JP5507378B2 - Electronic component mounting equipment - Google Patents

Description

  The present invention relates to an electronic component mounting apparatus for mounting a component supplied from an electronic component supply device on a substrate.

As a conventional electronic component mounting apparatus, for example, there is a technique described in Patent Document 1. In this technology, the same part of the suction component is photographed from two different angles with two cameras, and the three-dimensional image obtained by combining these images is used to determine which lead of the suction component is floating. It is.
However, since this apparatus is configured to direct two cameras to a common point, the degree of freedom of camera arrangement is limited. For example, in the case where the image on the lower surface of the component is particularly important among the three-dimensional images, there are the following problems. That is, if an image of the lower surface of a component is to be obtained with high accuracy, it is inevitably required to reduce the angle formed by both cameras (that is, the angle formed by the optical axes of both cameras). Then, depending on the structure of the camera, the actual arrangement may be difficult due to physical interference between the cameras. Further, if the distance between the imaging position and both cameras is increased to avoid this, not only the apparatus becomes large, but there is a problem that the sharpness of the image is impaired and the component recognition accuracy is lowered.

In other words, in order to obtain an accurate image of the bottom surface of a component with high accuracy, a camera is inevitably required near the component. Therefore, when the camera is placed on the head unit, it does not interfere with the vertical movement of the head. A mechanism for avoiding operation is required. Therefore, problems such as a decrease in accuracy due to the operating mechanism, a complicated control, and an increase in the weight of the head unit corresponding to the operating mechanism occur.
In addition, there is also a technique that uses a line sensor as a means for imaging the suction component and scans the lower side of the suction component to acquire an image of the lower surface of the component (for example, see Patent Documents 2 to 4). Thereby, it is possible to recognize the image of the suction component without interrupting (temporarily stopping) the conveyance of the component.

JP-A-7-151522 Japanese Patent No. 4213292 Japanese Patent No. 4216114 Japanese Patent No. 4361312

However, in the conventional apparatus described in each of the above-mentioned patent documents, the recognition process of the suction component is performed using a three-dimensional image, so that the processing time increases as the processing becomes complicated and the memory used increases. This will lead to performance degradation.
In addition, when a line sensor is used as a means for imaging a suction component, it is necessary to scan the lower part of the component with high speed and high accuracy, and it takes time to acquire only one image. Particularly on a head unit having a plurality of heads, it is extremely difficult to prevent interference between each head operating individually and the line sensor.
Accordingly, an object of the present invention is to provide an electronic component mounting apparatus that can efficiently and accurately perform component recognition.

  In order to solve the above problems, an electronic component mounting apparatus according to claim 1 is an electronic component mounting apparatus that sucks an electronic component by a suction nozzle in a head unit and mounts the electronic component at a predetermined position on a substrate. Illuminating means for irradiating light to the electronic component sucked by the suction nozzle, and provided from the illuminating means in a state corresponding to the lighting means and sucking the electronic component from above by the suction nozzle. By receiving the received light, an imaging means for imaging the imaging surface of the electronic component opposite to the suction surface by the suction nozzle from an obliquely lower side, and an angle around the axis of the suction nozzle by the imaging means A plurality of two-dimensional images obtained by imaging the imaging surface from a plurality of different directions are converted into two-dimensional images obtained by viewing the imaging surface from directly below by geometric conversion, respectively. It is characterized by comprising a recognition means for recognizing the electronic component by comparing a plurality of 2-dimensional images of the.

  In this way, the two-dimensional image obtained by imaging the imaging surface of the electronic component from obliquely below is converted into a two-dimensional image obtained by viewing the imaging surface from directly below by geometric conversion to perform component recognition. Therefore, it is not necessary to dispose the imaging means (camera) directly below or near the electronic component, and even when a plurality of cameras are disposed, physical interference between the cameras can be prevented. Further, even when the camera is fixed to the head unit, it is possible to realize a camera arrangement that does not hinder the vertical movement of the suction head. In addition, since component recognition is performed using a two-dimensional image, it is possible to realize simplification of processing and shortening of processing time compared to the case of using a conventional three-dimensional image.

According to a second aspect of the present invention, in the electronic component mounting apparatus according to the first aspect of the present invention, the imaging unit includes a plurality of cameras, and simultaneously images the imaging surface of the electronic component from the plurality of directions. It is a feature.
In this way, since a plurality of images are acquired simultaneously, the plurality of images are images of the suction component under exactly the same situation. Accordingly, it is possible to eliminate the influence of the suction position shift and suction angle shift of the suction component accompanying the movement of the head unit, and it is possible to suppress the component recognition error.

