the control device 9 can automatically identify the work of the component mounting device 2 that caused the mounting defect, thereby reducing the workload of the worker when identifying the cause of the mounting defect. In addition, since the control device 9 can automatically identify the work of the component mounting device 2 that caused the mounting defect, the time required to identify the cause of the mounting defect can be shortened compared to when the worker identifies the cause of the mounting defect. In addition, by automatically identifying the work of the component mounting device 2 that caused the mounting defect in the control device 9, unlike when the worker identifies the work of the component mounting device 2 that caused the mounting defect, the judgment does not differ for each worker, so the variation in judgment when identifying the work of the component mounting device 2 that caused the mounting defect can be suppressed. In addition, since the identified work of the component mounting device 2 can be recorded in the memory unit 94, data on the work of the component mounting device 2 that caused the mounting defect can be accumulated.

In the first embodiment, as described above, the control device 9 is configured to correct the first component recognition image Gc1 obtained by capturing an image of the first component E1 by the imaging unit based on the amount of correction of the mounting position by the nozzle N when mounting the first component E1, and to perform control to identify the work of the component mounting device 2 that caused the mounting failure based on a comparison between the first recognition image Gr1 cut out based on the component size frame Fc that matches the dimensions of the first component E1 and the second recognition image Gr2 based on the imaging unit. As a result, by correcting the first component recognition image Gc1, the first recognition image Gr1 and the second recognition image Gr2 can be aligned, so that the first recognition image Gr1 and the second recognition image Gr2 can be easily compared. As a result, control to identify the work of the component mounting device 2 that caused the mounting failure can be efficiently performed.

In the first embodiment, as described above, the imaging unit has a component imaging unit 22 that images the component E adsorbed to the nozzle N. The control device 9 is configured to determine whether the amount of correction of the mounting position by the nozzle N when mounting the component E at the mounting position of the printed circuit board Pb is normal or not based on a comparison between the first recognition image Gr1, which is obtained by correcting and cutting out the first component recognition image Gc1 captured by the component imaging unit 22, and the second recognition image Gr2 based on the image captured by the component imaging unit 22, and to perform control to identify the cause of the mounting failure as the component recognition work if the amount of correction of the mounting position by the nozzle N is incorrect. As a result, when the cause of the mounting failure is the component recognition work in which the nozzle N cannot correctly correct the mounting position due to a recognition failure of the component E captured by the component imaging unit 22, the cause of the mounting failure can be easily identified as the component recognition work.

In the first embodiment, as described above, the imaging unit has a component imaging unit 22 that images the component E adsorbed on the nozzle N. The control device 9 is configured to determine whether the component E was adsorbed in an inverted state when the nozzle N adsorbed the component E based on a comparison between the first recognition image Gr1, which is obtained by correcting and cutting out the first component recognition image Gc1 imaged by the component imaging unit 22, and the second recognition image Gr2 based on the image taken by the component imaging unit 22, and to perform control to identify the cause of the mounting failure as the component supply operation or the component recognition operation if the component E was adsorbed in an inverted state. This makes it easy to identify the cause of the mounting failure as the component supply operation or the component recognition operation when the cause of the mounting failure is the component supply operation in which the component E was supplied in an inverted state when the component E was supplied, or the component recognition operation in which the inverted component E was erroneously recognized as a normal component E.

In the first embodiment, as described above, the imaging unit includes a board imaging unit 23 that images the FI mark Fi1 (Fi2) for adjusting the position of the board. The control device 9 corrects the first board recognition image Gs1 imaged by the board imaging unit 23 based on a correction amount based on the positional deviation amount of the FI mark Fi1, and determines whether the correction amount based on the positional deviation amount of the FI mark Fi1 is normal or not based on a comparison between the first recognition image Gr1 cut out based on the mark dimension frame Fs aligned with the FI mark Fi1 and the second recognition image Gr2 based on the image taken by the board imaging unit 23. If the correction amount of the mounting position by the nozzle N is defective, the control device 9 is configured to perform control to identify the cause of the mounting defect as the recognition work of the FI mark Fi1. As a result, if the cause of the mounting defect is the recognition work of the FI mark Fi1 that erroneously recognized that the FI mark Fi1 is normal despite being misaligned, the cause of the mounting defect can be easily identified as the recognition work of the FI mark Fi1.

