Component mounting device, component mounting system, and method for acquiring height information

Abstract

To obtain highly reliable height information.SOLUTION: A component mounting device 10 includes a board imaging camera 22 capable of imaging a board P from a plurality of directions, and executes first pre-mounting extraction processing (S204) for extracting candidates for matching points by performing stereo matching on each image captured in pre-mounting imaging processing (S103), first post-mounting extraction processing (S207) for extracting candidates for matching points by performing stereo matching on each image captured in post-mounting imaging processing (S109), first detection processing (S208) for detecting a first highly reliable matching point by comparing the candidates for matching points extracted in the first pre-mounting extraction processing with the candidates for matching points extracted in the first post-mounting extraction processing, and first acquisition processing (S209) for acquiring height information on the basis of the first highly reliable matching point.SELECTED DRAWING: Figure 8

Description

Technical Field

[0001] The technology disclosed in this specification relates to a component mounting device for mounting components on a substrate, a component mounting system, and a height information acquisition method.

Background Art

[0002] Conventionally, in a component mounting device for mounting components on a substrate, there is known one provided with an imaging unit for imaging the substrate from a plurality of directions (see, for example, Patent Document 1). Specifically, the component mounting device described in Patent Document 1 images a predetermined area including the component mounting position from a plurality of imaging directions by the imaging unit before mounting the component, and performs stereo matching of each of the captured images to obtain the height information of the mounting position before mounting. Similarly, after mounting the component, the component mounting device images the predetermined area from a plurality of imaging directions by the imaging unit, and performs stereo matching of each of the captured images to obtain the height information of the mounting position after mounting.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the component mounting device described in Patent Document 1 had room for improvement in obtaining highly reliable height information. This specification discloses a technology capable of obtaining highly reliable height information.

Means for Solving the Problems

[0005] A component mounting apparatus comprising: a mounting unit for mounting components on a substrate; an imaging unit capable of imaging the substrate from multiple directions; and a control unit, wherein the control unit performs: a pre-mounting imaging process in which the imaging unit images the substrate from multiple directions before mounting the components; a post-mounting imaging process in which the imaging unit images the substrate from multiple directions after mounting the components; a first pre-mounting extraction process for extracting candidate matching points by performing stereo matching on each image captured in the pre-mounting imaging process; a first post-mounting extraction process for extracting candidate matching points by performing stereo matching on each image captured in the post-mounting imaging process; a first detection process for detecting a first highly reliable matching point by comparing the candidate matching points extracted in the first pre-mounting extraction process with the candidate matching points extracted in the first post-mounting extraction process; and a first acquisition process for acquiring height information based on the first highly reliable matching point. [Effects of the Invention]

[0006] With the above configuration, highly reliable height information can be obtained. [Brief explanation of the drawing]

[0007] [Figure 1] Schematic diagram of the component mounting system according to Embodiment 1 [Figure 2] Top view of a component mounting device [Figure 3] Schematic diagram of the head unit and circuit board imaging camera. [Figure 4] Block diagram showing the electrical configuration of a component mounting device. [Figure 5] Schematic diagrams of the upper and lower camera images. [Figure 6] A schematic diagram illustrating general matching point detection (before implementation). [Figure 7] A schematic diagram illustrating general matching point detection (after implementation). [Figure 8] Schematic diagram illustrating the detection of high-reliability matching points. [Figure 9] A schematic diagram to explain the conversion of parallax to height. [Figure 10] Schematic diagram for explaining correction of target mounting height [Figure 11] Schematic diagram for explaining processing using highly reliable matching points [Figure 12] Flowchart for mounting a component at the first mounting position [Figure 13] Flowchart of acquisition processing [Figure 14] Flowchart for mounting a component at the mounting position where the component is mounted after the first mounting position [Figure 15] Schematic diagram for explaining processing using highly reliable matching points according to Embodiment 2 [Figure 16] Schematic diagram for explaining processing using highly reliable matching points according to Embodiment 3 [Figure 17A] Schematic diagram for explaining processing using highly reliable matching points according to Embodiment 4 (method of dividing into multiple regions) [Figure 17B] Schematic diagram for explaining processing using highly reliable matching points (method of linear interpolation) [Figure 17C] Schematic diagram for explaining processing using highly reliable matching points (method of curve interpolation) [Figure 18] Schematic diagram for explaining processing using highly reliable matching points according to Embodiment 5 [Figure 19] Schematic diagram for explaining processing using highly reliable matching points according to Embodiment 6 [Figure 20] [[ID=3​​​​​​​​​​​​​​​​​​ (Summary of the embodiment) (1) The component mounting apparatus according to the embodiment includes a mounting unit for mounting components on a substrate, an imaging unit capable of imaging the substrate from multiple directions, and a control unit, wherein the control unit performs a pre-mounting imaging process in which the imaging unit images the substrate from multiple directions before mounting the components, a post-mounting imaging process in which the imaging unit images the substrate from multiple directions after mounting the components, a first pre-mounting extraction process for extracting candidate matching points by performing stereo matching on each image captured in the pre-mounting imaging process, a first post-mounting extraction process for extracting candidate matching points by performing stereo matching on each image captured in the post-mounting imaging process, a first detection process for detecting a first highly reliable matching point by comparing the candidate matching points extracted in the first pre-mounting extraction process with the candidate matching points extracted in the first post-mounting extraction process, and a first acquisition process for acquiring height information based on the first highly reliable matching point.

[0009] The component mounting apparatus described in Patent Document 1 acquires height information based on matching points detected by a general matching point detection method described later. As will be explained in detail later, the inventors of this application have found that in the case of images of a substrate, detecting matching points using a general matching point detection method may result in false detection of matching points. Acquiring height information based on falsely detected matching points reduces the reliability of the height information. For this reason, the component mounting apparatus described in Patent Document 1 had room for improvement in acquiring highly reliable height information.

[0010] The first pre-implementation extraction process and the first post-implementation extraction process described above are processes for extracting candidate matching points using conventional general matching point detection methods. According to the component mounting apparatus described in (1) above, in the first detection process, the candidate matching points extracted in the first pre-implementation extraction process and the candidate matching points extracted in the first post-implementation extraction process are compared to detect the first highly reliable matching point. In other words, the component mounting apparatus detects highly reliable matching points (first highly reliable matching points) by utilizing the fact that the image changes depending on the presence or absence of the component before and after component mounting. Since the first high-reliability matching point is more reliable than matching points detected by general matching point detection methods, high-reliability height information can be obtained by acquiring height information based on the first high-reliability matching point.

[0011] (2) The component mounting apparatus described in (1) above, wherein the control unit extracts candidate matching points by obtaining the value of an evaluation function that evaluates the similarity of a plurality of images in the first pre-mounting extraction process and the first post-mounting extraction process, and in the first detection process, detects the candidate matching point with the smallest difference between the value of the evaluation function obtained in the first pre-mounting extraction process and the value of the evaluation function obtained in the first post-mounting extraction process as the first high-reliability matching point.

[0012] According to the component mounting device described in (2) above, a highly reliable matching point (first highly reliable matching point) can be detected by detecting the candidate matching point with the smallest difference in the evaluation function values ​​before and after mounting as the first highly reliable matching point.

[0013] (3) A component mounting apparatus as described in (1) or (2) above, wherein the control unit performs stereo matching of a first region on each of the images captured in the pre-mounting imaging process to obtain height information of the mounting position of the component in the first pre-mounting extraction process to extract candidate matching points, performs stereo matching of a first region on each of the images captured in the post-mounting imaging process to extract candidate matching points, and in the first acquisition process, obtains pre-mounting height information of the mounting position based on the first high-reliability matching points.

[0014] The "height information of the mounting position before mounting" mentioned above refers to information indicating the height of the circuit board surface at the mounting position before the component is mounted. According to the component mounting device described in (3) above, the height information of the mounting position before mounting is obtained based on the first highly reliable matching point, so highly reliable height information of the mounting position before mounting can be obtained.

[0015] (4) A component mounting apparatus as described in (3) above, wherein the control unit may perform, before the first pre-mounting extraction process, a second pre-mounting extraction process for extracting candidate matching points by performing stereo matching of a second region on each of the images captured in the pre-mounting imaging process for obtaining height information of the substrate surface near the mounting position of the component; a second post-mounting extraction process for extracting candidate matching points by performing stereo matching of the second region on each of the images captured in the post-mounting imaging process; a second detection process for detecting a second high-reliability matching point by comparing the candidate matching point extracted in the second pre-mounting extraction process with the candidate matching point extracted in the second post-mounting extraction process; a second acquisition process for obtaining pre-mounting height information of the substrate surface near the mounting position based on the second high-reliability matching point; and a pre-mounting first region correction process for correcting the first region on each of the images captured in the pre-mounting imaging process based on the pre-mounting height information of the substrate surface near the mounting position obtained in the second acquisition process.

[0016] If the substrate is warped, a discrepancy will occur between the first region recognized by the control unit on the image captured during the pre-mount imaging process and the actual first region on that image. Performing stereo matching using the image of the discrepant region will reduce the reliability of the pre-mount height information for the mounting position. According to the component mounting apparatus described in (4) above, pre-mounting height information of the substrate surface near the mounting position is acquired based on a second highly reliable matching point, thus enabling the acquisition of highly reliable pre-mounting height information of the substrate surface near the mounting position. Furthermore, according to the component mounting apparatus described in (4) above, the first region is corrected on each image captured in the pre-mounting imaging process based on the highly reliable pre-mounting height information of the substrate surface near the mounting position, thereby reducing the discrepancy between the first region recognized by the control unit on the image and the actual first region on the image. This enables the acquisition of highly reliable pre-mounting height information of the mounting position.

[0017] (5) A component mounting apparatus as described in (4) above, wherein the control unit may perform, before the first post-mount extraction process, a third acquisition process for acquiring post-mount height information of the substrate surface near the mounting position based on the second high-reliability matching point, and a post-mount first region correction process for correcting the first region on each of the images captured in the post-mount imaging process based on the post-mount height information of the substrate surface near the mounting position acquired in the third acquisition process.

[0018] According to the component mounting apparatus described in (5) above, the first region is corrected on each image captured in the post-mounting imaging process based on the post-mounting height information of the substrate surface near the mounting position, which is highly reliable. Therefore, post-mounting height information of the mounting position can be obtained. The "height information after mounting at the mounting location" mentioned above refers to information indicating the height of the top surface of the component mounted at the mounting location if the component mounting is successful, and information indicating the height of the circuit board surface at the mounting location if the component mounting fails.

[0019] (6) A component mounting apparatus as described in (3) above, wherein the control unit may perform the first acquisition process after mounting the component to one of the mounting positions to acquire the height information of the one of the mounting positions before mounting, and perform a first correction process to correct the target mounting height when mounting the component to the mounting position where the component will be mounted after the one of the mounting positions based on the height information of the one of the mounting positions before mounting.