Furthermore, the electronic component mounting apparatus according to a third aspect is the invention according to the first aspect, wherein the imaging means is constituted by a single camera, and the suction nozzle is rotated around an axis of the suction nozzle. It is characterized by imaging the imaging surface of a component multiple times.
As described above, since a plurality of images are acquired by one camera, the number of parts can be reduced as compared with the case where a plurality of cameras are arranged. Therefore, cost reduction and space saving can be realized.

According to a fourth aspect of the present invention, in the electronic component mounting apparatus according to any one of the first to third aspects, the imaging unit receives reflected light from the light applied to the electronic component from the illumination unit. Thus, the imaging surface is imaged.
As described above, the imaging surface of the electronic component can be appropriately imaged by applying the camera that captures the reflected image as the imaging unit.

Still further, in the electronic component mounting apparatus according to claim 5, in the invention according to claim 4, the illumination unit and the imaging unit are arranged with respect to an axis of the suction nozzle when the electronic component is imaged by the imaging unit. It is characterized by being arranged in a line-symmetrical position.
Accordingly, the illumination unit and the imaging unit can be arranged so that light is irradiated only to the imaging surface among the surfaces (the lower surface and the imaging unit side surface) of the electronic component that can be imaged by the imaging unit. Can be imaged correctly.

Furthermore, in the electronic component mounting apparatus according to a sixth aspect of the present invention, in the invention according to any one of the first to third aspects, the imaging unit receives transmitted light of the light irradiated to the electronic component from the illumination unit. Thus, the imaging surface is imaged.
As described above, the imaging surface of the electronic component can be appropriately imaged by applying the camera that captures the transmission image as the imaging unit.

According to a seventh aspect of the present invention, in the electronic component mounting apparatus according to the sixth aspect of the present invention, the illumination unit and the imaging unit are configured so that the tip position of the suction nozzle is at the time of imaging of the electronic component by the imaging unit. It is characterized by being arranged at a point-symmetrical position.
Accordingly, the illumination unit and the imaging unit can be arranged so that the imaging unit is irradiated with light from the back side of the electronic component, and the imaging surface of the electronic component can be captured as a shadow image.

An electronic component mounting apparatus according to an eighth aspect is the invention according to any one of the first to seventh aspects, wherein the imaging unit and the illumination unit are fixed to the head unit. .
Accordingly, since the camera can be recognized even during the XY operation from when the electronic component is picked up to when it is mounted, there is no restriction on the mounting locus of the head unit, and the component mounting efficiency can be improved. Further, it is possible to suppress the positional deviation between the focal position of the camera and the electronic component sucked by the suction head, and it is possible to image the suction component with high accuracy.

Furthermore, the electronic component mounting apparatus according to claim 9 is the invention according to any one of claims 1 to 7, wherein the imaging means and the illumination means are fixed to a base of the electronic component mounting apparatus. It is characterized by.
Thereby, compared with the case where an imaging means is fixed to a head unit, the increase in the weight of a head unit can be suppressed and the operativity of component mounting can be improved.

  An electronic component mounting apparatus according to a tenth aspect is the invention according to any one of the first to ninth aspects, wherein the imaging means is constituted by a CCD camera, and the CCD of the CCD camera is on the imaging surface of the electronic component. The recognizing means is arranged to incline with respect to the elevation angle correction for correcting distortion of the captured image with respect to the actual shape of the electronic component generated by the inclination of the CCD as the geometric transformation, and the suction nozzle. Rotation correction in which the elevation-corrected image is rotated in the reverse direction around the center position of the image by the difference in imaging angle of the imaging means around the suction nozzle axis from a predetermined specific reference angle around the axis. It is characterized by performing.

  Thus, for example, when a rectangular electronic component is imaged, even if the CCD acquired image is distorted in a trapezoidal shape, this distortion is corrected and the rectangular image obtained when the electronic component is imaged from directly below is obtained. Can be converted. In addition, since images captured from different directions can be unified with images captured from a specific reference angle, comparison between a plurality of images can be easily performed.

  Furthermore, an electronic component mounting apparatus according to an eleventh aspect is the invention according to any one of the first to ninth aspects, wherein the imaging means is constituted by a CCD camera, and the CCD included in the CCD camera is an imaging surface of the electronic component. The recognition means, as the geometric transformation, calculates a difference between an imaging angle of the imaging means around the suction nozzle axis from a predetermined specific reference angle around the suction nozzle axis as the geometric transformation. Only the rotation correction of rotating the image picked up by the image pickup means in the reverse direction around the center position of the image is performed.

As described above, since the CCD is arranged in parallel to the imaging surface (horizontal plane), the CCD acquired image is not distorted unlike the case where the CCD is arranged to be inclined with respect to the imaging surface. Therefore, processing for correcting the distortion is not necessary. In addition, since images captured from different directions can be unified with images captured from a specific reference angle, comparison between a plurality of images can be easily performed.
An electronic component mounting apparatus according to a twelfth aspect of the present invention is the electronic component mounting apparatus according to any one of the first to eleventh aspects, wherein the recognition unit is configured to superimpose a plurality of two-dimensional images after geometric transformation. When the distance between the feature points is outside the allowable range, it is recognized that an abnormality has occurred in the electronic component.

