Parts recognition device and parts recognition method

Abstract

The present invention provides a component recognition device comprising an extraction unit and a selection unit. The extraction unit performs image processing on image data obtained by capturing a component to be mounted onto a board and extracts a plurality of candidate blobs, which are closed areas of a portion of the component and which can be used for recognition processing for the component. The selection unit sets a rectangular area circumscribing at least one of the plurality of candidate blobs extracted by the extraction unit, compares the tilt angle of the rectangular area with an ideal angle, which is the tilt angle of the component in an ideal state in which the component is positioned correctly, and makes a first selection of selecting, as the object of the recognition processing, the blob candidate in the rectangular area that is close to the ideal angle of the component.

Description

Technical Field

[0001] This specification discloses a technology related to a component recognition device and a component recognition method.

Background Art

[0002] The component mounting machine described in Patent Document 1 includes an image processing unit and a state recognition unit. The image processing unit divides, by luminance, at least a partial region of a component and a background region in image data acquired by imaging with a camera. The state recognition unit recognizes the supply state of a component based on at least one of the area and the shape of a component region occupied by at least a part of the component in the image data on which the image processing has been performed.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] A component mounting machine performs image processing on image data of a component imaged on a substrate to perform a recognition process of the component. For example, chip components such as rectangular parallelepiped resistors and capacitors are supplied in large quantities and need to be mounted on a substrate at high speed as compared with lead components and the like. Therefore, it is assumed that a recognition process of a component is performed using blob candidates extracted by performing image processing on image data. In this case, depending on the method of selecting blob candidates used for the recognition process of a component or the order of using blob candidates, the time required for the recognition process of a component may increase.

[0005] In view of such circumstances, this specification discloses a component recognition device and a component recognition method capable of reducing the time required for the recognition process of a component.

Means for Solving the Problems

[0006] This specification discloses a component recognition device comprising an extraction unit and a selection unit. The extraction unit processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component. The selection unit sets a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, compares the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and performs a first selection to select the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process.

[0007] This specification also discloses a component recognition device comprising an extraction unit, an assignment unit, and a recognition unit. The extraction unit processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component. The assignment unit assigns a priority order for use in the recognition process to the plurality of blob candidates extracted by the extraction unit. The recognition unit attempts the recognition process in order of the blob candidates with the highest priority assigned by the assignment unit.

[0008] Furthermore, this specification discloses a component recognition method comprising an extraction step and a selection step. The extraction step performs image processing on image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component. The selection step sets a rectangular region that circumscribes at least one of the plurality of blob candidates extracted in the extraction step, compares the inclination angle of the rectangular region with an ideal angle which is the inclination angle of the component in an ideal state where the component is correctly positioned, and performs a first selection in which the blob candidate of the rectangular region that is close to the ideal angle of the component is selected as the target of the recognition process.

[0009] This specification also discloses a component recognition method comprising an extraction step, an assignment step, and a recognition step. The extraction step extracts a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component by image processing of image data of a component to be mounted on a substrate. The assignment step assigns a priority order for use in the recognition process to the plurality of blob candidates extracted by the extraction step. The recognition step attempts the recognition process in order of the blob candidates with the highest priority assigned by the assignment step.

[0010] Furthermore, this specification discloses a technical concept in which, in claim 10 of the claims initially attached to the application (hereinafter referred to as the "original claims"), "the parts recognition device described in claim 1 or claim 6" is changed to "the parts recognition device described in any one of claims 1 to 9". In addition, this specification discloses a technical concept in which, in claim 11 of the original claims, "the parts recognition device described in claim 2 or claim 6" is changed to "the parts recognition device described in any one of claims 2 to 10". [Effects of the Invention]

[0011] According to the first part recognition device described above, blob candidates in a rectangular region close to the ideal angle of the part can be selected as the target of recognition processing, thereby reducing the time required for part recognition processing compared to cases where no appropriate blob candidate is selected. Furthermore, according to the second part recognition device, recognition processing can be attempted in order from the highest priority blob candidate, thereby reducing the time required for part recognition processing compared to cases where the order of use of blob candidates is not determined. The same applies to the part recognition method as described above regarding the part recognition device. [Brief explanation of the drawing]

[0012] [Figure 1] This is a plan view showing an example configuration of a parts mounting machine. [Figure 2] This is a schematic diagram showing an example of image data captured by a parts camera of a part held by a holding member. [Figure 3] This is a block diagram showing an example of a control block for a component recognition device. [Figure 4] This flowchart shows an example of a control procedure using a component recognition device. [Figure 5] This is a schematic diagram showing an example of a coordinate system setting. [Figure 6] This is a schematic diagram showing an example of a blob candidate extracted by the extraction unit. [Figure 7] This is a schematic diagram showing another example of a blob candidate extracted by the extraction unit. [Figure 8] This is another example of a control block for a component recognition device. [Figure 9] This flowchart shows another example of a control procedure using a component recognition device. [Modes for carrying out the invention]

[0013] 1. Embodiment 1-1. Example configuration of the parts mounting machine 10 The component recognition devices 50a and 50b can be applied to the component mounting machine 10. The component mounting machine 10 mounts a plurality of components 91 onto a substrate 90. As shown in Figure 1, the component mounting machine 10 of this embodiment includes a substrate transport device 11, a component supply device 12, a component transfer device 13, a component camera 14, a substrate camera 15, and a control device 16.

[0014] The substrate transport device 11 is composed of, for example, a belt conveyor and transports the substrate 90 in the transport direction (X-axis direction). The substrate 90 is a circuit board on which electronic circuits, electrical circuits, magnetic circuits, etc., are formed. The substrate transport device 11 carries the substrate 90 into the component mounting machine 10 and positions the substrate 90 at a predetermined position inside the machine. After the component mounting machine 10 has finished mounting multiple components 91, the substrate transport device 11 transports the substrate 90 out of the component mounting machine 10.