[0020] The target mounting height for components is initially set to a height that is approximately the same as the top surface of a flat circuit board (or slightly higher, taking into account the thickness of the solder paste). Therefore, when mounting components to a circuit board, the mounting unit lowers the component until its bottom surface (in other words, the surface that will be soldered to the circuit board) is approximately the same as the top surface of a flat circuit board (the target mounting height). In this case, if the circuit board is warped upwards, the component will come into contact with the circuit board and be subjected to impact before it reaches the target mounting height.

[0021] If we can obtain reliable pre-mounting height information for the mounting location, we can mitigate the impact on the component by correcting the target mounting height based on the obtained height information. However, the first highly reliable matching point is detected based on images before and after mounting, so it can only be detected after the component has been mounted. In other words, pre-mounting height information for a highly reliable mounting location can only be obtained after the component has been mounted.

[0022] The inventors of this invention, after investigating this matter, found that even if the substrate is warped, the pre-mount height of each mounting position is approximately the same. They discovered that for each mounting position, after a component is mounted at that position, reliable pre-mount height information for that mounting position is obtained based on a first highly reliable matching point. When mounting components at subsequent mounting positions, the target mounting height can be corrected based on this height information, thereby mitigating the impact on the components.

[0023] According to the component mounting apparatus described in (6) above, for each mounting position, after mounting a component to that position, the pre-mounting height information of the mounting position is acquired based on a first highly reliable matching point, and the target mounting height when mounting a component to a mounting position where a component will be mounted after that initial mounting position is corrected based on the pre-mounting height information of that initial mounting position, thereby mitigating the impact when mounting a component to a mounting position where a component will be mounted later.

[0024] (7) A component mounting apparatus as described in (3) above, wherein the control unit may, for each mounting position, after mounting the component to the mounting position, perform the first acquisition process to acquire the height information of the mounting position before mounting, and perform a second correction process to correct the target mounting height when mounting the component to the mounting position based on the height information of the mounting position before mounting, in which the component was mounted immediately before the mounting position.

[0025] As mentioned above, the inventors of the present invention have found that even if the substrate is warped, the height of each mounting position before mounting is approximately the same. According to the component mounting apparatus described in (7) above, reliable pre-mounting height information for each mounting position is acquired based on a first highly reliable matching point, and the target mounting height when mounting a component at the mounting position is corrected based on the pre-mounting height information of the mounting position where a component was mounted immediately before that mounting position, thereby mitigating the impact applied to the component mounted at that mounting position.

[0026] (8) A component mounting apparatus as described in (3) above, wherein the control unit may, for each of the substrates, perform a first map creation process for each of the mounting positions on the substrate, after mounting the component to the mounting position, to obtain the height information of the mounting position before mounting, and to create a height map in which the mounting position and the height information of the mounting position before mounting are associated; and a third correction process for mounting the component to the substrate, to correct the target mounting height when mounting the component to each of the mounting positions on the substrate based on the height map created on the substrate on which the component was mounted immediately before the substrate.

[0027] The inventors of this application have found that when producing multiple substrates, the degree of warping of each substrate is generally the same. According to the component mounting apparatus described in (8) above, the target mounting height when mounting components at each mounting position on the board is corrected based on a height map created on a board on which components were mounted immediately before the board in question, thereby mitigating the impact applied to the components mounted on the board.

[0028] (9) A component mounting apparatus as described in (3) above, comprising a fixing part for fixing the edge of the substrate, wherein the control unit may perform the first acquisition process after mounting the component to one of the mounting positions to acquire the height information of the one of the mounting positions before mounting, and a setting process to set a virtual substrate surface based on the acquired height information and the height information of the substrate edge fixed by the fixing part, and a fourth correction process to acquire the height information of the mounting position before mounting for the mounting position to which the component is mounted after the one of the mounting positions from the virtual substrate surface, and correct the target mounting height when mounting the component to the mounting position based on the height information acquired from the virtual substrate surface.

[0029] The inventors of this application have found that when a substrate is warped, the substrate surface of the substrate can be approximated by a virtual substrate surface based on the height information of one mounting position before mounting and the height information of the substrate edge. According to the component mounting apparatus described in (9) above, for each mounting position, high-reliability pre-mounting height information of the mounting position is acquired based on a first high-reliability matching point, and a virtual board surface is set based on the acquired height information and the height information of the board edge. The component mounting apparatus then corrects the target mounting height when mounting a component to a mounting position where a component will be mounted later, based on the height information acquired from the virtual board surface, thereby mitigating the impact applied to the component mounted to the later mounting position.

[0030] (10) A component mounting apparatus as described in (3) above, wherein the control unit, when mounting the component on one substrate, may perform a second map creation process for each of the mounting positions on the one substrate, after mounting the component to the mounting position, to obtain the height information of the mounting position before mounting, and to create a height map in which the mounting position and the height information of the mounting position before mounting are associated; and a fifth correction process for correcting the target mounting height when mounting the component to the mounting position on the substrate on which the component is mounted after the one substrate, based on the height map created on the one substrate.

[0031] The inventors of this application have found that when producing multiple substrates, the degree of warping of each substrate is generally the same. According to the component mounting apparatus described in (10) above, when mounting components to the mounting position on a board on which components will be mounted after one board, the target mounting height is corrected based on the height map created on the first board, thereby mitigating the impact on components mounted on subsequent boards.

[0032] (11) A component mounting apparatus as described in (3) above, wherein the control unit may, in the first acquisition process, acquire height information of the mounting position before mounting and height information of the mounting position after mounting based on the first high-reliability matching point, and execute a first success / failure determination process that determines whether the component has been mounted by comparing the change in height information of the mounting position before and after mounting with a first determination threshold.

[0033] The "height information of the mounting position after mounting" mentioned above refers to information indicating the height of the top surface of the component mounted at the mounting position if the component mounting is successful, and information indicating the height of the substrate surface at the mounting position if the component mounting fails. The height information before and after implementation of the implementation location, obtained in the first acquisition process, is highly reliable because it is acquired based on the first high-reliability matching point. According to the component mounting apparatus described in (11) above, the success or failure of component mounting is determined by comparing the change in height information before and after mounting of a highly reliable mounting position, obtained based on the first highly reliable matching point, with the first judgment threshold, thereby improving the reliability of the determination of mounting success or failure.

[0034] (12) A component mounting apparatus as described in (11) above, wherein the control unit performs a fourth acquisition process to obtain pre-mounting height information of the mounting position by performing stereo matching on each of the images captured in the pre-mounting imaging process; a fifth acquisition process to obtain post-mounting height information of the mounting position by performing stereo matching on each of the images captured in the post-mounting imaging process; and a second success / failure determination process which determines that mounting is successful if the change in the height information of the mounting position before and after mounting is greater than the first determination threshold plus a predetermined margin value, and determines that determination is impossible if it is less than the value, and if the second determination process determines that determination is impossible, the first success / failure determination process may be executed.

[0035] The fourth and fifth acquisition processes described above are processes that detect matching points using conventional general matching point detection methods and acquire height information. In the second success / failure determination process, the control unit determines that the implementation is successful if the change between the height information of the implementation position before implementation obtained in the fourth acquisition process and the height information of the implementation position after implementation obtained in the fifth acquisition process is greater than the value obtained by adding a predetermined margin value to the first determination threshold. The reason for this is that if the success or failure of the implementation is determined based on height information obtained based on general matching points, there is a possibility that the matching points may be misdetected, so the reliability of the determination is not necessarily high. For this reason, in order to ensure the certainty of the determination, the control unit determines success if the change in the height information of the implementation position before and after implementation obtained based on general matching points is greater than the first determination threshold (i.e., greater than the value obtained by adding a margin value to the first determination threshold).

[0036] In the component mounting apparatus described in (12) above, if the change in height information before and after mounting is smaller than the value obtained by adding a margin value to the first judgment threshold, the success or failure of mounting cannot be clearly determined (determination impossible), and the first success or failure determination process is executed to re-determine the success or failure of mounting. In other words, the component mounting apparatus described in (12) above determines the success or failure of mounting from the change in height information of the mounting position before and after mounting, which is obtained based on a general matching point, but if the success or failure of mounting cannot be clearly determined, the success or failure of mounting is re-determined from the change in height information of the mounting position before and after mounting, which is obtained based on a highly reliable mounting point, which is obtained based on the first highly reliable matching point. This improves the reliability of the determination of the success or failure of mounting.

[0037] In the component mounting apparatus described in (12) above, if the result of determining mounting success or failure based on the change in height information before and after mounting of the mounting position, which is obtained based on a general matching point, is that mounting is successful, the first success / failure determination process is not executed. Therefore, it is not necessary to execute the process for detecting the first high-reliability matching point (first pre-mounting extraction process, first post-mounting extraction process, first detection process, etc.). For this reason, the processing load on the control unit can be reduced compared to the case where the first high-reliability matching point is detected for all mounting positions to determine mounting success or failure.

[0038] (13) A component mounting apparatus as described in (11) above, wherein the control unit performs a sixth acquisition process to acquire height information of the substrate surface near the mounting position of the component by performing stereo matching of the second region on each of the images captured in the pre-mounting imaging process to acquire height information of the substrate surface near the mounting position of the component, and a seventh acquisition process to acquire height information of the substrate surface near the mounting position after mounting by performing stereo matching of the second region on each of the images captured in the post-mounting imaging process, and if the change in height information of the substrate surface near the mounting position before and after mounting is greater than or equal to a second determination threshold, the first success or failure determination process may be performed.

[0039] The sixth and seventh acquisition processes described above are processes that detect matching points using conventional, general matching point detection methods and acquire height information. Since the substrate surface near the mounting location is not an area where components are mounted, the change in height information before and after mounting on the substrate surface near the mounting location is usually small. However, if the height information before and after mounting on the substrate surface near the mounting location changes significantly, it is possible that a common matching point was misdetected in the sixth or seventh acquisition process. Determining the success or failure of component mounting based on height information acquired based on a misdetected matching point reduces the reliability of the determination.

[0040] In the component mounting apparatus described in (13) above, if the change between the height information of the substrate surface near the mounting position before mounting, acquired in the sixth acquisition process, and the height information of the substrate surface near the mounting position after mounting, acquired in the seventh acquisition process, is greater than or equal to the second judgment threshold (in other words, if the reliability of the change in height information is low), the first success / failure judgment process determines whether mounting is successful or not, thereby improving the reliability of the determination of mounting success or failure.

