WO2020183735A1 - Quality determination device and quality determination method - Google Patents

Abstract

This quality determination device is equipped with a storage unit, an acquisition unit, and a quality determination unit. The distribution of a plurality of base feature amounts at the time of extraction, from each base image, of a base feature amount that is an overall image feature amount is set as a feature amount distribution. At such time, the storage unit associates together and stores, in a storage device, image-related information that is at least one among the base images and the feature amount distribution, and search information that enables searching of the image-related information and includes the acquisition conditions at the time of acquisition of the base images. The acquisition unit acquires, from the storage device, image-related information that is associated with the search information that includes acquisition conditions matching the acquisition conditions of an object image captured of an object during substrate work by an imaging device of a substrate work machine. The quality determination unit uses the image-related information acquired by the acquisition unit to determine the quality of substrate work performed at the time of acquisition of the object image.

Description

Good / bad judgment device and good / bad judgment method

This specification discloses a technique relating to a quality determination device and a quality determination method.

In the image processing system described in Patent Document 1, the control device of each component mounting machine learns whether or not the learning result regarding the component supplied from the feeder at the start of production is stored in the storage device of the learning computer. Contact your computer. Then, when the learning result is stored in the storage device, the control device fetches the learning result from the storage device and performs the learning type super-resolution processing using the learning result from the beginning of production.

The visual inspection support system described in Patent Document 2 refers to the accuracy of inspection data among a plurality of visual inspection devices. Specifically, the visual inspection support system includes an evaluation unit, a grade data holding unit, and a grade providing unit. The evaluation unit evaluates the inspection accuracy based on the inspection data for each part based on the results of the inspection using the inspection data. The grade data holding unit holds information on the evaluated inspection accuracy as grade data for each part. The grade providing unit provides grade data to other inspection devices connected via a network.

Japanese Unexamined Patent Publication No. 2018-097731 JP-A-2002-277407

However, in the image processing system described in Patent Document 1, the learning result to be searched is related to the relationship between the low-resolution image and the high-resolution image, and does not judge the quality of the board-to-board work of the board-to-board working machine. .. Further, the visual inspection support system described in Patent Document 2 refers to information on inspection accuracy among a plurality of visual inspection devices.

In view of such circumstances, the present specification acquires image-related information which is at least one of the reference image and the feature amount distribution, and uses the acquired image-related information to determine the quality of the work on the substrate. A pass / fail judgment device and a pass / fail judgment method capable of this are disclosed.

This specification discloses a quality determination device including a storage unit, an acquisition unit, and a quality determination unit. Here, a plurality of images taken by the image pickup device of the board-to-board work machine that performs a predetermined work on the substrate with the same image-taking conditions for the same type of object in the board-to-board work are used as reference images. The distribution of a plurality of the reference feature amounts when the reference feature amount, which is the feature amount of the entire image, is extracted for each image of the reference image is defined as the feature amount distribution. At this time, the storage unit acquires image-related information that is at least one of the reference image and the feature amount distribution, and information that can search the image-related information when the reference image is acquired. The search information including the condition is associated and stored in the storage device. The acquisition unit is the image-related information associated with the search information including acquisition conditions that match the acquisition conditions of the target image in which the object is imaged by the image pickup device of the board-to-board work machine. Is obtained from the storage device. The quality determination unit determines the quality of the work on the substrate when the target image is acquired by using the image-related information acquired by the acquisition unit.

Further, the present specification discloses a quality determination method including a storage process, an acquisition process, and a quality determination process. Here, a plurality of images taken by the image pickup device of the board-to-board work machine that performs a predetermined work on the substrate with the same image-taking conditions for the same type of object in the board-to-board work are used as reference images. The distribution of a plurality of the reference feature amounts when the reference feature amount, which is the feature amount of the entire image, is extracted for each image of the reference image is defined as the feature amount distribution. At this time, the storage step acquires the image-related information that is at least one of the reference image and the feature amount distribution, and the information that can search the image-related information when the reference image is acquired. The search information including the condition is associated and stored in the storage device. The acquisition step is the image-related information associated with the search information including acquisition conditions that match the acquisition conditions of the target image in which the object is imaged by the image pickup device of the board-to-board work machine. Is obtained from the storage device. In the quality determination step, the quality of the work on the substrate when the target image is acquired is determined by using the image-related information acquired by the acquisition process.

According to the above-mentioned pass / fail judgment device, it has a storage unit, an acquisition unit, and a pass / fail judgment unit. As a result, the quality determination device can acquire image-related information and determine the quality of the work on the substrate using the acquired image-related information. The same can be said for the pass / fail determination device for the pass / fail determination method.

It is a block diagram which shows the structural example of the to-substrate work line WML. It is a top view which shows the structural example of the component mounting machine WM3. It is a block diagram which shows an example of the control block of the quality determination apparatus 40. It is a bottom view which shows an example of a part 91. FIG. 5 is a schematic view showing an example of an image of the component 91 shown in FIG. 4A held by the holding member 30 taken by the component camera 14. It is a schematic diagram which shows an example of a feature amount distribution FD1. It is a histogram which shows an example of the relationship between the work result of the work on a substrate and the Mahalanobis distance MD1. It is a flowchart which shows an example of the control procedure by a change part 44. It is a schematic diagram which shows an example of the image-related information 50a, the search information 50b, and the work result information 50c stored in the storage device 50m. It is a schematic diagram which shows the arrangement example of the quality determination apparatus 40.

1. 1. Embodiment 1-1. Configuration example of the substrate work line WML In the substrate work line WML, a predetermined substrate work is performed on the substrate 90. The type and number of anti-board work machines WM constituting the anti-board work line WML are not limited. As shown in FIG. 1, the board-to-board work line WML of this embodiment is a plurality (five) board-to-board work of a printing machine WM1, a printing inspection machine WM2, a component mounting machine WM3, a reflow furnace WM4, and an appearance inspection machine WM5. The machine WM is provided, and the substrate 90 is conveyed in this order by a substrate conveying device (not shown).

The printing machine WM1 prints solder on the substrate 90 at each mounting position of the plurality of parts 91. The printing inspection machine WM2 inspects the printing state of the solder printed by the printing machine WM1. The component mounting machine WM3 mounts a plurality of components 91 on the substrate 90 (on the solder printed by the printing machine WM1). The number of component mounting machines WM3 may be one or more. When a plurality of component mounting machines WM3 are provided, the plurality of component mounting machines WM3 can share and mount the plurality of parts 91.

