Image processing apparatus, component mounting system, and image processing method

Abstract

An image processing apparatus that processes a color image having gray scale values of RGB three primary colors includes: an image acquisition section that acquires an image including an identification target and a similar portion that is different from the identification target in a main color component and similar in brightness to the identification target as a color image; a difference image generation section that generates a difference image having a gray scale value based on a difference obtained by subtracting a gray scale value of a second primary color image from a gray scale value of a first primary color image, the first primary color image being an image in which a gray scale value of a first primary color of the RGB three primary colors close to the main color component of the similar portion is extracted from the color image, the second primary color image being an image in which a gray scale value of a second primary color of the RGB three primary colors other than the first primary color is extracted from the color image; an identification image generation section that generates an identification image having a gray scale value obtained by subtracting a gray scale value of the difference image from a gray scale value of an image in which a gray scale value of any one of the RGB three primary colors is extracted from the color image; and an identification processing section that performs an identification process of the identification target using the identification image.

Description

TECHNICAL FIELD

[0001] The present specification discloses an image processing apparatus, a component mounting system, and an image processing method. BACKGROUND

[0002] In the past, as such an image processing apparatus, there has been proposed an apparatus that processes an image that contains an identification target and in which RGB values of each pixel (RGB pixel information) are defined (for example, refer to Patent Literature 1). In this apparatus, an RGB value of an identification target that should be extracted from an image and an exclusion RGB value of a color or the like that should be actively excluded because it is likely to be mistaken for an identification target are set in advance. Then, by removing pixels included in the exclusion RGB value from the image, pixels included in the RGB value of the identification target are extracted, and thus only an image corresponding to a region of the identification target remains, and a check of a state of the identification target or the like is performed.

[0003] PRIOR ART DOCUMENTS

[0004] Patent Literature 1: Japanese Patent Application Publication No. 2010-175483 SUMMARY

[0005] PROBLEMS TO BE SOLVED BY THE INVENTION

[0006] However, in the above-described image processing apparatus, because the RGB value of the identification target and the exclusion RGB value need to be set in advance, the burden of the setting process becomes large. In addition, even if it is set in advance, if the RGB value changes due to a change in a light irradiation condition or the like because of a posture or a positional deviation of an object at the time of actually capturing an image, or the like, it is sometimes difficult to extract the RGB value of the identification target with high precision.

[0007] The main object of the present disclosure is to identify an identification target with high precision without setting in advance an RGB value of a part that should be excluded.

[0008] TECHNICAL SOLUTION FOR SOLVING THE PROBLEMS

[0009] The present disclosure adopts the following solution in order to achieve the above-described main object.

[0010] The image processing apparatus of the present disclosure is an image processing apparatus that processes a color image having gray values of RGB three primary colors, and the gist thereof is to include: an image acquisition unit that acquires an image including an identification target and a similar portion that is different from the identification target in a main color component and similar in brightness to the identification target as the color image; a difference image generation unit that generates a difference image having a gray value based on a difference obtained by subtracting a gray value of a second primary color image from a gray value of a first primary color image, the first primary color image being an image in which a gray value of a first primary color of the RGB three primary colors close to a main color component of the similar portion is extracted from the color image, the second primary color image being an image in which a gray value of a second primary color of the RGB three primary colors other than the first primary color is extracted from the color image; an identification image generation unit that generates an identification image having a gray value obtained by subtracting a gray value of the difference image from a gray value of an image in which a gray value of any one of the RGB three primary colors is extracted from the color image; and an identification processing unit that performs identification processing of the identification target using the identification image.

[0011] The image processing apparatus of the present disclosure generates a difference image having a gray value based on a difference obtained by subtracting a gray value of a second primary color image from a gray value of a first primary color image, and therefore, the gray value of the difference image is a value based on a difference between a first primary color close to a main color component of a similar portion and a second primary color other than the first primary color. In addition, because an identification image is generated by subtracting a gray value of the difference image from a gray value of an image in which a gray value of any one of the primary colors is extracted, it is possible to reduce the brightness of the similar portion within the identification image and clearly identify a difference in brightness between the identification target and the similar portion. Therefore, it is possible to suppress misidentification of the similar portion as the identification target. In addition, because each image used in the processing is extracted or generated from an actually acquired color image, it is not necessary to be set in advance. Therefore, it is possible to identify the identification target with high accuracy without setting in advance a gray value of RGB that should be excluded. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 is an explanatory diagram showing one example of an element mounting system 10.

[0013] Figure 2 is a structure diagram showing an outline of a structure of the mounting apparatus 20.