[0041] In the component mounting apparatus described in (13) above, if the change in height information of the substrate surface near the mounting position before and after mounting is smaller than the second determination threshold (in other words, if the reliability of the height information is high), the first success / failure determination process is not executed. Therefore, it is not necessary to execute the processes for obtaining the first high-reliability matching point (first pre-mounting extraction process, first post-mounting extraction process, first detection process, first acquisition process, etc.). For this reason, the processing load on the control unit can be reduced compared to the case where the first high-reliability matching point is detected for all mounting positions to determine mounting success or failure.

[0042] (14) A component mounting apparatus as described in (1) or (2) above, wherein the control unit mounts the components on a test board before starting production of the board, and for each mounting position on the test board, after mounting the components on the mounting position, it executes the first acquisition process to acquire height information and executes a third map creation process to create a height map of the board.

[0043] The inventors of this application have found that when producing multiple substrates, the degree of warping of each substrate is generally the same. According to the component mounting device described in (14) above, components are mounted on a test board to create a height map before starting board production. Therefore, after starting board production, height information is obtained from the height map, which reduces the processing load on the control unit during board production compared to creating the height map during board production.

[0044] (15) The component mounting system according to the embodiment is a component mounting system in which a plurality of component mounting devices described in (1) or (2) above are arranged in the transport direction of the substrate, wherein the control unit of one of the component mounting devices, when mounting the component on the substrate, executes the first acquisition process for each mounting position to acquire the height information of the mounting position before mounting, and executes a fourth map creation process to create a height map in which the mounting position and the height information of the mounting position before mounting are associated, and the control unit of a component mounting device located downstream of the one component mounting device in the transport direction of the substrate executes a sixth correction process to correct the target mounting height of the component when mounting the component at the mounting position based on the height map created by the one component mounting device.

[0045] According to the component mounting system described in (15) above, the downstream component mounting device corrects the target mounting height when mounting components at the mounting position based on the height map created by the upstream component mounting device, thereby mitigating the impact applied to the components mounted by the downstream component mounting device.

[0046] [Details of the embodiments of this disclosure] Embodiments of the present disclosure are described below. The present disclosure is not limited to these examples, but is indicated by the claims, and all modifications within the meaning and scope of the claims are intended to be included. Embodiments of the present disclosure can be implemented in various forms, such as apparatus, methods, computer programs for realizing the functions of such apparatus or methods, and recording media on which such computer programs are stored.

[0047] <Embodiment 1> Embodiment 1 will be described with reference to Figures 1 to 14. In the following description, the left-right direction shown in Figure 2 will be referred to as the X direction, the front-back direction as the Y direction, and the up-down direction shown in Figure 3 as the Z direction. In the following description, the left side shown in Figure 1 will be referred to as the upstream side, and the right side as the downstream side. In addition, in the following description, the reference numerals in the drawings may be omitted for identical components, with some exceptions.

[0048] (1) Component mounting system Referring to Figure 1, the component mounting system 1 according to Embodiment 1 will be described. The component mounting system 1 is a system for mounting components E (see Figure 2), such as electronic components, onto a substrate P. The component mounting system 1 has one or more mounting lines L (L1 to L3). Each mounting line L has one or more component mounting devices 10. In addition to the component mounting devices 10, the mounting line L also has other devices that perform work on the substrate P (loader, screen printer, print inspection machine, dispenser, post-mounting visual inspection machine, reflow machine, post-curing visual inspection machine, unloader, etc.), but these other devices are omitted in Figure 1. The component mounting apparatus 10 according to Embodiment 1 is an example of the component mounting apparatus (1) to (6) described in the summary of the embodiments above.

[0049] (1-1) Configuration of a component mounting device Referring to Figure 2, the component mounting apparatus 10 will be described. The component mounting apparatus 10 includes a stand 11, a transport conveyor 12, four tape component supply devices 13, a head unit 16 (an example of a mounting unit), a head moving unit 17 (an example of a mounting unit), two component imaging cameras 21, a substrate imaging camera 22 (an example of an imaging unit), a control unit 30 (see Figure 4), and an operation unit 31 (see Figure 4).

[0050] The frame 11 has a rectangular shape in plan view and a flat top surface. The area A shown by the dashed line is the work position (hereinafter referred to as work position A) where the substrate P is fixed when components E are mounted on the substrate P. Below work position A is a backup device 70 (see Figure 17A, an example of a fixing part) which fixes the substrate P to work position A when it is transported to work position A. The conveyor belt 12 is a device for transporting substrates P. The conveyor belt 12 is equipped with a pair of conveyor belts 14 (front conveyor belt 14A and rear conveyor belt 14B) that circulate in the X direction, a plurality of rollers on which the conveyor belts are wrapped, and a belt drive motor 44 (see Figure 4) that drives the conveyor belts 14. The conveyor belt 12 transports substrates P brought in from the upstream side to the work position A, and then transports the substrates P with components E mounted on them to the downstream side at the work position A.

[0051] The tape component supply devices 13 are located in four places in total, two on each side of the component mounting device 10 in the X direction, on the Y direction. Multiple tape feeders 15 are mounted on the tape component supply devices 13, aligned horizontally in the X direction. Each tape feeder 15 is equipped with a reel around which component tape holding component E is wound, and an electrically operated feeding device that pulls the component tape from the reel, supplying component E one at a time.

[0052] The head unit 16 is equipped with multiple mounting heads 18 for picking up and releasing components E. The head unit 16 is a so-called inline type, with multiple mounting heads 18 arranged in the X direction. The head unit 16 will be described later. Here, an inline type head unit 16 is used as an example, but the head unit 16 may also be a so-called rotary head, where multiple mounting heads 18 are arranged around a circumference. The head movement unit 17 is a mechanism that moves the head unit 16 in the X and Y directions within a predetermined range of motion. The head movement unit 17 includes a beam 19 that supports the head unit 16 so that it can reciprocate in the X direction, a pair of Y-axis guide rails 20 that support the beam 19 so that it can reciprocate in the Y direction, an X-axis servo motor 40 that moves the head unit 16 back and forth in the X direction, and a Y-axis servo motor 41 that moves the beam 19 back and forth in the Y direction.

[0053] The two component imaging cameras 21 are each positioned between two tape component supply devices 13 that are aligned in the X direction. The component imaging cameras 21 are mounted on a stand 11 and image the component E that is held in place by the mounting head 18 from below. The substrate imaging camera 22 is a camera that images the substrate P from multiple directions. The substrate imaging camera 22 is attached to the front of the head unit 16 and moves along the head unit 16 on the frame 11 in the X and Y directions.

[0054] Referring to Figure 3, the head unit 16 and the substrate imaging camera 22 will be described. The head unit 16 is equipped with a Z-axis servo motor 42 (see Figure 4) that individually raises and lowers a plurality of mounting heads 18, and an R-axis servo motor 43 (see Figure 4) that rotates these mounting heads 18 simultaneously around their axes. Each mounting head 18 has a nozzle shaft 18A and a suction nozzle 18B that is detachably attached to the lower end of the nozzle shaft 18A. Negative and positive pressure are supplied to the suction nozzle 18B via the nozzle shaft 18A from an air supply device (not shown). The suction nozzle 18B attracts a component E when negative pressure is supplied and releases the component E when positive pressure is supplied.

[0055] The substrate imaging camera 22 is equipped with a stereo camera 23. A stereo camera 23 is provided for each mounting head 18. The stereo camera 23 has an upper camera 23A, a lower camera 23B, two illumination units 23C, and two optical systems (lenses, mirrors, etc.) not shown. The upper camera 23A and the lower camera 23B are cameras that image a predetermined region 81 including the mounting position Pa of component E almost simultaneously. The predetermined region 81 also includes the substrate surface Pb near the mounting position. The upper camera 23A images the predetermined region 81 from an imaging direction inclined at an angle θH (0 degrees < θH < 90 degrees) with respect to the upper surface of the substrate (hereinafter simply referred to as the substrate surface). The lower camera 23B images the predetermined region 81 from an imaging direction inclined at an angle θL (0 degrees < θL < θH) with respect to the substrate surface. The two illumination units 23C illuminate the predetermined region 81 when the upper camera 23A and the lower camera 23B image the predetermined region 81 almost simultaneously. An optical system (not shown) projects the reflected light image emitted from the illumination unit 23C and reflected from the predetermined region 81 onto the light-receiving surface of either the upper camera 23A or the lower camera 23B.

[0056] (1-2) Electrical configuration of component mounting equipment Referring to Figure 4, the electrical configuration of the component mounting device 10 will be described. The component mounting device 10 includes a control unit 30 and an operation unit 31. The control unit 30 includes an arithmetic processing unit 32, a motor control unit 33, a storage unit 34, an image processing unit 35, an external input / output unit 36, a feeder communication unit 37, and the like.

[0057] The arithmetic processing unit 32 is equipped with a CPU, RAM, etc., and controls each part of the component mounting device 10 by executing the control program stored in the memory unit 34. The motor control unit 33 controls the rotation of each motor, such as the X-axis servo motor 40 and the Y-axis servo motor 41, under the control of the arithmetic processing unit 32. The memory unit 34 stores various programs and data that are executed by the arithmetic processing unit 32. The data stored in the memory unit 34 includes data indicating the shape of each component E and data indicating the mounting position (XY coordinates) of each component E.

[0058] The image processing unit 35 is configured to receive image signals output from the component imaging camera 21 and the stereo camera 23. The external input / output unit 36 ​​is a so-called interface and is configured to receive detection signals output from various sensors 38 provided on the main body of the component mounting device 10. Furthermore, the external input / output unit 36 ​​is configured to perform operational control of various actuators 39 (such as an air supply device) based on control signals output from the calculation processing unit 32. The feeder communication unit 37 is connected to the tape feeder 15 and controls the tape feeder 15 comprehensively.

[0059] The control unit 31 is equipped with a display device such as an LCD, and input devices such as a touch panel, keyboard, and mouse. The operator can operate the control unit 31 to make various settings and give instructions for operation.

[0060] (2) Acquisition of substrate height information and processing using height information Referring to Figures 3 and 5, the acquisition of height information of the substrate P and the processing using the height information, which are performed by the control unit 30, will be explained. In Figure 3, the reference plane Ps is a virtual plane that serves as the height reference. The reference plane Ps coincides with the top surface of the substrate P that is not warped.

[0061] The control unit 30 acquires height information of the substrate P, specifically the height information of the mounting position Pa of the component E relative to the reference plane Ps, and the height information of the substrate surface Pb near the mounting position relative to the reference plane Ps (hereinafter simply referred to as the height information of the substrate surface Pb), by stereo matching using the stereo camera 23. Here, the height of the mounting position Pa refers to the height of the substrate surface at the mounting position Pa when no component E is mounted there (before mounting). After component E is mounted at the mounting position Pa (after mounting), if the mounting of component E is successful, it refers to the height of the top surface of component E; if the mounting fails, it refers to the height of the substrate surface at the mounting position Pa.