In the first embodiment, as described above, the control device 9 is configured to determine whether the suction state of the nozzle N is normal or not based on the suction position when the nozzle N picks up the component E and the measurement value of the negative pressure of the nozzle N in each case where the mounting of the component E is determined to be defective in the inspection of the board and the mounting of the component E is determined to be normal in the inspection of the board, and to perform control to identify the cause of the mounting defect as the suction work if the suction state of the nozzle N is defective. This makes it possible to easily identify the cause of the mounting defect as the suction work in a case where the cause of the mounting defect is the suction work, in which the measurement value of the negative pressure of the nozzle N is defective due to the component E being picked up by a nozzle N with an incompatible aperture.

In addition, in the first embodiment, as described above, the control device 9 is configured to control the display unit 101 to associate the mounting position of the first component E1 on the board with the work of the component mounting device 2 that caused the defect. This allows the work of the component mounting device 2 that caused the defect corresponding to the mounting position of the first component E1 on the board to be visually displayed on the display unit 101, so that the worker can easily confirm the work of the component mounting device 2 that caused the defect corresponding to the mounting position of the first component E1 on the board.

In the first embodiment, as described above, the control device 9 is configured to control the display of the first recognition image Gr1, the second recognition image Gr2, and the defect information display screen Gf including the work of the component mounting device 2 that caused the defect on the display unit 101 based on receiving an operation to select the mounting position of the first component E1 displayed on the display unit 101. As a result, by checking the defect information display screen Gf displayed on the display unit 101, the worker can check the first recognition image Gr1, the second recognition image Gr2, and the work of the component mounting device 2 that caused the defect, and therefore can easily check records related to the work of the component mounting device 2 that caused the defect.

In the first embodiment, as described above, the component mounting system 100 is provided with a first inspection device 3 (second inspection device 5) that is arranged downstream of the component mounting device 2 and inspects the board on which the component E is mounted by the component mounting device 2. When a first component E1 is determined to be mounted poorly in the inspection by the first inspection device 3 (second inspection device 5), the control device 9 is configured to control the component mounting device 2 to identify the operation that caused the mounting failure based on a comparison between the first recognition image Gr1 and the second recognition image Gr2. This makes it possible to realize a component mounting system 100 that can identify the operation of the component mounting device 2 that caused the mounting failure by utilizing the inspection by the first inspection device 3 (second inspection device 5).

In addition, in the first embodiment, as described above, the component mounting method includes steps S1 to S3 of identifying the operation of the component mounting device 2 that caused the mounting failure based on a comparison between the first recognition image Gr1 and the second recognition image Gr2 when a first component E1 is determined to be a mounting failure in the board inspection. This makes it possible to automatically identify the operation of the component mounting device 2 that caused the mounting failure, thereby providing a component mounting method that can reduce the workload of the worker when identifying the cause of the mounting failure.

[Second embodiment]
The configuration of a component mounting system 200 according to the second embodiment will be described with reference to Figures 28 to 31. In the second embodiment, unlike the first embodiment, a control device 209 predicts whether or not a mounting defect will occur based on a component recognition image Gc captured by a component imaging unit 22 of a component mounting device 2. Note that in the second embodiment, detailed descriptions of the same configuration as in the first embodiment will be omitted.

With reference to Figures 28 to 30, the configuration of the component mounting system 200 according to the second embodiment of the present invention will be described.

(Component mounting system configuration)
28, the component mounting system 200 includes a printing device 1, a component mounting device 2, a first inspection device 3, a reflow furnace 4, a second inspection device 5, an image server 6, an device data server 7, an inspection data server 8, a control device 209, and a client terminal 10. The component mounting device 2 includes a communication unit 25 capable of communicating with the control device 209.