In this way, when images obtained by picking up electronic components from different directions are superimposed, the fact that only the images on a plane with a predetermined height are coincident and the distance between feature points is outside the allowable range is used. , It is determined that a deviation in the height direction has occurred at that portion.
At this time, if a point on the outline of the electronic component (for example, the four corner points of the component) is designated as the feature point, the electronic component is inclined with respect to the horizontal plane when the distance between the feature points is outside the allowable range. It is possible to determine that it is adsorbed. In this way, it is possible to recognize the suction angle deviation of the electronic component.

  Also, when the suction component is a lead terminal component, if the point of the outer edge of the lead terminal (for example, the center of the tip) is specified as the feature point, the lead between the feature points is outside the allowable range. It can be determined that the terminal is deformed. Further, at this time, when the deviation between the feature points occurs only in some of the feature points, it can be determined that the coplanarity error is generated in the lead terminal corresponding to the feature points. . On the other hand, when the amount of deviation between feature points gradually increases or decreases in a predetermined side, the tip surface of the lead terminal located on that side is inclined with respect to the tip surface of the lead terminal on the other side. It can be determined that this is a collinearity error.

Similarly, if the suction component is a ball terminal component, specify the center point of the ball terminal as a feature point, and if the suction component is a land terminal component, specify the center point or corner of the land terminal as a feature point, Ball terminal errors and land terminal errors can be recognized.
The electronic component mounting apparatus according to claim 13 is the electronic component mounting apparatus according to any one of claims 1 to 11, wherein the recognizing unit is configured to superimpose a plurality of two-dimensional images after geometric transformation. When the non-overlapping pixel area out of the pixel areas forming the image is outside the allowable range, it is recognized that an abnormality has occurred in the electronic component.

  In this way, when images obtained by picking up electronic components from different directions are overlapped, the fact that only the images of planar objects of a predetermined height match each other is used, and the pixel areas of non-overlapping component images are outside the allowable range. In some cases, it is determined that a deviation in the height direction has occurred at that portion. Thereby, similarly to the comparison between the above feature points, it is possible to recognize an electronic component suction angle shift and a terminal error.

According to a fourteenth aspect of the present invention, in the electronic component mounting apparatus according to the twelfth or thirteenth aspect of the present invention, the recognizing unit may detect a deviation in the suction angle of the electronic component from a horizontal plane and the electronic component as an abnormality of the electronic component. It is characterized by recognizing at least one of the deformation of the terminal.
Thus, by recognizing the deviation of the electronic component suction angle and the deformation of the terminals of the electronic component, it is possible to appropriately determine the electronic component that cannot be normally mounted on the circuit board. Therefore, it is possible to select whether or not to mount a component according to the component recognition result.

  According to the present invention, in order to perform component recognition by correcting a two-dimensional image obtained by capturing an electronic component from diagonally below to a two-dimensional image captured from directly below, physical interference between cameras that capture the electronic component, The camera can be arranged at a position where physical interference with the suction head is prevented, and safety and work efficiency of component mounting can be improved. In addition, since component recognition is performed using a two-dimensional image, it is possible to simplify processing and shorten processing time.

It is a top view which shows the electronic component mounting apparatus in this invention. It is a figure which shows the structure of the principal part of a head unit. It is a top view which shows arrangement | positioning of a CCD camera and illumination. It is a block diagram which shows the structure of the control system of an electronic component mounting apparatus. It is explanatory drawing of the terminal error of an electronic component. It is a flowchart which shows the components recognition process procedure performed with a main controller. It is a figure which shows arrangement | positioning of the illumination at the time of imaging, an electronic component, and a CCD camera. It is a figure explaining the influence by the difference in arrangement | positioning of CCD. It is a figure which shows the difference in the CCD acquisition image by the difference in arrangement | positioning of CCD. It is a figure for demonstrating the horizontal direction component of CCD. It is a figure which shows the image on CCD. It is a figure for demonstrating the vertical direction component of CCD. It is a figure which shows the trapezoid correction | amendment of a CCD acquisition image. It is a figure explaining lead terminal comparison. It is a figure which shows an example of the comparison result of a lead terminal. It is a figure which shows the allowable distance range in a lead terminal comparison. It is a figure explaining ball terminal comparison. It is a figure which shows the allowable distance range in a ball terminal comparison. It is a figure explaining land terminal comparison. It is a figure explaining the external comparison of components. It is a figure which shows an example of the external shape comparison result of components. It is a figure for demonstrating the operation | movement of this embodiment. It is a figure for demonstrating the operation | movement (rotation correction) of this embodiment. It is a figure which shows the modification of a CCD camera. It is a figure which shows the modification of the imaging method of an electronic component. It is a figure for demonstrating the eccentric amount of a head. It is a figure which shows the error of the CCD acquisition image by the eccentric amount of a head. It is a figure which shows the modification of an imaging means. It is a figure for demonstrating the image correction using the components for calibration. It is a figure for demonstrating the influence by component height difference. It is a figure which shows the modification of arrangement | positioning of a CCD camera and illumination.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(First embodiment)
(Constitution)
FIG. 1 is a plan view showing an electronic component mounting apparatus according to the present invention.
In the figure, reference numeral 1 denotes an electronic component mounting apparatus. The electronic component mounting apparatus 1 includes a pair of transport rails 11 extending in the X direction on the upper surface of the base 10. The transport rail 11 supports both sides of the circuit board 5 and is driven by a transport motor (not shown) to transport the circuit board 5 in the X direction.