The reflow furnace WM4 heats the substrate 90 on which a plurality of parts 91 are mounted by the parts mounting machine WM3, melts the solder, and performs soldering. The visual inspection machine WM5 inspects the mounting state of a plurality of parts 91 mounted by the component mounting machine WM3. In this way, the board-to-board work line WML uses a plurality of (five) board-to-board work machines WM to sequentially convey the boards 90 and execute a production process including an inspection process to produce the board product 900. Can be done. In addition, the work line to board WML includes, for example, a work machine WM such as a function inspection machine, a buffer device, a substrate supply device, a substrate reversing device, a shield mounting device, an adhesive coating device, and an ultraviolet irradiation device as necessary. You can also prepare.

The plurality of (five) anti-board work machines WM and the management device WMC constituting the anti-board work line WML are electrically connected by the communication unit LC. The communication unit LC can be connected to be communicable by wire, and can also be connected to be communicable by wireless. In addition, various communication methods can be adopted. In the present embodiment, a premises information communication network (LAN: Local Area Network) is configured by a plurality of (five) anti-board work machines WM and management device WMC. As a result, the plurality (five) anti-board working machines WM can communicate with each other via the communication unit LC. Further, the plurality (five) anti-board working machines WM can communicate with the management device WMC via the communication unit LC.

The management device WMC controls a plurality of (five) anti-board work machines WM constituting the anti-board work line WML, and monitors the operating status of the anti-board work line WML. The management device WMC stores various control data for controlling a plurality of (five) anti-board working machines WM. The management device WMC transmits control data to each of a plurality (five) anti-board working machines WM. Further, each of the plurality (five) anti-board working machines WM transmits the operating status and the production status to the management device WMC.

For example, a data server 70 can be provided in the management device WMC. The data server 70 can store, for example, the acquired data acquired by the board-to-board work machine WM regarding the board-to-board work. For example, various image data captured by the anti-board working machine WM are included in the acquired data. The record (log data) of the operating status acquired by the board working machine WM is included in the acquired data.

The data server 70 can also store various production information related to the production of the substrate 90. For example, component data such as information on the shape of each type of component 91, information on electrical characteristics, and information on how to handle the component 91 is included in the production information. The inspection results by inspection machines such as the print inspection machine WM2 and the appearance inspection machine WM5 are included in the production information.

1-2. Configuration example of the component mounting machine WM3 The component mounting machine WM3 mounts a plurality of components 91 on the substrate 90. As shown in FIG. 2, the component mounting machine WM3 includes a board transfer device 11, a component supply device 12, a component transfer device 13, a component camera 14, a board camera 15, and a control device 16.

The substrate transfer device 11 is composed of, for example, a belt conveyor or the like, and conveys the substrate 90 in the transfer direction (X-axis direction). The substrate 90 is a circuit board, and at least one of an electronic circuit and an electric circuit is formed. The board transfer device 11 carries the board 90 into the component mounting machine WM3 and positions the board 90 at a predetermined position in the machine. The board transfer device 11 carries out the board 90 out of the component mounting machine WM3 after the mounting process of the plurality of components 91 by the component mounting machine WM3 is completed.

The component supply device 12 supplies a plurality of components 91 mounted on the substrate 90. The component supply device 12 includes a plurality of feeders 121 provided along the transport direction (X-axis direction) of the substrate 90. Each of the plurality of feeders 121 is pitch-feeded with a carrier tape (not shown) in which the plurality of parts 91 are housed, so that the parts 91 can be collected at a supply position located on the tip side of the feeder 121. Further, the component supply device 12 can also supply electronic components (for example, lead components) that are relatively large in size as compared with chip components and the like in a state of being arranged on the tray.

The parts transfer device 13 includes a head drive device 131 and a moving table 132. The head drive device 131 is configured to be able to move the movable table 132 in the X-axis direction and the Y-axis direction by a linear motion mechanism. A mounting head 20 is detachably (replaceable) provided on the moving table 132 by a clamp member (not shown). The mounting head 20 uses at least one holding member 30 to collect and hold the component 91 supplied by the component supply device 12, and mounts the component 91 on the substrate 90 positioned by the substrate transfer device 11. As the holding member 30, for example, a suction nozzle or a chuck can be used.

A known imaging device can be used for the component camera 14 and the substrate camera 15. The component camera 14 is fixed to the base of the component mounting machine WM3 so that the optical axis is upward in the Z-axis direction (vertical upward direction). The component camera 14 can take an image of the component 91 held by the holding member 30 from below. The substrate camera 15 is provided on the moving table 132 of the component transfer device 13 so that the optical axis faces downward (vertically downward) in the Z-axis direction. The substrate camera 15 can image the substrate 90 from above. The component camera 14 and the substrate camera 15 perform imaging based on a control signal transmitted from the control device 16. Image data captured by the component camera 14 and the board camera 15 is transmitted to the control device 16.

The control device 16 includes a known arithmetic unit and a storage device, and constitutes a control circuit (both are not shown). Information, image data, and the like output from various sensors provided in the component mounting machine WM3 are input to the control device 16. The control device 16 sends a control signal to each device based on a control program and a predetermined mounting condition set in advance.

For example, the control device 16 causes the board camera 15 to image the board 90 positioned by the board transfer device 11. The control device 16 processes the image captured by the substrate camera 15 to recognize the positioning state of the substrate 90. Further, the control device 16 causes the holding member 30 to collect and hold the component 91 supplied by the component supply device 12, and causes the component camera 14 to image the component 91 held by the holding member 30. The control device 16 processes the image captured by the component camera 14 to recognize the suitability of the component 91 and the holding posture of the component 91.

The control device 16 moves the holding member 30 toward the upper side of the expected mounting position preset by a control program or the like. Further, the control device 16 corrects the planned mounting position based on the positioning state of the board 90, the holding posture of the component 91, and the like, and sets the mounting position where the component 91 is actually mounted. The planned mounting position and the mounting position include the rotation angle in addition to the position (X-axis coordinate and Y-axis coordinate).

The control device 16 corrects the target position (X-axis coordinate and Y-axis coordinate) and the rotation angle of the holding member 30 in accordance with the mounting position. The control device 16 lowers the holding member 30 at the corrected rotation angle at the corrected target position, and mounts the component 91 on the substrate 90. The control device 16 executes a mounting process of mounting the plurality of components 91 on the substrate 90 by repeating the above pick-and-place cycle.

