JPWO2020065698A1 - Production optimization system for component data, component data creation system, component mounting machine and component mounting line - Google Patents

Abstract

In the component data including the shape data showing the shape characteristics of the component used when the image of the component captured by the component imaging camera (56) of the component mounting machine (54) is processed and the component is recognized. In addition to the shape data, low-resolution image processing that processes one component image captured by the component imaging camera to recognize the component, and a plurality of images captured by the component imaging camera at different positions. Data that specifies whether to perform image processing by super-resolution image processing that recognizes the component by processing the super-resolution image created by super-resolution processing the component image is included. The component data including the data that specifies the super-resolution image processing includes at least one of the magnification of the super-resolution image, the number of component images used for the super-resolution image processing, and the amount of movement between the positions where each component image is imaged. Contains data that specifies super-resolution image processing conditions.

Description

This specification discloses technologies related to component data used for image recognition of components, a component data creation system that automatically creates the component data, a component mounting machine that uses the component data, and a production optimization system for a component mounting line. It was done.

When a camera captures an image of a minute component sucked on a suction nozzle while the component mounting machine is in operation and recognizes the image, it may be difficult to recognize the minute component due to the low resolution of the component image captured by the camera. .. Therefore, as described in Patent Document 1 (International Publication WO2015 / 083220), it is said that the component to be imaged is a small component that is difficult to recognize by ordinary image processing by acquiring the dimensional information of the component to be imaged. When the judgment is made, one super-resolution image (high-resolution image) is obtained from a plurality of low-resolution component images acquired by photographing the component multiple times while moving the camera using the super-resolution technology. There is a product created so that the super-resolution image is processed to recognize the component.

International Publication WO2015 / 083220

In Patent Document 1 described above, whether or not to create a super-resolution image is determined based on whether or not the component to be imaged is a small component. However, there are various types of parts to be mounted on the circuit board, and not only the part size differs for each part type, but also the shape and size of terminals such as leads and bumps, and the terminal arrangement pattern for each part type. It is difficult to correctly judge the necessity of a super-resolution image only by the component size because of various differences. Therefore, as in Patent Document 1, in a component mounting machine that determines the necessity of a super-resolution image based only on the component size, the terminal size may be small or the terminal may be small even if the component size does not create a super-resolution image. When the shape and arrangement pattern of are complicated, it may be necessary to create a super-resolution image in order to correctly recognize the terminals. On the contrary, even if the component size is for creating a super-resolution image, if the terminal size is relatively large and the terminal shape and arrangement pattern are relatively simple, it is not necessary to create a super-resolution image. However, in some cases, the recognition accuracy of the terminal can be ensured from one low-resolution component image.

When creating a super-resolution image with a component mounting machine, the processing from imaging to recognition of the component takes longer than normal image processing, so the more parts that create a super-resolution image, the more production The sex will be reduced. Therefore, in order to suppress the decrease in productivity as much as possible, it is necessary to minimize the types of parts that produce super-resolution images.

However, as in Patent Document 1 described above, in a component mounting machine that determines the necessity of a super-resolution image only by the component size, the super-resolution image is wasted with a component type that originally does not require a super-resolution image. There is a problem that productivity may be reduced as a result of low-resolution image processing for parts that originally require super-resolution images and failure to recognize terminals. It was.

In order to solve the above problems, component data including shape data indicating the shape characteristics of the component used when processing the image of the component captured by the component imaging camera of the component mounting machine to recognize the component. In the low-resolution image processing for recognizing the component by processing one component image captured by the component imaging camera in addition to the shape data, and a plurality of images captured by the component imaging camera at different positions. It includes data that specifies whether to perform image processing by super-resolution image processing that recognizes the part by processing the super-resolution image created by super-resolution processing of a piece of component image. ..

When recognizing an image of a component, the component mounting machine uses component data including shape data indicating the shape characteristics of the component. Since this part data is created for each part type, in addition to the shape data of that part type, data that specifies whether to perform image processing by low-resolution image processing or super-resolution image processing is also used for that part type. If it is included in the part data of, not only the part size of the part type but also the shape and size of the terminal, the arrangement pattern of the terminal, etc. are taken into consideration, and the image processing method suitable for the part type is included in the part data. Can be specified. As a result, when the component mounting machine recognizes the image of the component, the image is processed by the image processing method specified by the component data of the component type, and the image is processed by the image processing method suitable for the component type. Can be done. As a result, the component mounting machine does not wastefully create a super-resolution image with a component type that originally does not require a super-resolution image, and the number of component types for creating a super-resolution image can be minimized. At the same time, it is possible to perform low-resolution image processing on parts that originally require super-resolution images and to prevent terminal recognition from failing, enabling stable image recognition of parts and production. The sex can be improved.

FIG. 1 is a vertical sectional front view showing the configuration of the camera stand for creating component data according to the first embodiment. FIG. 2 is a plan view of a mounting table showing an example of the reference position mark. FIG. 3 is a block diagram showing an electrical configuration of a camera stand for creating component data. FIG. 4 is a diagram showing a first component image acquired in the first imaging. FIG. 5 is a diagram showing a second component image acquired in the second imaging. FIG. 6 is a block diagram showing a configuration of a component mounting line. FIG. 7 is a flowchart showing a processing flow of the first half of the component data creation program of the first embodiment. FIG. 8 is a flowchart showing a processing flow of the latter half of the component data creation program of the first embodiment. FIG. 9 is a flowchart showing a processing flow of the image processing program of the component mounting machine of the first embodiment. FIG. 10 is a flowchart showing a processing flow of the first half of the component data creation program of the second embodiment. FIG. 11 is a flowchart showing a processing flow of the latter half of the component data creation program of the second embodiment.

Hereinafter, two Examples 1 and 2 disclosed in the present specification will be described.

The first embodiment will be described with reference to FIGS. 1 to 9.
First, the configuration of the camera stand 11 for creating component data will be described with reference to FIGS. 1 and 2.

The camera 13 is attached downward to the upper part of the main body frame 12 of the camera stand 11 for creating component data. The camera 13 is a camera of the same type as the component imaging camera 56 mounted on the component mounting machine 54 shown in FIG. 6, and is configured by using, for example, an image pickup element that captures a grayscale image (monochrome image). However, a camera using an image sensor that captures a color image may be used.