The electronic component mounting apparatus 1 includes a head unit 12. The head unit 12 includes a plurality of suction nozzles that suck electronic components at the bottom, and is configured to be horizontally movable on the base 10 in the XY directions by an X-axis gantry 13 and a Y-axis gantry 14.
In this electronic component mounting apparatus 1, electronic component supply devices 15 that supply electronic components by a tape feeder or the like are mounted on both sides of the transport rail 11 in the Y direction. The electronic component supplied from the electronic component supply device 15 is vacuum-sucked by the suction nozzle of the head unit 12 and mounted on the circuit board 5.

The head unit 12 is also provided with a component recognition mechanism 20 for recognizing an electronic component sucked by the suction nozzle.
Furthermore, the electronic component mounting apparatus 1 is provided with a nozzle changer 16 for exchanging the suction nozzle according to the size and shape of the component to be sucked. A plurality of types of nozzles are stored and managed in the nozzle changer 16.

Next, the configuration of the main part of the head unit 12 will be described with reference to FIG.
The head unit 12 is movable in the X direction by attaching its base to the X-axis gantry 13 shown in FIG.
The head unit 12 includes a plurality of suction heads 12a (only one is shown in FIG. 2), and a suction nozzle 12b that sucks and holds the electronic component 3 is provided at the tip (lower end) of each suction head 12a. . Each suction nozzle 12b is configured to be rotatable about a nozzle axis and to be movable up and down in the Z direction (height direction).

  The component recognition mechanism 20 is a pair of support columns 21a and 21b integrally formed with the head unit 12, and the suction surface of the electronic component 3 attached to the tip end (lower end) of the support column 21a and sucked by the suction nozzle 12b. A CCD camera 22 that images the opposite imaging surface from obliquely below, and an illumination 23 that illuminates the lower surface of the electronic component 3 that is attached to the tip (lower end) of the column 21b and is adsorbed by the adsorption nozzle 12b. And comprising.

  The columns 21a and 21b have a shape that does not hinder the vertical movement of the suction head 12a. The CCD camera 22 has a predetermined focal position (part recognition position) on the vertical line extending downward from the center of the tip hole of the suction nozzle 12b and above the lower ends of the columns 21a and 21b. ) Are arranged at a predetermined elevation angle. Then, after the component is picked up, the CCD camera 22 picks up the image of the lower surface of the electronic component 3 from the lower side while the suction head 12a is raised until the lower surface of the electronic component 3 reaches the component recognition position.

  At this time, the illumination 23 irradiates the electronic component 3 obliquely from below with respect to the CCD camera 22 from a line-symmetric position around the axis of the suction head 12a. Reference numeral 24 denotes a path of light emitted from the illumination 23. The CCD camera 22 is arranged so that the lower surface of the electronic component 3 and the side surface on the CCD camera 22 side can be imaged, but the light emitted from the illumination 23 hits only the lower surface of the electronic component 3 and the side surface on the illumination 23 side. That is, the illumination 23 irradiates light only to the lower surface of the electronic component 3 among the surfaces of the electronic component 3 that can be imaged by the CCD camera 22, and the CCD camera 22 reflects off the lower surface of the electronic component 3. The lower surface of the electronic component 3 is imaged by receiving the reflected light.

When mounting the components, the suction head 12 a is lowered below the lower ends of the columns 21 a and 21 b to mount the electronic component 3 on the circuit board 5.
FIG. 3 is a plan view showing the arrangement of the CCD camera 22 and the illumination 23.
As shown in FIG. 3, the electronic component 3 sucked by one suction nozzle 12 b is imaged by two CCD cameras 22 while being irradiated with two illuminations 23. That is, the electronic component 3 is simultaneously imaged from two directions with different angles about the nozzle axis by the two CCD cameras 22. Here, it is assumed that the angles around the nozzle axes of the two CCD cameras 22 differ by, for example, 90 °.