1-3. Configuration example of the quality determination device 40 For example, as described above, the component mounting machine WM3 processes the image captured by the component camera 14 to recognize the suitability of the component 91 and the holding posture of the component 91. .. At this time, for example, based on the feature amount (for example, the color, outer shape, area, etc.) of the component 91 extracted from the image, the component 91 is held, such as the suitability of the component 91 and the holding posture of the component 91. When trying to judge the quality of work, there is a possibility of erroneous judgment due to omission of judgment conditions.

For example, it is assumed that the regular part 91 has a white circular shape, and the area of the part 91 (area of a circle) is a predetermined value. At this time, if the component mounting machine WM3 sets only the color and area of the component 91 as determination conditions, for example, the color of the component 91 is white, the area is a predetermined value, but the external shape of the component 91 is quadrangular. , It is not possible to determine the suitability of the component 91. Further, if an attempt is made to reduce erroneous determination, the number of determination conditions to be set increases, and the work of setting the determination conditions becomes complicated.

Therefore, in the present embodiment, the quality determination device 40 is provided. The quality determination device 40 determines the quality of the work on the substrate based on the feature amount of the entire image of each image captured by the image pickup device 80 of the substrate work machine WM. The pass / fail determination device 40 includes a distribution acquisition unit 41 and a pass / fail determination unit 42 when regarded as a control block. It is preferable that the quality determination device 40 further includes at least one of the threshold value setting unit 43 and the change unit 44.

As shown in FIG. 3, the quality determination device 40 of the present embodiment includes a distribution acquisition unit 41, a quality determination unit 42, a threshold value setting unit 43, and a change unit 44. Further, although the quality determination device 40 of the present embodiment is provided in the control device 16 of the component mounting machine WM3, it can also be provided in another board-to-board work machine WM. Further, the quality determination device 40 can be provided outside the board working machine WM (for example, the management device WMC).

1-3-1. Distribution acquisition unit 41
The distribution acquisition unit 41 extracts the reference feature amount BF1 which is the feature amount of the entire image for each image of the reference image BP1 and acquires the feature amount distribution FD1 which is the distribution of the extracted plurality of reference feature amounts BF1. The reference image BP1 refers to a plurality of images taken by the image pickup device 80 of the board-to-board work machine WM that performs a predetermined work on the board 90 with the same type of imaging conditions for the same type of object in the work on the board.

When the anti-board working machine WM is the component mounting machine WM3, the image pickup device 80 is, for example, the component camera 14, and the object at this time is the component 91 held by the holding member 30. In this case, the board-to-board work is the holding work of the component 91. Further, the reference image BP1 may be an image captured under the same imaging conditions for the component 91 having the same component type. The imaging conditions include, for example, the type of light source, the irradiation direction of light, the exposure time, the aperture value, and the like. It should be noted that, for example, it is difficult to completely match the imaging conditions due to the influence of natural light or the like, so the imaging conditions may be any acquisition conditions that can be specified by the substrate working machine WM (component mounting machine WM3).

The reference image BP1 may be, for example, an image in which the region of each component 91 is extracted from one image simultaneously captured by the component camera 14 on a plurality of components 91 held by the rotary head or the line head. Further, the reference image BP1 may be an image obtained by sequentially capturing a plurality of parts 91 by the component camera 14. Further, the reference image BP1 may be an image in which these images are mixed.

When the board-to-board working machine WM is the component mounting machine WM3, the image pickup device 80 may be, for example, a side camera (not shown) provided on the mounting head 20. The side camera is different from the component camera 14 in that the component 91 is imaged from the side surface side of the component 91, but the same can be said for the above.

Further, when the board working machine WM is the component mounting machine WM3, the image pickup device 80 may be, for example, the board camera 15. The object at this time is the substrate 90 positioned by the substrate transfer device 11. In this case, the board-to-board work is the positioning work of the board 90. Further, when the substrate working machine WM is the appearance inspection machine WM5, the image pickup device 80 is a camera (not shown) provided in the appearance inspection machine WM5. The object at this time is the component 91 mounted on the substrate 90. In this case, the board-to-board work is the mounting work of the component 91 by the component mounting machine WM3.

Further, when the substrate working machine WM is the printing inspection machine WM2, the image pickup device 80 is a camera (not shown) provided in the printing inspection machine WM2. The object at this time is the solder printed on the substrate 90. In this case, the board-to-board work is a solder printing work by the printing machine WM1. As described above, the substrate working machine WM, the image pickup apparatus 80, the object, and the substrate working are not limited. Moreover, what has been described about the imaging conditions can be said in the same manner in any of the above cases. For convenience of explanation, this specification describes a reference image BP1 in which the component camera 14 of the component mounting machine WM3 captures the component 91 held by the holding member 30.

FIG. 4A is a bottom view showing an example of the component 91. FIG. 4B shows an example of an image of the component 91 shown in FIG. 4A held by the holding member 30 taken by the component camera 14. In FIG. 4B, a state in which a plurality of pixels constituting the image are arranged in a grid pattern is also shown. For example, when the component 91 is a chip resistor, a chip capacitor, or the like, the component 91 includes an electrode region AR11 and an electrode region AR12, which are regions of the electrode portion, and a main body region AR13, which is a region of the main body portion.

The electrode region AR11 and the electrode region AR12 are silver. Further, it is assumed that the main body region AR13 on the back surface (bottom surface) side of the component 91 is, for example, white, and the main body region AR13 on the front surface side of the component 91 is, for example, black. At this time, for example, if the holding member 30 mistakenly sucks the back surface (bottom surface) side of the component 91 (reversal of the component 91), the component camera 14 images the front surface side of the component 91.

In this case, the electrode region AR11 and the electrode region AR12 are silver, and the main body region AR13 on the surface side is black. Therefore, the difference in brightness between these regions is that the holding member 30 adsorbs the surface side of the component 91. Compared with the case (in the normal adsorption state, the main body region AR13 to be imaged is white), it is larger. In this way, when at least one of the brightness and the color of the component 91 in the image changes, the characteristics of the entire image change as compared with the case of the normal adsorption state.

Further, for example, if the holding member 30 mistakenly sucks the end portion of the component 91 (standing suction of the component 91), the holding posture of the component 91 changes as compared with the normal suction state. Therefore, in the image, the electrode region AR11, The shapes (rectangular in FIG. 4B) of each of the electrode region AR12 and the main body region AR13 are deformed. Further, the area of each region of the component 91 changes as compared with the normal suction state. Further, the length of the outer circumference of each region of the component 91 changes as compared with the normal suction state.