A lens unit 14 is attached below the camera 13, a coaxial epi-illumination device 15 is attached below the lens unit 14, and a side-illumination device 16 is attached below the coaxial epi-illumination device 15. .. Although not shown, the coaxial epi-illumination device 15 has a light emitting source such as an LED that radiates light horizontally from the side to the center side, and the light of the light emitting source is perpendicular to the image pickup object below. It is configured to have a half mirror that reflects light.

On the other hand, the side-illuminating lighting device 16 has, for example, a configuration in which upper and lower three-stage side-emitting light sources 21, 22, and 23 are arranged in a bowl shape or a polygonal pyramid shape. The side-emitting light sources 21, 22, and 23 of each stage are configured by, for example, assembling a plurality of LED mounting substrates on which a large number of LEDs are mounted in a polygonal ring shape such as an octagon, and the side-emitting light sources 21 of each stage. The side-emitting light sources 21, 22, 23 of each stage with respect to the image-imaging object so that the illumination lights of 22 and 23 are emitted from diagonally above the image-imaging object (part) located on the optical axis of the camera 13. The angle of side lighting is set.

The side-illuminating light sources 21, 22, 23 of each stage of the side-illuminating illuminating device 16 are configured to be individually switchable between lighting and extinguishing, and the side-illuminating illuminating device 16 is configured by an image processing device 41 (see FIG. 3) described later. The lighting pattern, which is a combination pattern of lighting / extinguishing of the side-emitting light sources 21, 22, 23 in each stage of the above, is switched according to the type (size, shape, etc.) of the component to be imaged. As a result, it is configured to be able to reproduce the same lighting conditions as the lighting device (not shown) mounted on the component mounting machine 54 that mounts the same type of component as the component to be imaged.

Below the side-illuminated lighting device 16, a mounting table 25 on which a component to be imaged is mounted is installed. The mounting table 25 is horizontally supported by a moving table 27 of a moving device 26 that moves the mounting table 25 in the horizontal direction. The moving device 26 is configured to use an actuator 28 such as a ball screw device, a linear motor, or a linear solenoid as a drive source, and to move the mounting base 25 in the horizontal direction by the actuator 28. The moving direction of the mounting table 25 is, for example, a direction in which the component to be imaged moves obliquely in the field of view of the camera 13. The moving device 26 may be configured so that the moving direction of the mounting table 25 can be adjusted.

At least the portion of the mounting table 25 on which the parts are placed is anti-slip processed in order to prevent the parts from slipping during movement. As the anti-slip treatment, for example, coating of a material having a high coefficient of friction such as silicone, rough surface treatment for increasing the friction coefficient of the surface, or chemical surface treatment may be used. Alternatively, at least a portion of the mounting table 25 on which the parts are placed may be formed of a material having a high coefficient of friction such as silicone resin or rubber.

Alternatively, the mounting table 25 may be provided with a holder (not shown) for holding the mounted parts. This holder may hold the parts on the mounting table 25 by, for example, a negative pressure attraction force or a magnetic attraction force, or may hold the parts on the mounting table 25 by the adhesive force of the adhesive tape. There may be.

Further, as shown in FIG. 2, one or a plurality of reference position marks 29 are provided as reference position portions on the upper surface of the mounting table 25 at a predetermined position that does not overlap with the parts to be mounted. The reference position mark 29 may have any shape as long as it can be specified by image recognition. The direction may be specified by the positional relationship of a plurality of reference position marks 29, and if the shape of the reference position marks 29 is a shape whose direction can be specified by image recognition, the number of reference position marks 29 is 1. It may be only one. Further, when the moving direction of the mounting table 25 by the moving device 26 is fixed exactly in a fixed direction, the number of reference position marks 29 may be only one.

In the example of FIG. 2, four reference position marks 29 are arranged near one corner of the upper surface of the mounting table 25, but the positions of the reference position marks 29 are on the mounting table 25 within the field of view of the camera 13. If the position is such that both the component and the reference position mark 29 can be accommodated and imaged, and the position is included in a plurality of component images (see FIGS. 4 and 5) captured by moving the mounting table 25, the reference is used. The position of the position mark 29 may be changed as appropriate. The image processing device 41 also functions as a super-resolution image creating unit 44 (see FIG. 3), and when performing super-resolution processing, parts on the mounting table 25 and a reference position mark 29 are placed in the field of view of the camera 13. A plurality of component images (FIGS. 4 and 5) in which the component on the platform 25 and the reference position mark 29 are imaged at different positions by taking images a plurality of times while moving the platform 25 by the moving device 26 within the fitting range. (See) is acquired, the amount of movement between the positions where each part image is imaged is measured based on the position of the reference position mark 29 in each part image, and the plurality of part images are positioned based on the measured value. A super-resolution image is created by performing super-resolution processing that integrates them together.

The component image captured by the camera 13 is taken into the image processing device 41 (see FIG. 3) and subjected to super-resolution image processing or low-resolution image processing described later. The image processing device 41 is mainly composed of a computer, and as shown in FIG. 3, an input device 42 composed of a keyboard, a mouse, a touch panel, etc., a component image captured by the camera 13, and a super-resolution process created by super-resolution processing. A display device 43 or the like that displays a resolution image (high resolution image) or the like is connected.

The lighting mode (illumination pattern) for illuminating the components on the mounting table 25 can be switched depending on the type of components imaged by the camera 13. For example, an all-lighting lighting mode in which parts are illuminated by both the coaxial epi-illumination device 15 and the side-illumination illumination device 16, a coaxial epi-illumination mode in which parts are coaxial epi-illumination by only the coaxial epi-illumination device 15, and a side-illumination illumination device 16 There is a side-illumination mode that only side-illuminates the part. Since the reference position mark 29 needs to be coaxial epi-illumination by the coaxial epi-illumination device 15, when the component is side-illuminated and imaged only by the side-illuminating illumination device 16, the mounting table 25 is moved for each imaging position. Is stopped, and the image of the component illuminated by the side illumination device 16 at that position and the image of the reference position mark 29 illuminated by the coaxial epi-illumination device 15 are performed separately. In this case, the amount of movement between the positions where each component image is imaged is measured based on the position of the reference position mark 29 in each component image imaged by coaxial epi-illumination with the coaxial epi-illumination device 15, and the measured value is used. Based on this, a super-resolution image is created by aligning and integrating a plurality of component images captured by side-illuminating with the side-illuminating illuminating device 16.