FIG. 4 is a block diagram showing the configuration of the control system of the electronic component mounting apparatus 1.
Image data captured by the CCD camera 22 is input to the main controller 50 that controls the entire electronic component mounting apparatus 1.
The main control device 50 includes an image calculation device 51, an image recognition device 52, and a controller 53 including a microcomputer including a CPU, a RAM, a ROM, and the like. Each drive source for moving the head unit 12 is connected to the controller 53, and each drive source of the head unit 12 is controlled by the controller 53.

The image calculation device 51 inputs image data captured by the CCD camera 22, performs image processing to be described later, and generates a two-dimensional image obtained by capturing the electronic component 3 from obliquely below, with a specific reference angle (around a preset nozzle axis). The electronic component 3 is converted into a two-dimensional image viewed from directly below (elevation angle correction, rotation correction). The image conversion will be described later in detail.
The image recognition device 52 recognizes the electronic component 3 sucked by the suction nozzle 12b using the image data converted by the image calculation device 51. Here, the size of the electronic component 3 by the suction nozzle 12b, the suction position shift, the suction angle shift, the terminal error of the electronic component 3, and the like are recognized. Here, the suction position shift is an error between the center position of the electronic component 3 and the suction position by the suction nozzle 12b, and the suction angle shift is an inclination of the lower surface of the electronic component 3 with respect to the horizontal plane.

FIG. 5 is an explanatory diagram of a terminal error of an electronic component. FIG. 5 shows a case where the electronic component 3 is a lead terminal component (QFP or the like).
In a normal state in which no error has occurred in the lead terminal 3a, as shown in FIG. 5A, the tips of all the lead terminals 3a protruding from the four sides are located on the same plane (lead tip plane). However, as shown in FIG. 5B, when a specific lead terminal 3a is deformed upward or downward (hereinafter referred to as a coplanarity error), or as shown in FIG. In some cases, the leading end surface of the lead terminal 3a in the second row is inclined with respect to the other row (hereinafter referred to as a collinearity error). When the terminal errors shown in FIGS. 5B and 5C occur, there are terminals that cannot contact the substrate surface when the electronic component 3 is mounted on the substrate. Therefore, in this embodiment, a coplanarity error and a collinearity error are detected as the terminal error by component recognition.

The controller 53 performs component mounting processing based on the recognition result of the electronic component 3 by the image recognition device 52. The controller 53 stores the type of electronic component 3, the external dimensions, the number of terminals, the lead length (in the case of lead terminals), the pitch between terminals, and the like.
FIG. 6 is a flowchart showing a component recognition processing procedure executed by the main controller 50. This component recognition process is started every time the electronic component 3 is sucked by the suction nozzle 12b, and is performed after the electronic component 3 is picked up and mounted on the circuit board 5.

First, in step S1, the main controller 50 reads the height of the electronic component 3 currently sucked from the component data stored in the memory. Then, as shown in FIG. 7, the suction nozzle 12 b is moved up and down so that the lower surface of the suction component matches the component recognition position that is the focal position of the CD camera 22.
In the present embodiment, the position where the focal position of the CCD camera 22 appears on the CCD is called the CCD center position. A straight line connecting the focal position of the CCD camera 22 and the CCD center position is called a recognition center line.
In step S <b> 2, main controller 50 acquires image data captured by CCD camera 22. Here, as described above, two pieces of image data captured from an angle different by 90 ° around the nozzle axis are acquired.

  Next, in step S3, main controller 50 performs elevation angle correction on the two image data acquired in step S2. Here, the elevation angle correction is a process for correcting the distortion of the CCD acquired image with respect to the actual shape of the electronic component 3 caused by the elevation angle (tilt with respect to the horizontal plane) of the CCD. First, a difference between a component image captured by a CCD camera arranged directly below the component and a component image captured by a CCD camera disposed obliquely below will be described.

  As shown in FIG. 8A, when the CCD is arranged directly below the part and parallel to the lower surface of the part, when the part is a rectangular parallelepiped, the part image is as shown in FIG. The horizontal direction (left-right direction in FIG. 9) and the vertical direction (up-down direction in FIG. 9) are rectangles that are reduced at an equal ratio. On the other hand, as shown in FIG. 8B, when the CCD is arranged obliquely below the component and inclined with respect to the lower surface of the component, the horizontal direction is equal as shown in FIG. 9B. The component image is reduced to a trapezoidal shape by being reduced and deformed and reduced in the vertical direction. This is because, as shown in FIG. 8, when a line segment obtained by dividing the lower surface of the component into three equal parts is acquired by the CCD directly below, it is arranged at equal intervals, whereas it is not equal on the inclined CCD. In the elevation correction, a process of converting the trapezoidal image shown in FIG. 9B into a rectangular image shown in FIG. 9A is performed, which is also referred to as trapezoidal correction in the following description.