[0014] Figure 3 is an explanatory diagram showing an electrical connection relationship of the mounting apparatus 20, the management apparatus 40, and the mounting inspection apparatus 30.

[0015] Figure 4 is a flowchart showing one example of a post-mounting inspection processing program.

[0016] Figure 5 is an explanatory diagram showing one example of the element P as a recognition target.

[0017] Figure 6 is an explanatory diagram showing one example of a situation in which the outer shape of the element P is misrecognized in the comparative example.

[0018] Figure 7 is a flowchart showing one example of a generation process of the recognition image.

[0019] Fig. 8 is an explanatory diagram showing one example of a situation in which the recognition image Gi is generated. DETAILED DESCRIPTION

[0020] Next, the embodiments of the present disclosure will be described using the drawings. Figure 1 is a schematic configuration diagram showing the configuration of the component mounting system 10, Figure 2 is a schematic configuration diagram showing the configuration of the mounting device 20, Figure 3 is an explanatory diagram showing the electrical connection relationship of the mounting device 20, the management device 40, and the mounting inspection device 30. In addition, in the Figure 1 , Figure 2 , the left-right direction is set as the X direction, the front-rear direction is set as the Y direction, and the up-down direction is set as the Z direction.

[0021] As shown in Figure 1 , the component mounting system 10 is provided with a printing device 12, a printing inspection device 14, a plurality of mounting devices 20, a mounting inspection device 30, and a management device 40, which are connected to a LAN 18 as a network. The printing device 12 performs printing on a substrate S (refer to Figure 2 ) by pressing solder into pattern holes formed in a screen mask. The printing inspection device 14 inspects the state of the solder printed by the printing device 12. The mounting devices 20 are arranged along the conveyance direction (X direction) of the substrate S, mount components on the substrate S, and inspect the mounting state of the mounted components. The mounting inspection device 30 inspects the mounting state of the components mounted on the substrate S by each of the mounting devices 20. The management device 40 manages the entire component mounting system 10. The printing device 12, the printing inspection device 14, the plurality of mounting devices 20, and the mounting inspection device 30 are arranged in this order along the conveyance direction of the substrate S to constitute a production line. In addition, the production line can be provided with a reflow soldering device or the like that performs reflow soldering processing on the substrate S on which components are mounted, in addition to these devices, and the mounting inspection device 30 can be arranged on the downstream side of the reflow soldering device.

[0022] As shown in Figure 2 , Figure 3As shown, the mounting apparatus 20 is provided with a substrate conveying apparatus 21 that conveys the substrate S, an element supply apparatus 22 that supplies elements, a head 23 that has a suction nozzle 24 that suctions the elements arranged so as to be able to be raised and lowered, and a head moving apparatus 25 that moves the head 23 in the XY directions. The substrate conveying apparatus 21 has two pairs of conveying belts arranged at intervals in front of and behind each other and erected in the left-right direction, and conveys the substrate S from the left to the right in the figure by each of the conveying belts. The element supply apparatus 22 is, for example, a tape feeder that supplies elements by feeding out a tape in which elements are housed at predetermined intervals, and is provided in the mounting apparatus 20 in a manner that enables a plurality of kinds of elements to be supplied. Figure 2

[0023] In addition, the mounting apparatus 20 is provided with, in addition to these, a mark camera 26, a parts camera 27, a storage section 28, and a mounting control apparatus 29 that controls the entire mounting apparatus 20. The mark camera 26 is mounted to the head 23 and moves in the XY directions together with the head 23 by the head moving apparatus 25. The mark camera 26 generates an image from a photographic object such as an element supplied by the element supply apparatus 22 and an element mounted to the substrate S in addition to a mark marked on the substrate S, and outputs the generated image to the mounting control apparatus 29. In addition, the parts camera 27 is arranged between the element supply apparatus 22 and the substrate conveying apparatus 21, generates an image from an element held (sucked) to the suction nozzle 24 from below, and outputs the generated image to the mounting control apparatus 29. In this embodiment, at least the mark camera 26 is provided with a color photographic element and is configured so as to be able to photograph a color image in which each pixel has a gray scale value (R, G, B) of RGB three primary colors. In addition, the gray scale values (R, G, B) each take a value of 256 levels from 0 to 255. The storage section 28 is a device such as an HDD that stores a processing program, information related to the mounting position of an element, information of a mounting result, and the like.

