WO2017068638A1

WO2017068638A1 - Image processing apparatus and parts mounter

Info

Publication number: WO2017068638A1
Application number: PCT/JP2015/079522
Authority: WO
Inventors: 石川　信幸
Original assignee: 富士機械製造株式会社
Priority date: 2015-10-20
Filing date: 2015-10-20
Publication date: 2017-04-27
Also published as: JP6795512B2; JPWO2017068638A1

Abstract

This image processing apparatus stores, in a storage unit, information on two lead existence regions, each including one existence location of two leads (L) and (R) in one part P (PA01). Each time the part P is imaged, on the basis of gray-scale values of pixels acquired from two partial region images A, B corresponding to the lead existence regions, the image processing apparatus sets, for each partial region image, a threshold for determining a boundary of the lead included in the partial region image. Thus, image processing is implemented with a threshold suited for each lead existence location and the threshold is optimized for each captured image, and hence the boundary of each lead can be properly determined.

Description

Image processing apparatus and component mounting machine

The present invention relates to an image processing apparatus and a component mounting machine.

Conventionally, an image is picked up while irradiating light from a light source onto a component having a plurality of leads (points of interest), the picked-up image is binarized to detect the position of each lead, and the detected lead position Based on this, a component mounter that corrects the mounting position and orientation of components and mounts them on a board has been proposed. For example, the component mounter of Patent Document 1 provides a reference pin formed of a material equivalent to a component lead in a robot hand that holds a component, and images the reference pin before imaging the component. Then, based on the captured image of the reference pin, a threshold for binarizing the captured image of the component is set. Thus, it is possible to set an appropriate threshold value by eliminating the influence of the deterioration of the light source.

Japanese Patent Laid-Open No. 08-330797

However, the above-described component mounter does not take into consideration the uneven irradiation of the light irradiated to the component and the uneven reflection of the irradiated light. Here, depending on the size, shape, material, etc. of the part, there may be a partial difference in the light irradiation from the light source and the reflection of the irradiated light, resulting in uneven light irradiation and uneven reflection. is there. In that case, in the captured image of one component, there is a mixture of parts with greatly different brightness tendencies, so even if the threshold value set from the captured image of the reference pin is uniformly used for the entire image, the binarization process is appropriate. May not be able to be processed.

The main object of the present invention is to more appropriately perform image processing when detecting a point of interest of a part.

The present invention adopts the following means in order to achieve the main object described above.

The image processing apparatus of the present invention
An image processing apparatus that processes a captured image of the component imaged before mounting in a component mounter that mounts the component on a board,
A storage unit for storing region information regarding a plurality of regions each including a plurality of points of interest in one part;
An extraction unit that extracts a plurality of partial region images corresponding to the region information from the captured image;
For each of the extracted partial region images, an acquisition unit that acquires a gradation value of a pixel included in the partial region image;
Setting that sets a threshold for determining the boundary of the point of interest included in the partial area image based on the acquired gradation value of the pixel for each partial area image every time the part is imaged And
It is a summary to provide.

In the image processing apparatus of the present invention, region information relating to a plurality of regions each including a plurality of points of interest in one component is stored, and a plurality of partial region images corresponding to the region information each time a component is imaged. Based on the gradation value of each pixel acquired from the above, a threshold value for determining the boundary of the point of interest included in the partial area image is set for each partial area image. For this reason, it is possible to perform image processing using an appropriate threshold value for each point of interest. Moreover, since a threshold value is set every time a part is imaged, an optimum threshold value can be used for each captured image. Therefore, it is possible to more appropriately perform image processing when detecting the part of interest of the component.

In the image processing apparatus of the present invention, the component has a plurality of leads that are respectively inserted into a plurality of insertion holes provided in the substrate, and the plurality of points of interest are each of the plurality of leads. The storage unit may store information on a plurality of regions including one each of the presence locations of each lead as the region information. In this way, it is possible to improve the accuracy of detecting the position of the lead by appropriately performing image processing when detecting the position of the lead.