FIG. 10 is a diagram for explaining the horizontal component of the CCD, and is a diagram when the electronic component 3 and the CCD are viewed from above. In FIG. 10, the vertical direction corresponds to the horizontal direction of the CCD.
When attention is paid to one horizontal row in the image taken on the inclined CCD, the correspondence between the triangle formed by connecting the two points on the image (line) and the lens center point, and the electronic component 3 as the subject. The triangle formed by connecting the two points on the line to the lens center point is a similar shape.

The length ratio of each side of the triangle is the ratio between the distance between the lower surface of the component and the lens and the distance between the lens and the CCD. That is, when viewed only in the horizontal direction, an image with an equal ratio reduction is drawn on the CCD, and the size is determined by the positional relationship between the cross-sectional image on the line on the lower surface of the component and the camera.
By the way, in FIG. 8, when the lower surface of the component is extended infinitely in the left direction, the image reaches α, and the position of α becomes a position indicating infinity. Assuming that there are infinitely long parts in the horizontal direction, the image on the CCD is as shown in FIG. Thus, the image on the CCD is deformed and reduced in the vertical direction of the CCD (left and right direction in FIG. 11). In step S3 of FIG. 6, the keystone correction is performed by correcting the distortion of the horizontal component and the distortion of the vertical component of the CCD acquired image.

FIG. 12 is a view for explaining the vertical component of the CCD, and is a view when the electronic component 3 and the CCD are viewed from the side.
In FIG. 12, the center of the lower end of the CCD is Pccd1, and an arbitrary point in the CCD tilt direction is Pccd2. Further, the distance between the point Pccd1 and the lens center point in the Y direction is Ly, and the distance in the X direction is Lx. Furthermore, the actual CCD inclination with respect to a line parallel to the component lower surface (usually a horizontal line since the component lower surface is horizontal) is assumed to be θccd.

  As described above, when the CCD is arranged in parallel with the lower surface of the component, the image on the CCD is an equal ratio reduced image of the lower surface of the component. Therefore, it is assumed that the CCD arranged in parallel with the lower surface of the component is on a plane passing through the point Pccd1, and this is called a virtual CCD. On the virtual CCD, the points corresponding to the points Pccd1 and Pccd2 are P1 and P2, respectively. Note that the point Pccd1 and the point P1 are the same point. Then, a ratio between the distance between the points Pccd1 and Pccd2 and the distance between the points P1 and P2 is calculated, and the trapezoid correction is performed using the obtained ratio.

In the following description, a distance between a certain point A and a certain point B is denoted as L {AB}.
When calculating the distance L {P2-P1}, a point P2, a point Pccd2, and a triangle connecting the point Pccd2 'where the perpendicular line dropped from the point Pccd2 to the virtual CCD and the virtual CCD intersect, the point P2, the lens center point, and the lens It is utilized that the perpendicular line dropped from the center point to the virtual CCD and the triangle connecting the intersection of the virtual CCD are similar. The ratio of each side of these triangles is the same as the ratio of the distance L {Pccd2-Pccd2 '} to the distance Ly.
The ratio of the distance L {P2-Pccd2 '}, L {Pccd2'-P1}, the distance Lx, and the distance Ly is the distance L {P2-Pccd2'} and the distance {Pccd2'-Pccd2 } And the ratio.

Therefore, the distance L {P2-Pccd2 '} is expressed by the following equation.
L {P2-Pccd2 '} = L {Pccd2'-Pccd2}. (L {Pccd2'-P1} + Lx) / (Ly-L {Pccd2'-Pccd2})
= (L {Pccd2-Pccd1} .Sin (.theta.ccd). (L {Pccd2-Pccd1} .Cos (.theta.ccd) + Lx)) / (Ly-L {Pccd2-Pccd1} .Sin (.theta.ccd)) (1) )

Therefore, L {P2-P1} is expressed by the following equation.
L {P2-P1} = L {P2-Pccd2 '} + L {Pccd2'-P1}
= L {Pccd2-Pccd1} .Sin (.theta.ccd). (L {Pccd2-Pccd1} .Cos (.theta.ccd) + Lx)) / (Ly-L {Pccd2-Pccd1} .Sin (.theta.ccd)) + L {Pccd2-Pccd1}・ Cos (θccd) (2)

  As described above, the horizontal component of the CCD acquired image is reduced at an equal ratio, but the reduction ratio changes depending on the position of the CCD in the vertical direction. The ratio of the image size on the actual CCD to the image on the virtual CCD is equal to the ratio of the distance between the lens center point and the point Pccd2 to the distance between the lens center point and the point P2 in FIG. This is also equal to the ratio of the distance L {Pccd2, Pccd2 '} to the distance Ly from the two similar triangles.