Here, a procedure for creating a super-resolution image from a plurality of component images captured by the camera 13 will be described with reference to FIGS. 4 and 5.

First, the component on the mounting table 25 and the reference position mark 29 are placed in the field of view of the camera 13 and imaged to acquire the first component image (first imaging). After that, the mounting table 25 is moved by the moving device 26 within the range in which the parts on the mounting table 25 and the reference position mark 29 are within the field of view of the camera 13, and then the parts and the reference position on the mounting table 25 are moved by the camera 13. The mark 29 is imaged to acquire a second component image (second imaging). After that, the same operation is repeated to acquire a predetermined number (n) of component images.

In parallel with or after the imaging of the predetermined number of component images, the image processing device 41 performs image processing on each component image, and the positions (X1, Y1) of the reference position mark 29 in each component image, ( X2, Y2), ..., (Xn, Yn) are measured. In the examples of FIGS. 4 and 5, the positions (X1, Y1), (X2, Y2) of the reference position mark 29 in each part image are set with the upper right corner of each part image as the origin (0,0). , (Xn, Yn) are measured, but the reference position mark in each part image is set with the other corners of each part image (for example, the lower left corner) and the center point of each image as the origin (0,0). The positions (X1, Y1), (X2, Y2), ..., (Xn, Yn) of 29 may be measured.

After that, the amount of movement (ΔX1, ΔY1), (ΔX2, ΔY2), ..., (ΔXn-1, ΔYn-1) between the positions where the image of each component is captured is calculated.

[First movement amount]
ΔX1 ＝ ｜ X1 −X2 ｜
ΔY1 ＝ ｜ Y1 −Y2 ｜
[Second movement amount]
ΔX2 ＝ ｜ X2 −X3 ｜
ΔY2 ＝ ｜ Y2 −Y3 ｜
[Last movement amount]
ΔXn-1 ＝ ｜ Xn-1 −Xn ｜
ΔYn-1 ＝ ｜ Yn-1 −Yn ｜

After that, the reference position mark 29 in each part image is based on the amount of movement (ΔX1, ΔY1), (ΔX2, ΔY2), ..., (ΔXn-1, ΔYn-1) between the positions where the image of each part is imaged. A super-resolution image is created by performing a super-resolution process of aligning and integrating a predetermined number (n) of component images so that the positions of the above are the same.

As shown in FIG. 3, the component data creation system 48 is composed of a camera stand 11 for creating component data, an image processing device 41, and the like, and creates component data based on the image processing result of the image processing device 41. This component data includes shape data indicating the shape characteristics of the component used when the image of the component captured by the component imaging camera 56 of the component mounting machine 54 is processed to recognize the component. This shape data is, for example, data related to the size of the component body (dimensions in the X, Y, Z directions, etc.) and terminals. For example, in the case of a component with a lead, the lead width, length, lead pitch, and lead It is data such as position and number of leads, and in the case of BGA parts, it is data such as bump diameter, bump pitch, bump position, and number of bumps.

The component data of the first embodiment includes "low resolution image processing" in which the component mounting machine 54 processes one component image captured by the component imaging camera 56 to recognize the component, and components at different positions. Specify which of "super-resolution image processing" that recognizes the component by processing the super-resolution image created by super-resolution processing a plurality of component images captured by the imaging camera 56. Contains data to be processed. Further, when the component data includes data that specifies super-resolution image processing, the component data includes the magnification of the super-resolution image, the number of component images used for super-resolution processing, and each component image is captured. Data is included that specifies at least one super-resolution image processing condition for the amount of movement between positions. Further, the component data also includes data for designating imaging conditions and / or lighting conditions for imaging the component. Here, the imaging conditions are, for example, the shutter speed (exposure time), the imaging magnification, the imaging distance, the aperture value, etc. of the component imaging camera 56, and the illumination conditions are, for example, the brightness of the illumination light of the lighting device of the component mounting machine 54. The incident angle of the illumination light to the component (illumination angle), the illumination mode (illumination pattern), the lighting time of the pulse illumination (duty ratio), and the like.

As shown in FIG. 3, the image processing device 41 of the component data creation system 48 functions as the above-mentioned super-resolution image creation unit 44 by executing the component data creation programs of FIGS. 7 and 8 described later. From the first shape data creation unit 45 that creates shape data from the super-resolution image created by the super-resolution image creation unit 44, and one component image captured by the camera 13 of the component data creation camera stand 11. It also functions as a second shape data creation unit 46 that creates shape data. Further, in this image processing device 41, when the second shape data creation unit 46 succeeds in creating shape data from the one part image, the second shape data creation unit 46 starts from the one part image. If part data including the created shape data and data for designating low-resolution image processing is created, while the second shape data creation unit 46 fails to create shape data from the one part image, The first shape data creation unit 45 creates component data including shape data created from the super-resolution image and data for designating super-resolution image processing.

Here, since the second shape data creation unit 46 creates shape data from one component image captured by the camera 13, the number of component images is only one (the number of imagings of the camera 13 is only once). However, in the first embodiment, in order to reduce the error occurrence rate of low-resolution image processing, the second shape data creation unit 46 also uses the camera 13 as in the first shape data creation unit 45. The number of component images to be imaged is set to a plurality of images (the number of times the camera 13 is imaged is a plurality of times), and the shape data is sequentially created from one component image for the plurality of component images. If the creation of shape data from any of the component images fails, the component including the shape data created by the first shape data creation unit 45 from the super-resolution image and the data for designating the super-resolution image processing. I try to create data.

Further, when both the first shape data creation unit 45 and the second shape data creation unit 46 fail to create the shape data, the image processing device 41 takes an image of the parts on the mounting table 25 and / Alternatively, change the lighting conditions to create shape data. In this way, when the creation of shape data fails due to improper imaging conditions and / or lighting conditions, the imaging conditions and / or lighting conditions are optimized and the shape data is successfully created. be able to. Here, the imaging conditions are, for example, the shutter speed (exposure time) of the camera 13, the imaging magnification, the imaging distance, the aperture value, and the like, and the illumination conditions are, for example, the illumination of the coaxial epi-illumination device 15 and / or the side-illumination illumination device 16. The brightness of the light, the incident angle of the illumination light on the component (illumination angle), the illumination mode (illumination pattern), the lighting time of the pulse illumination (duty ratio), and the like.