(Control device)
As shown in FIG. 28, the control device 209 is configured to perform control to determine whether or not the amount of correction of the mounting position by the nozzle N is appropriate based on the third component recognition image Gc3 captured by the component imaging unit 22 of the component E being mounted when the component mounting device 2 is mounting the component E on the printed circuit board Pb. The control device 209 is also configured to determine whether or not the amount of correction of the mounting position by the nozzle N is appropriate based on a difference image between the third recognition image Gr3 based on the third component recognition image Gc3 captured by the component imaging unit 22 and the first recognition image Gr1 of the first component E1 of the same type as the component E captured by the component imaging unit 22 to obtain the amount of correction of the mounting position by the nozzle N and the third component recognition image Gc3, and to perform control to stop the component mounting device 2 if the amount of correction of the mounting position by the nozzle N is not appropriate. The third component recognition image Gc3 is an example of a "second component recognition image" in the claims.

Specifically, as shown in FIG. 29, the control device 209 is configured to correct the third component recognition image Gc3 based on the correction amount, and to perform control to acquire the third recognition image Gr3 cut out based on the component dimension frame Fc that is matched to the dimensions of the first component E1 of the same type as the component E imaged by the component imaging unit 22 to acquire the correction amount and the third component recognition image Gc3.

As shown in FIG. 30, the control device 209 is configured to control the acquisition of the mounting position of the first component E1 of the same type from the inspection data server 8, and the acquisition of the first component recognition image Gc1 associated with the mounting position from the image server 6. The control device 209 is configured to control the movement (correction) of the first component recognition image Gc1 based on the amount of correction of the mounting position by the nozzle N when mounting the first component E1. The control device 209 is configured to control the cutting out of the portion of the moved first board recognition image Gs1 that corresponds to the component dimension frame Fc, based on the component dimension frame Fc. This cut-out image is the converted first recognition image Gr1.

The control device 209 is configured to perform control to determine whether the amount of correction of the mounting position by the nozzle N is appropriate based on a comparison between the third recognition image Gr3 and the first recognition image Gr1. Here, if the third recognition image Gr3 and the first recognition image Gr1 match, the amount of correction of the mounting position by the nozzle N is not appropriate, so it is predicted that a mounting failure of the component E will occur. In this case, the control device 209 is configured to perform control to stop the component mounting device 2 if the amount of correction of the mounting position by the nozzle N is not appropriate. Here, if the third recognition image Gr3 and the first recognition image Gr1 do not match, the amount of correction of the mounting position by the nozzle N is appropriate. Such a determination is performed each time the third component recognition image Gc3 is obtained from the component mounting device 2 while performing the FI mark recognition image determination, component recognition image determination, and suction position deviation amount determination of the component mounting method of the first embodiment. Note that other configurations of the second embodiment are similar to those of the first embodiment, so description will be omitted.

(Method for predicting mounting defects)
With reference to FIG. 31, a mounting failure prediction method implemented in the control device 209 will now be described.

31, in step S201, the third component recognition image Gc3 and the correction amount based on the third component recognition image Gc3 are acquired from the component mounting device 2. In step S202, the third component recognition image Gc3 is moved (corrected) by the correction amount based on the third component recognition image Gc3, and then the third recognition image Gr3 is acquired by cutting out the third component recognition image Gc3 moved based on the component dimension frame Fc. In step S203, the first component recognition image Gc1 of the first component E1 of the same type and the correction amount based on the first component recognition image Gc1 are acquired from the inspection data server 8. In step S204, the first component recognition image Gc1 is moved (corrected) by the correction amount based on the first component recognition image Gc1, and then the first recognition image Gr1 is acquired by cutting out the first component recognition image Gc1 moved based on the component dimension frame Fc.

In step S205, a difference image between the third recognition image Gr3 and the first recognition image Gr1 is created. In step S206, it is determined whether the number of pixels in which differences occur in the difference image is within a predetermined range.

If the number of pixels is within the predetermined range, the process proceeds to step S207. Then, in step S207, it is determined that the amount of correction of the mounting position by nozzle N based on the recognition of component E is defective. As a result, there is a high possibility that a mounting defect will occur, and therefore, in step S208, the component mounting device 2 is stopped, and the mounting defect prediction method ends.