In this way, when at least one of the shape and the area of the shape of the component 91 in the image changes, the characteristics of the entire image change as compared with the case of the normal suction state. The shape of each region is formed by at least one of the brightness and color of a plurality of pixels constituting the image. Further, the inversion of the component 91 and the standing suction of the component 91 are examples of cases where the work result of the work on the substrate (in this case, the holding work of the component 91) is defective, and the cause of the defect is not limited to these. Further, the chip resistor and the chip capacitor are examples of the component 91, and the component 91 is not limited thereto.

As described above, when the work result of the work on the board is poor, the characteristics of the entire image change as compared with when the work result is good. Therefore, the distribution acquisition unit 41 extracts the reference feature amount BF1 which is the feature amount of the entire image for each image of the reference image BP1 and acquires the feature amount distribution FD1 which is the distribution of the extracted plurality of reference feature amounts BF1. To do. As shown in FIG. 4B, the outer edge of the reference image BP1 is aligned with the outer edge of the object (for example, one component 91). Further, as described above, the feature amount of the entire image is preferably at least one of the brightness and color of a plurality of pixels constituting the image, the shape formed by these, and the area of the shape. Is.

The distribution acquisition unit 41 can acquire the feature quantity distribution FD1 by, for example, a method known in multivariate analysis (for example, principal component analysis). FIG. 5 shows an example of the feature amount distribution FD1. In the figure, the brightness and color of a plurality of pixels constituting the image shown in FIG. 4B, the shape formed by these (rectangle in the example shown in FIG. 4B), and the area of the shape are defined as the feature amount of the entire image. An example of the feature amount distribution FD1 at the time of

Further, the plurality of points shown in the feature region FR1 of FIG. 5 image the reference feature amount BF1 of each image of the reference image BP1. As shown in the figure, the feature amount distribution FD1 can be represented by a two-dimensional feature region FR1 or a three-dimensional or higher feature region FR1. The feature region FR1 indicates the outer edge of a plurality of reference feature quantities BF1 and is also referred to as “unit space”.

Note that the work on the board by the board-to-board work machine WM usually has good work results, and the rate of poor work results is extremely low. For example, the adsorption rate of the component 91 in the component mounting machine WM3 is extremely high. Therefore, even if the reference image BP1 includes an image when the work result of the work on the substrate is poor, the influence is small. Therefore, in the present specification, the work result of the work on the substrate when the reference image BP1 is acquired is treated as being good. In order to improve the determination accuracy by the quality determination device 40, the reference image BP1 may be limited to the image when the work result of the work on the substrate is good.

1-3-2. Good / bad judgment unit 42
The quality determination unit 42 extracts the target feature amount OF1 which is the feature amount of the entire image for each image of the target image OP1, and is based on the degree of deviation of the target feature amount OF1 with respect to the feature area FR1 defined by the feature amount distribution FD1. Then, the quality of the work on the substrate when the target image OP1 is acquired is determined.

The target image OP1 is at least one image acquired after the reference image BP1 and refers to an image related to the reference image BP1. The target image OP1 is related to the reference image BP1 when the acquisition conditions that can be defined by the substrate working machine WM match the acquisition of the reference image BP1, and the acquisition conditions are the type of the object and the type of the image pickup apparatus 80. , And, it is preferable to include at least the type of the object among the types of imaging conditions of the image pickup apparatus 80.

In order to compare the reference image BP1 and the target image OP1, the type of the object when the reference image BP1 is acquired (in the example shown in FIG. 4B, the component type of the part 91) and the type when the target image OP1 is acquired. The type of the object (part type of the part 91) must at least match. Further, if at least one of the type and imaging conditions of the imaging device 80 (component camera 14 in the example shown in FIG. 4B) is different, there is a possibility that the acquired image will be different even if the same object is imaged. is there. Therefore, it is preferable that at least one of the type of the image pickup apparatus 80 and the image pickup condition is the same.

As described above for the reference image BP1, the imaging conditions include, for example, the type of light source, the irradiation direction of light, the exposure time, the aperture value, and the like. Further, for example, since it is difficult to completely match the imaging conditions due to the influence of natural light or the like, the imaging conditions may be any acquisition conditions that can be specified by the substrate working machine WM.

The quality determination unit 42 can extract the target feature amount OF1 which is the feature amount of the entire image for each image of the target image OP1 in the same manner as the distribution acquisition unit 41. Then, the quality determination unit 42 determines the quality of the work on the substrate when the target image OP1 is acquired, based on the degree of deviation of the target feature amount OF1 with respect to the feature region FR1 defined by the feature amount distribution FD1.

The quality determination unit 42 can determine the quality of the work on the substrate when the target image OP1 is acquired by using various known methods. The quality determination unit 42 can use, for example, a method such as a neural network, deep learning, or a support vector. Further, the quality determination unit 42 can also use a method such as the Mahalanobis Taguchi method or the subspace method. The Mahalanobis Taguchi method requires a smaller number of reference image BP1s corresponding to training data than an artificial intelligence method such as a neural network.

Therefore, it is preferable that the quality determination unit 42 determines the quality of the work on the substrate based on whether or not the Mahalanobis distance MD1 from the feature region FR1 exceeds a predetermined threshold value TH1. Further, it is preferable that the quality determination unit 42 determines the quality of the work on the substrate by the RT method, which is one method of the Mahalanobis Taguchi method. The RT method (Recognition Taguchi method) is suitable for good / bad judgment using image data, and is suitable for use in the present embodiment.

The quality determination unit 42 of the present embodiment determines the quality of the work on the substrate when the target image OP1 is acquired by the RT method. The quality determination unit 42 determines the quality of the work on the substrate based on whether or not the Mahalanobis distance MD1 from the feature region FR1 exceeds a predetermined threshold value TH1. Specifically, the quality determination unit 42 determines that the work result of the work on the substrate when the target image OP1 is acquired is good when the Mahalanobis distance MD1 is equal to or less than the predetermined threshold value TH1. Further, the quality determination unit 42 determines that the work result of the work on the substrate when the target image OP1 is acquired is defective when the Mahalanobis distance MD1 is larger than the predetermined threshold value TH1.