[0015] The component supply device 12 supplies a plurality of components 91 to be mounted on the substrate 90. The component supply device 12 includes a plurality of feeders 12a provided along the conveyance direction (X-axis direction) of the substrate 90. Each of the plurality of feeders 12a is equipped with a reel. A carrier tape in which a plurality of components 91 are stored is wound around the reel. The feeder 12a pitch-feeds the carrier tape and supplies the component 91 so that it can be picked up at the supply position located on the tip side of the feeder 12a.

[0016] In addition, the component supply device 12 may also include a bulk feeder that supplies the components 91 stored in a bulk state (a state where the postures of the plurality of components 91 are irregular) so that they can be picked up. Furthermore, the component supply device 12 can also supply relatively large electronic components (e.g., lead components, etc.) compared to chip components, etc. in a state where they are arranged on a tray.

[0017] The component transfer device 13 includes a head drive device 13a and a moving stage 13b. The head drive device 13a is configured to move the moving stage 13b in the X-axis direction and the Y-axis direction (a direction orthogonal to the X-axis direction in the horizontal plane) by a linear motion mechanism. A mounting head 20 is detachably (replaceably) provided on the moving stage 13b by a clamp member. The mounting head 20 picks up and holds the component 91 supplied by the component supply device 12 using at least one holding member 30, and mounts the component 91 on the substrate 90 positioned by the substrate conveyance device 11. The holding member 30 can use, for example, a suction nozzle, a chuck, etc.

[0018] The component camera 14 and the substrate camera 15 can use known imaging devices. The component camera 14 is fixed to the base of the component mounting machine 10 so that the optical axis is upward in the vertical direction (Z-axis direction orthogonal to the X-axis direction and the Y-axis direction). The component camera 14 can image the component 91 held by the holding member 30 from below. The substrate camera 15 is provided on the moving table 13b of the component transfer device 13 so that the optical axis is downward in the vertical direction (Z-axis direction). The substrate camera 15 can image the substrate 90 from above. The component camera 14 and the substrate camera 15 perform imaging based on a control signal sent from the control device 16. The image data of the captured images captured by the component camera 14 and the substrate camera 15 is transmitted to the control device 16.

[0019] The control device 16 includes a known arithmetic device and a storage device, and a control circuit is configured. Information output from various sensors provided in the component mounting machine 10, image data, etc. are input to the control device 16. The control device 16 sends a control signal to each device based on a control program and predetermined mounting conditions set in advance.

[0020] For example, the control device 16 causes the substrate camera 15 to image the substrate 90 positioned by the substrate transfer device 11. The control device 16 performs image processing on the image captured by the substrate camera 15 to recognize the positioning state of the substrate 90. Also, the control device 16 causes the component supply device 12 to collect and hold the component 91 supplied by the component supply device 12 on the holding member 30, and causes the component camera 14 to image the component 91 held by the holding member 30. The control device 16 performs image processing on the image captured by the component camera 14 to recognize the holding posture of the component 91.

[0021] The control device 16 moves the holding member 30 upwards towards the planned mounting position, which is predetermined by a control program or the like. The control device 16 also corrects the planned mounting position based on the positioning state of the substrate 90, the holding posture of the component 91, etc., to set the actual mounting position for the component 91. The planned mounting position and the mounting position include rotation angles in addition to position (X-axis coordinates and Y-axis coordinates).

[0022] The control device 16 corrects the target position (X-axis coordinates and Y-axis coordinates) and rotation angle of the holding member 30 to match the mounting position. The control device 16 lowers the holding member 30 at the corrected rotation angle at the corrected target position and mounts the component 91 onto the substrate 90. The control device 16 repeats the above pick-and-place cycle to perform a mounting process in which multiple components 91 are mounted onto the substrate 90.

[0023] 1-2. Example of the configuration of the component recognition device 50a The component mounting machine 10 performs image processing on image data of components 91 to be mounted on the substrate 90 to perform component recognition processing. The recognition processing includes recognizing the position of the component 91 relative to a reference member based on the image data. For example, if the image data is of a component 91 held by a holding member 30, the component mounting machine 10 at least recognizes the position of the component 91 relative to the holding member 30 based on the image data. For example, chip components such as rectangular parallelepiped resistors and capacitors are supplied in large quantities and need to be mounted on the substrate 90 at high speed compared to leaded components. Therefore, it is envisioned that the recognition processing of the component 91 is performed using a closed region (blob 80) extracted by image processing of the image data.

[0024] Figure 2 shows an example of image data captured by the component camera 14 of a component 91 held by the holding member 30. For example, if the component 91 is a chip component, the component 91 comprises electrode regions A11 and A12, which are the electrode areas, and a body region A13, which is the body area. Electrode regions A11 and A12 are metallic (silver). If the component 91 is a chip resistor, the body region A13 on the back side of the component 91 (the side captured by the component camera 14) is, for example, black. Furthermore, if the component 91 is a chip capacitor, the body region A13 on the back side of the component 91 is, for example, white.

[0025] For example, when the image data shown in Figure 2 is processed by binarization, if component 91 is a chip resistor, the main body region A13 and the background of component 91 are black, making it easier to extract the closed region (blob 81) corresponding to electrode region A11 and the closed region (blob 82) corresponding to electrode region A12. If component 91 is a chip capacitor, the main body region A13 is white, and the brightness of the main body region A13 is close to the brightness of the metallic color (silver) of electrode regions A11 and A12, making it easier to extract the closed region (blob 83) corresponding to the entire component 91.

[0026] If the component mounting machine 10 extracts multiple (two in the above example) closed regions (blobs 80), it can combine them to form a single closed region (blob 83) and perform the recognition process for component 91. In this way, if there are multiple closed regions (multiple blob candidates 60) that can be used for the recognition process of component 91, depending on the method of selecting the blob candidates 60, the recognition process for component 91 may fail, and the time required for the recognition process of component 91 may increase.

[0027] Furthermore, if there are multiple closed regions (multiple blob candidates 60) available for the recognition process of component 91, depending on the order in which the blob candidates 60 are used, the recognition process of component 91 may fail, potentially increasing the time required for the recognition process of component 91. Therefore, in this embodiment, the component mounting machine 10 is equipped with a component recognition device 50a. As will be described later, the component mounting machine 10 may also be equipped with a component recognition device 50b.