In the image processing apparatus of the present invention, for each of the plurality of partial area images, a binary value that binarizes the gradation value of each pixel included in the partial area image using a threshold set by the setting unit. It is good also as a thing provided with a conversion processing part. In this way, it is possible to accurately perform the process of determining the boundary (edge) of the part of interest of the component by binarization.

The component mounter of the present invention is
A component mounter comprising the above-described image processing apparatus of the present invention,
A supply device for supplying parts;
A mounting device for holding the component supplied by the supply device with a holder and mounting the held component on the substrate;
An imaging device for imaging the component;
When the captured image captured by the imaging device is processed by the image processing device, the position of the component held by the holder is acquired based on the processing result, and the component is acquired based on the acquired position. A control device for controlling the mounting device so that is mounted on the substrate;
It is a summary to provide.

Since the component mounter according to the present invention includes the image processing apparatus according to the present invention described above, the same effect as that exhibited by the image processing apparatus is achieved. For this reason, it is possible to more appropriately perform image processing when detecting the part of interest of the component, and to improve the mounting accuracy of the component. Note that the imaging apparatus may capture an image of the component while the component is held by the holder, or may capture an image of the component while the component is not held by the holder.

In the component mounter of the present invention, the supply device supplies a component having a plurality of leads respectively inserted into a plurality of insertion holes provided in the substrate as the component, and at least one of the plurality of leads. A cutting device that partially cuts the leads, and the image processing device includes a device in which the storage unit stores region information regarding a plurality of regions each including the plurality of leads. The part that is held by the holding tool in a state in which at least one lead is partially cut by the cutting device is imaged, and the control device holds the holding device based on the processing result of the image processing device. And acquiring the positions of the plurality of leads of the component being controlled, and controlling the mounting apparatus so that the leads are inserted into the insertion holes based on the acquired positions. It can also be a thing. In this way, it is possible to improve the accuracy of detecting the position of the lead by appropriately performing image processing when detecting the position of the lead. In addition, since the lead can be accurately inserted into the insertion hole of the substrate based on the acquired lead position, the mounting accuracy can be improved.

1 is a configuration diagram showing an outline of the configuration of a component mounter 10. FIG. FIG. 3 is a block diagram showing a configuration relating to control of the component mounter 10. The block diagram which shows the outline of a structure of the parts camera. The flowchart which shows an example of a component mounting process. The flowchart which shows an example of an image process. Explanatory drawing which shows an example of a partial region image. Explanatory drawing which shows an example of a mode that the binarization threshold value is set for every partial area image. Explanatory drawing which shows an example of a mode that the binarization threshold value is set from the target image of a comparative example. The perspective view of the components P of a modification. Explanatory drawing of a mode that the boundary of several attention location is discriminated in the component P of a modification.

FIG. 1 is a block diagram showing an outline of the configuration of the component mounter 10, and FIG. 2 is a block diagram showing a configuration related to control of the component mounter 10. 1 is the X-axis direction, the front-rear direction is the Y-axis direction, and the vertical direction is the Z-axis direction.

As shown in FIG. 1, the component mounter 10 includes a component supply device 12 that supplies a component P, a substrate transfer device 16 that transfers a flat substrate S, and a substrate holding device 18 that holds the transferred substrate S. And a head 20 to which a component chuck device 22 that grips the component P is attached, and a moving mechanism 24 that moves the head 20 in the XY directions. In addition, the component mounter 10 includes a mark camera 26 that captures a mark attached to the substrate S, a part camera 30 that captures a component P gripped by the component chuck device 22, and image processing that performs image processing of various images. The apparatus 40 (refer FIG. 2) and the control apparatus 50 (refer FIG. 2) which control the whole component mounting machine 10 are provided. The component chuck device 22 grips the component P using an openable / closable chuck claw, and is detachably attached to the head 20. In addition to the component chuck device 22, the head 20 can be attached with a component suction device that sucks the component P using a nozzle.