As described above, for example, the main body region AR13 shown in FIG. 4B is white if the work result of the work on the substrate (in this case, the holding work of the component 91) is good (if it is in a normal suction state). is there. When the color of the main body region AR13 is white in most of the images, for example, when the component 91 is inverted and the color of the main body region AR13 becomes black, the Mahalanobis distance MD1 shown in FIG. 5 is significantly increased. Therefore, the quality determination unit 42 can determine that the work result of the work on the board (holding work of the component 91) is defective. The same can be said for other features.

1-3-3. Threshold setting unit 43
The threshold setting unit 43 sets the threshold TH1 of the Mahalanobis distance MD1. It is preferable that the threshold value setting unit 43 sets the threshold value TH1 by using the outlier information acquired by the Smirnov-Grabs test. The Smirnov-Grabs test detects outliers in a data when it follows a normal distribution. The threshold value setting unit 43 of the present embodiment detects an outlier by the Smirnov-Grabs test, and sets the outlier as the threshold value TH1 of the Mahalanobis distance MD1. As a result, the threshold value setting unit 43 can easily set the threshold value TH1 of the Mahalanobis distance MD1.

Further, the threshold value setting unit 43 is used when the pass / fail determination unit 42 determines that the work with the substrate is good and the actual work with the substrate is poor, and the threshold setting unit 43 determines that the work with the substrate is good. The threshold TH1 can be modified so that the threshold TH1 is smaller than the Mahalanobis distance MD1 of.

For example, it is assumed that the quality determination unit 42 determines that the work on the substrate is good based on the target feature amount OF1 shown in FIG. When the actual work on the substrate is poor, the threshold setting unit 43 corrects the threshold TH1 so that the threshold TH1 is smaller than the Mahalanobis distance MD1 of the target feature amount OF1. As a result, the threshold value setting unit 43 can optimize the threshold value TH1 of the Mahalanobis distance MD1.

The threshold value setting unit 43 uses the Mahalanobis of the target feature amount OF1 used when the pass / fail determination unit 42 determines that the work with the substrate is defective and the actual work with the substrate is good, and determines that the work with the substrate is defective. The threshold TH1 can also be modified so that the threshold TH1 is larger than the distance MD1. In this case as well, the threshold setting unit 43 can optimize the threshold TH1 of the Mahalanobis distance MD1.

Further, the threshold value setting unit 43 uses the target feature amount OF1 used when the pass / fail determination unit 42 determines that the work on the substrate is good when the number of times the work on the substrate is determined to be defective reaches a predetermined number of times. The threshold value TH1 can be corrected by the discriminant analysis method using the target feature amount OF1 used when the work on the substrate is judged to be defective. As a result, the threshold value setting unit 43 can optimize the threshold value TH1 of the Mahalanobis distance MD1.

FIG. 6 is a histogram showing an example of the relationship between the work result of the work on the substrate and the Mahalanobis distance MD1. The horizontal axis of the figure shows the Mahalanobis distance MD1, and the vertical axis shows the frequency of the work result of the work on the substrate. The data D1 to the data D13 are the frequencies of the data when the work result of the work on the substrate is good. The data D14 to the data D23 are the frequencies of the data when the work result of the work on the substrate is defective.

The work result of the work on the board shows the actual work result (good or bad). For example, it is assumed that the holding work of the component 91 shown in FIG. 4A is good. At this time, the Mahalanobis distance MD1 is calculated based on the image (image shown in FIG. 4B) of the component 91 held by the holding member 30 taken by the component camera 14. As a result, at a distance corresponding to the Mahalanobis distance MD1, one data is acquired when the work result of the work on the substrate is good. By repeating this, the histogram shown in the figure can be obtained.

The predetermined number of times is the number of data when the work result of the work on the board required in the discriminant analysis method is defective, and can be set arbitrarily. When the number of times that the quality determination unit 42 determines that the work on the substrate is defective reaches a predetermined number of times, a predetermined number of data when the work result of the work on the substrate is defective is accumulated. At this time, the threshold value setting unit 43 determines that the target feature amount OF1 used when the work with the board is good (data D1 to data D13) and the work with the board is bad (data D14 to data D23). The threshold value TH1 is corrected by the discriminant analysis method using the target feature amount OF1 used in. The threshold value TH1 shown in FIG. 6 indicates that the threshold value TH1 has been modified by the discriminant analysis method. The methods themselves such as the Mahalanobis-Taguchi method, the RT method, the Smirnov-Grabs test, and the discriminant analysis method are known, and detailed description thereof is omitted in the present specification.

1-3-4. Change part 44
Even if the component type of the component 91 is the same, if the vendor (manufacturer) is different, the external shape, external dimensions, color, etc. of the component 91 may be slightly different. Therefore, it is preferable that the quality determination device 40 further includes a change unit 44.

The change unit 44 switches the feature amount distribution FD1 used by the pass / fail determination unit 42 to the feature amount distribution FD1 that matches the part 91 after the vendor change at the timing when the vendor of the part 91 supplied to the component mounting machine WM3 is switched. .. However, it is assumed that the vendor of the component 91 supplied to the component mounting machine WM3 is changed, and the feature quantity distribution FD1 that matches the vendor-changed component 91 is held by the quality determination device 40. As a result, the quality determination unit 42 can determine the quality of the work on the substrate by using the appropriate feature amount distribution FD1 regardless of the change of the vendor of the component 91.

Specifically, the change unit 44 determines whether or not it is the timing when the vendor of the component 91 is switched (step S11 shown in FIG. 7). The change unit 44 can know the timing at which the vendor of the component 91 is switched from, for example, the identification information of the feeder 121. At the vendor switching timing (in the case of “Yes” in step S11), the changing unit 44 switches the feature amount distribution FD1 used by the quality determination unit 42 to the feature amount distribution FD1 that matches the component 91 after the vendor change. (Step S12). Then, the control ends once. When it is not the vendor switching timing (when “No” in step S11), the control is temporarily terminated without executing the process shown in step S12.

1-4. Configuration example of the quality determination device 40 when the substrate working machine WM does not hold the image-related information 50a Here, at least one of the reference image BP1 and the feature amount distribution FD1 is defined as the image-related information 50a. As described above, the reference image BP1 is a plurality of images captured by the image pickup device 80 of the board-to-board work machine WM that performs a predetermined anti-board work on the board 90 with the same type of imaging conditions for the same type of object in the anti-board work. To say.