Therefore, the ratio rx of the horizontal component of the CCD acquired image is expressed by the following equation.
rx = L {Pccd2'-Pccd2} / Ly
= L {Pccd2−Pccd1} · Sin (θccd) / Ly (3)
On the other hand, the ratio ry of the vertical component of the CCD acquired image is equal to the ratio of the distance L {Pccd2-Pccd1} to the distance L {P2-P1}. Therefore, the ratio ry of the vertical component of the CCD acquired image is expressed by the following equation.
ry = L {P2-P1} / L {Pccd2-Pccd1} (4)

Here, if Lx is Lx0, Ly is Ly0, Sin (θccd) is S0, Cos (θccd) is C0, and L {Pccd2-Pccd1} is ΔY, then rx and ry are the above (2) and (3), respectively. ) Is expressed as follows based on the formula.
rx = ΔY · S0 / Ly0 (5)
ry = (ΔY · S0 · (ΔY · C0 + Lx0) / (Ly0−ΔY · S0) + ΔY · C0) / ΔY (6)
Here, the distances Lx, Ly and the angle θccd are obtained in advance by calculation or measurement. Thereby, ratio rx and ry can be calculated | required by calculation.

Next, each pixel data is arranged on a RAM area that is regarded as a virtual CCD image storage area, and the pixel data on the actual CCD coordinates obtained by dividing by rx, ry with respect to each coordinate of the area, The RAM area is sequentially captured as each pixel.
Virtual CCD pixel data (x, y) = actual CCD pixel data (x / rx, y / ry) (7)

  As a result, the acquired image on the actual CCD shown in FIG. 13A is developed on the RAM as shown in FIG. 13B, and the keystone correction is performed. In the present embodiment, the image subjected to the trapezoidal correction is handled as an image obtained by imaging the lower surface of the component from directly below. Note that the format of this RAM area is the same as that used when an actual CCD acquired image is expanded on the RAM, and allows existing image processing to be performed.

Next, in step S4 of FIG. 6, the main controller 50 performs rotation correction for rotating the two images having different imaging angles corrected in step S3 to specific reference angles, and proceeds to step S5. Here, the corrected image is rotated in the reverse direction around the center position (CCD center position) of the image by the difference in the imaging angle from the specific reference angle. That is, for example, when the imaging angle is shifted 45 ° counterclockwise with respect to the specific reference angle, the corrected image is rotated clockwise around the CCD center position.
In step S5, main controller 50 calculates the actual dimensions of each part in the final lower surface of the component (enlarges the ratio). Here, the actual size is calculated by multiplying the size of the virtual image by the ratio of LLy to Ly in FIG. 8, that is, (LLy / Ly).

Next, in step S6, main controller 50 performs image recognition processing using the image acquired in step S5. Here, since the electronic component 3 is simultaneously photographed by two cameras, image recognition processing is performed using one of the two images. As the image recognition processing, processing for recognizing the component adsorption position shift, component size, and the like is performed. Here, the displacement of the suction position is detected by comparing the image center position of the component in the image acquired in step S5 with the CCD center position.
In step S7, main controller 50 superimposes the two images acquired in step S5. At this time, the CCD center positions are overlapped in each image.

Next, in step S8, the main controller 50 reads the component data stored in the memory and acquires the type of picked-up suction component (chip component, lead terminal component, ball terminal component, land terminal component, etc.). The process proceeds to step S9.
In step S9, the main controller 50 determines whether or not the picked-up suction component is a lead terminal component. If the picked-up component is a lead terminal component, the main control device 50 proceeds to step S10. The process proceeds to step S11.

  In step S10, main controller 50 detects the deformation of the lead terminal and the amount of deformation from the two images superimposed in step S8, and proceeds to step S15 described later. Here, for example, edge images in two images are extracted by extracting edge points in the image by using generally known image processing, and the outer edge portion of the lead terminal is generated from the generated edge images. To figure out. Next, as shown in FIG. 14, the positions of the outer edge portions of the same lead terminal in the two images are compared. Based on the comparison result, a lead terminal error is detected.

FIG. 15 is a diagram illustrating an example of a comparison result of lead terminals.
In a normal state in which no terminal error has occurred in the lead terminals, as shown in FIG. 15A, the positions of all the terminals of both images are the same coordinates, and no deviation of the outer edge portion occurs even when the two images are superimposed. . On the other hand, when a coplanarity error has occurred, as shown in FIG. 15B, there are terminals that do not match in both images. Further, when a collinearity error has occurred, as shown in FIG. 15C, there is a row of terminals that do not match in both images, and the displacement amount of the outer edge portion of the terminals between the images changes in position. Gradually increase or decrease.