The component supply device 12 includes a tape feeder 13 that supplies a component P by sending out a tape to which a component P with a lead (a radial component shown in FIG. 1 or an axial component not shown) is attached. The tape feeder 13 includes a lead cutter 14 that cuts the lead L of the component P into a predetermined length, a bending mechanism (not shown) that bends the lead L of the axial component downward, and the like. As shown in the enlarged view in FIG. 1, the lead cutter 14 has a fixed blade 14 a and a movable blade 14 b that can be operated by driving an air cylinder (not shown) opposed to each other. The lead cutter 14 shears the lead L by advancing to a position where the tip of the movable blade 14b overlaps the fixed blade 14a, for example, with the lead L of the radial part interposed therebetween. When the component P is mounted on the substrate S, the lead L is inserted into the insertion hole formed in the substrate S. The lead L inserted into the insertion hole is subjected to processing such as bending from the lower surface side of the substrate S by a lead processing device (not shown).

The head 20 includes an elevating mechanism and a rotating mechanism (not shown), and moves the attached component chuck device 22 and the component suction device up and down in the Z-axis direction and rotates (rotates) around the axis. An air flow path connected to a negative pressure source such as a vacuum pump (not shown) is provided in the head 20. The head 20 supplies the negative pressure through the air flow path from the negative pressure source, thereby causing the component chuck device 22 to grip the component P or causing the component suction device to suck the component P.

The part camera 30 captures the part P from below when the part P gripped by the part chuck device 22 passes above the part camera 30, and outputs the obtained captured image to the image processing apparatus 40. FIG. 3 is a configuration diagram showing an outline of the configuration of the parts camera 30. As shown in FIG. 3, the parts camera 30 includes an imaging element 32 having a rectangular imaging region in which a plurality of light receiving elements such as CCDs are two-dimensionally arranged, a lens 34 provided above the imaging element 32, and And an illumination device 36 that emits light from the side of the lead L when imaging the component P. The illumination device 36 irradiates light from the side of the lead L, so that it is on the lower surface of the component P (the background portion of the captured image) and the front end surface of the lead L, compared to the light irradiating from below the lead L. The degree of reflection of the irradiation light is moderate, and the image processing apparatus 40 can easily detect the boundary (edge) of the lead L in the captured image. Here, as described above, since the lead L is sheared by the lead cutter 14, depending on the wear state of the fixed blade 14a and the movable blade 14b, the posture (inclination) of the part P, etc. ) And the angle of the surface to be imaged differ for each component P. For this reason, the degree of light reflection differs for each component P (for each lead L), and the brightness in the captured image may change partially.

The control device 50 is constituted by a CPU, ROM, RAM, HDD, and the like, and controls the entire device. As shown in FIG. 2, the control device 50 includes a drive signal to each device such as the component supply device 12, the substrate transport device 16, and the substrate holding device 18, a drive signal to the head 20 (component chuck device 22), and a moving mechanism. 24, a drive signal to the mark camera 26 and the parts camera 30, and the like are output. The control device 50 is communicably connected to the image processing device 40, and outputs information related to the execution of image processing to the image processing device 40, and inputs information related to the result of image processing from the image processing device 40. Or

The image processing apparatus 40 includes a CPU, a ROM, a RAM, and the like, and performs image processing by inputting a captured image captured by the mark camera 26 or the part camera 30. The image processing apparatus 40 stores an image processing program and various types of information in a storage unit 42 such as a ROM. The storage unit 42 stores component information regarding the component P with leads. The component information includes a component type that indicates the type of the component P, the number of leads L that the component P has, and information that can identify a lead presence area that includes a position where each lead L exists. The lead existence area is a rectangular area including the tip position of the lead L in the captured image, and is an area in which position coordinates of the upper left corner and the lower right corner are determined. The position coordinates are represented by XY coordinates based on the position of the center of the captured image (the center of the part P imaged by the parts camera 30). For example, of the two leads of the component type “PA01”, the lead existing area of the left lead L (L) is the upper left corner position coordinates (XL1s, YL1s) and the lower right corner position coordinates (XL1e, YL1e). Is stipulated.