Further, the feature amount distribution FD1 refers to the distribution of a plurality of reference feature amounts BF1 when the reference feature amount BF1 which is the feature amount of the entire image is extracted for each image of the reference image BP1. It is preferable that the feature amount of the entire image is at least one of the brightness and color of a plurality of pixels constituting the image, the shape formed by these, and the area of the shape.

When the anti-board working machine WM does not hold the image-related information 50a, it is difficult for the anti-board working machine WM to judge the quality of the anti-board work when the image is acquired by using the image-related information 50a. is there. Therefore, the quality determination device 40 of the present embodiment acquires the image-related information 50a and determines the quality of the work on the substrate by using the acquired image-related information 50a. The pass / fail determination device 40 includes a storage unit 51, an acquisition unit 52, and a pass / fail determination unit 42 when regarded as a control block.

It is preferable that the quality determination device 40 further includes at least one of the providing unit 53, the distribution acquisition unit 41, the threshold value setting unit 43, and the changing unit 44. As shown in FIG. 3, the quality determination device 40 of the present embodiment includes a storage unit 51, an acquisition unit 52, a provision unit 53, a distribution acquisition unit 41, a quality determination unit 42, a threshold value setting unit 43, and the like. It is provided with a change unit 44.

1-4-1. Memory 51
The storage unit 51 stores the image-related information 50a and the search information 50b in the storage device 50m in association with each other. As described above, the image-related information 50a is at least one of the reference image BP1 and the feature amount distribution FD1. The search information 50b is information that can search the image-related information 50a, and includes acquisition conditions when the reference image BP1 is acquired. The search information 50b includes, for example, information such as a management number, a name, a creation date, a data capacity, and a creator of the image-related information 50a. As described above, the acquisition condition is a condition that can be specified by the substrate working machine WM, and is at least the object of the information about the object, the information about the image pickup device 80, and the information about the image pickup condition of the image pickup device 80. It is preferable to include information about.

For example, when the object is a part 91, the information about the object includes, for example, information about the part type, model, vendor (manufacturer), production lot, etc. of the part 91. Further, the information about the object may include, for example, information about the shape of the part 91, information about how to handle the part 91, and the like. The information about the image pickup apparatus 80 includes, for example, information about the model of the image pickup apparatus 80, the vendor (manufacturer), and the like. The imaging conditions of the imaging device 80 include, for example, the type of light source, the irradiation direction of light, the exposure time, the aperture value, and the like.

Further, it is preferable that the storage unit 51 stores the work result information 50c regarding the quality of the work on the substrate when the reference image BP1 is acquired in association with the image-related information 50a and the search information 50b. As a result, the storage device 50m can store the image-related information 50a, the search information 50b, and the work result information 50c in association with each other. The work result information 50c may include the work status (for example, work time, work place, equipment used, work conditions, etc.) of the work on the board in addition to the work result (good or bad) of the work on the board. ..

The storage device 50m only needs to be able to store the image-related information 50a, the search information 50b, and the work result information 50c in association with each other, and a known storage device can be used. As the storage device 50m, a non-volatile storage device capable of rewriting such information can be used. Further, as will be described later, the storage device 50 m may be provided so as to be accessible from the board-to-board work machine WM connected to the network.

1-4-2. Acquisition unit 52
The acquisition unit 52 obtains the image-related information 50a associated with the search information 50b including the acquisition conditions that match the acquisition conditions of the target image OP1 in which the image pickup device 80 of the board-to-board work machine WM imaged the object in the board-to-board work. , Obtained from the storage device 50 m.

FIG. 8 schematically shows an example of image-related information 50a, search information 50b, and work result information 50c stored in the storage device 50m. No. Reference numeral 1 denotes a reference image BP1 represented by data 5D1 obtained by capturing an object represented by data 5A1 by the imaging device 80 indicated by data 5B1 under the imaging conditions indicated by data 5C1 and stored in the storage device 50m. It is shown that.

Also, No. 1 indicates that the feature amount distribution FD1 shown in the data 5E1 acquired from the reference image BP1 is stored in the storage device 50m together with the reference image BP1. Furthermore, No. 1 indicates that the work result information 50c of the work on the substrate when the reference image BP1 is acquired is shown in the data 5F1. No. The above-mentioned thing about No. 1 is No. The same can be said for 2 and later.

However, No. Reference numeral 3 shows that the reference image BP1 shown in the data 5D3 is stored in the storage device 50m, but the feature amount distribution FD1 acquired from the reference image BP1 is not stored in the storage device 50m. On the contrary, No. 4 indicates that the feature amount distribution FD1 shown in the data 5E3 is stored in the storage device 50 m, but the reference image BP1 when the feature amount distribution FD1 is acquired is not stored in the storage device 50 m. ing.

For example, the acquisition condition of the target image OP1 when the image pickup device 80 of the substrate working machine WM images the target image OP1 is No. It is assumed that the object, the imaging device, and the imaging conditions of 1 are matched. Further, it is assumed that the substrate working machine WM does not hold the reference image BP1 or the feature amount distribution FD1 corresponding to the acquisition condition. At this time, the acquisition unit 52 can acquire the reference image BP1 represented by the data 5D1 from the storage device 50m.

Further, the acquisition unit 52 can also acquire the feature amount distribution FD1 shown in the data 5E1 from the storage device 50m. Further, the acquisition unit 52 can also acquire both the reference image BP1 shown in the data 5D1 and the feature amount distribution FD1 shown in the data 5E1 from the storage device 50m. Further, in any case, the acquisition unit 52 can acquire the work result information 50c indicated by the data 5F1 from the storage device 50m.

In the above example, for example, it is assumed that among the acquisition conditions of the target image OP1, the imaging conditions match the imaging conditions shown in the data 5C2. In this case, the acquisition unit 52 can acquire the reference image BP1 represented by the data 5D2 from the storage device 50m. Further, the acquisition unit 52 can also acquire the feature amount distribution FD1 shown in the data 5E2 from the storage device 50m.

Further, the acquisition unit 52 can also acquire both the reference image BP1 shown in the data 5D2 and the feature amount distribution FD1 shown in the data 5E2 from the storage device 50m. Further, in any case, the acquisition unit 52 can acquire the work result information 50c indicated by the data 5F2 from the storage device 50m.

In this way, the acquisition unit 52 can acquire the image-related information 50a associated with the search information 50b including the acquisition condition matching the acquisition condition of the target image OP1 from the storage device 50m. If there is no image-related information 50a that matches all the acquisition conditions, the acquisition unit 52 can acquire at least the image-related information 50a that matches the type of the object from the storage device 50m.