[0062] (2-1) Acquisition of height information by stereo matching As shown in Figure 5, the control unit 30 sets a first region AR1 for acquiring height information of the mounting position Pa in the image 61 captured by the upper camera 23A (hereinafter referred to as the upper camera image 61) and the image 62 captured by the lower camera 23B (hereinafter referred to as the lower camera image 62). The first region AR1 is set so as to include the mounting position Pa on which the component E is mounted in the vertical direction of the captured image.

[0063] The control unit 30 performs stereo matching between the first region AR1 of the upper camera image 61 and the first region AR1 of the lower camera image 62 to detect matching points. The detection of matching points will be explained later. Based on the detected matching points, the control unit 30 calculates the parallax of those images and obtains height information of the mounting position Pa by converting the calculated parallax into height.

[0064] Similarly, the control unit 30 sets a second region AR2 in the upper camera image 61 and the lower camera image 62, respectively, for acquiring height information of the substrate surface Pb. The second region AR2 is set in the vertical direction of the captured image on the opposite side from where the mounting head 18 is captured.

[0065] The control unit 30 performs stereo matching between the second region AR2 of the upper camera image 61 and the second region AR2 of the lower camera image 62 to detect matching points. Based on the detected matching points, the control unit 30 calculates the parallax of those images and obtains height information of the substrate surface Pb by converting the calculated parallax into height. As will be explained in more detail later, the control unit 30 corrects the first region AR1 based on the height information of the substrate surface Pb, and then acquires the height information of the mounting position Pa based on the corrected first region AR1. For this reason, the height information of the substrate surface Pb is acquired before the height information of the mounting position Pa.

[0066] (2-1-1) Detection of matching points The control unit 30 performs both conventional general matching point detection and highly reliable matching point detection (hereinafter referred to as high-reliability matching point). This will be explained in detail below.

[0067] (2-1-1-1) General detection of matching points Refer to Figure 6 to explain the general detection of matching points. In the lower camera image 62 shown in Figure 6, regions R1 to R4 each contain the mounting position Pa where component E will be mounted. Solder paste 80 printed on the substrate P is visible in regions R1 to R4. Although not clearly visible in Figure 6, component E has already been mounted in regions R1 and R2, and component E is also visible. The solder paste 80 in regions R1 and R2 appears to be crushed by component E. Component E will be mounted in regions R3 and R4.

[0068] In stereo matching, the control unit 30 acquires a template image 63 from either the upper camera image 61 or the lower camera image 62, and moves the template image 63 on the other image (hereinafter referred to as the search image) to detect matching points. In the example shown in Figure 6, the lower camera image 62 is the search image, and the template image 63 is acquired from the upper camera image 61.

[0069] In Figure 6, the dashed line 64 is the epipolar line. When the relative positions of two cameras are known, a point on the image captured by one camera lies on a straight line on the image captured by the other camera. This straight line is called the epipolar line. The control unit 30 moves the template image 63 one pixel at a time along the epipolar line on the search image 62, and for each pixel movement, it calculates the value of an evaluation function (evaluation value) that evaluates the similarity of multiple images.

[0070] Evaluation functions such as SAD (Sum of Absolute Difference) and SSD (Sum of Squared Difference) can be used. SAD is an evaluation function that evaluates similarity by "the sum of the absolute values ​​of the differences in pixel values." Pixel values ​​can also be called density. For example, if two images are identical, the pixel value of a pixel in one image will match the pixel value of the pixel at the same position in the other image, so the difference in pixel values ​​will be 0. Therefore, the more similar the two images are, the closer the evaluation value will be to 0. In other words, the more similar the two images are, the smaller the evaluation value will be. SSD is an evaluation function that assesses similarity using the "sum of squared differences in pixel values ​​(squared error)." With SSD, the more similar the two images are, the smaller the evaluation value. The evaluation function is not limited to SAD or SSD; other evaluation functions may also be used.

[0071] The graph shown in Figure 6 is an example of a graph of evaluation values ​​obtained each time the template image 63 is moved by one pixel. The control unit 30 detects the point with the smallest evaluation value (i.e., the point with the highest similarity to the template image 63) as a general matching point.

[0072] The upper camera image 61 and lower camera image 62 shown in Figure 7 are images after component E has been mounted in region R3. The graph shown in Figure 7 is an example of a graph of evaluation values ​​obtained each time the template image 65 is moved by one pixel.

[0073] (2-1-1-2) Detection of high-reliability matching points As shown in Figures 6 and 7, images of the substrate P often contain multiple similarly shaped components E and solder paste 80, unlike images of the outdoor environment. Therefore, multiple points with small evaluation values ​​may exist. When multiple points with small evaluation values ​​exist, there is a risk of misdetection of matching points using general matching point detection methods.

[0074] For example, in the case shown in Figure 6, since regions R3 and R4 are similar, if evaluation values ​​are calculated while moving the template image 63, the evaluation values ​​will be small at the positions overlapping with region R3 and region R4. Therefore, if matching points are detected only from the image before implementation, region R4 may be mistakenly detected as a matching point when the matching point that should be detected is actually region R3.

[0075] Therefore, the control unit 30 extracts points in the upper camera image 61 and lower camera image 62 captured before implementation where the evaluation value is less than or equal to a predetermined value, as candidates for general matching points. Similarly, the control unit 30 extracts points in the upper camera image 61 and lower camera image 62 captured after implementation where the evaluation value is less than or equal to a predetermined value, as candidates for general matching points. The control unit then compares the evaluation values ​​of the candidates for general matching points before implementation with the evaluation values ​​of the candidates for general matching points after implementation to detect highly reliable matching points (hereinafter referred to as high-reliability matching points).

[0076] Specifically, as shown in Figure 8, region R3 shows a small difference in evaluation values ​​before and after implementation (i.e., high similarity). In contrast, regions R1 and R2 show a large difference in evaluation values ​​before implementation (i.e., low similarity), but a large difference in evaluation values ​​after implementation (i.e., high similarity). Region R4 shows a large difference in evaluation values ​​before implementation (i.e., high similarity), but a large difference in evaluation values ​​after implementation (i.e., low similarity). Therefore, the control unit 30 detects region R3, which is the region with the smallest difference in evaluation values ​​before and after implementation, as a highly reliable matching point. In other words, the control unit 30 detects a highly reliable matching point by utilizing the fact that the image changes depending on the presence or absence of component E before and after implementation.

[0077] (2-1-2) Conversion to parallax height As shown in Figures 6 and 7, the difference (offset amount) between the position of the template image 63 in the upper camera image 61 and the matching point is obtained as the parallax p (pixel). As shown in Figure 9, if the resolution of the stereo camera 23 is R (μm / pixel), the distance A (μm) can be calculated using the parallax p and the following equation 1. A=p×R / sin(θH-θL)...Equation 1 Then, using the distance A obtained by equation (1), the height h (μm) of the object relative to the reference plane Ps can be determined by the following equation 2. h=A×sin(θL)...Equation 2

[0078] This allows us to obtain height information for the mounting position Pa relative to the reference plane Ps, and height information for the substrate surface Pb relative to the reference plane Ps. The height information can be any information that correlates with height h. For example, the height information h shown in Figure 9 may be used as height information, or distance A or disparity p, which are correlated with height h, may be used as height information. The method of obtaining height information by stereo matching is not limited to the above example, and any method may be used.

[0079] (2-2) Processing using height information Height information is used in various processes. Here, we will explain the processes using height information, including correction of the target mounting height using the height information of the mounting position Pa, correction of the first region AR1 using the height information of the substrate surface Pb, and determination of mounting success or failure.

[0080] (2-2-1) Correction of target implementation height using height information of the implementation position Referring to Figure 10, the correction of the target mounting height using the height information of the mounting position Pa will be explained. As shown in Figure 10, if the height of the mounting position Pa before mounting (i.e., the height of the substrate surface at the mounting position Pa before the component E is mounted) is higher than the reference surface Ps, there is a risk that the component E may collide with the substrate P when the component E is mounted on the substrate P, causing an impact to the component E. For this reason, when mounting the component E on the substrate P, the control unit 30 corrects the target mounting height of the component E based on the height information of the mounting position Pa before mounting, and lowers the mounting head 18 to match the corrected target mounting height. This reduces the impact applied to the component E.

[0081] (2-2-2) Correction of the first region AR1 using substrate surface height information Referring to Figure 5, the correction of the first region AR1 will be explained. The correction of the first region AR1 includes vertical correction and horizontal correction. The control unit 30 recognizes the suction state of the component E based on the image captured by the component imaging camera 21, and corrects the horizontal position of the first region AR1 based on the recognition result. Then, the control unit 30 corrects the vertical position of the first region AR1 based on the height information of the substrate surface Pb. As a result, even when the component E is attracted to the mounting head 18 in a misaligned state, or when there is warping (positional misalignment in the height direction) of the substrate P, the first region AR1 can be set to an appropriate position.

[0082] (2-2-3) Determining the success or failure of component mounting If the mounting of component E is successful, a difference will occur in the height information of the mounting position Pa before and after mounting, corresponding to the thickness of component E. Conversely, if mounting fails, there will be no significant difference in the height information of the mounting position Pa before and after mounting. Therefore, the control unit 30 determines whether the change in the height information of the mounting position Pa before and after mounting is greater than or equal to a first determination threshold, as shown in Equation 3 below. The first determination threshold is set according to the thickness of the component E to be mounted. Change in height information before and after implementation at implementation position Pa > First judgment threshold ... Equation 3 The control unit 30 determines that the implementation is successful if the change in height information is greater than or equal to a first determination threshold, and determines that it is a failure if it is less than the first determination threshold.

[0083] (3) Processing using high-reliability matching points As described above, the control unit 30 performs processing using height information, such as correcting the target mounting height using the height information of the mounting position Pa, correcting the first region AR1 using the height information of the substrate surface Pb, and determining whether mounting is successful or not. There are various methods for using high-reliability matching points in these processes. The following describes the processing using high-reliability matching points according to Embodiment 1.

[0084] In the example shown in Figure 11, there are mounting positions Pa1, Pa2, and Pa3 where component E is mounted, and component E is mounted in the order of mounting positions Pa1, Pa2, and Pa3. In Embodiment 1, the control unit 30 first mounts component E at mounting position Pa1 (an example of one mounting position), and then acquires highly reliable pre-mounting height information for mounting position Pa1 based on a high-reliability matching point (an example of the first acquisition process). Then, the control unit 30 corrects the target mounting height when mounting component E at mounting positions Pa2 and Pa3, where component E is mounted after mounting position Pa1, based on the pre-mounting height information for mounting position Pa1.