The following is a description of the mounting operation of the component mounter 10 thus configured. FIG. 4 is a flowchart showing an example of the component mounting process. This process is executed by the control device 50.

When the component mounting process is executed, the control device 50 first causes the component chuck device 22 to grip the component P supplied from the component supply device 12 (S100). In S100, the control device 50 controls the component supply device 12 to send out the component P, controls the moving mechanism 24 so that the head 20 moves to the component supply position, and the component chuck device 22 can grip the component P. The head 20 is controlled so as to be in the height position and orientation, and a negative pressure from a negative pressure source is supplied to cause the component chuck device 22 to perform a gripping operation. Subsequently, the control device 50 controls the moving mechanism 24 so that the component P gripped by the component chuck device 22 moves onto the substrate S via the upper part of the parts camera 30 (S110). The parts camera 30 is controlled so that the part P is imaged when it is on (S120). The control device 50 also instructs the image processing device 40 to execute image processing. Next, the control apparatus 50 acquires the result of the image processing performed by the image processing apparatus 40 on the captured image obtained by the imaging in S120 (S130). Although details of the image processing will be described later, since the image processing device 40 performs a process of detecting the position (tip position) of the lead L from the captured image, the control device 50 acquires the position of the lead L in S130.

Then, the control device 50 sets a correction value related to the mounting position of the component P based on the acquired position of the lead L (S140), and mounts the component P at a position on the substrate S reflecting the correction value. (S150), the component mounting process is terminated. Here, there is a case where the component P gripped by the component chuck device 22 is tilted or deviated from a normal state and has a poor posture. For this reason, the control device 50 determines the mounting position and axis in the XY directions based on the position (tip position) of the lead L in S140 so that the lead L can be correctly inserted into the insertion hole even if the component P has a poor posture. Set the correction value for the mounting direction around. In S150, the control device 50 controls the moving mechanism 24 and the head 20 so that the tip of the lead L of the component P held by the component chuck device 22 is directly above the insertion hole of the substrate S. After controlling the head 20 so that the component chuck device 22 is lowered until the component chuck device 22 is inserted into the insertion hole, the supply of the negative pressure from the negative pressure source is released and the component chuck device 22 is caused to perform a grip releasing operation. Implement P.

Next, image processing executed by the image processing apparatus 40 will be described. FIG. 5 is a flowchart illustrating an example of image processing. This process is executed when a captured image is input after an image processing execution instruction from the control device 50 is input. Note that the image processing execution instruction includes the component type to be processed. The captured image is input as a grayscale image in which the gradation value of each pixel is 8 bits.

When the image processing is executed, the image processing apparatus 40 first determines whether or not the component type of the current processing target has been changed from the previous processing target (S200). When determining that the component type has been changed, the image processing apparatus 40 reads information on the lead existing area of the corresponding component type from the component information stored in the storage unit 42 (S210). Then, the image processing apparatus 40 sets a range of the partial area image corresponding to the read lead existing area information (S220). For example, when information on two areas is stored as the lead existence area, the image processing apparatus 40 reads the information on the two lead existence areas and sets the ranges of the two partial area images corresponding to the respective lead existence areas. To do. On the other hand, if the image processing apparatus 40 determines that the component type has not been changed, the image processing apparatus 40 skips the processes of S210 and S220.