[0024] The mounting control apparatus 29 is configured from a known CPU, ROM, RAM, and the like. The mounting control apparatus 29 outputs a drive signal to the substrate conveying apparatus 21, the head 23, the head moving apparatus 25, and the like. The mounting control apparatus 29 inputs an image from the mark camera 26 and the parts camera 27. The mounting control apparatus 29 recognizes the position of the substrate S by processing an image of the substrate S before mounting of an element photographed by the mark camera 26 to recognize the position of the substrate S, or recognizes a mounting state of an element by processing an image of the substrate S after mounting of an element photographed by the mark camera 26. In addition, the mounting control apparatus 29 determines whether or not the suction nozzle 24 is sucking an element on the basis of an image photographed by the parts camera 27, or determines a suction posture of the element.

[0025] As shown in FIG. 2, the mounting apparatus 20 is provided with a substrate conveying apparatus 21 that conveys the substrate S, an element supply apparatus 22 that supplies elements, a head 23 that has a suction nozzle 24 that suctions the elements arranged so as to be able to be raised and lowered, and a head moving apparatus 25 that moves the head 23 in the XY directions. The substrate conveying apparatus 21 has two pairs of conveying belts arranged at intervals in front of and behind each other and erected in the left-right direction, and conveys the substrate S from the left to the right in the figure by each of the conveying belts. The element supply apparatus 22 is, for example, a tape feeder that supplies elements by feeding out a tape in which elements are housed at predetermined intervals, and is provided in the mounting apparatus 20 in a manner that enables a plurality of kinds of elements to be supplied. Figure 3 ​As shown, the mounting inspection apparatus 30 includes a substrate conveyor 32 for conveying substrates S with components mounted thereon by the mounting apparatuses 20; an inspection camera 34 for capturing images for inspecting the mounting status of the components; a camera moving apparatus 36 for moving the inspection camera 34 in the X and Y directions; and an inspection control device 39 for controlling the entire mounting inspection apparatus 30. The substrate conveyor 32 and camera moving apparatus 36 are identical in structure to the substrate conveyor 21 and head moving apparatus 25 of the mounting apparatus 20, respectively.

[0026] The inspection control device 39 is composed of a well-known CPU, ROM, RAM, and other components. It outputs drive signals to the substrate conveyor 32 and camera moving device 36, and outputs imaging signals to the inspection camera 34. The inspection control device 39 also receives images from the inspection camera 34 and processes these images to inspect the mounting status of components. Furthermore, the inspection control device 39 is communicatively connected to the mounting control device 29 and the management control device 42 of the management device 40 via the LAN 18, transmitting information related to inspection status and results.

[0027] like Figure 3 As shown, the management device 40 includes: a management control device 42, a storage unit 44, an input device 46 and a display 48. The management control device 42 is composed of a well-known CPU, ROM, RAM, etc. The storage unit 44 is a device such as an HDD that stores various information such as processing programs. The input device 46 includes a keyboard and a mouse for the operator to input various commands. The display 48 is a liquid crystal display device that displays various information. In addition, a production program is stored in the storage unit 44. The production program specifies information on the types of components to be installed on the substrate S, the installation order of each component, the installation position of each component, the installation angle, the component supply device 22 that supplies each component, the suction nozzle 24 that adsorbs the component, and the number of substrates S produced. The management control device 42 is connected to the installation control device 29 via the LAN 18 in a communicative manner, receives information related to the installation status from the installation control device 29, or sends the production program to the installation control device 29. The management control device 42 is communicatively connected to the inspection control device 39 via the LAN 18, and receives information regarding the inspection status and inspection results from the inspection control device 39, or transmits information regarding the substrate S to be inspected to the inspection control device 39. The management control device 42 is also communicatively connected to each control device (not shown) of the printing device 12 and the print inspection device 14 via the LAN 18, and receives information regarding the job status from each device, or transmits job instructions.

[0028] The following describes the operation of the mounting device 20 thus configured, with respect to the mounting process of components and the inspection process of the board S after mounting of components. In the mounting process of components, the mounting control device 29 first controls the head moving device 25 to move the head 23 above the component supply position of the component supply device 22, and lowers the suction nozzle 24 to cause the suction nozzle 24 to pick up the component supplied to the component supply position. Next, the mounting control device 29 controls the head moving device 25 to move the head 23 above the part camera 27, and controls the part camera 27 to take an image of the component picked up by the suction nozzle 24. Then, the mounting control device 29 processes the taken image to determine the positional deviation and the like of the component picked up by the suction nozzle 24, and corrects the target mounting position of the component in such a manner as to eliminate the positional deviation. Further, the mounting control device 29 controls the head moving device 25 to move the head 23 above the board S and lowers the suction nozzle 24 to mount the component on the target mounting position on the board S. In addition, the mounting control device 29 of the present embodiment performs the inspection process of inspecting the mounting state of components before the board S is carried out, when the mounting of components on the board S is completed. Figure 4 is a flowchart showing one example of the post-mounting inspection process program.