When the acquisition unit 52 acquires at least the reference image BP1 which is the image-related information 50a from the storage device 50m, the quality determination device 40 includes the distribution acquisition unit 41 described above. For example, No. 8 shown in FIG. As in the case of 3, it is assumed that the acquisition unit 52 acquires the reference image BP1 or the reference image BP1 and the work result information 50c, and does not acquire the feature amount distribution FD1.

In this case, as described above, the distribution acquisition unit 41 acquires the feature quantity distribution FD1 by using the reference image BP1 acquired by the acquisition unit 52. As a result, the quality determination unit 42 uses the image-related information 50a (feature amount distribution FD1 acquired from the reference image BP1) acquired by the acquisition unit 52 to acquire the quality of the work on the substrate when the target image OP1 is acquired. Can be judged.

When the acquisition unit 52 acquires at least the feature amount distribution FD1 which is the image-related information 50a from the storage device 50m, the quality determination device 40 does not need the distribution acquisition unit 41 described above. For example, No. 8 shown in FIG. As in the case of 4, it is assumed that the acquisition unit 52 acquires the feature amount distribution FD1 or the feature amount distribution FD1 and the work result information 50c.

In this case, the quality determination unit 42 can determine the quality of the work on the substrate when the target image OP1 is acquired by using the image-related information 50a (feature amount distribution FD1) acquired by the acquisition unit 52. In either case, the pass / fail determination unit 42 extracts the target feature amount OF1 which is the feature amount of the entire image for each image of the target image OP1, and the target feature amount OF1 with respect to the feature area FR1 defined by the feature amount distribution FD1. It is preferable to judge the quality of the work on the substrate when the target image OP1 is acquired, based on the degree of deviation of.

Further, as described above, it is preferable that the quality determination unit 42 determines the quality of the work on the substrate based on whether or not the Mahalanobis distance MD1 from the feature region FR1 exceeds a predetermined threshold value TH1. Further, it is preferable that the quality determination unit 42 determines the quality of the work on the substrate by the RT method, which is one of the Mahalanobis Taguchi methods. Further, as described above, it is preferable that the quality determination device 40 includes the threshold value setting unit 43.

When the board working machine WM is the component mounting machine WM3, when the vendor of the component 91 supplied to the component mounting machine WM3 is changed and the image-related information 50a matching the component 91 after the vendor change is not retained. In addition, it is preferable that the acquisition unit 52 acquires the image-related information 50a from the storage device 50m. As a result, the quality determination unit 42 can determine the quality of the work on the substrate by using the appropriate feature amount distribution FD1 regardless of the change of the vendor of the component 91.

In this case, as described above, it is preferable that the quality determination device 40 includes the change unit 44. The change unit 44 switches the feature amount distribution FD1 used by the quality determination unit 42 to the feature amount distribution FD1 that matches the part 91 after the vendor change at the timing when the vendor of the part 91 supplied to the component mounting machine WM3 is switched. ..

1-4-3. Provision 53
The providing unit 53 stores the target image OP1 and at least one of the feature amount distribution FD1 acquired by using the target image OP1 and the reference image BP1 in the storage device 50m. When the providing unit 53 stores the feature amount distribution FD1 in the storage device 50m, the distribution acquisition unit 41 acquires the feature amount distribution FD1 by using the target image OP1 and the reference image BP1.

No. shown in FIG. Reference numeral 101 denotes a state in which the providing unit 53 stores both the target image OP1 and the feature amount distribution FD1 acquired by using the target image OP1 and the reference image BP1 in the storage device 50 m. Specifically, the target image OP1 represented by the data 5Dx obtained by capturing the object represented by the data 5A1 by the imaging device 80 indicated by the data 5B1 under the imaging conditions shown by the data 5C3 is stored in the storage device 50m. ing.

Further, the feature amount distribution FD1 shown by the data 5Ex is stored in the storage device 50 m together with the target image OP1. Further, the work result information 50c indicated by the data 5Fx when the target image OP1 is acquired is stored in the storage device 50m. In addition, when the image-related information 50a associated with the search information 50b including the acquisition condition matching the acquisition condition of the target image OP1 is already stored in the storage device 50m, the providing unit 53 stores the stored image. It is preferable to store the image-related information 50a and the like in addition to the related information 50a. As a result, the learning data is increased, and the determination accuracy by the quality determination device 40 is improved.

1-4-4. Example of Arrangement of Pass / Fail Judgment Device 40 As shown in FIG. 9, the storage unit 51 and the storage device 50 m are provided on the data server 70 capable of communicating with the board working machine WM, and the acquisition unit 52 and the pass / fail judgment unit 42 are paired with each other. It is preferable that it is provided in the substrate working machine WM. Further, the data server 70 is preferably a shared server capable of sharing information among a plurality of board-to-board work machines WM. As shown in FIG. 3, the providing unit 53, the distribution acquisition unit 41, the threshold value setting unit 43, and the changing unit 44 can be appropriately provided on the substrate working machine WM.

As shown in the figure, the plurality of board-to-board work machines WM and the data server 70 are communicably connected by the communication unit 70N. The communication unit 70N can connect them by wire, and can also connect them wirelessly. In addition, various communication methods can be adopted. In the present embodiment, a wide area network (WAN: Wide Area Network) is configured by a plurality of board-to-board work machines WM and a data server 70. Further, the data server 70 is a shared server capable of sharing information among a plurality of board-to-board work machines WM.

For example, when the shared server is a cloud server that can be accessed by a plurality of companies, the storage amount of the image-related information 50a is likely to increase as compared with the shared server configured by one company. Therefore, the acquisition unit 52 can easily acquire the desired image-related information 50a. It also improves the reliability and stability of the data. However, there is a demerit that acquisition conditions (object, imaging device, imaging conditions, etc.), image-related information 50a, work result information 50c, and the like are disclosed to other companies.

On the contrary, in the case of a shared server configured by one company, the confidentiality of information such as acquisition conditions (object, imaging device, imaging conditions, etc.), image-related information 50a, work result information 50c, etc., as compared with a cloud server. Is high. However, since the storage amount of the image-related information 50a tends to be small, there is a demerit that it becomes difficult for the acquisition unit 52 to acquire the image-related information 50a.

When the providing unit 53 stores the target image OP1 in the storage device 50m and does not store the feature amount distribution FD1 in the storage device 50m, for example, the data server 70 may include a control unit corresponding to the distribution acquisition unit 41. it can. The control unit acquires the feature amount distribution FD1 by using the target image OP1 and the reference image BP1.