[0085] Here, since a highly reliable matching point has not been detected when the component E is first mounted at mounting position Pa1, the control unit 30, when mounting component E at mounting position Pa1, acquires height information of mounting position Pa1 based on a general matching point and corrects the target mounting height.

[0086] (3-1) Component mounting flow when mounting components at the initial mounting location Referring to Figure 12, the component mounting flow when mounting component E at the initial mounting position Pa will be explained. In step S101, the control unit 30 uses the component imaging camera 21 to capture an image of the component E held by the mounting head 18, and recognizes the suction state of the component E based on the captured image. Subsequently, the control unit 30 moves the head unit 16 to move the mounting head 18, which is holding the component E to be mounted at the initial mounting position Pa, to above the mounting position Pa.

[0087] In S102, when the mounting head 18 reaches above the mounting position Pa, the control unit 30 lowers the mounting head 18 toward the mounting position Pa. The lowering of the mounting head 18 may start before it reaches above the mounting position Pa. In S103, the control unit 30 captures a predetermined region 81 including the mounting position Pa with the stereo camera 23 from the time the mounting head 18 starts to descend until it completes its descent (an example of pre-mounting imaging). This acquires an upper camera image 61 and a lower camera image 62 before mounting.

[0088] In S104, the control unit 30 sets a second region AR2 in the upper camera image 61 and the lower camera image 62, respectively, before mounting. The control unit 30 then performs stereo matching between the second region AR2 of the upper camera image 61 and the second region AR2 of the lower camera image 62 to extract candidate matching points (general matching points) of the second region AR2 before mounting (an example of the second pre-mount extraction process), and detects the matching point with the smallest evaluation function value among the extracted candidates (corresponding to the matching point detected by the general matching point detection method). Based on the detected matching point, the control unit 30 acquires the height information of the substrate surface Pb before mounting (an example of the fourth acquisition process).

[0089] In S105, the control unit 30 sets the first region AR1 in the upper camera image 61 and the lower camera image 62, respectively, before mounting. At this time, the control unit 30 corrects the first region AR1 based on the recognition result of the suction state of the component E based on the imaging result of the component imaging camera 21, and the height information of the substrate surface Pb before mounting acquired in S104.

[0090] In S106, the control unit 30 performs stereo matching between the first region AR1 of the upper camera image 61 and the first region AR1 of the lower camera image 62 before implementation to extract candidate matching points (general matching points) of the first region AR1 before implementation, and detects the matching point with the smallest evaluation function value among the extracted candidates (corresponding to the matching point detected by the general matching point detection method). Based on the detected matching point, the control unit 30 acquires the height information of the implementation position Pa before implementation (an example of the fifth acquisition process).

[0091] In S107, the control unit 30 mounts the component E, which is held in place by the mounting head 18, to the mounting position Pa. At this time, the control unit 30 corrects the target mounting height based on the pre-mounting height information of the mounting position Pa acquired in S106. In S108, the control unit 30 raises the mounting head 18. In S109, the control unit 30 uses the stereo camera 23 to image a predetermined region 81 including the mounting position Pa from the time the mounting head 18 starts to rise until it completes its rise (an example of post-mounting imaging processing).

[0092] In S110, the control unit 30 sets a second region AR2 in the upper camera image 61 and the lower camera image 62 after mounting. The control unit 30 then performs stereo matching between the second region AR2 of the upper camera image 61 and the second region AR2 of the lower camera image 62 to extract candidate matching points (general matching points) of the second region AR2 after mounting (an example of a second post-mount extraction process), and detects the matching point with the smallest evaluation function value among the extracted candidates (corresponding to the matching point detected by a general matching point detection method). Based on the detected matching point, the control unit 30 obtains the height information of the substrate surface Pb after mounting. In S111, the control unit 30 sets the first region AR1 in the upper camera image 61 and the lower camera image 62 after mounting. At this time, the control unit 30 corrects the first region AR1 based on the recognition result of the suction state of the component E based on the imaging result of the component imaging camera 21, and the height information of the substrate surface Pb after mounting acquired in S110.

[0093] In S112, the control unit 30 performs stereo matching between the first region AR1 of the upper camera image 61 and the first region AR1 of the lower camera image 62 to extract candidate matching points (general matching points) for the first region AR1 after implementation. Among the extracted candidates, the control unit 30 detects the matching point with the smallest evaluation function value (corresponding to the matching point detected by the general matching point detection method). Based on the detected matching point, the control unit 30 obtains the height information of the implementation position Pa after implementation.

[0094] In S114, the control unit 30 determines whether the mounting of component E is successful or not using the aforementioned equation 3. In S115, the control unit 30 proceeds to S115 if it determined in S114 that the implementation was successful, and proceeds to S116 if it determined that it was unsuccessful. In S116, the control unit 30 performs an acquisition process to acquire the height information of the highly reliable mounting position Pa before mounting, based on the first highly reliable matching point. In S117, the control unit 30 performs error handling. During error handling, the control unit 30 performs at least one of the following four processes. • Immediately stop the component mounting device 10 and prompt the operator to confirm whether the mounting was successful or not. • The success or failure of the mounting process is re-evaluated using a different method, such as the substrate imaging camera 22. The post-mounting visual inspection machine located downstream of the component mounting device 10 is instructed to perform additional inspections around the mounting position Pa that was determined to be a failure. The upper camera image 61 and the lower camera image 62 are output to an external device.

[0095] (3-2) Acquisition process Referring to Figure 13, the acquisition process performed in S116 will be explained. In S201, the control unit 30 compares the candidate matching points of the second region AR2 before implementation, extracted in S104, with the candidate matching points of the second region AR2 after implementation, extracted in S110, to detect a high-reliability matching point (hereinafter referred to as the second high-reliability matching point) (an example of the second detection process).

[0096] In S202, the control unit 30 acquires highly reliable pre-mounting height information of the substrate surface Pb based on the second highly reliable matching point detected in S201 (an example of the second acquisition process). In S203, the control unit 30 corrects the first region AR1 of the upper camera image 61 and the first region AR1 of the lower camera image 62 before mounting, based on the height information of the substrate surface Pb acquired in S202 (an example of pre-mounting first region correction processing).

[0097] In S204, the control unit 30 performs stereo matching between the first region AR1 of the upper camera image 61 and the first region AR1 of the lower camera image 62 before implementation, and extracts candidate matching points for the first region AR1 before implementation (an example of the first pre-implementation extraction process). In S205, the control unit 30 acquires highly reliable height information of the substrate surface Pb after mounting based on the second highly reliable matching point detected in S201 (an example of a third acquisition process).

[0098] In S206, the control unit 30 corrects the first region AR1 of the upper camera image 61 and the first region AR1 of the lower camera image 62 after mounting, based on the height information of the substrate surface Pb acquired in S205 (an example of post-mounting first region correction processing). In S207, the control unit 30 performs stereo matching between the first region AR1 of the upper camera image 61 after implementation and the first region AR1 of the lower camera image 62 to extract candidate matching points for the first region AR1 after implementation (an example of the first post-implementation extraction process).

[0099] In S208, the control unit 30 compares the candidate matching points of the first region AR1 before implementation extracted in S204 with the candidate matching points of the first region AR1 after implementation extracted in S207 to detect a high-reliability matching point (hereinafter referred to as the first high-reliability matching point) (an example of the first detection process). In S209, the control unit 30 acquires the height information of the highly reliable mounting position Pa before mounting based on the detected first highly reliable matching point (an example of the first acquisition process). In S210, the control unit 30 stores the acquired height information of the mounting position Pa before mounting in RAM.

[0100] (3-3) Component mounting flow when mounting a component at a mounting position where component E is mounted after the initial mounting position. Referring to Figure 14, the component mounting flow when component E is mounted at mounting position Pa after the initial mounting position Pa is described. In this component mounting flow, S301 is executed additionally, and S302 is executed instead of S107. In this component mounting flow, S116 (acquisition process) is not executed.

[0101] In S301, the control unit 30 reads the height information stored in RAM in S210 (i.e., the height information of the initial mounting position Pa before mounting) from RAM. In S302, the control unit 30 mounts the component E, which is held in place by the mounting head 18, to the mounting position Pa. At this time, the control unit 30 corrects the target mounting height based on the height information read from RAM in S301 (an example of the first correction process).

[0102] In this component mounting flow, the control unit 30 executes S106 to obtain pre-mount height information for mounting position Pa based on a general matching point. However, in S302, the control unit 30 corrects the target mounting height based on the height information read from RAM in S301, rather than the height information obtained in S106. The pre-mount height information for mounting position Pa obtained in S106 is used to determine the success or failure of mounting in S114.

[0103] (4) Effects of the Embodiment According to the component mounting apparatus 10 of Embodiment 1, in S208 (first detection process), the candidate matching points (matching points detected by a general matching point detection method) extracted in S204 (first pre-mounting extraction process) are compared with the candidate matching points (matching points detected by a general matching point detection method) extracted in S207 (first post-mounting extraction process) to detect a first high-reliability matching point. Since the first high-reliability matching point is more reliable than the matching points detected by a general matching point detection method, high-reliability height information can be obtained by acquiring height information based on the first high-reliability matching point (S209).

[0104] According to the component mounting device 10, a highly reliable matching point (first highly reliable matching point) can be detected by detecting the candidate matching point with the smallest difference in evaluation values ​​before and after mounting as the first highly reliable matching point.

[0105] According to the component mounting device 10, the height information of the mounting position Pa before mounting is acquired based on the first highly reliable matching point (S209), so highly reliable height information of the mounting position Pa before mounting can be obtained.

[0106] According to the component mounting device 10, the height information of the substrate surface Pb before mounting is acquired based on a second highly reliable matching point (S202), so highly reliable height information of the substrate surface Pb before mounting can be obtained. Furthermore, according to the component mounting device 10, the first region AR1 is corrected on each image captured in S103 (pre-mounting imaging processing) based on the highly reliable height information of the substrate surface Pb before mounting, so the discrepancy between the first region AR1 recognized by the control unit 30 on the image and the actual first region AR1 on the image can be reduced. As a result, highly reliable height information of the mounting position Pa before mounting can be obtained.

[0107] According to the component mounting device 10, the first region AR1 is corrected on each image captured in S109 (post-mounting imaging processing) based on highly reliable post-mounting height information of the substrate surface Pb (S206), so that highly reliable post-mounting height information of the mounting position Pa can be obtained.

[0108] According to the component mounting device 10, for the first mounting position Pa, after mounting component E to that mounting position Pa, highly reliable pre-mounting height information for the mounting position Pa is obtained based on the first highly reliable matching point (S209). When mounting component E to the mounting position Pa where component E will be mounted after the first mounting position Pa, the target mounting height is corrected based on the pre-mounting height information for the first mounting position Pa (S302). This makes it possible to mitigate the impact when mounting component E to the mounting position Pa where component E will be mounted later.