[0029] When the program is started, the mounting control device 29 first acquires the color image of the board S after mounting of components, taken by the marker camera 26 (S100). As described above, each pixel of the color image taken by the marker camera 26 has the gray scale values (R, G, B) of the RGB three primary colors. Next, the mounting control device 29 performs a generation process of a recognition image for recognizing components from the acquired color image (S110). Details of the generation process will be described later. Then, the mounting control device 29 recognizes components using the generated recognition image to inspect the mounting state (S120), and registers (stores) the inspection result in the storage section 28 (S130), and ends the post-mounting inspection process program. In S120, the mounting control device 29 acquires, for example, the deviation of the mounting position of components and the deviation of the mounting angle of components, to inspect whether or not the amount of deviation in the X direction and the amount of deviation in the Y direction with respect to the target mounting position, and the amount of rotational deviation (angle) with respect to the target mounting angle, are within the reference values, and the like. In addition, the mounting control device 29 also inspects whether or not the components are defective and the like.

[0030] Here, Figure 5 is an explanatory diagram showing one example of the component P to be recognized. As illustrated, the component P is, for example, a component having a rectangular shape in plan view, and having electrodes Pe at both end portions and a central portion formed of a non-glossy material such as resin. Therefore, in the image processing, the electrodes Pe are recognized as the recognition object to recognize the outer shape and position of the component P. In addition, Figure 6This is an explanatory diagram showing an example of a situation in which the outer shape of a component P is erroneously recognized in a comparative example. Figure 6 In the figure, it is shown that Figure 5 The situation when processing an image of a component P on a substrate S. When the image is captured, the electrode Pe of the component P reflects light and emits white light, so it appears white in the image. Therefore, the grayscale values ​​(R, G, B) of the electrode Pe are, for example, (255, 255, 255). According to the following equation (1), the brightness Y is 255.

[0031] Y=0.30×R+0.59×G+0.11×B···(1)

[0032] In addition, the image also shows the copper foil Cf provided on the substrate S and the solder So (shown as a dotted line) printed on the substrate S by the printing device 12. Among them, the copper foil Cf reflects light more easily than the solder So and appears in the image in orange or yellow. The grayscale values ​​(R, G, B) of the copper foil Cf are, for example, (239, 160, 154), and according to formula (1), the brightness Y is 183. In this way, for the copper foil Cf, the main color component is red (R), which is close to orange and yellow, and is different from the electrode Pe, but the brightness Y is relatively high and similar to the brightness Y of the electrode Pe. ■Here, the main color component is different from the electrode Pe to be identified, which means that since the electrode Pe is reflected in white, the three primary colors all have high grayscale values, while one of the three primary colors has a higher grayscale value than the other primary colors. In addition, if the main color component of the identification object, that is, the primary color with a higher grayscale value, is one of the three primary colors, then when the primary color different from the primary color has a higher grayscale value, it is sufficient to regard the main color component as different. In addition, the brightness Y similar to the electrode Pe refers to a predetermined range within a few tens of % such as about 30% or less in difference with respect to the brightness Y of the electrode Pe. Therefore, when the installation control device 29 performs image processing on the color image captured by the marking camera 26, it is difficult to distinguish between the electrode Pe and the copper foil Cf, and sometimes the copper foil Cf is mistakenly identified as a part of the electrode Pe. In this case, the installation control device 29 recognizes the shape of the component P including a part of the copper foil Cf, and therefore misjudges the deviation of the installation position of the component P or the deviation of the installation angle. For example, in Figure 6 The above is an example of a situation where the mounting control device 29 misidentifies the outer shape of the component P and mistakenly determines that the mounting angle has deviation, even though there is actually little deviation in the mounting angle of the component P. To prevent such misidentification, in this embodiment, the generation process of the recognition image in S110 is performed as follows. Figure 7is a flowchart showing one example of the generation processing of the recognition-use image, and FIG. 8 is an explanatory diagram showing one example of the state of generating the recognition-use image Gi. In addition, in FIG. 8, the hatching of the gray scale values is shown differently from the actual gray scale values of each image, and, for example, even if the portion of the copper foil Cf and the like is shown with a uniform gray scale value, actually, a portion with a high gray scale value and a portion with a low gray scale value are mixed.