If the number of pixels is outside the specified range, proceed to step S209, and after it is determined that the amount of correction of the mounting position by nozzle N based on the recognition of component E is normal, the mounting defect prediction method is terminated.

(Effects of the Second Embodiment)
In the second embodiment, similarly to the first embodiment, the component mounting system 200 includes a control device 209 that performs control to identify the operation of the component mounting device 2 that caused the defect based on a comparison between the first recognition image Gr1 and the second recognition image Gr2. This reduces the workload of the worker when identifying the cause of the mounting defect.

In the second embodiment, as described above, the imaging unit has a component imaging unit 22 that images the component E adsorbed to the nozzle N. The control device 209 is configured to perform control to determine whether the amount of correction of the mounting position by the nozzle N is appropriate based on the amount of correction of the mounting position by the nozzle N based on the third component recognition image Gc3 captured by the component imaging unit 22 of the component E being mounted when the component mounting device 2 is mounting the component E on the printed circuit board Pb, and the third component recognition image Gc3 acquired from the component mounting device 2. As a result, by utilizing the process for identifying the work of the component mounting device 2 that caused the mounting failure, it is possible to predict the risk of occurrence of a mounting failure by determining that the amount of correction of the mounting position by the nozzle N is not appropriate based on the amount of correction of the mounting position and the component recognition image Gc acquired while the component mounting device 2 is mounting the component E on the board.

In the second embodiment, as described above, the control device 209 is configured to determine whether the amount of correction of the mounting position by the nozzle N is appropriate based on the third recognition image Gr3 based on the third component recognition image Gc3 captured by the component imaging unit 22, and the difference image between the amount of correction of the mounting position by the nozzle N and the first recognition image Gr1 of the first component E1 of the same type as the component E captured by the component imaging unit 22 to obtain the third component recognition image Gc3, and to control the component mounting device 2 to stop if the amount of correction of the mounting position by the nozzle N is not appropriate. As a result, it is possible to predict whether a mounting defect will occur by determining whether the amount of correction of the mounting position by the nozzle N is appropriate while mounting the component E on the board using the process for identifying the work of the component mounting device 2 that caused the mounting defect. As a result, if it is predicted that there is a risk of a mounting defect, the component mounting device 2 can be stopped, thereby suppressing the occurrence of a mounting defect. Note that other effects of the second embodiment are similar to those of the first embodiment.

[Third embodiment]
The configuration of a component mounting system 300 according to the third embodiment will be described with reference to Fig. 32 and Fig. 33. In the third embodiment, unlike the first embodiment, the control device 309 is configured to perform control for determining the cause of mounting failure based on the number of mismatched pixels. Note that in the third embodiment, detailed description of the same configuration as in the first embodiment will be omitted.

The component mounting system 300 includes a printing device 1, a component mounting device 2, a first inspection device 3, a reflow furnace 4, a second inspection device 5, an image server 6, an apparatus data server 7, an inspection data server 8, a control device 309, and a client terminal 10. The other configurations of the second embodiment are the same as those of the first embodiment, and therefore will not be described.

(Parts recognition image judgment)
Referring to FIG. 33, the component recognition image determination performed by the control device 309 will be described below.

As shown in FIG. 33, steps S321 to S324 are the same as steps S21 to S24 in the first embodiment, so a description will be omitted. In step S25, a difference image of the second recognition image Gr2 relative to a preset component recognition reference image during normal mounting is created. Here, since there are multiple second recognition images Gr2, multiple difference images are obtained. In step S326, the average and standard deviation of the number of pixels where differences occur in the difference images are obtained. In step S327, a difference image of the first recognition image Gr1 relative to the component recognition reference image is created. In step S328, this difference image is binarized, and the number of mismatched pixels (pixels that are set to 1) is obtained.