[0028] When considered as a control block, the component recognition device 50a comprises an extraction unit 51 and a selection unit 52a. The component recognition device 50a may also include a recognition unit 53. As shown in Figure 3, the component recognition device 50a of the embodiment comprises an extraction unit 51, a selection unit 52a, and a recognition unit 53. The component recognition device 50a can be installed in various control devices.

[0029] For example, the component recognition device 50a can be installed in the control device 16 of the component mounting machine 10, a line management device that manages the production line including the component mounting machine 10, or a management device higher than the line management device. The component recognition device 50a can also be formed on the cloud. As shown in Figure 3, in the component recognition device 50a of this embodiment, the extraction unit 51, the selection unit 52a, and the recognition unit 53 are installed in the control device 16 of the component mounting machine 10.

[0030] Furthermore, the component recognition device 50a of the embodiment performs control according to the flowchart shown in Figure 4. The extraction unit 51 performs the processing and decision shown in steps S11 to S14. The selection unit 52a performs the processing and decision shown in steps S15a to S18a. The recognition unit 53 performs the processing shown in steps S19 to S21.

[0031] 1-2-1. Extraction part 51 The extraction unit 51 processes the image data of the component 91 to be mounted on the substrate 90 to extract multiple blob candidates 60 that are closed regions of a part of the component 91 and can be used for the recognition process of the component 91. The blob candidates 60 are synonymous with the closed regions (blobs 80) described above, but differ from the blobs 80 in that the shape of the closed region actually extracted is taken into consideration.

[0032] For example, due to the lighting conditions when component 91 is imaged, it may be difficult to extract the entire closed region (blob 81) corresponding to electrode region A11. As shown in Figures 6 and 7, the actually extracted closed region may be only a part of the region corresponding to electrode region A11 (shown as an ellipse in the figures for illustrative purposes). The same applies to the closed region (blob 82) corresponding to electrode region A12, and also to a single closed region (blob 83) formed by combining blobs 81 and 82.

[0033] The extraction unit 51 only needs to be able to extract multiple blob candidates 60 and can take various forms. For example, the extraction unit 51 can extract a maximum blob 61 and a combined blob 63. The maximum blob 61 refers to the blob candidate 60 with the largest area of ​​the closed region among the multiple blob candidates 60. The combined blob 63 refers to the blob candidate 60 formed by combining the next largest blob 62 (which has the next largest area of ​​the closed region after the maximum blob 61) and the maximum blob 61. This allows the extraction unit 51 to suppress misrecognition, which would cause relatively small foreign objects, such as electrodes, to be mistakenly identified as part of the component 91.

[0034] Specifically, the extraction unit 51 processes the image data and performs blob processing (step S11 shown in Figure 4). The extraction unit 51 can process the image data and perform blob processing in various ways. For example, as previously described, the extraction unit 51 can process the image data by binarization and perform blob processing. Blob processing analyzes the shape characteristics of the part 91 based on at least one of the following: presence, number, area, position, length, and direction of closed regions.

[0035] For example, the threshold for the binarization process can be set to a value between the brightness of the component 91 and the brightness of the background. Furthermore, the binarization process does not necessarily have to divide the image into white and black regions; for example, the threshold can be set to the average brightness of each region or to a predetermined brightness value. In addition, the extraction unit 51 can change the threshold for the binarization process for each type of component 91, and can also change the type of image processing.

[0036] Next, the extraction unit 51 determines whether the number of blob candidates 60 obtained by the blob processing is two or more (step S12). If the number of blob candidates 60 is two or more (Yes in step S12), the extraction unit 51 extracts the largest blob 61 and the next largest blob 62 (step S13). The extraction unit 51 also extracts a combined blob 63 (step S14).

[0037] In the examples shown in Figures 6 and 7, for illustrative purposes, each of the two blob candidates 60 is represented by an ellipse. The curve of the ellipse indicates the outer edge 60e of the blob candidate 60. Of the two blob candidates 60, the blob candidate 60 at the bottom of the page is designated as the largest blob 61, and the blob candidate 60 at the top of the page is designated as the second-largest blob 62. The combined blob 63 is represented by a dashed-dotted line. The dashed-dotted line indicates the outer edge 60e of the region formed by combining the largest blob 61 and the second-largest blob 62.

[0038] 1-2-2. Selection unit 52a and recognition unit 53 If there are multiple blob candidates 60 to be used for the recognition process of component 91, as described later, depending on the method of selecting the blob candidates 60, the recognition process of component 91 may fail, and the time required for the recognition process of component 91 may increase.

[0039] Therefore, the selection unit 52a performs a first selection. Specifically, the selection unit 52a sets a rectangular region 70 that circumscribes at least one of the multiple blob candidates 60 extracted by the extraction unit 51. Then, the selection unit 52a compares the ideal angle θ0, which is the inclination angle of the part 91 in the ideal state where the part 91 is correctly positioned, with the inclination angles θ1 and θ2 of the rectangular region 70, and selects the blob candidate 60 of the rectangular region 70 that is close to the ideal angle θ0 of the part 91 as the target for recognition processing. In the above example, the ideal state refers to the state in which the part 91 is held by the holding member 30 at a predetermined inclination angle.

[0040] Figure 5 shows an example of a coordinate system setting. The coordinate system shown in the figure is defined by the X0 axis and the Y0 axis which is perpendicular to the X0 axis. The origin G0 is the intersection of the X0 and Y0 axes. As shown in the figure, the inclination angle of part 91 is set to 0 degrees when the longitudinal direction of part 91 is parallel to the X0 axis. Also, for the sake of explanation, the angles described herein are the angles obtained by rotating the longitudinal direction of part 91 counterclockwise from the X0 axis. Note that the angles described herein can also be defined by the angles obtained by rotating the longitudinal direction of part 91 clockwise from the X0 axis.