Further, when the image processing device 41 creates component data including data that specifies super-resolution image processing, the super-resolution image when the super-resolution image creation unit 44 creates the super-resolution image The component data includes data that specifies at least one super-resolution image processing condition among the magnification, the number of component images used for super-resolution processing, and the amount of movement between the positions where each component image is captured. In this way, when the component mounting machine 54 executes the super-resolution image processing, the component data creation system 48 creates the component data from the super-resolution image processing conditions of the component mounting machine 54. It is possible to set the same conditions as when the above is performed, and the accuracy of super-resolution image processing can be improved.

Further, when the image processing device 41 creates component data including data that specifies super-resolution image processing, the imaging conditions and imaging conditions when a plurality of component images used for creating the super-resolution image are captured. / Or include data that specifies the lighting conditions in the part data. On the other hand, when creating component data including data that specifies low-resolution image processing, data that specifies the imaging conditions and / or lighting conditions when one component image used to create the shape data is captured. Is included in the part data. In this way, when the component mounting machine 54 executes the super-resolution image processing or the low-resolution image processing, the imaging condition and / or the lighting condition of the component mounting machine 54 is set to the component data by the component data creation system 48. Can be set to the same conditions as when the image was created, and the image processing accuracy can be improved.

As shown in FIG. 6, the image processing device 41 of the component data creation system 48 is connected to the network 51 of the component mounting line 50. The component mounting line 50 is configured by arranging a plurality of component mounting machines 54 on a transport path 53 that conveys the circuit board 52, and each component mounting machine 54 sucks the components supplied from the feeder 55 with a suction nozzle to form a circuit. A component mounting board is produced by mounting it on the board 52. Each component mounting machine 54 includes a component imaging camera 56 that images a component sucked on a suction nozzle, a lighting device (not shown) that illuminates a component imaged by the component imaging camera 56, and a component mounting machine 54. It is configured to include a control device 57 and the like for controlling the operation of the function. The control device 57 of the component mounting machine 54 also functions as an image processing device that processes a component image captured by the component imaging camera 56.

The control device 57 of each component mounting machine 54 of the component mounting line 50 acquires the component data created by the image processing device 41 of the component data creation system 48 via the network 51 of the component mounting line 50. In the network 51 of the component mounting line 50, in addition to the image processing device 41 of the component data creation system 48, a production control device 61 that manages the production of the component mounting line 50 and a production that optimizes the production of the component mounting line 50 The optimization device 62 and the like are connected. The component data created by the image processing device 41 of the component data creation system 48 is also transmitted to the production control device 61 and the production optimization device 62 via the network 51 of the component mounting line 50. In this case, the production control device 61 receives the component data transmitted from the image processing device 41 of the component data creation system 48, and transmits the component data from the production control device 61 to the control device 57 of each component mounting machine 54 and / Alternatively, the data may be transmitted to the production optimization device 62.

The control device 57 of each component mounting machine 54 specifies low-resolution image processing for the component data of the component sucked on the suction nozzle by executing the image processing program of the component mounting machine of FIG. 9 described later during production. When data is included, when recognizing the part as an image, the part is recognized by processing one part image captured by the part imaging camera 56 and referring to the shape data included in the part data. Perform low-resolution image processing. Further, when the component data of the component sucked by the suction nozzle includes data for specifying super-resolution image processing, the super-resolution processing is performed on a plurality of component images captured by the component imaging camera 56 at different positions. The super-resolution image processing created in the above process is processed, and the super-resolution image processing for recognizing the part is executed by referring to the shape data included in the part data.

Further, the control device 57 of each component mounting machine 54 includes data for designating super-resolution image processing in the component data of the component to be imaged, and is used for the magnification of the super-resolution image and the super-resolution processing. When data that specifies at least one super-resolution image processing condition of the number of component images and the amount of movement between positions where each component image is imaged is included, when executing super-resolution image processing, Super-resolution image processing is executed under the super-resolution image processing conditions specified in the component data.

Further, when the control device 57 of each component mounting machine 54 includes data for designating an imaging condition and / or a lighting condition for imaging the component in the component data of the component to be imaged, the component data is used. Performs super-resolution image processing or low-resolution image processing under the specified imaging conditions and / or illumination conditions.

On the other hand, the production optimizing device 62 distributes the feeder 55 necessary for supplying all the component types used for producing the component mounting board to the plurality of component mounting machines 54 constituting the component mounting line 50, and each of them. A process of optimizing the feeder arrangement of the component mounting machine 54 and optimizing the production job (production program) to be executed by each component mounting machine 54 is performed.

In the first embodiment, the production optimization device 62 uses the image processing device 41 of the component data creation system 48 to collect component data used for image processing for each component type for all component types used in the production of the component mounting board. The feeder arrangement of each component mounting machine 54 is arranged in consideration of the image processing method specified in the component data and the image processing capability of each component mounting machine 54, which is acquired directly from or via the production control device 61. Optimize. By doing so, when there is a component mounting machine 54 that cannot secure the required accuracy of super-resolution image processing due to insufficient image processing capability among the plurality of component mounting machines 54 constituting the component mounting line 50. Allocates only the feeder 55 that supplies the component type that performs low-resolution image processing, and allocates the feeder 55 that supplies the component type that performs super-resolution image processing for the component mounting machine 54 that lacks image processing capability. This can be prevented in advance, and the occurrence of an error in super-resolution image processing due to insufficient image processing capability can be prevented in advance.

Next, the processing contents of the component data creation programs of FIGS. 7 and 8 executed by the image processing device 41 of the component data creation system 48 will be described.

When the image processing device 41 of the component data creation system 48 activates the component data creation programs of FIGS. 7 and 8, first, in step 101, the preparatory work for image processing (the worker places the component on the mounting table 25). Then, the input device 42 performs the preparation completion operation) and waits until the completion is completed.