In step S329, it is determined whether the number of mismatched pixels is within a predetermined range. Here, the predetermined range is between the average number of difference-occurring pixels minus three times the standard deviation of the number of difference-occurring pixels and the average number of difference-occurring pixels plus three times the standard deviation of the number of difference-occurring pixels. In step S330, it is determined whether the number of mismatched pixels is within a predetermined range. If the number of mismatched pixels is within the predetermined range, the process proceeds to step S331, where it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the first component E1 is normal, and then the component recognition image determination is terminated. If the number of mismatched pixels is outside the predetermined range, the process proceeds to step S332, where it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the first component E1 is defective, and then the component recognition image determination is terminated.

(Effects of the Third Embodiment)
In the third embodiment, similarly to the first embodiment, the component mounting system 300 includes a control device 309 that performs control to identify the operation of the component mounting device 2 that caused the defect based on a comparison between the first recognition image Gr1 and the second recognition image Gr2. This reduces the workload of the worker when identifying the cause of the mounting defect. Other effects of the third embodiment are similar to those of the first embodiment.

[Modification]
It should be noted that the embodiments disclosed herein are illustrative and not restrictive in all respects. The scope of the present invention is indicated by the claims, not by the description of the embodiments above, and further includes all modifications (variations) within the meaning and scope of the claims.

For example, in the above first to third embodiments, the component recognition image judgment and the component inversion recognition image judgment are separate judgments, but the present invention is not limited to this. In the present invention, the component recognition image judgment and the component inversion recognition image judgment may be combined and performed as a single judgment.

That is, a part recognition image determination may be performed as in the first modified example shown in Figure 34.

Specifically, steps S421 to S424 are the same as steps S21 to S24 of the component recognition image judgment in the first embodiment, so the description will be omitted. In step S425, the second recognition image Gr2 is learned by machine learning, and the gradation of the learned second recognition image Gr2 is averaged (flattened). In step S426, the averaged second recognition image Gr2 and the first recognition image Gr1 are compared to obtain a difference pixel ratio, which is the ratio of the number of pixels with a difference. In step S427, it is determined whether the difference pixel ratio is equal to or less than the first threshold value. If the difference pixel ratio is equal to or less than the first threshold value, the process proceeds to step S428, where the amount of correction of the mounting position by the nozzle N based on the recognition of the first component is determined to be normal, and then the component image recognition judgment is terminated. If the difference pixel ratio is equal to or less than the first threshold value, the process proceeds to step S429. In step S429, it is determined whether the difference pixel ratio exceeds the second threshold value. If the difference pixel ratio exceeds the second threshold value, the process proceeds to step S430, in which the first component E1 picked up by the nozzle N was inverted, so it is determined that the pickup posture of the first component E1 is defective, and the component image recognition judgment ends. If the difference pixel ratio is equal to or less than the second threshold value, the process proceeds to step S431, in which it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the first component E1 is defective, and the component image recognition judgment ends.

It is also possible to perform part recognition image judgment as in the second modified example shown in Figure 35.

Specifically, steps S521 to S524 are the same as steps S21 to S24 of the component recognition image judgment in the first embodiment, so the description will be omitted. In step S525, the second recognition image Gr2 is learned by machine learning, and the gradation of the learned second recognition image Gr2 is averaged (flattened). In step S526, the averaged hash value of the second recognition image Gr2 and the hash value of the first recognition image Gr1 are acquired. In step S527, the Hamming distance is acquired based on the acquired averaged hash value of the second recognition image Gr2 and the hash value of the first recognition image Gr1. In step S528, it is determined whether the Hamming distance is equal to or less than the first threshold value. If the Hamming distance is equal to or less than the first threshold value, the process proceeds to step S529, and after it is determined that the amount of correction of the mounting position by the nozzle N based on the recognition of the first component is normal, the component image recognition judgment is terminated. If the difference pixel ratio is equal to or less than the first threshold value, the process proceeds to step S429. In step S429, if the Hamming distance exceeds the first threshold value, the process proceeds to step S530. In step S530, it is determined whether the hash value bit pattern, which is the difference between the hash value of the second recognition image Gr2 and the hash value of the first recognition image Gr1, is the first pattern. If the hash value bit pattern is the first pattern, the process proceeds to step S531, where the first component E1 picked up by the nozzle N is inverted, and the pickup posture of the first component E1 is determined to be poor, and then the component image recognition determination is terminated. If the hash value bit pattern is not the first pattern, the process proceeds to step S532, where the amount of correction of the mounting position by the nozzle N based on the recognition of the first component E1 is determined to be poor, and then the component image recognition determination is terminated.