2. 2. Pass / Fail Judgment Method The same applies to the pass / fail determination method as described above for the pass / fail determination device 40. Specifically, the pass / fail judgment method includes a storage step, an acquisition step, and a pass / fail judgment step. The storage process corresponds to the control performed by the storage unit 51. The acquisition process corresponds to the control performed by the acquisition unit 52. The quality determination step corresponds to the control performed by the quality determination unit 42. Further, it is preferable that the pass / fail determination method further includes at least one of a providing step, a distribution acquisition step, a threshold setting step, and a changing step. The providing process corresponds to the control performed by the providing unit 53. The distribution acquisition step corresponds to the control performed by the distribution acquisition unit 41. The threshold value setting step corresponds to the control performed by the threshold value setting unit 43. The change step corresponds to the control performed by the change unit 44.

3. 3. An example of the effect of the embodiment According to the quality determination device 40, a storage unit 51, an acquisition unit 52, and a quality determination unit 42 are provided. As a result, the quality determination device 40 can acquire the image-related information 50a and determine the quality of the work on the substrate using the acquired image-related information 50a. The same can be said for the quality determination device 40 as for the quality determination method.

40: Good / bad judgment device, 51: Storage unit, 52: Acquisition unit, 53: Provision unit,
50a: Image-related information, 50b: Search information, 50c: Work result information,
50m: storage device, 41: distribution acquisition unit, 42: pass / fail judgment unit,
70: Data server, 80: Imaging device, 90: Board, 91: Parts,
BP1: Reference image, BF1: Reference feature amount, FD1: Feature amount distribution,
FR1: Feature region, MD1: Mahalanobis distance, TH1: Threshold,
OP1: Target image, OF1: Target feature amount,
WM: anti-board work machine, WM3: parts mounting machine.

Claims (13)

1. A plurality of images taken by the image pickup device of the board-to-board work machine that performs a predetermined work on the board under the same imaging conditions for the same type of object in the board-to-board work are used as reference images, and each image of the reference image is an image. When the distribution of a plurality of the reference feature amounts when the reference feature amount, which is the total feature amount, is extracted as the feature amount distribution,
   Image-related information that is at least one of the reference image and the feature distribution,
   Search information that is searchable for the image-related information and includes acquisition conditions when the reference image is acquired, and
   A storage unit that associates and stores in a storage device,
   The storage device stores the image-related information associated with the search information including acquisition conditions that match the acquisition conditions of the target image in which the object is imaged by the image pickup device of the board-to-board work machine. With the acquisition department to acquire from
   Using the image-related information acquired by the acquisition unit, a quality determination unit that determines the quality of the work on the substrate when the target image is acquired, and a quality determination unit.
   A pass / fail judgment device.
2. The quality determination device according to claim 1, wherein the feature amount of the entire image is at least one of the brightness and color of a plurality of pixels constituting the image, the shape formed by these, and the area of the shape. ..
3. The acquisition condition is a condition that can be defined by the substrate working machine, and includes information on the object, information on the imaging device, and at least information on the object among information on imaging conditions of the imaging device. The pass / fail determination device according to claim 1 or 2.
4. Any one of claims 1 to 3 in which the storage unit stores work result information regarding the quality of the work on the substrate when the reference image is acquired in association with the image-related information and the search information. The pass / fail judgment device described in.
5. The acquisition unit acquires at least the reference image, which is the image-related information, from the storage device.
   A distribution acquisition unit for acquiring the feature amount distribution using the reference image acquired by the acquisition unit is further provided.
   The pass / fail determination unit extracts the target feature amount, which is the feature amount of the entire image, for each image of the target image, and based on the degree of deviation of the target feature amount with respect to the feature area defined by the feature amount distribution. The quality determination device according to any one of claims 1 to 4, which determines the quality of the work on the substrate when the target image is acquired.
6. The acquisition unit acquires at least the feature amount distribution, which is the image-related information, from the storage device.
   The pass / fail determination unit extracts the target feature amount, which is the feature amount of the entire image, for each image of the target image, and based on the degree of deviation of the target feature amount with respect to the feature area defined by the feature amount distribution. The quality determination device according to any one of claims 1 to 4, which determines the quality of the work on the substrate when the target image is acquired.
7. The quality determination device according to claim 5 or 6, wherein the quality determination unit determines the quality of the work on the substrate based on whether or not the Mahalanobis distance from the feature region exceeds a predetermined threshold value.
8. The quality determination device according to claim 7, wherein the quality determination unit determines the quality of the work on the substrate by the RT method, which is a method of the Mahalanobis Taguchi method.
9. Any of claims 1 to 8, further comprising a providing unit that stores the target image and at least one of the feature amount distribution acquired by using the target image and the reference image in the storage device. The pass / fail judgment device described in item 1.
10. The board-to-board working machine is a component mounting machine that mounts the component that is the object on the board.
    When the vendor of the component supplied to the component mounting machine is changed and the image-related information matching the component after the vendor change is not retained.
    The quality determination device according to any one of claims 1 to 9, wherein the acquisition unit acquires the image-related information from the storage device.
11. The storage unit and the storage device are provided in a data server capable of communicating with the board working machine.
    The quality determination device according to any one of claims 1 to 10, wherein the acquisition unit and the quality determination unit are provided on the substrate working machine.
12. The pass / fail determination device according to claim 11, wherein the data server is a shared server capable of sharing information between a plurality of the board-to-board working machines.
13. A plurality of images taken by the image pickup device of the board-to-board work machine that performs a predetermined work on the board under the same imaging conditions for the same type of object in the board-to-board work are used as reference images, and each image of the reference image is an image. When the distribution of a plurality of the reference feature amounts when the reference feature amount, which is the total feature amount, is extracted as the feature amount distribution,
    Image-related information that is at least one of the reference image and the feature distribution,
    Search information that is searchable for the image-related information and includes acquisition conditions when the reference image is acquired, and
    And the storage process of associating and storing in the storage device,
    The storage device stores the image-related information associated with the search information including acquisition conditions that match the acquisition conditions of the target image in which the object is imaged by the image pickup device of the board-to-board work machine. The acquisition process to be acquired from
    Using the image-related information acquired in the acquisition step, a quality determination step of determining the quality of the work on the substrate when the target image is acquired, and a quality determination step.
    A pass / fail judgment method.

Images