[0041] The ideal angle θ0 of part 91 is predetermined based on mounting conditions and other factors. As shown in Figures 6 and 7, the ideal angle θ0 of part 91 is, for example, 90 degrees. Normally, part 91 is held by the holding member 30 at an angle close to the ideal angle θ0. Therefore, part 91 held by the holding member 30 at an angle far from the ideal angle θ0, such as 0 degrees, is likely to be different from the intended part 91.

[0042] Figure 6 shows an example of a blob candidate 60 extracted by the extraction unit 51, in which a rectangular region 70 circumscribing the combined blob 63 is defined. The inclination angle θ1 of the rectangular region 70 is indicated by the X1 axis and the Y1 axis perpendicular to the X1 axis. The X1 axis is along the longitudinal direction of the rectangular region 70, and the Y1 axis is along the short direction of the rectangular region 70. Figure 7 shows another example of a blob candidate 60 extracted by the extraction unit 51, in which a rectangular region 70 circumscribing the largest blob 61 is defined. The inclination angle θ2 of the rectangular region 70 is indicated by the X2 axis and the Y2 axis perpendicular to the X2 axis. The X2 axis is along the longitudinal direction of the rectangular region 70, and the Y2 axis is along the short direction of the rectangular region 70.

[0043] The selection unit 52a compares the ideal angle θ0 of the component 91 with the inclination angles θ1 and θ2 of the rectangular region 70, and selects the blob candidate 60 of the rectangular region 70 that is close to the ideal angle θ0 of the component 91 as the target for recognition processing (steps S15a to S17a shown in Figure 4). In the example shown in Figures 6 and 7, the selection unit 52a determines whether the inclination angle θ1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal angle θ0 than the inclination angle θ2 of the rectangular region 70 circumscribing the largest blob 61 (step S15a).

[0044] If the tilt angle θ1 is closer to the ideal angle θ0 than the tilt angle θ2 (if Yes in step S15a), the selection unit 52a selects the combined blob 63 as the target for recognition processing (step S16a). If the tilt angle θ2 is closer to the ideal angle θ0 than the tilt angle θ1 (if No in step S15a), the selection unit 52a selects the largest blob 61 as the target for recognition processing (step S17a). If there is only one blob candidate 60 (if No in step S12), the selection unit 52a selects the single blob candidate 60 as the target for recognition processing (step S18a).

[0045] In the example shown in Figures 6 and 7, since the inclination angle θ1 is closer to the ideal angle θ0 than the inclination angle θ2, the selection unit 52a selects the combined blob 63 as the target for recognition processing (step S16a). The selection unit 52a can also set a predetermined offset in the rectangular region 70. Specifically, the selection unit 52a can set a rectangular region 70 that is separated by a predetermined offset amount from the rectangular region 70 that circumscribes at least one of the multiple blob candidates 60 extracted by the extraction unit 51, to the outside of the blob candidate 60.

[0046] The recognition unit 53 attempts recognition processing using the blob candidate 60 selected by the selection unit 52a. The recognition unit 53 can attempt various recognition processes. For example, the recognition unit 53 can perform a first recognition process in which it sets a rectangular region 70 circumscribing the blob candidate 60, estimates the center C0 of the rectangular region 70 as the center of the part 91, and estimates the inclination angles θ1 and θ2 of the rectangular region 70 as the inclination angles of the part 91 (step S19). This allows the recognition unit 53 to recognize the approximate position and inclination angle of the part 91. In the above example, the selection unit 52a selects the combined blob 63 as the target of recognition processing. The recognition unit 53 estimates the center C0 of the rectangular region 70 shown in Figure 6 as the center of the part 91, and estimates the inclination angle θ1 of the rectangular region 70 shown in Figure 6 as the inclination angle of the part 91.

[0047] Furthermore, after performing the first recognition process, the recognition unit 53 can detect the outer edge 60e of the blob candidate 60 and perform a second recognition process to estimate the center and tilt angle of the part 91 based on the detected outer edge 60e of the blob candidate 60 (step S20). As a result, the recognition unit 53 can recognize the position and tilt angle of the part 91 more accurately than in the first recognition process.

[0048] The recognition unit 53 can detect the outer edge 60e of the blob candidate 60 using a known edge detection method. For example, the recognition unit 53 can determine that the outer edge 60e of the blob candidate 60 is present when the amount of change in brightness of adjacent pixels in the image data exceeds a predetermined amount. In the above example, the recognition unit 53 detects the outer edge 60e of the combined blob 63, which is shown by the dashed line in Figure 6. The recognition unit 53 then estimates the center of the combined blob 63 to be the center of the component 91, and estimates the inclination angle θ1 of the combined blob 63 to be the inclination angle of the component 91.

[0049] Furthermore, the recognition unit 53 can estimate the external dimensions of the part 91 based on the outer edge 60e of the blob candidate 60 and perform a verification process to confirm that the estimated external dimensions of the part 91 are within the dimensional tolerances specified for the part 91 (step S21). This allows the recognition unit 53 to confirm that the part 91 is within the dimensional tolerances. In the above example, the recognition unit 53 estimates the width dimension W1 of the part 91 based on the width dimension W1 of the connecting blob 63 shown by the dashed line in Figure 6. The recognition unit 53 also estimates the depth dimension D1 of the part 91 based on the depth dimension D1 of the connecting blob 63.

[0050] The dimensional tolerance for the width dimension specified for part 91 is a range that is increased or decreased by a predetermined amount, centered on the width dimension W0 of the design part 91 shown in Figure 5. The recognition unit 53 confirms that the estimated width dimension W1 of part 91 falls within this range. Similarly, the dimensional tolerance for the depth dimension specified for part 91 is a range that is increased or decreased by a predetermined amount, centered on the depth dimension D0 of the design part 91 shown in Figure 5. The recognition unit 53 confirms that the estimated depth dimension D1 of part 91 falls within this range. Then, the control by the part recognition device 50a is temporarily terminated.