In the third embodiment, the part recognition image judgment is performed based on the number of pixels, but the present invention is not limited to this. In the present invention, the FI mark recognition image judgment and the part inversion recognition image judgment may also be performed based on the number of pixels.

In the second embodiment, the control device 209 determines whether or not a mounting failure is predicted based on the component recognition image Gc captured by the component imaging unit 22 of the component mounting device 2, but the present invention is not limited to this. In the present invention, the component mounting device may determine whether or not a mounting failure occurs based on the component recognition image captured by the component imaging unit. That is, the control device is configured to acquire multiple second recognition images and control the transmission of learning results based on learning the multiple second recognition images to the component mounting device. In addition, the component mounting device is configured to predict whether or not a mounting failure will occur based on the acquired learning results and the recognition images captured by the imaging unit, and to control the component mounting device to stop if the occurrence of a mounting failure is predicted. As a result, if it is predicted that there is a risk of a mounting failure based on the learning results and the recognition images, the component mounting device can be stopped, so that the occurrence of a mounting failure can be suppressed. In addition, the image server may determine whether or not a mounting failure is predicted based on the component recognition image captured by the component imaging unit.

In addition, in the above first to third embodiments, for convenience of explanation, an example is shown in which the control processing of the control device 9 (209, 309) is explained using a flow-driven flowchart in which processing is performed in order according to a processing flow, but the present invention is not limited to this. In the present invention, the control processing of the control device may be performed by event-driven processing in which processing is performed on an event-by-event basis. In this case, the control processing may be performed completely event-driven, or may be a combination of event-driven and flow-driven.

2 Component mounting device 3 First inspection device (inspection device)
5. Second inspection device (inspection device)
9, 209, 309 Control device 21 Mounting unit 22 Component imaging unit (imaging unit)
23 Substrate imaging unit (imaging unit)
100, 200, 300 Component mounting system 101 Display unit E Component E1 First component E2 Second component Fc Component dimension frame Fi1 FI mark (board recognition mark)
Fi2 FI mark (board recognition mark)
Fs: Mark dimension frame Gc: Part recognition image Gc1: First part recognition image Gc2: Second part recognition image Gc3: Third part recognition image (second part recognition image)
Gf Defect information display screen Gr1 First recognition image Gr2 Second recognition image Gr3 Third recognition image Gs Board recognition image Gs1 First board recognition image Gs2 Second board recognition image N Nozzle P Board

Claims (13)