FIG. 6 is an explanatory diagram showing an example of a partial area image. FIG. 6 illustrates a partial region image set in the captured image of the component P of the component type PA01. The component P of the component type PA01 includes two left and right leads L as shown in the component information of FIG. As shown in the figure, the image processing device 40 reads based on the position coordinates (XL1s, YL1s) of the upper left corner, which is the lead existing area of the left lead L (L), and the position coordinates (XL1e, YL1e) of the lower right corner. A partial area image A including L (L) is set. The image processing apparatus 40 also determines the right lead L based on the upper left corner position coordinates (XR1s, YR1s) and the lower right corner position coordinates (XR1e, YR1e), which are the lead existing areas of the right lead L (R). A partial area image B including (R) is set. In this way, a plurality of partial area images are set in one captured image as areas corresponding to the areas where each lead L exists. Note that the lead existence area is defined as an area including the lead L (a captured image of the lead L can be acquired) even if the lead L is misaligned. The image processing apparatus 40 is not limited to setting the partial area image using the coordinates of the existing area of the lead L as it is, and the partial area image may be set by performing a coordinate conversion process.

Next, the image processing apparatus 40 extracts the partial area image (each pixel constituting the image) set in S220 from the captured image (S230), and calculates the gradation value of each pixel of one partial area image to be processed. Obtaining (S240), a binarization threshold value for detecting the boundary (edge) of the lead L in the partial area image is set based on the gradation value of each pixel (S250). The binarization threshold is suitable for a partial region image to be processed using a known method such as a P-tile method, a mode method, or a discriminant analysis method based on a gradation value distribution (histogram) of each pixel. Can be set to a value.

When the binarization threshold is set, the image processing apparatus 40 performs binarization processing that binarizes each pixel of the partial area image using the binarization threshold (S260). Subsequently, the image processing apparatus 40 detects the edge of the lead L (boundary between the lead L and the background portion) from the binarized image (S270), and acquires the lead position based on the detected edge (S280). Then, the image processing apparatus 40 determines whether or not there is an unprocessed partial area image (S290). If it is determined that there is an unprocessed partial area image, the image processing apparatus 40 returns to S240 and targets the unprocessed partial area image. Then, the processes of S240 to S280 are performed. On the other hand, if the image processing apparatus 40 determines that there is no unprocessed partial area image, the image processing ends. As described above, in the present embodiment, the binarization threshold is set for each partial region image, the binarization process is performed, and the edge of the lead L is detected.

FIG. 7 is an explanatory diagram showing an example of a state in which a binarization threshold is set for each partial region image. Here, the lead L is easier to reflect the irradiation light than the body part of the component P. For this reason, the captured image (partial region image) obtained by imaging the component P is an image in which the portion of the lead L appears white and the other background portion (body portion of the component P) appears black. Therefore, the pixel of the lead L portion usually has a gradation value with higher brightness than the pixel of the background portion. For this reason, as shown in the histogram of the partial area image A, the gradation value gradation of each pixel in one partial area image usually has a distinct difference in brightness. On the other hand, as described above, the brightness of the captured image may change due to the light reflection state being different for each component P (for each lead L). For this reason, the captured image may become darker or brighter as a whole, or the difference in brightness between the lead L and the background portion may not appear clearly. For example, as exemplified by the histogram of the partial region image B, there may be a case where the difference between brightness and darkness does not appear clearly because the gradation value of each pixel is biased toward a smaller side. As described above, the tendency of the gradation value of each pixel may be greatly different at the location where each lead L exists. In this embodiment, since the binarization threshold is set for each partial area image, an appropriate binarization threshold can be set according to the tendency of the gradation value of each pixel of each partial area image. For this reason, it is possible to appropriately detect the edge of the lead L by appropriately performing the binarization process in each partial region image.

On the other hand, FIG. 8 is an explanatory diagram showing an example of how the binarization threshold is set from the target image of the comparative example. FIG. 8 shows an example in which one binarization threshold is set with one rectangular area including the partial area images A and B as a processing target image. In this case, the image processing apparatus creates a histogram obtained by adding the histograms of the partial area images A and B shown in FIG. 7, and sets a threshold value based on the histogram. When the brightness of the leads L (L) and (R) in the captured image is greatly different, the threshold value may not be an appropriate threshold value for at least one of the leads L. For example, considering the histogram of FIG. 7, it can be said that the threshold value is not appropriate for at least the lead L (R). For this reason, when the image processing apparatus 40 performs binarization processing by setting one threshold value from the captured image (part of the target image of the captured image), the appropriate binarization processing is not performed, and a plurality of leads L Edges may not be detected properly.