[0051] The selection unit 52a can also perform a second selection. Specifically, similar to the first selection, the selection unit 52a sets a rectangular region 70 that circumscribes at least one of the multiple blob candidates 60 extracted by the extraction unit 51. Then, in the second selection, the selection unit 52a compares the ideal dimensions (width W0, depth D0), which are the design external dimensions of the part 91, with the external dimensions (width W1, depth D1, width W2, depth D2) of the rectangular region 70. The selection unit 52a selects the blob candidate 60 of the rectangular region 70 that is close to the ideal dimensions (width W0, depth D0) of the part 91 as the target for recognition processing.

[0052] In the second selection, the selection unit 52a determines, in the judgment shown in step S15a of Figure 4, whether the external dimensions (width W1, depth D1) of the rectangular region 70 circumscribing the combined blob 63 are closer to the ideal dimensions (width W0, depth D0) than the external dimensions (width W2, depth D2) of the rectangular region 70 circumscribing the largest blob 61. The selection unit 52a can perform other judgments and processes in the flowchart shown in Figure 4 in the same manner as in the first selection. In this specification, redundant explanations have been omitted.

[0053] In the examples shown in Figures 6 and 7, the width dimension W1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal dimension (width dimension W0) than the width dimension W2 of the rectangular region 70 circumscribing the largest blob 61. Also, the depth dimension D1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal dimension (depth dimension D0) than the depth dimension D2 of the rectangular region 70 circumscribing the largest blob 61. Therefore, the selection unit 52a selects the combined blob 63 as the target for recognition processing (processing corresponding to step S16a).

[0054] Thus, the selection unit 52a can perform the second selection in the same manner as the first selection. However, if, for example, the vendor of part 91 is different, it is necessary to change the ideal dimensions (width dimension W0, depth dimension D0) of part 91. The more types of part 91 there are, the more complicated the modification process becomes, and there is a possibility that the setting of the ideal dimensions (width dimension W0, depth dimension D0) will not be changed. In this case, the recognition unit 53 may fail to recognize part 91. Therefore, it is preferable to use the second selection in conjunction with the first selection.

[0055] For example, the recognition unit 53 attempts to perform recognition processing using the blob candidate 60 selected by the second selection. The selection unit 52a can then perform the first selection if the recognition processing by the recognition unit 53 is unsuccessful. The quality of the recognition processing by the recognition unit 53 can be determined using, for example, inspection results from other imaging devices such as the substrate camera 15, or inspection results from an appearance inspection machine located downstream of the component mounting machine 10. In the above example, for example, the selection unit 52a can determine that the recognition processing is unsuccessful if the recognized position of the component 91 relative to the holding member 30 is not included in a predetermined range that includes the actual position of the component 91 relative to the holding member 30.

[0056] If the recognition position of component 91 by the recognition unit 53 is not included in the predetermined range, the mounting position of component 91 mounted on the substrate 90 is likely to be outside the planned mounting range. Based on the above inspection results, the selection unit 52a can determine whether or not the mounting position of component 91 is included in the planned mounting range, and can determine whether or not the recognition position of component 91 is included in the predetermined range (good or bad recognition processing). The selection unit 52a can also perform a first selection if the number of defects resulting from the recognition processing by the recognition unit 53 exceeds a predetermined number. The predetermined number can be set arbitrarily. This allows the selection unit 52a to perform a first selection if the number of recognition processing defects exceeds a predetermined number.

[0057] The selection unit 52a can first perform both the first and second selections. Then, the recognition unit 53 attempts both recognition processing using the blob candidate 60 selected by the first selection and recognition processing using the blob candidate 60 selected by the second selection. For the sake of explanation, in this specification, the recognition result in which the part 91 is correctly recognized by attempting recognition processing using the blob candidate 60 selected by the first selection is referred to as the first recognition result. Similarly, the recognition result in which the part 91 is correctly recognized by attempting recognition processing using the blob candidate 60 selected by the second selection is referred to as the second recognition result.

[0058] The first and second recognition results only need to represent recognition results in which part 91 was correctly recognized, and can take various forms. For example, the first and second recognition results can be represented by the number of recognition processes in which part 91 was correctly recognized, or by the percentage (recognition rate) in which part 91 was correctly recognized relative to the number of recognition processes attempted. The selection unit 52a can make a first selection for subsequent recognition processes if the first recognition result is better than the second recognition result. This allows the selection unit 52a to make a first selection based on past recognition results.

[0059] 1-3. Example of the configuration of the component recognition device 50b For example, due to a faulty holding state of the part 91 by the holding member 30, the part 91 may rotate more than the expected angle within the holding member 30. In this case, the selection unit 52a selects another blob candidate 60 that is close to the ideal angle θ0 of the part 91 as the target for recognition processing. As a result, a blob candidate 60 different from the one that should have been selected may be selected, potentially leading to a failure in the recognition processing of the part 91. Similarly, if the inclination angles θ1 and θ2 of the rectangular region 70 are similar, a blob candidate 60 different from the one that should have been selected may be selected, potentially leading to a failure in the recognition processing of the part 91.

[0060] Thus, if there are multiple blob candidates 60 used for the recognition process of part 91, depending on the order in which the blob candidates 60 are used, the recognition process of part 91 may fail, and the time required for the recognition process of part 91 may increase. Therefore, the part mounting machine 10 can be equipped with a part recognition device 50b. As shown in Figure 8, the part recognition device 50b, when considered as a control block, comprises an extraction unit 51, an assignment unit 52b, and a recognition unit 53.

[0061] The component recognition device 50b, like the component recognition device 50a, can be installed in various control devices. In the component recognition device 50b shown in Figure 8, the extraction unit 51, the application unit 52b, and the recognition unit 53 are installed in the control device 16 of the component mounting machine 10. The component recognition device 50b performs control according to the flowchart shown in Figure 9. The extraction unit 51 performs the processing and decisions shown in steps S11 to S14. The application unit 52b performs the processing and decisions shown in steps S15b to S18b. The recognition unit 53 performs the processing and decisions shown in steps S19 to S23.