After that, when the preparatory work for image processing is completed, the processes of steps 102 to 106 are executed to acquire a predetermined number of component images and create a super-resolution image as follows. First, in step 102, the component on the mounting table 25 and the reference position mark 29 are placed in the field of view of the camera 13 and imaged to acquire the first component image (first imaging). After that, the process proceeds to step 103, the mounting table 25 is moved by the moving device 26 within the range in which the parts on the mounting table 25 and the reference position mark 29 are within the field of view of the camera 13, and then the camera 13 proceeds to step 104. The component on the mounting table 25 and the reference position mark 29 are imaged to acquire a second component image (second imaging).

After that, the process proceeds to step 105, and it is determined whether or not the number of images has reached the predetermined number (n times). If the number of images has not yet reached the predetermined number, steps 102 to 104 described above are performed until the number of images reaches the predetermined number. The process is repeated to acquire a predetermined number (n) of component images.

As a result, when a predetermined number of component images are acquired, the process proceeds to step 106, and the predetermined number of component images are subjected to super-resolution processing to create a super-resolution image. After that, the process proceeds to step 107, and the super-resolution image is processed to create shape data showing the shape characteristics of the part to be used when the part mounting machine 54 recognizes the part.

After that, the process proceeds to step 108, and the component images used for the low-resolution image processing are imaged a predetermined number of times. As the component image used for this low-resolution image processing, a predetermined number of component images used for creating the super-resolution image described above may be used as they are, but in the low-resolution image processing, the reference position mark 29 is used for the component image. It is not necessary to include the image of the component, and only the component on the mounting table 25 needs to be included in the field of view of the camera 13 for imaging. Therefore, the component image may be captured by increasing the imaging magnification. Further, a predetermined number of component images used for this low-resolution image processing may be captured at the same position, or the mounting table 26 may be used to fit the parts on the mounting table 25 within the field of view of the camera 13. 25 may be moved to image at different positions.

After that, the process proceeds to step 109 to create shape data indicating the shape characteristics of the component from one component image in order for a predetermined number of component images. The process of creating the shape data may be performed in parallel with the operation of capturing the component image a predetermined number of times.

After that, the process proceeds to step 110 of FIG. 8, and it is determined from any of the predetermined number of component images whether or not the creation of shape data has failed, and as a result, the shape data of the predetermined number of component images is determined. If it is determined that none of the parts have failed to be created (that is, the shape data has been successfully created for all the predetermined number of part images), the process proceeds to step 114, and the shape data created from any of the part images and the low value are obtained. Create component data that includes data that specifies resolution image processing.

In this case, in step 109, since shape data is created from each of a predetermined number of component images, a predetermined number of shape data is created, but since only one shape data is included in the component data, for example, 1 The part data may include shape data created from a specific part image of the first piece or the second and subsequent pieces. Alternatively, the component data may include the averaged shape data obtained by calculating the average value of a predetermined number of shape data. Alternatively, an intermediate shape data thereof may be selected from a predetermined number of shape data and included in the part data. Further, the component data also includes data that specifies imaging conditions and / or lighting conditions when the component image is captured.

On the other hand, if it is determined in step 110 that the creation of shape data has failed from any of the predetermined number of component images, the process proceeds to step 111 to create shape data from the super-resolution image. If it is determined whether or not the result is successful, and as a result, it is determined that the shape data is successfully created from the super-resolution image, the process proceeds to step 112, and the shape data and the super-resolution image created from the super-resolution image are obtained. Create part data including data that specifies processing. Further, at least one super-resolution of the magnification of the super-resolution image when the super-resolution image is created, the number of component images used for the super-resolution processing, and the amount of movement between the positions where each component image is imaged. The data that specifies the image processing conditions is included in the component data, and the data that specifies the imaging conditions and / or the lighting conditions when a predetermined number of component images used for creating the super-resolution image are captured is also the component data. Include in.

On the other hand, if it is determined in step 111 that the creation of shape data from the super-resolution image has failed, the process proceeds to step 113, and the imaging conditions and / or lighting conditions for imaging the parts on the mounting table 25 are changed. Then, the processes after step 102 of FIG. 7 are executed again, and the process of creating super-resolution images from a predetermined number of component images to create shape data and the process of creating shape data from each component image are performed again. Execute. The changes in the imaging conditions and / or the lighting conditions may be automatically set by the image processing device 41, or may be set by the operator by operating the input device 42.

If the imaging conditions and / or the lighting conditions are changed and the shape data is successfully created for all the predetermined number of component images, the process proceeds to step 114, and the shape data and the low solution created from any of the component images are obtained. Create component data that includes data that specifies image image processing. This component data includes data that specifies imaging conditions and / or lighting conditions when the component image is captured.

Further, even if the imaging condition and / or the lighting condition is changed and the shape data cannot be created from any of the predetermined number of component images, the shape data can be successfully created from the super-resolution image. For example, the process proceeds to step 112, and component data including shape data created from the super-resolution image and data for designating super-resolution image processing is created. This component data includes data that specifies the above-mentioned super-resolution image processing conditions, and data that specifies imaging conditions and / or lighting conditions when a predetermined number of component images used for creating the super-resolution image are captured. To include.

On the other hand, the control device 57 of each component mounting machine 54 directly acquires or produces component data used for image recognition of components to be mounted on the circuit board 52 from the image processing device 41 of the component data creation system 48 before the start of production. Acquired via the management device 61. The control device 57 of each component mounting machine 54 executes the image processing program of the component mounting machine of FIG. 9 at each timing of imaging the component sucked on the suction nozzle during production, and is designated by the component data of the component. Image processing is performed by the image processing method.

When the control device 57 of each component mounting machine 54 starts the image processing program of the component mounting machine of FIG. 9, first, in step 201, whether or not super-resolution image processing is specified in the component data of the component to be imaged. If it is determined that the super-resolution image processing is specified as a result, the process of steps 202 to 206 is executed to create a super-resolution image by the component data creation system 48. A predetermined number of component images are acquired and a super-resolution image is created by the same method as in. At this time, if the component data of the component to be imaged includes data for specifying super-resolution image processing conditions and data for specifying imaging conditions and / or lighting conditions, the component mounting machine 54 is super A super-resolution image is created by setting the resolution image processing condition, the imaging condition, and / or the illumination condition to the super-resolution image processing condition, the imaging condition, and / or the illumination condition specified in the component data.