Here, the correspondence between the components of the present embodiment and the components of the present invention will be clarified. The component mounter 10 of this embodiment corresponds to the component mounter of the present invention, and the image processing device 40 corresponds to the image processing device. Further, the storage unit 42 corresponds to the storage unit, the image processing device 40 that executes S230 of the image processing of FIG. 5 corresponds to the extraction unit, and the image processing device 40 that executes S240 of the image processing corresponds to the acquisition unit. The image processing apparatus 40 that executes S250 of the image processing corresponds to the setting unit. The image processing apparatus 40 that executes the image processing S260 corresponds to a binarization processing unit. The component supply device 12 (tape feeder 13) corresponds to a supply device, the head 20 to which the component chuck device 22 is attached and the moving mechanism 24 correspond to a mounting device, and the parts camera 30 corresponds to an imaging device. The control device 50 that executes S130 to S150 of the component mounting process 4 corresponds to the control device. Further, the lead cutter 14 corresponds to a cutting device.

The image processing apparatus 40 described above stores in the storage unit 42 information related to a plurality of lead existing areas each including a plurality of leads L existing in one component P. The image processing device 40 sets a threshold value for each partial region image based on the gradation value of each pixel acquired from a plurality of partial region images corresponding to the lead existing region each time the component P is imaged. . Therefore, the image processing apparatus 40 can perform image processing using an appropriate threshold value for each location where the lead L exists, and can use an optimal threshold value for each captured image. Therefore, the image processing apparatus 40 can more appropriately perform image processing when detecting the location where the lead L is present, and detect the position of the lead L with high accuracy.

Further, since the image processing device 40 binarizes the gradation value of each pixel included in each partial region image using a threshold set for each partial region image, the accuracy of the binarization process can be improved. it can. Since the component mounter 10 includes the image processing device 40, the effect of the image processing device 40 is achieved. Further, in the component mounter 10, since the image processing apparatus 40 detects the position of the lead L with high accuracy, the lead L of the component P is smoothly inserted into the insertion hole of the substrate S to improve the mounting accuracy of the component P. be able to.

It should be noted that the present invention is not limited to the above-described embodiment, and it goes without saying that the present invention can be implemented in various modes as long as it belongs to the technical scope of the present invention.

For example, in the above-described embodiment, the image processing apparatus 40 performs image processing on the captured image of the radial part. However, the present invention is not limited to this, and the captured image of the axial part may be processed. The lead L of the axial component is bent so that the tip is directed downward after being cut by the lead cutter 14. For this reason, the memory | storage part 42 should just memorize | store the information including the front-end | tip position of the lead L as information of a lead presence area | region.

In the above-described embodiment, the image processing apparatus 40 performs image processing on a captured image captured from below the component P by the parts camera 30. However, the present invention is not limited to this, and a captured image captured from above the component P by the mark camera 26. May be image processed. In this case, the part P at the supply position of the part P may be imaged, or the part P at a position different from the supply position may be imaged.

In the above-described embodiment, the threshold value for binarization processing when the image processing device 40 detects the position of the lead L is set. However, the present invention is not limited to this, and the image processing device 40 has a plurality of points of interest of the component P. What is necessary is just to set the threshold value for discriminating a boundary (boundary with a background part). For example, the threshold value may be a threshold value for determining the boundary between the component P and the background portion, or may be a threshold value for determining the boundary between members in the component P formed of different members.