[0033] In Figure 7 the generation processing of the recognition-use image, the mounting control device 29 first generates an R image Gr (a first primary color image) that extracts the gray scale value R of red and a B image Gb (a second primary color image) that extracts the gray scale value B of blue, based on the gray scale values (R, G, B) of each pixel of the color image acquired in S100 described above (S200). As described above, the main color component of the copper foil Cf is orange or yellow, and therefore the mounting control device 29 generates the R image Gr by extracting the gray scale value R of red, which is close to orange and yellow, among the RGB three primary colors (see Figure 8A ). In addition, the mounting control device 29 generates the B image Gb by extracting the gray scale value B of blue, which is close to the complementary color of orange or yellow, among the RGB three primary colors (see Figure 8B ). In addition, the gray scale value R of red, which is close to the main color component of the copper foil Cf, becomes the largest value among the gray scale values (R, G, B) of the copper foil Cf, and the gray scale value B of blue, which is close to the complementary color, becomes the smallest value among the gray scale values (R, G, B) of the copper foil Cf.

[0034] Next, the mounting control device 29 generates a difference image Gd that has a gray scale value obtained by subtracting the gray scale value B of each pixel of the B image Gb from the gray scale value R of each pixel of the R image Gr (S210). For example, the gray scale values (R, G, B) of the electrode Pe are (255, 255, 255), and the gray scale values (R, G, B) of the copper foil Cf are (239, 160, 154), and therefore the gray scale value that is the difference in the portion of the electrode Pe is a value 0 or the like, and the gray scale value that is the difference in the portion of the copper foil Cf is a value 85 or the like. Therefore, as shown in Figure 8C , the difference image Gd has a difference that is almost a value 0 (black) in the portion of the electrode Pe and the portion of the background, and a difference of the gray scale value R and the gray scale value B of the copper foil Cf appears in the portion of the copper foil Cf. In addition, the gray scale value R of red, which is the largest among the gray scale values (R, G, B) of the copper foil Cf, and the gray scale value B of blue, which is the smallest, are greatly expressed as the difference in the portion of the copper foil Cf.

[0035] Next, the mounting control device 29 performs a smoothing process on the difference image Gd to generate a difference image Gd' (a difference smoothed image, a post-smoothing image, see Figure 8D). The smoothing processing of S220 is performed by blurring the difference between the gray scale value of the gray scale value R and the gray scale value B by smoothing the difference using, for example, a known moving average filter, a Gaussian filter, or the like, and can eliminate the influence of noise and the like. Also, the mounting control device 29 generates an identification image Gi having a gray scale value obtained by subtracting the gray scale value of each pixel of the difference image Gd' from the gray scale value of each pixel of the B image Gb (S230), and ends the generation processing of the identification image. Here, the gray scale value of the portion of the copper foil Cf in the B image Gb is not as high as the R image Gr, and the identification image Gi is generated by subtracting the difference image Gd' from the B image Gb, so the gray scale value of the portion of the copper foil Cf can be further suppressed in the identification image Gi. Therefore, as shown in FIG. 10, the identification image Gi becomes an image in which the brightness of the portion of the copper foil Cf is lower than the original color image (refer to FIG. 1), the R image Gr of Figure 8E Figure 6 Figure 8A Figure 8B On the other hand, the portion of the electrode Pe also becomes a higher gray scale value (B = 255) in the B image Gb, and the difference also becomes a value of 0 or the like in the difference image Gd (Gd' ), so almost no difference is generated. Therefore, in the identification image Gi, the brightness of the portion of the electrode Pe is substantially unchanged from the original color image. Therefore, the mounting control device 29 can prevent the copper foil Cf from being included in the electrode Pe and misidentifying the outer shape of the element P when identifying the element P using the identification image Gi in S120, so the electrode Pe can be appropriately identified.

[0036] ​​​As mentioned above, the copper foil Cf appears orange or yellow within the color image. However, depending on the light reflection conditions, a portion of the copper foil Cf, such as the center portion, may appear white, resulting in grayscale values ​​(R, G, B) of (255, 255, 255) or values ​​close thereto. In this case, the difference image Gd generated in S210 will have a value of 0 for a portion of the copper foil Cf, similar to the electrode Pe. The remaining portion of the copper foil Cf will exhibit the aforementioned difference between grayscale values ​​R and B. Furthermore, if the mounting control device 29 does not perform the smoothing process in S220 and instead subtracts the difference image Gd from the B image Gb to generate the identification image Gi, the brightness of the remaining portion of the copper foil Cf will decrease, leaving a portion of the copper foil Cf at the same brightness as the electrode Pe. This could lead the mounting control device 29 to mistakenly identify the boundary between a portion of the copper foil Cf and the remaining portion as the boundary of the electrode Pe. In contrast, if the mounting control device 29 performs the smoothing process in S220, the difference generated in the remaining portion of the copper foil Cf is dispersed to a portion of the copper foil Cf. This prevents the portion of the copper foil Cf from having a brightness that is significantly different from the remaining portion. Thus, the mounting control device 29 can reduce the possibility of misidentifying the boundary between the portion of the copper foil Cf and the remaining portion. Thus, in order to reduce the influence of a sudden change in the difference and prevent misidentification, the smoothing process in S220 is performed in this embodiment.