[0062] In the flowchart shown in Figure 9, the same processes and decisions as those in the component recognition device 50a shown in Figure 4 are assigned the same reference numerals, and redundant explanations are omitted in this specification. Specifically, the component recognition device 50b is equipped with an assignment unit 52b instead of a selection unit 52a, and the processes and decisions shown in steps S15b to S18b differ from those in the flowchart shown in Figure 4. Furthermore, the component recognition device 50b differs from the flowchart shown in Figure 4 in that the recognition unit 53 performs the processes and decisions shown in steps S22 and S23. Note that the matters previously described for the component recognition device 50a can be applied to the component recognition device 50b as appropriate.

[0063] The extraction unit 51 processes the image data of the component 91 to be mounted on the substrate 90 to extract a plurality of blob candidates 60 that are closed regions of a part of the component 91 and can be used for the recognition process of the component 91 (steps S11 to S14 shown in Figure 9). The extraction unit 51 can extract a plurality of blob candidates 60 in the same manner as the extraction unit 51 of the component recognition device 50a.

[0064] The assignment unit 52b assigns a priority order to the multiple blob candidates 60 extracted by the extraction unit 51 for use in recognition processing. The assignment unit 52b only needs to be able to assign a priority order and can take various forms. For example, the assignment unit 52b sets a rectangular region 70 that circumscribes at least one of the multiple blob candidates 60 extracted by the extraction unit 51. The assignment unit 52b then compares the ideal dimensions (width W0, depth D0), which are the design external dimensions of the part 91, with the external dimensions (width W1, depth D1, width W2, depth D2) of the rectangular region 70. The assignment unit 52b assigns a higher priority order to blob candidates 60 whose rectangular region 70 is closer to the ideal dimensions (width W0, depth D0) of the part 91.

[0065] The assigning unit 52b can recognize blob candidates 60 of the rectangular region 70 that are close to the ideal dimensions (width W0, depth D0) of the part 91 by performing the second selection as described above in the selection unit 52a of the part recognition device 50a, and can assign a higher priority to such blob candidates 60. In the example shown in Figures 6 and 7, the assigning unit 52b determines whether the external dimensions (width W1, depth D1) of the rectangular region 70 circumscribing the combined blob 63 are closer to the ideal dimensions (width W0, depth D0) than the external dimensions (width W2, depth D2) of the rectangular region 70 circumscribing the largest blob 61 (step S15b shown in Figure 9).

[0066] If the above condition is met (Yes in step S15b), the assignment unit 52b gives a higher priority to the combined blob 63 than to the maximum blob 61 (step S16b). If the above condition is not met (No in step S15b), the assignment unit 52b gives a higher priority to the maximum blob 61 than to the combined blob 63 (step S17b). If there is only one blob candidate 60 (No in step S12), the assignment unit 52b assigns priority to use that one blob candidate 60 for recognition processing (step S18b).

[0067] In the example above, the width dimension W1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal dimension (width dimension W0) than the width dimension W2 of the rectangular region 70 circumscribing the largest blob 61. Also, the depth dimension D1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal dimension (depth dimension D0) than the depth dimension D2 of the rectangular region 70 circumscribing the largest blob 61. Therefore, the assigning unit 52b gives a higher priority to the combined blob 63 compared to the largest blob 61 (step S16b).

[0068] The recognition unit 53 attempts recognition processing in order of the blob candidates 60 assigned by the assignment unit 52b, starting with those with the highest priority. The recognition unit 53 can attempt recognition processing in the same manner as the recognition unit 53 of the component recognition device 50a (steps S19 to S21). If the recognition processing using one blob candidate 60 fails, the recognition unit 53 attempts recognition processing using the blob candidate 60 with the next highest priority (steps S22 and S23).

[0069] In the example above, the assignment unit 52b gives a higher priority to the coupling blob 63 compared to the maximum blob 61. Therefore, the recognition unit 53 first attempts to recognize the component 91 using the coupling blob 63. Then, the recognition unit 53 determines whether the result of the verification process is good or not (step S22). If the result of the verification process is good (Yes in step S22), the control by the component recognition device 50b is terminated for the time being.

[0070] If the result of the verification process is unsatisfactory (No in step S22), the recognition unit 53 attempts the recognition process using the next highest priority blob candidate 60 (in this case, the largest blob 61) (step S23). The recognition unit 53 attempts the recognition process in the same manner as shown in steps S19 to S21. Then, the control by the component recognition device 50b returns to the decision shown in step S22, and the decisions and processes shown in steps S22 and S23 are repeated until the result of the verification process is satisfactory.

[0071] Furthermore, the assigning unit 52b sets a rectangular region 70 that circumscribes at least one of the multiple blob candidates 60 extracted by the extraction unit 51. The assigning unit 52b also compares the ideal angle θ0, which is the inclination angle of the part 91 in the ideal state where the part 91 is correctly positioned, with the inclination angles θ1 and θ2 of the rectangular region 70, and assigns a higher priority to the blob candidates 60 in the rectangular region 70 that are closer to the ideal angle θ0 of the part 91.

[0072] The assigning unit 52b can recognize blob candidates 60 of the rectangular region 70 that are close to the ideal angle θ0 of the part 91 by performing the first selection as described above in the selection unit 52a of the part recognition device 50a, and can assign a higher priority to such blob candidates 60. For example, the assigning unit 52b determines whether the inclination angle θ1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal angle θ0 than the inclination angle θ2 of the rectangular region 70 circumscribing the largest blob 61 (a determination that replaces step S15b shown in Figure 9).

[0073] If the above conditions are met, the assignment unit 52b gives a higher priority to the combined blob 63 than to the maximum blob 61 (step S16b). If the above conditions are not met, the assignment unit 52b gives a higher priority to the maximum blob 61 than to the combined blob 63 (step S17b). If there is only one blob candidate 60 (the answer is No in step S12), the assignment unit 52b assigns priority to use that one blob candidate 60 for recognition processing (step S18b).