Specifically, first, in step 202, the component adsorbed on the suction nozzle and the reference position mark (not shown) provided in a fixed positional relationship with the suction nozzle are placed in the field of view of the component imaging camera 56. And take an image to acquire the first component image (first image). After that, the process proceeds to step 203, the suction nozzle is moved within the range in which the component and the reference position mark fit within the field of view of the camera 13, and then the process proceeds to step 204, and the component and the reference position mark are imaged by the component imaging camera 56. Then, the second component image is acquired (second imaging).

After that, the process proceeds to step 205, and it is determined whether or not the number of images has reached the predetermined number (n times). If the number of images has not yet reached the predetermined number, steps 202 to 204 described above are performed until the number of images reaches the predetermined number. The process is repeated to acquire a predetermined number (n) of component images. As a result, when a predetermined number of component images are acquired, the process proceeds to step 206, and the predetermined number of component images are subjected to super-resolution processing to create a super-resolution image. After that, the process proceeds to step 208, and the super-resolution image processing for recognizing the part by processing the super-resolution image and referring to the shape data included in the part data is executed.

On the other hand, in step 201 described above, when it is determined that the super-resolution image processing is not specified in the component data of the component to be imaged, that is, when it is determined that the low-resolution image processing is specified. Proceeds to step 207, the component sucked by the suction nozzle is housed in the field of view of the component imaging camera 56, and the image is taken to acquire one component image. At this time, if the imaging condition and / or the lighting condition is specified in the component data, the imaging condition and / or the lighting condition of the component mounting machine 54 is set to the imaging condition and / or the lighting condition specified in the component data. Set and capture one component image. After that, the process proceeds to step 208, and low-resolution image processing for recognizing the component by processing one component image and referring to the shape data included in the component data is executed.

As described above, the component mounting machine 54 uses component data including shape data indicating the shape characteristics of the component when recognizing the image of the component. Since this part data is created for each part type, in addition to the shape data of that part type, data that specifies whether to perform image processing by low-resolution image processing or super-resolution image processing is also used for that part type. If it is included in the part data of, not only the part size of the part type but also the shape and size of the terminal, the arrangement pattern of the terminal, etc. are taken into consideration, and the image processing method suitable for the part type is included in the part data. Can be specified. As a result, when the component mounting machine 54 recognizes the image of the component, the component mounting machine 54 performs image processing by the image processing method specified by the component data of the component type, thereby performing image processing by the image processing method suitable for the component type. be able to. As a result, the component mounting machine 54 does not wastefully create a super-resolution image with a component type that originally does not require a super-resolution image, and minimizes the number of component types for creating a super-resolution image. In addition to being able to do so, it is possible to perform low-resolution image processing on parts that originally require super-resolution images and to eliminate the failure of terminal recognition, and to perform stable image recognition of parts. Productivity can be improved.

Next, Example 2 will be described with reference to FIGS. 10 and 11. However, the parts substantially the same as those in the first embodiment are designated by the same reference numerals to omit or simplify the description, and mainly different parts will be described.

In the component data creation programs of FIGS. 7 and 8 described in the first embodiment, even if the shape data is successfully created for all the predetermined number of component images, the super-resolution image is created before that. , The shape data is created from the super-resolution image, but in the second embodiment shown in FIGS. 10 and 11, when the shape data is successfully created for all the predetermined number of component images (that is, low resolution). (When image processing is specified), the process of creating a super-resolution image and the process of creating shape data from the super-resolution image are not performed. Hereinafter, the processing contents of the component data creation programs of FIGS. 10 and 11 executed by the image processing device 41 of the component data creation system 48 will be described.

When the image processing device 41 of the component data creation system 48 activates the component data creation programs of FIGS. 10 and 11, first, in step 301, the preparatory work for image processing (the worker places the component on the mounting table 25). Then, the input device 42 performs the preparation completion operation) and waits until the completion is completed.

After that, when the preparatory work for image processing is completed, the process proceeds to step 302, and the component images used for the low-resolution image processing are imaged a predetermined number of times to acquire a predetermined number of component images. After that, the process proceeds to step 303 to create shape data indicating the shape characteristics of the parts from one part image in order for a predetermined number of part images. The process of creating the shape data may be performed in parallel with the operation of capturing the component image a predetermined number of times.

After that, the process proceeds to step 304 to determine whether or not the creation of shape data has failed from any of the predetermined number of component images, and as a result, the shape data is created for the predetermined number of component images. When it is determined that there is no failure (that is, the shape data has been successfully created for all the predetermined number of component images), the process proceeds to step 305, and any component image is obtained in the same manner as in the first embodiment. Create component data including shape data created from and data that specifies low-resolution image processing.

On the other hand, if it is determined in step 304 described above that the creation of shape data has failed from any of the component images of a predetermined number of parts, the processes of steps 306 to 309 are executed to carry out the above-described embodiment. In the same manner as in 1, a predetermined number of component images required to create a super-resolution image are imaged while the mounting table 25 is moved by the moving device 26.

As a result, when a predetermined number of component images are acquired, the process proceeds to step 310 of FIG. 11 and super-resolution processing is performed on the predetermined number of component images to create a super-resolution image. After that, the process proceeds to step 311 to process the super-resolution image to create shape data indicating the shape characteristics of the part to be used when the part mounting machine 54 recognizes the part.

After that, the process proceeds to step 312, and it is determined whether or not the shape data has been successfully created from the super-resolution image. As a result, if it is determined that the shape data has been successfully created from the super-resolution image, the step is performed. Proceeding to 313, component data including shape data created from the super-resolution image and data for designating super-resolution image processing is created. Further, at least one super-resolution of the magnification of the super-resolution image when the super-resolution image is created, the number of component images used for the super-resolution processing, and the amount of movement between the positions where each component image is imaged. The data that specifies the image processing conditions is included in the component data, and the data that specifies the imaging conditions and / or the lighting conditions when a predetermined number of component images used for creating the super-resolution image are captured is also the component data. Include in.

On the other hand, if it is determined in step 312 described above that the creation of shape data has failed from the super-resolution image, the process proceeds to step 314, and the parts on the mounting table 25 are imaged by the same method as in the first embodiment. The imaging conditions and / or illumination conditions to be performed are changed, and the processes of steps 302 to 304 in FIG. 10 are executed again.

If the imaging conditions and / or the lighting conditions are changed and the shape data is successfully created for all the predetermined number of component images, the process proceeds to step 305, and the shape data and the low solution created from any of the component images are obtained. Create component data that includes data that specifies image image processing. This component data includes data that specifies imaging conditions and / or lighting conditions when the component image is captured.