[0037] Here, the correspondence between the structural elements of this embodiment and the structural elements of the present disclosure is clarified. Figure 4 The installation control device 29 of S100 of the post-installation inspection processing program corresponds to the image acquisition unit and executes Figure 7 The mounting control device 29 performing the recognition image generation process in steps S200 to S220 corresponds to the differential image generation unit, the mounting control device 29 performing the recognition image generation process in step S230 corresponds to the recognition image generation unit, and the mounting control device 29 performing the post-mount inspection process in step S120 corresponds to the recognition processing unit. Furthermore, the mark camera 26 corresponds to the imaging device, the mounting control device 29 corresponds to the image processing device, and the component mounting system 10 corresponds to the component mounting system. Furthermore, in this embodiment, the operation of the mounting control device 29 is described to clarify an example of the image processing method disclosed herein.

[0038] The mounting control device 29 of the embodiment described above generates a difference image Gd obtained by subtracting the B image Gb (second primary color image) from the R image Gr (first primary color image). Therefore, the gray scale value of the difference image Gd becomes a difference between a red color (first primary color) similar to the main color component of the copper foil Cf in the similar portion of the electrode Pe and a blue color (second primary color) other than the red color. In addition, because the difference image Gd is subtracted from the B image Gb (Gd') to generate the recognition image Gi, the brightness of the copper foil Cf can be reduced in the recognition image Gi, so that the copper foil Cf can be prevented from being erroneously recognized as the electrode Pe. Therefore, the electrode Pe as a recognition target can be recognized with high accuracy without previously setting the gray scale value of RGB that should be excluded before image capturing.

[0039] In addition, the mounting control device 29 can make the difference between the red color similar to the main color component and the blue color similar to the complementary color of the main color component larger and further reduce the brightness of the copper foil Cf in the recognition image Gi, compared to the case where the gray scale value of green is extracted, because the B image Gb is used in which the gray scale value of the blue color similar to the complementary color of the main color component of the copper foil Cf is extracted.

[0040] In addition, the mounting control device 29 can reduce the influence caused by the sharp change of the difference in the recognition image Gi and prevent the copper foil Cf from being erroneously recognized as the electrode Pe, because the difference image Gd' having the gray scale value obtained by smoothing the difference obtained by subtracting the B image Gb from the R image Gr is generated.

[0041] In addition, the present disclosure is not limited to any of the above-described embodiments, and can be implemented in various ways as long as it belongs to the technical scope of the present disclosure.

[0042] For example, in the above-described embodiment, the recognition image Gi is generated using the difference image Gd' having the gray scale value obtained by smoothing the difference obtained by subtracting the B image Gb from the R image Gr, but is not limited thereto. For example, the difference image Gd obtained by subtracting the B image Gb from the R image Gr can be used as it is without smoothing to generate the recognition image Gi.

[0043] In the above-described embodiment, the difference image Gd is generated using the B image Gb in which the gray scale value of the blue color close to the complementary color of the main color component of the copper foil Cf is extracted, but it is not limited to using the complementary color, and a primary color image of another color can be used. That is, it can also be that the gray scale value G of green among the RGB three primary colors is extracted and the G image Gg is generated as the second primary color image, and the difference image Gd (Gd') is generated by subtracting the G image Gg from the R image Gr. In this case, the identification image Gi can be generated by subtracting the difference image Gd (Gd') from the G image Gg. In addition, in the case where the R image Gr is used as the first primary color image and the B image Gb is used as the second primary color image, the identification image Gi can be generated by subtracting the difference image Gd (Gd') from the B image Gb as the second primary color image, but it is not limited thereto, and the identification image Gi can be generated by subtracting the difference image Gd (Gd') from any one of the R image Gr, the G image Gg, and the like as the first primary color image.

[0044] In the above-described embodiment, the process of excluding the copper foil Cf provided to the substrate S as the similar site is exemplified, but it is not limited thereto, and the solder So and the like printed on the substrate S by the printing device 12 can be excluded as the similar site, and the substrate S itself can be excluded as the similar site.