[0074] In the example above, the inclination angle θ1 of the rectangular region 70 circumscribing the combined blob 63 is closer to the ideal angle θ0 of the component 91 than the inclination angle θ2 of the rectangular region 70 circumscribing the largest blob 61. Therefore, the assignment unit 52b gives a higher priority to the combined blob 63 compared to the largest blob 61 (step S16b). In this case as well, the recognition unit 53 can attempt recognition processing in order from the blob candidates 60 with the highest priority assigned by the assignment unit 52b.

[0075] 2. Others In the previously described embodiment, the explanation uses image data of a component 91 held by the holding member 30 as an example. However, the image data is not limited to images of a component 91 to be mounted on the substrate 90. For example, the image data may be images of a component 91 supplied in a bulk feeder, captured by an imaging device such as a substrate camera 15. In this case, the component mounting machine 10 recognizes, based on the image data, at least the position of the component 91 relative to the cavity (reference member) in which the component 91 is housed. The ideal state is a state in which the component 91 is housed in the cavity so that it can be supplied to the component mounting machine 10. The matters described herein can be combined and selected as appropriate.

[0076] 3. Part Recognition Method The same applies to the part recognition methods as to the part recognition devices 50a and 50b that have already been described. Specifically, the first part recognition method comprises an extraction step and a selection step. The extraction step corresponds to the control performed by the extraction unit 51. The selection step corresponds to the control performed by the selection unit 52a. The first part recognition method may also include a recognition step. The recognition step corresponds to the control performed by the recognition unit 53. Furthermore, the second part recognition method comprises an extraction step, an assignment step, and a recognition step. The extraction step corresponds to the control performed by the extraction unit 51. The assignment step corresponds to the control performed by the assignment unit 52b. The recognition step corresponds to the control performed by the recognition unit 53.

[0077] 4. An example of the effects of the embodiment According to the first part recognition device 50a, a blob candidate 60 in a rectangular region 70 close to the ideal angle θ0 of part 91 can be selected as the target of recognition processing, and the time required for the recognition processing of part 91 can be reduced compared to the case where no appropriate blob candidate 60 is selected. Furthermore, according to the second part recognition device 50b, recognition processing can be attempted in order from the blob candidate 60 with the highest priority, and the time required for the recognition processing of part 91 can be reduced compared to the case where the order of use of blob candidates 60 is not determined. The same applies to the part recognition method as described above for part recognition devices 50a and 50b. [Explanation of Symbols]

[0078] 50a, 50b: Part recognition device, 51: Extraction unit, 52a: Selection unit, 52b: Assignment part, 53: Recognition part, 60: Blob candidate, 60e: Outer edge part, 61: Largest blob, 62: Second largest blob, 63: Joined blob, 70: Rectangular area, 90: Substrate, 91: Component, θ0: Ideal angle, θ1, θ2: Inclination angle, W0, D0: Ideal dimensions (width dimension, depth dimension) W1, D1, W2, D2: External dimensions (width and depth), C0: Center.

Claims

1. An extraction unit that processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A selection unit performs a first selection by setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, comparing the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and selecting the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process. A recognition unit that attempts the recognition process using the blob candidates selected by the selection unit, Equipped with, The selection unit sets a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, compares the ideal dimensions, which are the design external dimensions of the component, with the external dimensions of the rectangular region, and performs a second selection to select the blob candidate in the rectangular region that is close to the ideal dimensions of the component as the target of the recognition process. The recognition unit attempts the recognition process using the blob candidate selected by the second selection, The selection unit is a component recognition device that performs the first selection when the recognition process by the recognition unit fails.

2. The component recognition device according to claim 1, wherein the selection unit performs the first selection when the number of defects resulting from the recognition process by the recognition unit exceeds a predetermined number.

3. An extraction unit that processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A selection unit performs a first selection by setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, comparing the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and selecting the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process. A recognition unit that attempts the recognition process using the blob candidates selected by the selection unit, Equipped with, The selection unit sets a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, compares the ideal dimensions, which are the design external dimensions of the component, with the external dimensions of the rectangular region, and performs a second selection to select the blob candidate in the rectangular region that is close to the ideal dimensions of the component as the target of the recognition process. The recognition unit attempts both the recognition process using the blob candidate selected by the first selection and the recognition process using the blob candidate selected by the second selection. The component recognition device, when the selection unit attempts to perform the recognition process using the blob candidate selected by the second selection and compares it with the second recognition result in which the component was correctly recognized, and the first recognition result in which the component was correctly recognized when the blob candidate selected by the first selection is good, performs the first selection for subsequent recognition processes.

4. An extraction unit that processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A assigning unit assigns a priority order to the plurality of blob candidates extracted by the extraction unit for use in the recognition process, A recognition unit attempts the recognition process in order of the blob candidates with the highest priority assigned by the assignment unit, Equipped with, The assigning unit sets a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, compares the ideal dimensions, which are the design external dimensions of the part, with the external dimensions of the rectangular region, and assigns a higher priority to the blob candidate whose rectangular region is closer to the ideal dimensions of the part.

5. An extraction unit that processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A assigning unit assigns a priority order to the plurality of blob candidates extracted by the extraction unit for use in the recognition process, A recognition unit attempts the recognition process in order of the blob candidates with the highest priority assigned by the assignment unit, Equipped with, The assigning unit sets a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, compares the ideal angle, which is the inclination angle of the part in an ideal state where the part is correctly positioned, with the inclination angle of the rectangular region, and assigns a higher priority to the blob candidate whose rectangular region is closer to the ideal angle of the part.

6. The part recognition device according to claim 4 or 5, wherein the recognition unit, when the recognition process using one of the blob candidates fails, attempts to perform the recognition process using the blob candidate with the next highest priority after the one blob candidate.

7. The extraction unit is The largest blob is the one with the largest area in the closed region, A blob with the next largest area of ​​the closed region after the largest blob, and a combined blob formed by combining the largest blob, A component recognition device according to any one of claims 1 to 5, which extracts a component.