Thus, the present invention can be applied to image processing for discriminating boundaries between a plurality of points of interest in one component P. The following is a description of the image processing of the modified example. FIG. 9 is a perspective view of a component P according to a modification. The component P of the modified example is an LED chip component including a body part Pb formed of a ceramic material and including an LED, and a cover part Pc formed of a transparent resin material that covers the LED. In this modification, the component supply device 12 supplies the component P in a state of being placed on a tray or the like, and the head 20 holds the component P by the component chuck device 22 and mounts it on the substrate S. Further, the mark camera 26 images the upper surface of the component P, and the image processing apparatus 40 performs image processing on the captured image. It is assumed that the component mounter 20 includes a temporary placement table (not shown), and the mark camera 26 covers the upper surface of the component P with the component P held by the component chuck device 22 at the component supply position temporarily placed on the temporary placement table. It is good also as what picks up an image.

In this modified example, a location that becomes a boundary between the cover portion Pc and the body portion Pb (a boundary between different materials) and a location that becomes the boundary between the body portion Pb and the background portion are taken as locations of interest. For this reason, the memory | storage part 42 memorize | stores the area | region information regarding area | region A1, A2 (refer FIG. 9) which contains each attention location one by one. Further, the image processing device 40 extracts partial area images corresponding to the areas A1 and A2 from the captured image of the component P. Then, the image processing device 40 sets a threshold value for discriminating the boundary between the cover part Pc and the body part Pb by binarization processing based on the gradation value of each pixel of the partial area image corresponding to the area A1. Based on the gradation value of each pixel of the partial area image corresponding to the area A2, a threshold value for determining the boundary between the body part Pb and the background part by binarization processing is set. Also in this modified example, the image processing apparatus 40 sets a threshold value for each partial region image that includes a point of interest, and thus can perform image processing using an appropriate threshold value as in the above-described embodiment.

Further, the image processing apparatus 40 is not limited to the one that performs binarization processing, and may perform the following processing for discriminating the boundary of the point of interest. FIG. 10 is an explanatory diagram showing how boundaries of a plurality of points of interest are determined in the component P of the modification, and illustrates an example in which boundaries are determined at a plurality of points (for example, four points) with the cover portion Pc as a point of interest. In the image processing of FIG. 10, the image processing apparatus 40 has a plurality of sets (four sets of T, U, V, and W) each including two determination points that are separated by a predetermined distance across the outer edge of the cover portion Pc. The boundary is determined using the gradation value of the pixel. The determination points of each set are a point (for example, Tin) inside (on the cover portion Pc) from the outer edge of the cover portion Pc, and a point (for example, Tout) outside the outer edge (on the body portion Pb) of the cover portion Pc. Consists of. In addition, the storage unit 42 stores region information regarding a plurality (four) regions (line segments) each including two determination points of each group. The image processing device 40 extracts four partial region images corresponding to the respective regions from the captured image of the component P, and calculates a difference ΔG between the gradation values of the pixels at the two determination points included in the partial region images. To do. Further, the image processing apparatus 40 sets a threshold value of the difference ΔG for each partial region image, compares the difference ΔG with the threshold value, and determines a boundary (presence / absence of a boundary) of the point of interest. FIG. 10 illustrates pixel gradation values and their differences at the determination points Uin and Uout and the determination points Vin and Vout.