[0045] In the above-described embodiment, the present disclosure is applied to the inspection process by the mounting device 20, but it is not limited thereto, and the present disclosure can be applied to the inspection process by the mounting inspection device 30. In this case, the inspection control device 39 can perform the same process as the image processing device. In addition, it is not limited to the mounting control device 29 and the inspection control device 39, and the management control device 42 of the management device 40 can perform the same process as the image processing device. Alternatively, it can be that the mounting control device 29 performs the acquisition of the color image and the generation of the difference image Gd (Gd'), the identification image Gi, the inspection control device 39 and the management control device 42 perform the identification process (inspection process) using the identification image Gi, and the like, and the process is performed by two or more devices. Alternatively, it is not limited to the inspection process of the substrate S after the component mounting to which the present disclosure is applied, and the inspection process of the inspection of the coating state of the solder So before the component mounting can be applied to the present disclosure. In this case, the control device of the printing inspection device 14 can perform the same process as the image processing device.

[0046] In the above-described embodiment, as the recognition target, the element P having a rectangular shape in plan view and provided with the electrodes Pe at both end portions is exemplified, but the present disclosure is not limited thereto, and an element not provided with the electrodes Pe at both end portions, an element having a shape other than a rectangular shape such as a circular shape in plan view, or the like can be a recognition target. In addition, the element P mounted to the substrate S is a recognition target, but the present disclosure is not limited thereto, and an element before being mounted to the substrate S, such as an element supplied to the element supply position by the element supply device 22 or an element temporarily placed at a predetermined position after being sucked by the nozzle 24 at the element supply position, can be a recognition target.

[0047] In the above-described embodiment, the present disclosure is applied to image processing required for inspection in mounting processing of the element P to the substrate S, but the present disclosure can be applied to, for example, image processing of a foreign matter as a recognition target for checking the presence or absence of the foreign matter attached to a product, or the like, not limited to image processing required for inspection in mounting processing.

[0048] In this regard, the image processing device of the present disclosure can also be configured as follows. For example, in the image processing device of the present disclosure, the above-described recognition image generation section can generate an image having a gray scale value obtained by subtracting a gray scale value of the above-described difference image from a gray scale value of the above-described second primary color image as the above-described recognition image. The gray scale value of the second primary color image of the similar site is not higher than the gray scale value of the first primary color image, and the brightness can be suppressed. Therefore, by subtracting the difference image from the second primary color image, the brightness of the similar site can be further reduced in the recognition image, and thus the similar site can be further suppressed from being erroneously recognized as the recognition target. In addition, since the second primary color image is used for generation of the difference image, the processing can be performed promptly without preparing a new primary color image.

[0049] In the image processing device of the present disclosure, the above-described difference image generation section can use an image in which a primary color close to a complementary color of a main color component of the above-described similar site among RGB three primary colors is set as the above-described second primary color as the above-described second primary color image. In this way, the difference between the gray scale value of the first primary color image and the gray scale value of the second primary color image, that is, the gray scale value of the difference image, is further increased, and thus the gray scale value obtained by subtracting the gray scale value of the difference image from the gray scale value of the second primary color image can be reduced. Therefore, the brightness of the similar site can be further reduced in the recognition image, and thus the similar site can be further suppressed from being erroneously recognized as the recognition target.

[0050] In the image processing apparatus of the present disclosure, the difference image generation section can generate, as the difference image, an image having a gray scale value obtained by smoothing a difference between a gray scale value of the first primary color image and a gray scale value of the second primary color image. This makes it possible to suppress a sharp change in the difference between pixels of the difference image, and thus makes it possible to reduce cases in which a sharp change in the gray scale value occurs in a similar portion in the recognition image. This makes it possible to suppress cases in which a portion in which a sharp change in the gray scale value occurs in the similar portion is erroneously recognized as a boundary or the like of the recognition target.

[0051] The component mounting system of the present disclosure is a component mounting system that mounts components on a board, and the gist thereof is,

[0052] The component mounting system includes: an imaging device that images a color image of the board on which the components are mounted; and

[0053] Any of the image processing apparatuses described above processes the color image that includes the components mounted on the board as the recognition target.

[0054] The component mounting system of the present disclosure includes: an imaging device that images a color image of a board on which components are mounted; and any of the image processing apparatuses of the present disclosure that processes the color image that includes the components mounted on the board as the recognition target. Thus, as with the image processing apparatuses described above, the recognition target can be recognized with high accuracy without previously setting the gray scale values of RGB that should be excluded.