[0109] <Embodiment 2> The component mounting apparatus 10 according to Embodiment 2 is an example of the component mounting apparatus described in (7) in the summary of the embodiments described above. Referring to Figure 15, the processing using the high-reliability matching point according to Embodiment 2 will be described. In Embodiment 2, for each mounting position Pa, the control unit 30 performs a first acquisition process (a process to acquire height information based on the first high-reliability matching point) after mounting the component E at the mounting position Pa to acquire the highly reliable pre-mounting height information of the mounting position Pa. That is, the control unit 30 acquires the highly reliable pre-mounting height information of the mounting position Pa for each mounting position Pa. Then, the control unit 30 corrects the target mounting height when mounting the component E at the mounting position Pa based on the pre-mounting height information of the mounting position Pa where the component E was mounted immediately before the said mounting position Pa (an example of the second correction process).

[0110] For example, when the control unit 30 mounts component E at mounting position Pa2, it corrects the target mounting height based on the pre-mounting height information of mounting position Pa1 obtained based on the first high-reliability matching point. When mounting component E at mounting position Pa3, it corrects the target mounting height based on the pre-mounting height information of mounting position Pa2 obtained based on the first high-reliability matching point. Since there is no mounting position Pa where component E was mounted immediately before at the initial mounting position Pa1, the target mounting height is corrected based on the height information obtained based on the matching point detected by a general matching point detection method.

[0111] According to the component mounting apparatus 10 of Embodiment 2, for each mounting position Pa, highly reliable pre-mounting height information of the mounting position Pa is acquired based on a first highly reliable matching point, and the target mounting height when mounting a component E at the mounting position Pa is corrected based on the pre-mounting height information of the mounting position Pa where a component E was mounted immediately before that mounting position Pa, thereby mitigating the impact applied to the component E mounted at the mounting position Pa.

[0112] <Embodiment 3> The component mounting apparatus 10 according to Embodiment 3 is an example of the component mounting apparatus described in (8) above in the summary of the embodiments. Referring to Figure 16, the processing using high-reliability matching points according to Embodiment 3 will be described. In Embodiment 3, for each substrate P, the control unit 30 performs a first acquisition process (a process to acquire height information based on a first high-reliability matching point) after mounting a component E at each mounting position Pa on the substrate P to obtain highly reliable pre-mounting height information of the mounting position Pa, and creates a height map in which the mounting position Pa and the pre-mounting height information of the mounting position Pa are associated (an example of the first map creation process). Then, when mounting components E onto the substrate P, the control unit 30 corrects the target mounting height for each mounting position Pa on the substrate P based on a height map created on the substrate P on which components E were mounted immediately before the substrate P (an example of a third correction process).

[0113] Here, when mounting component E onto the first substrate P, since there is no substrate P on which component E was previously mounted, the target mounting height is corrected based on the matching point detected by a general matching point detection method.

[0114] According to the component mounting apparatus 10 of Embodiment 3, when mounting components E at each mounting position Pa on the substrate P, the target mounting height is corrected based on a height map created on the substrate P on which components E were mounted immediately before the said substrate P. This makes it possible to mitigate the impact applied to the components E mounted on the said substrate P.

[0115] <Embodiment 4> The component mounting apparatus 10 according to Embodiment 4 is an example of the component mounting apparatus described in (9) above in the summary of the embodiments. Referring to Figures 17A, 17B, and 17C, the process using the high-reliability matching point according to Embodiment 4 will be described. In these figures, the clamp portion 71 that holds the substrate P and the pressing portion 72 that presses down on the edge of the substrate from above are part of the backup device 70 that fixes the substrate P at the work position A. The backup device 70 also includes a plurality of backup pins that support the substrate P from below, and a lifting portion that raises and lowers the backup pins. The substrate P is positioned vertically by the pressing portion 72 and the backup pins while being held by the clamp portion 71.

[0116] In Embodiment 4, the control unit 30 mounts component E at the first mounting position Pa1 (an example of one mounting position) and then performs a first acquisition process (a process to acquire height information based on a first highly reliable matching point) to acquire highly reliable pre-mounting height information for the first mounting position Pa1. The control unit 30 then sets a virtual board surface 73 based on the acquired height information and the height information of the board edge fixed by the backup device 70 (an example of a setting process). Then, the control unit 30 obtains the height information of the mounting position Pa where the component E is mounted after the first mounting position Pa1 (i.e., the mounting position Pa where the second and subsequent components E are mounted) from the virtual board surface 73, and corrects the target mounting height when mounting the component E at the mounting position Pa based on the height information obtained from the virtual board surface 73 (an example of the fourth correction process).

[0117] There are various methods for setting the virtual board surface 73. Here, we will illustrate three methods. • Method of dividing into multiple regions • Method of linear interpolation • Method of curve interpolation

[0118] (1) Method of dividing into multiple regions Referring to Figure 17A, a method for dividing the substrate into multiple regions will be described. As shown in Figure 17A, the substrate edge is fixed by the backup device 70, so the height information of the substrate edge is constant. The height information of the substrate edge is assumed to be stored in the storage unit 34 in advance.

[0119] The control unit 30 divides the area from the initial mounting position Pa1 to the edge of the board into multiple regions according to the difference between the height information of mounting position Pa1 before mounting and the height of the board edge. This sets up a virtual board surface 73. Then, for mounting positions Pa (mounting position Pa2 in Figure 7A) where the second and subsequent components E are mounted, the control unit 30 obtains the height information of the region to which that mounting position Pa belongs from the virtual board surface 73.

[0120] (2) Linear interpolation method Referring to Figure 17B, the linear interpolation method will be explained. The control unit 30 sets a line passing through the mounting position Pa1 and the edge of the board as a virtual board surface 73, based on the height information of the first mounting position Pa1 before mounting and the height information of the edge of the board. Then, for the mounting positions Pa where the second and subsequent components E are mounted, the control unit 30 obtains the height information of that mounting position Pa before mounting from the set line.

[0121] (3) Method of curve interpolation Referring to Figure 17C, the curve interpolation method will be explained. The control unit 30 sets a quadratic curve passing through the mounting position Pa1 and the board edge as a virtual board surface, based on the height information of the first mounting position Pa1 before mounting and the height information of the board edge. Then, for the mounting positions Pa where the second and subsequent components E are mounted, the control unit 30 obtains the height information of the mounting position Pa before mounting from the set quadratic curve.

[0122] (4) Effects of the Embodiment According to the component mounting apparatus 10 of Embodiment 4, for the first mounting position Pa1, the apparatus acquires highly reliable pre-mounting height information for the mounting position Pa1 based on a first highly reliable matching point, and sets a virtual substrate surface 73 based on the acquired height information and the height information of the substrate edge. Then, the component mounting apparatus 10 corrects the target mounting height when mounting component E to mounting position Pa after mounting position Pa1 based on the height information acquired from the virtual substrate surface 73, thereby mitigating the impact applied to component E when it is mounted to the later mounting position Pa.

[0123] <Embodiment 5> The component mounting apparatus 10 according to Embodiment 5 is an example of the component mounting apparatus (10) described in the summary of the embodiments above. Referring to Figure 18, the processing using high-reliability matching points according to Embodiment 5 will be described. In Embodiment 5, when mounting components E on the first substrate P (an example of a single substrate), the control unit 30, after mounting the components E at each mounting position Pa on the first substrate P, executes a first acquisition process (a process to acquire height information based on the first high-reliability matching points) to acquire highly reliable pre-mounting height information for the mounting positions Pa. Then, the control unit 30 creates a height map in which the mounting positions Pa and the pre-mounting height information for the mounting positions Pa are associated (an example of a second map creation process). Then, the control unit 30 corrects the target mounting height when mounting component E at the mounting position Pa of the second and subsequent substrates P (an example of a substrate on which component E is mounted after the first substrate) based on the height map created on the first substrate P (an example of a fifth correction process).

[0124] According to the component mounting apparatus 10 of Embodiment 5, when mounting components E at mounting positions Pa on the second and subsequent substrates P, the target mounting height is corrected based on the height map created on the first substrate P, thereby mitigating the impact applied to the components E mounted on the second and subsequent substrates P.

[0125] <Embodiment 6> The component mounting system 1 according to Embodiment 6 is an example of the component mounting system (15) described in the summary of the embodiments above. Referring to Figure 19, the processing using the high-reliability matching point according to Embodiment 6 will be described. In Embodiment 6, when mounting a component E on the substrate P, the control unit 30 of the component mounting device 10 (an example of one component mounting device) at the upstreammost point in the transport direction of the substrate P executes a first acquisition process (a process to acquire height information based on the first high-reliability matching point) for each mounting position Pa to acquire the height information of the highly reliable mounting position Pa before mounting, and creates a height map in which the mounting position Pa and the height information of the mounting position Pa before mounting are associated (an example of a fourth map creation process). A height map is created for each substrate P.

[0126] The control unit 30 of the component mounting device 10 located downstream of the upstream component mounting device 10 corrects the target mounting height of component E when mounting component E at mounting position Pa based on the height map created by the upstream component mounting device 10 (an example of a fifth correction process).

[0127] According to the component mounting system 1 of Embodiment 6, the downstream component mounting device 10 corrects the target mounting height when mounting component E at mounting position Pa based on the height map created by the upstream component mounting device 10, thereby mitigating the impact applied to component E mounted by the downstream component mounting device 10.

[0128] Here, we have illustrated the case where only the upstream component mounting device 10 creates a height map. However, each component mounting device 10 (except for the downstream component mounting device 10) may create a height map for each substrate P, and the downstream component mounting device 10 may correct the target mounting height based on the height map created by the immediately preceding component mounting device 10.

[0129] <Embodiment 7> The component mounting apparatus 10 according to Embodiment 7 is an example of the component mounting apparatus (11) to (12) described in the overview of the embodiments above. The component mounting apparatus 10 according to Embodiment 7 determines mounting success or failure based on the change in height information before and after mounting at mounting position Pa, which is obtained based on a general matching point. However, since the general matching point is not always highly reliable, it may not be possible to clearly determine mounting success or failure. If the control unit 30 cannot clearly determine mounting success or failure, it re-determines mounting success or failure based on the change in height information before and after mounting at a highly reliable mounting position Pa, which is obtained based on a first highly reliable matching point.

[0130] Specifically, the control unit 30 performs stereo matching on each image captured in S103 (pre-mounting imaging process) to obtain height information of the mounting position Pa before mounting (an example of a fourth acquisition process), and performs stereo matching on each image captured in S109 (post-mounting imaging process) to obtain height information of the mounting position Pa after mounting (an example of a fifth acquisition process). The fourth and fifth acquisition processes are processes that detect matching points using a conventional general matching point detection method and obtain height information.