a component mounting device including a mounting unit that picks up a component by a nozzle and mounts it on a board, and an imaging unit that images the board or the component to obtain a recognition image;
a control device that performs control to identify the operation of the component mounting device that caused the defective mounting when a first component is determined to be defective during inspection of a board on which components are mounted by the component mounting device, based on a comparison between a first recognition image related to the mounting of the first component and a second recognition image related to the mounting of a second component of the same type as the first component that was determined to be properly mounted during inspection. The component mounting system according to claim 1, wherein the control device is configured to correct a first component recognition image of the first component captured by the imaging unit based on a correction amount of the mounting position by the nozzle when mounting the first component, and to perform control to identify the operation of the component mounting device that caused the mounting failure based on a comparison between the first recognition image cut out based on a component dimension frame that matches the dimensions of the first component and the second recognition image captured by the imaging unit. the imaging unit includes a component imaging unit that images the component sucked by the nozzle,
3. The component mounting system according to claim 2, wherein the control device is configured to determine whether or not an amount of correction of the mounting position made by the nozzle when mounting the component at the mounting position on the board is normal based on a comparison between the first recognition image extracted by correcting the first component recognition image captured by the component imaging unit and the second recognition image based on imaging by the component imaging unit, and to perform control to identify that the cause of the mounting defect is the component recognition work if the amount of correction of the mounting position made by the nozzle is defective. the imaging unit includes a component imaging unit that images the component sucked by the nozzle,
3. The component mounting system according to claim 2, wherein the control device is configured to determine whether the component was picked up in an inverted state when picked up by the nozzle based on a comparison between the first recognition image extracted by correcting the first component recognition image captured by the component imaging unit and the second recognition image based on imaging by the component imaging unit, and to perform control to determine that the cause of the mounting defect is the component supply operation or the component recognition operation if the component was picked up in an inverted state. the imaging unit includes a board imaging unit that images a board recognition mark for adjusting a position of the board,
The control device is configured to correct the board recognition image captured by the board imaging unit based on a correction amount based on the positional misalignment amount of the board recognition mark, and to determine whether the correction amount based on the positional misalignment amount of the board recognition mark is normal or not based on a comparison between the first recognition image cut out based on a mark dimension frame aligned with the board recognition mark and the second recognition image captured by the board imaging unit, and if the correction amount of the mounting position by the nozzle is defective, to control the component mounting system to identify the cause of the mounting defect as the recognition work of the board recognition mark. The component mounting system according to claim 1, wherein the control device is configured to determine whether the suction state of the nozzle is normal or not based on the suction position when the nozzle picks up the component and the measurement value of the negative pressure of the nozzle in each case where the component mounting is determined to be defective in the board inspection and the component mounting is determined to be normal in the board inspection, and to perform control to identify the cause of the mounting failure as the suction work if the suction state of the nozzle is defective. the imaging unit includes a component imaging unit that images the component sucked by the nozzle,
2. The component mounting system according to claim 1, wherein the control device is configured to perform control to determine whether or not an amount of correction of the mounting position by the nozzle is appropriate based on a second component recognition image of the component being mounted captured by the component imaging unit when the component is being mounted on a board by the component mounting device, and based on the second component recognition image acquired from the component mounting device. The component mounting system according to claim 7, wherein the control device is configured to determine whether or not the amount of correction of the mounting position by the nozzle is appropriate based on a third recognition image based on the second component recognition image captured by the component imaging unit, and a difference image between the amount of correction of the mounting position by the nozzle and the first recognition image of the first component of the same type as the component captured by the component imaging unit to obtain the second component recognition image, and to control the component mounting device to stop if the amount of correction of the mounting position by the nozzle is not appropriate. the control device is configured to acquire a plurality of the second recognition images and to perform control to transmit a learning result based on learning of the plurality of the second recognition images to the component mounting device;
2. The component mounting system according to claim 1, wherein the component mounting device is configured to predict whether or not a mounting defect will occur based on the acquired learning result and the recognition image captured by the imaging unit, and to perform control to stop the component mounting device if a mounting defect is predicted to occur. The component mounting system according to claim 1, wherein the control device is configured to control the display unit to display the mounting position of the first component on the board in association with the operation of the component mounting device that caused the mounting failure. The component mounting system according to claim 10, wherein the control device is configured to control the display device to display the first recognition image, the second recognition image, and a defect information display screen including the operation of the component mounting device that caused the mounting defect, based on receiving an operation to select the mounting position of the first component displayed on the display device. Further, an inspection device is provided downstream of the component mounting device, and inspects a board on which components are mounted by the component mounting device.
2. The component mounting system according to claim 1, wherein the control device is configured to, when a first component is determined to have a mounting defect during inspection by the inspection device, perform control to identify the operation of the component mounting device that caused the mounting defect based on a comparison between the first recognition image and the second recognition image. Mounting components on a board by a mounting unit of a component mounting device;
inspecting the board on which the components are mounted by the mounting unit;
a step of, when a first component is determined to be mounted poorly during inspection of a board, identifying an operation of the component mounting device that caused the mounting failure based on a comparison between a first recognition image related to the mounting of the first component and a second recognition image related to the mounting of a second component of the same type as the first component determined to be mounted properly during inspection.

Images