[0055] The image processing method of the present disclosure is an image processing method that processes a color image having a gray scale value of RGB three primary colors, and the gist thereof is,

[0056] The image processing method includes the steps of:

[0057] (a) acquiring, as the color image, an image that includes a recognition target and a similar portion in which a main color component is different from the recognition target and a brightness is similar to the recognition target;

[0058] (b) generating, using a first primary color image and a second primary color image, a difference image having a gray scale value based on a difference between a gray scale value of the first primary color image and a gray scale value of the second primary color image, the first primary color image being an image in which a gray scale value of a first primary color of RGB three primary colors close to a main color component of the similar portion is extracted from the color image, the second primary color image being an image in which a gray scale value of a second primary color of RGB three primary colors other than the first primary color is extracted from the color image;

[0059] (c) generating an identification image having a gray scale value obtained by subtracting the gray scale value of the difference image from the gray scale value of the image in which any one of the RGB primary colors is extracted from the color image; and

[0060] (d) performing identification processing of the identification object using the identification image.

[0061] The image processing method of the present disclosure, like the image processing apparatus described above, generates a difference image having a gray scale value based on a difference obtained by subtracting the gray scale value of a second primary color image from the gray scale value of a first primary color image, and generates an identification image by subtracting the gray scale value of the difference image from the gray scale value of any one of the primary color images. Thus, the identification object can be identified with high accuracy without previously setting the gray scale value of RGB that should be excluded. In this image processing method, the various modes of the image processing apparatus described above can be employed, or steps for implementing each function of the image processing apparatus can be additionally implemented.

[0062] Industrial applicability

[0063] The present disclosure can be utilized in the technical field of image processing of color images and mounting processing of components, and the like.

[0064] Explanation of reference numerals

[0065] 10...component mounting system; 12...printing device; 14...print inspection device; 18...LAN; 20...mounting device; 21, 32...substrate conveyance device; 22...component supply device; 23...head; 24...suction nozzle; 25...head moving device; 26...mark camera; 27...part camera; 28...storage section; 29...mounting control device; 30...mounting inspection device; 34...inspection camera; 36...camera moving device; 39...inspection control device; 40...management device; 42...management control device; 44...storage section; 46...input device; 48...display; Cf...copper foil; Gd, Gd'...difference image; Gb...B image; Gi...identification image; Gr...R image; P...component; Pe...electrode; S...substrate; So...solder.

Claims

1. An image processing apparatus that processes a color image having gray scale values of RGB three primary colors for each pixel, the image processing apparatus comprising: an image acquisition section that acquires an image including an identification target and a similar portion having a main color component different from the identification target and a brightness similar to the identification target as the color image; a difference image generation section that generates a difference image having a gray scale value based on a difference obtained by subtracting a gray scale value of a second primary color image from a gray scale value of a first primary color image, the first primary color image being an image in which a gray scale value of a first primary color close to the main color component of the similar portion among RGB three primary colors is extracted from the color image, the second primary color image being an image in which a gray scale value of a second primary color other than the first primary color among RGB three primary colors is extracted from the color image; an identification image generation section that generates an identification image having a gray scale value obtained by subtracting a gray scale value of the difference image from a gray scale value of the second primary color image; and an identification processing section that performs an identification process of the identification target using the identification image, the identification image generation section generates an image having a gray scale value obtained by subtracting a gray scale value of the difference image from a gray scale value of the second primary color image as the identification image.

2. The image processing apparatus according to claim 1, wherein the difference image generation section uses an image in which a primary color close to a complementary color of the main color component of the similar portion among RGB three primary colors is set as the second primary color as the second primary color image.

3. The image processing apparatus according to claim 1 or 2, wherein the difference image generation section generates an image having a gray scale value obtained by smoothing a difference obtained by subtracting a gray scale value of the second primary color image from a gray scale value of the first primary color image as the difference image.

4. An component mounting system that mounts components on a board, the component mounting system comprising: a photographing apparatus that photographs a color image of the board on which the components are mounted; and the image processing apparatus according to any one of claims 1 to 3 that processes the color image including the components mounted on the board as the identification target.

5. An image processing method that processes a color image having gray scale values of RGB three primary colors for each pixel, the image processing method comprising steps of: (a) acquiring an image including an identification target and a similar portion having a main color component different from the identification target and a brightness similar to the identification target as the color image; (b) generating a difference image having a difference value based on a difference between a gray scale value of a first primary color image and a gray scale value of a second primary color image, the first primary color image being an image in which a gray scale value of a first primary color of RGB three primary colors close to a main color component of the similar part is extracted from the color image, the second primary color image being an image in which a gray scale value of a second primary color of RGB three primary colors other than the first primary color is extracted from the color image; (c) generating an identification image having a gray scale value obtained by subtracting a gray scale value of the difference image from a gray scale value of the second primary color image; and (d) performing an identification process of the identification object using the identification image.

Images