[0131] Then, as shown in Equation 4 below, the control unit 30 determines that the mounting is successful if the change in height information of the mounting position Pa before and after mounting is greater than the value obtained by adding a predetermined margin value to the first determination threshold, and determines that the success or failure of the mounting cannot be clearly determined (i.e., it is impossible to determine) if it is less than that value (an example of the second success or failure determination process). Change in height information before and after implementation at implementation position Pa > First judgment threshold + margin value ... Equation 4

[0132] For example, suppose the first judgment threshold is 150 μm and the margin value is 25 μm. In this case, if the height change of the mounting position Pa before and after mounting is 250 μm, then, as shown in Equation 5 below, the change in height information is greater than the value obtained by adding the margin value to the first judgment threshold (= 175 μm), and therefore the mounting is judged to be successful. 250 μm > Judgment threshold 150 μm + margin value 25 μm = 175 μm ... Equation 5

[0133] In contrast, if the change in height information before and after implementation at implementation position Pa is 170 μm, then, as shown in Equation 6 below, the change in height information is greater than the first judgment threshold (=150 μm), but smaller than the value obtained by adding a margin value to the first judgment threshold (=175 μm), so it is determined that it is impossible to determine. 170 μm < Judgment threshold 150 μm + margin value 25 μm = 175 μm ... Equation 6

[0134] If the control unit 30 determines that it cannot make a determination, it executes the first success / failure determination process described below to re-determine whether the implementation is successful or not. The first success / failure determination process will now be described. Based on the first high-reliability matching point, the control unit 30 performs stereo matching on each image captured in S103 (pre-implementation imaging process) to obtain highly reliable pre-implementation height information of the implementation position Pa, and also performs stereo matching on each image captured in S109 (post-implementation imaging process) to obtain highly reliable post-implementation height information of the implementation position Pa (an example of the first acquisition process). Then, as shown in Equation 3 above, the control unit 30 compares the change in height information of the mounting position Pa before and after mounting with a first determination threshold. If the change in height information of the mounting position Pa before and after mounting is greater than the first determination threshold, it determines that mounting is successful, and if it is less than or equal to the first determination threshold, it determines that it is a failure.

[0135] Referring to Figure 20, the component mounting flow according to Embodiment 7 will be described. The component mounting flow according to Embodiment 7 is performed for all mounting positions Pa. Here, processes that are substantially the same as those in Embodiment 1 are denoted by the same reference numerals and their descriptions are omitted. In S401, the control unit 30 determines whether the change in height information before and after mounting at mounting position Pa satisfies Equation 4. If Equation 4 is satisfied, the control unit 30 determines that mounting was successful; otherwise, it determines that it is impossible to determine. In S402, the control unit 30 terminates processing if the determination result in S401 is successful, and proceeds to S403 if it is not possible to make a determination.

[0136] In S403, the control unit 30 executes a first success / failure determination process to re-determine whether the implementation is successful or not. In S404, the control unit 30 terminates the process if the result of the first success / failure determination process is successful, and proceeds to S117 if it fails.

[0137] Referring to Figure 21, the flow of the first success / failure determination process executed in S403 will be described. Here, processes that are substantially the same as those in the acquisition process flow of Embodiment 1 are denoted by the same reference numerals and their explanations are omitted. In S501, the control unit 30 acquires post-mount height information of the highly reliable mounting position Pa based on the first highly reliable matching point. In S502, the control unit 30 determines that the mounting is successful if the change in height information before and after mounting at mounting position Pa satisfies equation 3, and fails if it does not.

[0138] According to the component mounting apparatus 10 of Embodiment 7, if the change in height information before and after mounting, obtained based on a general matching point, is smaller than the value obtained by adding a margin value to the first judgment threshold (Yes in S402), the success or failure of mounting cannot be clearly determined (unable to determine), and the first success or failure determination process is executed to re-determine the success or failure of mounting. This improves the reliability of the determination of the success or failure of mounting.

[0139] The component mounting device 10 does not perform the first success / failure determination process if the result of determining mounting success or failure based on the change in height information before and after mounting at the mounting position Pa, which is acquired based on a general matching point, is successful. Therefore, it does not need to perform the processes for detecting the first high-reliability matching point (first pre-mounting extraction process, first post-mounting extraction process, first detection process, first acquisition process, etc.). As a result, the processing load on the control unit 30 can be reduced compared to the case where the first high-reliability matching point is detected for all mounting positions Pa and mounting success or failure is determined.

[0140] <Embodiment 8> The component mounting apparatus 10 according to Embodiment 8 is an example of the component mounting apparatus (11) described in the summary of the embodiments above. In the previously described embodiment 7, the first high-reliability matching point is detected only when it is not possible to clearly determine whether the mounting is successful or not; therefore, the detection of the first high-reliability matching point is not necessarily performed for all mounting positions Pa. In contrast, in embodiment 8, the control unit 30 detects the first high-reliability matching point for all mounting positions Pa and determines whether the mounting is successful or not.

[0141] Referring to Figure 22, the component mounting flow according to Embodiment 8 will be described. Here, processes that are substantially the same as those in Embodiment 1 are denoted by the same reference numerals and their descriptions are omitted. In the component mounting flow according to Embodiment 8, S601 is executed instead of S110 to S112. In S601, the control unit 30 executes a process to acquire height information of the mounting position before and after mounting based on the high-reliability matching point. This process is substantially the same as S201 to S501 of the first success / failure determination process shown in Figure 21, so a detailed explanation is omitted.

[0142] According to the component mounting apparatus 10 of Embodiment 8, the success or failure of mounting of component E is determined by comparing the change in height information of the mounting position Pa before and after mounting, which is obtained based on the first highly reliable matching point, with the first determination threshold, thereby improving the reliability of the determination of mounting success or failure.

[0143] <Embodiment 9> The component mounting apparatus 10 according to Embodiment 9 is an example of the component mounting apparatus (13) described in the summary of the embodiments above. In Embodiment 9, the control unit 30 performs stereo matching of the second region AR2 on each image captured in S103 (pre-mounting imaging process) to acquire height information of the substrate surface Pb before mounting (an example of the sixth acquisition process), and performs stereo matching of the second region AR2 on each image captured in S109 (post-mounting imaging process) to acquire height information of the substrate surface Pb after mounting (an example of the seventh acquisition process). The sixth and seventh acquisition processes are processes that detect matching points using a conventional general matching point detection method and acquire height information. The control unit 30 then executes the first success / failure determination process if the change in height information of the substrate surface Pb before and after mounting, obtained based on general matching points, is greater than or equal to the second determination threshold.

[0144] Referring to Figure 23, the component mounting flow according to Embodiment 9 will be described. In the component mounting flow according to Embodiment 9, S701 and S403 are additionally executed. In S701, the control unit 30 determines whether the change in height information of the substrate surface Pb before and after mounting is greater than or equal to a second determination threshold. If the change is less than or equal to the second determination threshold (i.e., the determination is reliable), the control unit 30 proceeds to S111; if it is greater than the second determination threshold (i.e., the determination is unreliable), the control unit 30 proceeds to S403. In S403, the control unit 30 executes the first success / failure determination process described in Embodiment 7.

[0145] According to the component mounting apparatus 10 of Embodiment 9, if the change between the height information of the substrate surface Pb before mounting (height information of the substrate surface Pb before mounting acquired based on general matching points) acquired in S104 (sixth acquisition process) and the height information of the substrate surface Pb after mounting (height information of the substrate surface Pb after mounting acquired based on general matching points) acquired in S110 (seventh acquisition process) is greater than or equal to the second judgment threshold (in other words, if the reliability of the height information is low), the success or failure of mounting is determined by the first success or failure determination process, thereby improving the reliability of the determination of success or failure of mounting.

[0146] In the component mounting apparatus 10 according to Embodiment 9, if the change in height information before and after mounting is smaller than the second determination threshold (in other words, if the reliability of the height information is high), the first success / failure determination process is not executed. Therefore, it is not necessary to execute the process for obtaining the first high-reliability matching point (first pre-mounting extraction process, first post-mounting extraction process, first detection process, etc.). For this reason, the processing load on the control unit 30 can be reduced compared to the case where the first high-reliability matching point is detected for all mounting positions Pa to determine mounting success or failure.

[0147] <Embodiment 10> The component mounting apparatus 10 according to Embodiment 10 is an example of the component mounting apparatus (14) described in the summary of the embodiments above. In Embodiment 10, the control unit 30 mounts components E onto a test substrate P before starting production of the substrate P. For each mounting position Pa on the test substrate P, after mounting components E onto the mounting position Pa, it performs a first acquisition process (a process to acquire height information based on a first highly reliable matching point) to obtain highly reliable height information and create a height map of the substrate P (an example of a third map creation process).

[0148] The height map may be a map that associates the mounting position Pa with the height information of the mounting position Pa before mounting, or a map that associates the mounting position Pa with the height information of the mounting position Pa after mounting, or a map that associates the position of the substrate surface Pb with the height information of the substrate surface Pb before mounting, or a map that associates the position of the substrate surface Pb with the height information of the substrate surface Pb after mounting.

[0149] Referring to Figure 24, the component mounting flow after the start of production of the substrate P according to Embodiment 10 will be described. Here, we will explain using the example where a height map is created that associates the position of the substrate surface Pb with the height information of the substrate surface Pb before mounting, and a height map is created that associates the position of the substrate surface Pb with the height information of the substrate surface Pb after mounting. In S801, the control unit 30 obtains the pre-mount height information of the substrate surface Pb from a height map that associates the position of the substrate surface Pb with the pre-mount height information of the substrate surface Pb. In S802, the control unit 30 obtains the height information of the substrate surface Pb after mounting from a height map that associates the position of the substrate surface Pb with the height information of the substrate surface Pb after mounting.

[0150] According to the component mounting apparatus 10 of Embodiment 10, a height map is created by mounting components E onto a test substrate P before starting production of the substrate P. Therefore, after starting production of the substrate P, height information is obtained from the height map, which reduces the processing load on the control unit 30 during the production of the substrate P compared to when the height map is created during the production of the substrate P.

[0151] <Other Embodiments> The technology disclosed herein is not limited to the embodiments described above in the description and drawings, and the following embodiments, for example, are also included in the technical scope disclosed herein.

[0152] (1) In the above embodiment, an imaging unit having a stereo camera 23 was given as an example of an imaging unit capable of imaging the substrate P from multiple directions. In contrast, the imaging unit may also image the substrate P from multiple directions with a single camera. For example, the substrate P may be imaged from multiple directions with a single camera by arranging mirrors in the optical path to divide the field of view of a single camera.