If the imaging conditions and / or the lighting conditions are changed and the creation of shape data from any of the predetermined number of component images fails, the processes of steps 306 to 309 are executed again. After capturing a predetermined number of component images required to create a super-resolution image by the same method as in the first embodiment, the process proceeds to step 310 of FIG. 11, and the predetermined number of component images are subjected to super-resolution processing. To create a super-resolution image, and in the next step 311 to create shape data from the super-resolution image.

If the imaging conditions and / or the illumination conditions are changed and the shape data is successfully created from the super-resolution image, the process proceeds to step 313, and the shape data created from the super-resolution image and the data for specifying the super-resolution image processing are specified. Create part data including. This component data includes data that specifies the above-mentioned super-resolution image processing conditions, and data that specifies imaging conditions and / or lighting conditions when a predetermined number of component images used for creating the super-resolution image are captured. To include.

In the second embodiment described above, the same effect as that of the first embodiment can be obtained, and the super-resolution image creation process can be omitted for the component types capable of low-resolution image processing. , There is an advantage that the parts data creation process can be simplified and the processing time can be shortened.

[Other Examples]
In the first and second embodiments, even when the shape data is created from the component images captured by the camera 13, the number of component images captured by the camera 13 is set to a plurality of sheets (the number of times the camera 13 is imaged is a plurality of times), and a plurality of images are taken. When the process of creating shape data from one part image in order is performed for the part images of the above, and the creation of shape data from any of the plurality of part images fails, super-resolution is obtained. The component data including the shape data created from the image and the data for specifying the super-resolution image processing is created, but when the shape data is created from the component image captured by the camera 13, the image is captured by the camera 13. The number of component images may be only one (the number of times the camera 13 is imaged is only one), and the shape data may be created from the one component image.

Further, the reference position mark 29, which is a reference position portion provided on the upper surface of the mounting table 25 of the component data creation camera stand 11 described in the first embodiment, needs to be imaged by coaxial epi-illumination with the coaxial epi-illumination device 15. Therefore, when the parts on the mounting table 25 are side-fired and imaged only by the side-illuminating device 16, the movement of the mounting table 25 is stopped at each imaging position, and the side-illuminating device is located at that position. It is necessary to separately image the part illuminated by the side illumination in 16 and the reference position mark 29 illuminated by the coaxial epi-illumination device 15, which doubles the number of imagings required for the super-resolution processing. ..

Therefore, a light emitting element for displaying a reference position such as an LED may be provided on the upper surface of the mounting table 25 as a reference position portion. The position and number of light emitting elements for displaying the reference position may be the same as in the case of the reference position mark 29 of the first embodiment.

The parts on the mounting table 25 may be parts such as bump parts in which an array of bumps is formed, which are side-illuminated and imaged only by the side-illuminating device 16. In the case of bump parts, the bumps are placed on the mounting table 25 with the bumps facing upward. When the bump component is side-illuminated and imaged only by the side-illuminating device 16, the light emitting element for displaying the reference position provided on the upper surface of the mounting table 25 is made to emit light, and is placed on the mounting table 25 in the field of view of the camera 13. Both the bump component and the light emitting element for displaying the reference position are included in the image. Even if only side-illuminated illumination is used, if the light emitting element for displaying the reference position is made to emit light and both the bump component on the mounting table 25 and the light emitting element for displaying the reference position are contained in the field of view of the camera 13 for imaging. Both the bump component on the mounting table 25 and the light emitting element for displaying the reference position can be image-recognized by one imaging. Therefore, even if the parts on the mounting table 25 are parts such as bump parts that are imaged only by side illumination, the light emitting element for displaying the reference position is coaxially shot by causing the light emitting element for displaying the reference position to emit light. There is no need to illuminate and image, and the number of imagings required for super-resolution processing does not increase.

Alternatively, as a sensor unit that does not provide a reference position portion on the upper surface of the mounting table 25 and measures the moving amount of the mounting table 25, for example, X that measures the moving amount in the X direction (X coordinate of the position of the mounting table 25). A directional linear scale and a Y-direction linear scale for measuring the amount of movement in the Y direction (Y coordinate of the position of the mounting table 25) may be provided, and the amount of movement of the mounting table 25 may be measured by both linear scales.

Further, the camera stand 11 for creating component data described in the first embodiment captures the components on the mounting table 25 at different positions by fixing the position of the camera 13 and moving the mounting table 25 by the moving device 26. However, by fixing the position of the mounting table 25 and moving the camera 13 in the horizontal direction by the moving device, a plurality of parts on the mounting table 25 were imaged at different positions. You may try to acquire the image of one part.

Alternatively, the mounting table 25 or the camera 13 is configured to be manually moved by the operator, and the operator manually moves the mounting table 25 or the camera 13 to image the parts on the mounting table 25 at different positions. It is also possible to acquire a plurality of component images.

In addition, the present invention is not limited to the above embodiments, and for example, the configuration of the mobile device 26 may be changed, or the configuration of the lighting device for illuminating the parts on the mounting table 25 may be changed. Of course, it can be changed and implemented within a range that does not deviate.

11 ... Camera stand for creating parts data, 13 ... Camera, 14 ... Lens unit, 15 ... Coaxial epi-illumination device, 16 ... Side-illumination lighting device, 21, 22, 23 ... Upper and lower three-stage side-illumination light source, 25 ... Mounting stand , 26 ... Moving device, 29 ... Reference position mark (reference position part), 41 ... Image processing device, 44 ... Super-resolution image creation unit, 45 ... First shape data creation unit, 46 ... Second shape data creation unit, 48 ... Parts data creation system, 50 ... Parts mounting line, 51 ... Network, 52 ... Circuit board, 53 ... Transport path, 54 ... Parts mounting machine, 56 ... Parts imaging camera, 57 ... Parts mounting machine control device (image) Processing equipment), 61 ... Production control equipment, 62 ... Production optimization equipment

Claims (14)