Here, also in the component P of the modified example, the captured image is partially darkened or brightened depending on the surface condition (dirt, wrinkles, etc.), poor posture, light reflection condition, etc., and the cover part Pc The difference in brightness from the body part Pb may not appear clearly. In FIG. 10, since the gradation value of the determination point Vin exceeds the predetermined value Gref, the partial region image having the determination point V has normal brightness, and the gradation value of the determination point Uin does not exceed the predetermined value Gref. It is assumed that the brightness of the partial area image having the determination point U is not normal. The image processing apparatus 40 sets the threshold value G1 when the gradation value of the inner determination point exceeds the predetermined value Gref, and sets the threshold value G2 smaller than the threshold value G1 when the gradation value does not exceed the predetermined value Gref. Then, if the difference ΔG in gradation values exceeds the threshold value, the image processing apparatus 40 determines that there is a boundary of the cover portion Pc (boundary with the body portion Pb) between both determination points, and the difference in gradation values. If ΔG does not exceed the threshold value, it is determined that there is no boundary of the cover portion Pc between the two determination points (positional deviation or the like has occurred in the component P). In FIG. 10, since the difference ΔGv between the determination point Vin and the determination point Vout exceeds the threshold value G1, the image processing apparatus 40 determines that there is a boundary of the cover portion Pc between the two determination points. Further, the image processing device 40 determines that there is a boundary of the cover portion Pc between the determination points because the difference ΔGu between the determination point Uin and the determination point Uout exceeds the threshold value G2. Similar processing is performed on the partial area images of the determination points T and W. As described above, since the threshold value is set for each partial region image in this modified example, the boundary of the point of interest can be appropriately determined.

In the above-described embodiment, the image processing device 40 is included in the component mounter 10, but is not limited thereto, and the image processing device 40 may not be included in the component mounter 10. For example, a management apparatus that manages the component mounter 10 may have the function of an image processing apparatus.

The present invention can be used in the manufacturing industry of an image processing apparatus that processes an image of a component mounted on a substrate and a component mounting machine including the image processing apparatus.

10 component mounting machine, 12 component supply device, 13 tape feeder, 14 lead cutter, 14a fixed blade, 14b movable blade, 16 substrate transport device, 18 substrate holding device, 20 head, 22 component chuck device, 24 moving mechanism, 26 mark Camera, 30 parts camera, 32 image sensor, 34 lens, 36 illumination device, 40 image processing device, 42 storage unit, 50 control device, A1, A2 area, L lead, P part, Pb body unit, Pc cover unit, S substrate.

Claims

An image processing apparatus that processes a captured image of the component imaged before mounting in a component mounter that mounts the component on a board,
A storage unit for storing region information regarding a plurality of regions each including a plurality of points of interest in one part;
An extraction unit that extracts a plurality of partial region images corresponding to the region information from the captured image;
For each of the extracted partial region images, an acquisition unit that acquires a gradation value of a pixel included in the partial region image;
Setting that sets a threshold for determining the boundary of the point of interest included in the partial area image based on the acquired gradation value of the pixel for each partial area image every time the part is imaged And
An image processing apparatus comprising:
The image processing apparatus according to claim 1,
The component has a plurality of leads that are respectively inserted into a plurality of insertion holes provided in the substrate,
The plurality of points of interest are the respective locations of the plurality of leads,
The storage unit stores information on a plurality of regions including one location of each lead as the region information.
The image processing apparatus according to claim 1, wherein:
An image processing apparatus comprising: a binarization processing unit that binarizes a gradation value of each pixel included in the partial region image using a threshold set by the setting unit for each of the plurality of partial region images.
A component mounter comprising the image processing apparatus according to claim 1,
A supply device for supplying parts;
A mounting device for holding the component supplied by the supply device with a holder and mounting the held component on the substrate;
An imaging device for imaging the component;
When the captured image captured by the imaging device is processed by the image processing device, the position of the component held by the holder is acquired based on the processing result, and the component is acquired based on the acquired position. A control device for controlling the mounting device so that is mounted on the substrate;
A component mounting machine.
The component mounting machine according to claim 4,
The supply device supplies, as the component, a component having a plurality of leads respectively inserted into a plurality of insertion holes provided in the substrate,
A cutting device for partially cutting at least one of the plurality of leads,
As the image processing device, the storage unit includes a device that stores region information regarding a plurality of regions each including the plurality of leads.
The imaging device images the component held by the holder in a state where at least one lead is partially cut by the cutting device;
The control device acquires the positions of the plurality of leads of the component held by the holder based on a processing result by the image processing device, and the leads are inserted into the insertion hole based on the acquired positions. A component mounter that controls the mounting apparatus to be inserted into the component mounting machine.