[0153] (2) In the above embodiment, the imaging unit has illustrated the case in which the substrate P is imaged from two directions by the stereo camera 23, but the imaging unit may also image the substrate P from three or more directions.

[0154] (3) In the above embodiment, the stereo camera 23 is shown as imaging the substrate P between the time the mounting head 18 starts to descend and when it completes its descent, but the mounting head 18 may also image the substrate P before it starts to descend. Similarly, in the above embodiment, the stereo camera 23 was shown to image the substrate P between the time the mounting head 18 starts to rise and when it completes its rise, but it may also image after the rise is complete.

[0155] (4) In the above embodiment, an example was given in which both pre-implementation height information and post-implementation height information are obtained based on the first high-reliability matching point. However, it is also possible to obtain only pre-implementation height information or only post-implementation height information.

[0156] (5) In the above embodiment, an example was given in which a high-reliability matching point (second high-reliability matching point) is detected by comparing it with matching points extracted using a general matching point detection method in the second region for obtaining height information of the substrate surface Pb near the mounting position. In contrast, detection of a high-reliability matching point (second high-reliability matching point) is not required for the second region. Specifically, the height information of the substrate surface Pb near the mounting position may be obtained based on general matching points, and the first region AR1 may be corrected based on that height information. [Explanation of symbols]

[0157] 1: Component mounting system 10: Component mounting equipment 16: Head unit (an example of the mounting section) 17: Head movement section (an example of a mounting section) 22: Circuit board imaging camera (an example of an imaging unit) 30: Control Unit 70: Backup device (an example of a fixed part) 73: Virtual board surface AR1: 1st area AR2: 2nd area E: Parts P: Board Pa: Implementation location Pb: Substrate surface

Claims

1. A component mounting device, The mounting section for mounting components onto the circuit board, The imaging unit is capable of imaging the substrate from multiple directions, Control unit and Equipped with, The control unit, A pre-mounting imaging process is performed in which the imaging unit images the substrate from multiple directions before mounting the aforementioned components. A post-mounting imaging process is performed in which the imaging unit images the substrate from multiple directions after the mounting of the aforementioned components, A first pre-implementation extraction process extracts candidate matching points by performing stereo matching on each image captured in the aforementioned pre-implementation imaging process, A first post-implementation extraction process performs stereo matching on each image captured in the aforementioned post-implementation imaging process to extract candidate matching points, A first detection process that detects a first high-reliability matching point by comparing the candidate matching points extracted in the first pre-implementation extraction process with the candidate matching points extracted in the first post-implementation extraction process, A first acquisition process for acquiring height information based on the first highly reliable matching point, A component mounting device that performs this task.

2. A component mounting apparatus according to claim 1, The control unit, In the first pre-implementation extraction process and the first post-implementation extraction process, the value of an evaluation function that evaluates the similarity of multiple images is obtained and the candidate matching points are extracted. A component mounting apparatus that, in the first detection process, detects a candidate matching point as the first high-reliability matching point, which has the smallest difference between the value of the evaluation function obtained in the first pre-mounting extraction process and the value of the evaluation function obtained in the first post-mounting extraction process.

3. A component mounting apparatus according to claim 1 or claim 2, The control unit, In the first pre-mount extraction process, stereo matching is performed on the first region on each of the images captured in the pre-mount imaging process to obtain height information of the mounting position of the component, and candidate matching points are extracted. In the first post-implementation extraction process, stereo matching is performed on the first region on each of the images captured in the post-implementation imaging process to extract candidate matching points. A component mounting apparatus that, in the first acquisition process, acquires the height information of the mounting position before mounting based on the first high-reliability matching point.

4. A component mounting apparatus according to claim 3, The control unit, before the first pre-implementation extraction process, A second pre-mount extraction process is performed to extract candidate matching points by performing stereo matching on a second region of each image captured in the pre-mount imaging process, which is a second region in which height information of the substrate surface near the mounting position of the component is obtained. A second post-implementation extraction process is performed to extract candidate matching points by performing stereo matching on the second region on each of the images captured in the aforementioned post-implementation imaging process, A second detection process that detects a second high-reliability matching point by comparing the candidate matching points extracted in the second pre-implementation extraction process with the candidate matching points extracted in the second post-implementation extraction process, A second acquisition process is performed to acquire the height information of the substrate surface near the mounting position before mounting, based on the second highly reliable matching point, Based on the pre-mounting height information of the substrate surface near the mounting position acquired in the second acquisition process, a pre-mounting first region correction process is performed to correct the first region on each of the images captured in the pre-mounting imaging process, A component mounting device that performs this task.

5. A component mounting apparatus according to claim 4, The control unit, before the first post-implementation extraction process, A third acquisition process that acquires post-mounting height information of the substrate surface near the mounting position based on the second high-reliability matching point, Based on the post-mount height information of the substrate surface near the mounting position acquired in the third acquisition process, a post-mount first region correction process is performed to correct the first region on each of the images captured in the post-mount imaging process, A component mounting device that performs this task.

6. A component mounting apparatus according to claim 3, The control unit, after mounting the component at one of the mounting positions, executes the first acquisition process to obtain the height information of the one mounting position before mounting. A component mounting apparatus that performs a first correction process to correct the target mounting height when mounting the component at the mounting position where the component is mounted after the aforementioned mounting position, based on the height information of the aforementioned mounting position before mounting.

7. A component mounting apparatus according to claim 3, The control unit, For each of the aforementioned mounting positions, after mounting the component at the mounting position, the first acquisition process is executed to obtain the height information of the mounting position before mounting. A component mounting apparatus that performs a second correction process to correct the target mounting height when mounting the component at the aforementioned mounting position based on the pre-mounting height information of the mounting position where the component was mounted immediately before the said mounting position.

8. A component mounting apparatus according to claim 3, The control unit, For each of the aforementioned substrates, for each of the aforementioned mounting positions on the substrate, the first acquisition process is performed after mounting the component at the mounting position to obtain the height information of the mounting position before mounting, and a first map creation process is performed to create a height map in which the mounting position and the height information of the mounting position before mounting are associated. When mounting the component on the substrate, a third correction process is performed to correct the target mounting height when mounting the component at each mounting position on the substrate based on the height map created on the substrate on which the component was mounted immediately before the substrate. A component mounting device that performs this task.

9. A component mounting apparatus according to claim 3, The substrate is provided with a fixing part for fixing the substrate edge, The control unit, After mounting the component at one of the mounting positions, the first acquisition process is performed to obtain the height information of the one mounting position before mounting, and a setting process is performed to set a virtual board surface based on the acquired height information and the height information of the board edge fixed by the fixing part. A fourth correction process is performed to obtain the height information of the mounting position where the component is mounted after the aforementioned mounting position from the virtual board surface, and to correct the target mounting height when mounting the component at that mounting position based on the height information obtained from the virtual board surface. A component mounting device that performs this task.

10. A component mounting apparatus according to claim 3, The control unit, When mounting the components on a single substrate, a second map creation process is performed to obtain the height information of the mounting position before mounting by executing the first acquisition process after mounting the components on the single substrate, and to create a height map in which the mounting position and the height information of the mounting position before mounting are associated. A fifth correction process that corrects the target mounting height when mounting the component at the mounting position on the board on which the component is mounted after the one board, based on the height map created on the one board, A component mounting device that performs this task.

11. A component mounting apparatus according to claim 3, In the first acquisition process, the control unit acquires the height information of the mounting position before mounting and the height information of the mounting position after mounting based on the first high-reliability matching point. A component mounting apparatus that performs a first success / failure determination process to determine whether the component has been mounted by comparing the change in height information of the mounting position before and after mounting with a first determination threshold.

12. A component mounting apparatus according to claim 11, The control unit, A fourth acquisition process involves performing stereo matching on each of the images captured in the pre-implementation imaging process to obtain the pre-implementation height information of the implementation position, A fifth acquisition process involves performing stereo matching on each of the images captured in the post-implementation imaging process to obtain post-implementation height information of the implementation position, A second success / failure determination process determines that the implementation is successful if the change in height information of the implementation position before and after implementation is greater than the value obtained by adding a predetermined margin value to the first determination threshold, and that the determination is impossible if it is less than the value. Execute, A component mounting apparatus that executes the first success / failure determination process if the second success / failure determination process determines that it is impossible to determine the success or failure.

13. A component mounting apparatus according to claim 11, The control unit, A sixth acquisition process is performed to obtain the height information of the substrate surface near the mounting position of the component by performing stereo matching of the second region of each image captured in the pre-mounting imaging process, A seventh acquisition process, which involves performing stereo matching on the second region on each of the images captured in the post-implementation imaging process to obtain post-implementation height information of the substrate surface near the implementation position, Execute, A component mounting apparatus that executes the first success / failure determination process when the change in height information of the substrate surface near the mounting position before and after mounting is greater than or equal to a second determination threshold.

14. A component mounting apparatus according to claim 1 or claim 2, The control unit, Before commencing production of the aforementioned circuit board, the aforementioned components are mounted on the mounting positions of a test circuit board. A component mounting apparatus that, for each of the mounting positions on the test substrate, performs the first acquisition process to acquire height information after mounting the component at the mounting position, and then performs a third map creation process to create a height map of the substrate.

15. A component mounting system in which a plurality of component mounting devices according to claim 1 or claim 2 are arranged in the transport direction of the substrate, The control unit of one component mounting device, when mounting the component at the mounting position on the substrate, executes the first acquisition process for each mounting position to acquire the height information of the mounting position before mounting, and executes a fourth map creation process to create a height map in which the mounting position and the height information of the mounting position before mounting are associated. A component mounting system in which the control unit of a component mounting device located downstream of the aforementioned single component mounting device in the transport direction of the substrate performs a sixth correction process to correct the target mounting height of the component when mounting the component at the mounting position, based on the height map created by the aforementioned single component mounting device.

16. A method for obtaining height information to obtain height information of a circuit board on which components are mounted, A pre-mounting imaging step is performed in which the substrate is imaged from multiple directions by an imaging unit before the aforementioned components are mounted. A post-mounting imaging step is performed in which, after mounting the aforementioned components, the substrate is imaged from multiple directions by an imaging unit. A first pre-implementation extraction step involves performing stereo matching on each image captured in the aforementioned pre-implementation imaging step to extract candidate matching points, A first post-implementation extraction step involves performing stereo matching on each image captured in the post-implementation imaging step to extract candidate matching points. A first detection step that detects a first highly reliable matching point by comparing the candidate matching points extracted in the first pre-implementation extraction step with the candidate matching points extracted in the first post-implementation extraction step, A first acquisition step of acquiring height information based on the first highly reliable matching point, A method for obtaining height information, including the above.

Images