8. An extraction unit that processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A selection unit performs a first selection by setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction unit, comparing the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and selecting the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process. A recognition unit that attempts the recognition process using the blob candidates selected by the selection unit, Equipped with, The part recognition device includes a recognition unit which sets a rectangular region circumscribing the blob candidate, estimates the center of the rectangular region as the center of the part, and performs a first recognition process which estimates the inclination angle of the rectangular region as the inclination angle of the part; and a part recognition device which, after performing the first recognition process, detects the outer edge of the blob candidate and performs a second recognition process which estimates the center and inclination angle of the part based on the detected outer edge of the blob candidate.

9. An extraction unit that processes image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A assigning unit assigns a priority order to the plurality of blob candidates extracted by the extraction unit for use in the recognition process, A recognition unit attempts the recognition process in order of the blob candidates with the highest priority assigned by the assignment unit, Equipped with, The part recognition device includes a recognition unit which sets a rectangular region circumscribing the blob candidate, estimates the center of the rectangular region as the center of the part, and performs a first recognition process which estimates the inclination angle of the rectangular region as the inclination angle of the part; and a part recognition device which, after performing the first recognition process, detects the outer edge of the blob candidate and performs a second recognition process which estimates the center and inclination angle of the part based on the detected outer edge of the blob candidate.

10. The part recognition device according to claim 8 or 9, wherein the recognition unit estimates the external dimensions of the part based on the outer edge of the blob candidate, and performs a confirmation process to confirm that the estimated external dimensions of the part are within the dimensional tolerances specified for the part.

11. An extraction step of image processing image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A selection step is performed to set a rectangular region circumscribing at least one of the plurality of blob candidates extracted in the extraction step, compare the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and select the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process. A recognition step in which the recognition process is attempted using the blob candidate selected in the selection step, Equipped with, The selection step involves setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted in the extraction step, comparing the ideal dimensions, which are the design external dimensions of the component, with the external dimensions of the rectangular region, and performing a second selection to select the blob candidate in the rectangular region that is close to the ideal dimensions of the component as the target of the recognition process. The recognition step involves attempting the recognition process using the blob candidate selected by the second selection, The selection step is a component recognition method that performs the first selection when the recognition process by the recognition step fails.

12. An extraction step of extracting a plurality of blob candidates that are closed regions of a part of the component and can be used for recognition processing of the component by image processing image data of a component to be mounted on a substrate, A selection step is performed to set a rectangular region circumscribing at least one of the plurality of blob candidates extracted in the extraction step, compare the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and select the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process. A recognition step in which the recognition process is attempted using the blob candidate selected in the selection step, Equipped with, The selection step involves setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted in the extraction step, comparing the ideal dimensions, which are the design external dimensions of the component, with the external dimensions of the rectangular region, and performing a second selection to select the blob candidate in the rectangular region that is close to the ideal dimensions of the component as the target of the recognition process. The recognition step involves trying both the recognition process using the blob candidate selected by the first selection and the recognition process using the blob candidate selected by the second selection. The selection step is a component recognition method in which, if the first recognition result is better than the second recognition result in which the component was correctly recognized when the blob candidate selected by the first selection is used to try the recognition process and the component was correctly recognized, then the first selection is performed for subsequent recognition processes.

13. An extraction step of image processing image data of a component to be mounted on a substrate to extract a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component, A prioritization step is performed to assign a priority order to the plurality of blob candidates extracted by the extraction step for use in the recognition process. A recognition step in which the recognition process is attempted in order from the blob candidates with the highest priority assigned in the assignment step, Equipped with, The assignment step is a part recognition method which involves setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction step, comparing the ideal dimension, which is the design external dimension of the part, with the external dimension of the rectangular region, and assigning a higher priority to the blob candidate whose rectangular region is closer to the ideal dimension of the part.

14. An extraction step of extracting a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component by image processing image data of a component to be mounted on a substrate, A prioritization step is performed to assign a priority order to the plurality of blob candidates extracted by the extraction step for use in the recognition process. A recognition step in which the recognition process is attempted in order from the blob candidates with the highest priority assigned in the assignment step, Equipped with, The assignment step is a part recognition method which involves setting a rectangular region that circumscribes at least one of the plurality of blob candidates extracted by the extraction step, comparing the ideal angle, which is the tilt angle of the part in an ideal state where the part is correctly positioned, with the tilt angle of the rectangular region, and assigning a higher priority to the blob candidate whose rectangular region is closer to the ideal angle of the part.

15. An extraction step of extracting a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component by image processing image data of a component to be mounted on a substrate, A selection step is performed to set a rectangular region circumscribing at least one of the plurality of blob candidates extracted in the extraction step, compare the ideal angle, which is the tilt angle of the component in an ideal state where the component is correctly positioned, with the tilt angle of the rectangular region, and select the blob candidate in the rectangular region that is close to the ideal angle of the component as the target of the recognition process. A recognition step in which the recognition process is attempted using the blob candidate selected in the selection step, Equipped with, The recognition step is a part recognition method which involves performing a first recognition process to set a rectangular region circumscribing the blob candidate, estimating the center of the rectangular region as the center of the part, and estimating the inclination angle of the rectangular region as the inclination angle of the part, and after performing the first recognition process, detecting the outer edge of the blob candidate and performing a second recognition process to estimate the center and inclination angle of the part based on the detected outer edge of the blob candidate.

16. An extraction step of extracting a plurality of blob candidates that are closed regions of a part of the component and can be used for the recognition process of the component by image processing image data of a component to be mounted on a substrate, A prioritization step is performed to assign a priority order to the plurality of blob candidates extracted by the extraction step for use in the recognition process. A recognition step in which the recognition process is attempted in order from the blob candidates with the highest priority assigned in the assignment step, Equipped with, The recognition step is a part recognition method which involves performing a first recognition process to set a rectangular region circumscribing the blob candidate, estimating the center of the rectangular region as the center of the part, and estimating the inclination angle of the rectangular region as the inclination angle of the part, and after performing the first recognition process, detecting the outer edge of the blob candidate and performing a second recognition process to estimate the center and inclination angle of the part based on the detected outer edge of the blob candidate.

Images