In the component data including the shape data indicating the shape characteristics of the component used when the image of the component captured by the component imaging camera of the component mounting machine is processed and the component is recognized.
In addition to the shape data, low-resolution image processing that processes one component image captured by the component imaging camera to recognize the component, and a plurality of images captured by the component imaging camera at different positions. Part data including data that specifies whether to perform image processing by super-resolution image processing that recognizes the part by processing the super-resolution image created by super-resolution processing the part image. When the data specifying the super-resolution image processing is included, at least the magnification of the super-resolution image, the number of component images used for the super-resolution image, and the amount of movement between the positions where each component image is imaged are at least. The component data according to claim 1, wherein the component data includes data for designating one super-resolution image processing condition. The component data according to claim 1 or 2, which includes data for designating imaging conditions and / or lighting conditions for imaging the component. In the parts data creation system that creates the parts data according to any one of claims 1 to 3.
A mounting table for mounting components for mounting on a circuit board,
A camera that captures the parts placed on the above-mentioned stand from above,
A lighting device that illuminates the parts imaged by the camera,
It is equipped with an image processing device that processes the component image captured by the camera.
The above-mentioned stand or the camera is configured to be movable, and the amount of movement of the above-mentioned stand or the camera is movable.
The image processing device is
By moving the pre-described pedestal or the camera within the range in which the parts on the pre-described pedestal fit within the field of view of the camera and taking images multiple times, a plurality of parts that image the parts on the previously described pedestal at different positions. In addition to acquiring an image, the amount of movement between the positions where each component image is captured is measured, and based on the measured value, the plurality of component images are subjected to super-resolution processing to create a super-resolution image. Image creation department and
The first shape data creation unit that creates the shape data from the super-resolution image created by the super-resolution image creation unit, and
It is provided with a second shape data creation unit that creates the shape data from one component image captured by the camera.
When the second shape data creation unit succeeds in creating the shape data from the one part image, the shape data created by the second shape data creation unit from the one part image and the low Create the component data including the data that specifies the resolution image processing,
When the second shape data creation unit fails to create the shape data from the one part image, the shape data and the super solution created by the first shape data creation unit from the super-resolution image. A component data creation system that creates the component data including data that specifies image image processing. The second shape data creation unit sequentially performs a process of creating the shape data from one component image of a plurality of component images captured by the camera, and any one of the plurality of component images. When the creation of the shape data from the part image fails, the part data including the shape data created by the first shape data creation unit from the super-resolution image and data for designating the super-resolution image processing. The part data creation system according to claim 4, wherein the component data creation system is created. When both the first shape data creation unit and the second shape data creation unit fail to create the shape data, the shape data is obtained by changing the imaging conditions and / or the lighting conditions for imaging the component. The part data creation system according to claim 4 or 5, which is created. When the second shape data creation unit succeeds in creating the shape data from the one part image, the super-resolution image creation unit creates the super-resolution image and the first shape data creation unit. The part data creation system according to any one of claims 4 to 6, which does not create the shape data according to the above. When creating the component data including the data specifying the super-resolution image processing, the magnification of the super-resolution image when the super-resolution image creation unit creates the super-resolution image, the super Claims 4 to 4 to include data that specifies at least one super-resolution image processing condition among the number of component images used for resolution processing and the amount of movement between the positions where each component image is captured. The parts data creation system according to any one of 7. When the component data including the data specifying the super-resolution image processing is created, the imaging conditions and / or the lighting conditions when the plurality of component images used for creating the super-resolution image are captured. Is included in the component data to specify
When creating the component data including the data for designating the low-resolution image processing, the imaging conditions and / or the lighting conditions when the one component image used for creating the shape data is captured are specified. The component data creation system according to any one of claims 4 to 8, wherein the data to be used is included in the component data. A reference position is provided on the upper surface of the above-mentioned stand so as not to overlap with the parts on the above-mentioned stand.
The super-resolution image creating unit captures images at different positions by moving the pre-described table or the camera a plurality of times within a range in which the parts on the pre-described table and the reference position portion fit within the field of view of the camera. In addition to acquiring a plurality of component images obtained by imaging the parts on the table and the reference position portion described above, the amount of movement between the positions where the reference position portions are imaged based on the position of the reference position portion in each component image. The component data creation system according to any one of claims 4 to 9, wherein a super-resolution image is created by performing super-resolution processing on the plurality of component images based on the measured values. A camera for imaging components that captures components mounted on a circuit board,
A lighting device that illuminates the parts to be imaged by the component imaging camera, and
In a component mounting machine equipped with an image processing device that processes a component image captured by the component imaging camera.
The image processing apparatus acquires the component data according to claim 1 as component data used for image processing of the component.
When the component data includes data that specifies the low-resolution image processing, the one component image captured by the component imaging camera is processed and the shape data included in the component data is referred to. Performs low-resolution image processing to recognize the relevant part
When the component data includes data that specifies the super-resolution image processing, a super-resolution image created by super-resolution processing a plurality of component images captured by the component imaging camera at different positions. A component mounting machine that performs super-resolution image processing that recognizes the component by referring to the shape data included in the component data. The image processing apparatus includes data for designating the super-resolution image processing in the component data, and obtains the magnification of the super-resolution image, the number of component images used for the super-resolution processing, and each component image. When data that specifies at least one super-resolution image processing condition of the amount of movement between the positions to be imaged is included, the said-mentioned part data is specified when the super-resolution image processing is executed. The component mounting machine according to claim 11, wherein the super-resolution image processing is executed under super-resolution image processing conditions. When the component data includes data for specifying imaging conditions and / or lighting conditions for imaging the component, the image processing apparatus has the imaging conditions and / or lighting conditions specified in the component data. The component mounting machine according to claim 11 or 12, wherein the super-resolution image processing or the low-resolution image processing is executed in the above. A component mounting line in which a plurality of component mounting machines are arranged in a transport path for transporting a circuit board, and each component mounting machine picks up components supplied from a feeder and mounts them on the circuit board to produce a component mounting board. In the production optimization system
It is equipped with a production optimization device that distributes the feeders required to supply all the component types used for the production of the component mounting board to the plurality of component mounting machines and optimizes the feeder arrangement of each component mounting machine.
The production optimization device acquires the component data according to any one of claims 1 to 3 used for image processing for each component type for all component types used in the production of the component mounting board, and the component. A production optimization system for a component mounting line that optimizes the feeder arrangement of each component mounting machine in consideration of the image processing method specified by the data and the image processing capability of each component mounting machine.

Images