  In the present embodiment, as the characteristic points of the lead terminals, for example, the positions of the center points of the lead tips are compared. At this time, it is determined whether or not the distance between the feature points is within a preset allowable distance range. Here, as shown in FIG. 16, in one image (solid line), an ellipse centered on the center point of the tip of the lead is set, and this is set as an allowable distance range. Then, it is determined whether or not the center point of the leading end of the lead in the other image (broken line) is within this allowable distance range, and when it is within the allowable distance range as shown in FIG. Is determined not to be a terminal error, and if it is outside the allowable distance range as shown in FIG. 16B, it is determined to be a terminal error.

As a lead terminal comparison, a method of determining whether or not there is a pixel area where the lead terminal images do not overlap when two images are superimposed can be used. In this case, when a pixel area that does not overlap is outside the allowable range, it is determined that a terminal error has occurred.
In step S11, the main controller 50 determines whether or not the picked-up suction component is a ball terminal component. If the picked-up suction component is a ball terminal component, the main control device 50 proceeds to step S12. The process proceeds to step S13.

  In step S12, the main controller 50 determines the presence / absence and amount of deviation of the ball terminal from the two images superimposed in step S8, and proceeds to step S15 described later. Again, the two edge images are compared in the same manner as the lead terminal comparison described above. For example, as shown in FIG. 17, the center point of the ball terminal is designated as a feature point, and the positions of the centers of the same ball terminal in two images are compared. Based on the comparison result, a ball terminal error is detected.

Also in this case, as shown in FIG. 18, an ellipse centered on the center point of the ball terminal is set in one image (solid line), and this is set as the allowable distance range. Then, it is determined whether or not the center point of the ball terminal in the other image (broken line) is within this allowable distance range, and if it is within the allowable distance range as shown in FIG. It is determined that there is no error, and if it is outside the allowable distance range as shown in FIG.
In this case as well, a ball terminal error can be detected by detecting a pixel area where the ball terminal image does not overlap.

In step S13, the main controller 50 determines whether or not the picked-up suction component is a land terminal component. If the picked-up suction component is a land terminal component, the main control device 50 proceeds to step S14. The process proceeds to step S15.
In step S14, the main controller 50 determines whether or not the land terminal is shifted and the amount of shift from the two images superimposed in step S8, and proceeds to step S15 described later. Again, the two edge images are compared in the same manner as the terminal comparison described above. For example, as shown in FIG. 19, the center point (or corner) of the land terminal is designated as the feature point, and the positions of the centers (or corners) of the same land terminal are compared. Then, a land terminal error is detected based on the comparison result. Note that a land terminal error may be detected by detecting a pixel area where the land terminal images do not overlap.

  In step S15, main controller 50 compares the external shapes of the components from the two images superimposed in step S8, and proceeds to step S16. Here, as shown in FIG. 20, pixels having the same coordinates in the two images are compared. Then, based on the comparison result, the deviation of the outer shape is detected. For example, the coordinates of the pixels corresponding to the four corners of one component image are obtained, and the pixels having the coordinates are compared in both images. As a result, when the pixel data do not match, it is determined that the component outlines do not match.

FIG. 21 is a diagram illustrating an example of the external shape comparison result of components.
When the electronic component 3 is correctly sucked by the suction nozzle and the lower surface of the component is in a normal state, it is assumed that two CCD acquired images taken from different directions have the shape shown in FIG. On the other hand, when the electronic component 3 is inclined in the height direction, for example, as shown by the solid line in FIG. 21B, the two CCD acquired images have different shapes as compared with the normal time indicated by the broken line. Then, the images after both of the trapezoidal corrections at this time also have different shapes as compared with the normal time indicated by the broken lines, as indicated by the solid line in FIG.

  Therefore, when the electronic component 3 is tilted in the height direction, even if two corrected images are superimposed, their shapes do not match. Therefore, it is possible to detect an inclination in the height direction of the electronic component 3 (component adsorption angle deviation) by detecting the deviation of the component outer shape. In the present embodiment, when the amount of deviation of the component outer shape is outside the allowable range, it is determined that the component adsorption angle deviation has occurred.

In step S16, main controller 50 determines component mounting according to the comparison result of the two corrected images, and ends the component recognition process. Specifically, in each of the above-described processes, if the occurrence of the suction position deviation or suction angle deviation of the electronic component 3 or the occurrence of a terminal error of the electronic component 3 is recognized, the mounting of the component on the board is stopped. .
The CCD camera 22 corresponds to the imaging means, the illumination 23 corresponds to the illumination means, and the processing in FIG. 6 corresponds to the recognition means.

(Operation)
Next, the operation of this embodiment will be described.
When the electronic component mounting process is started, the controller 53 of the main controller 50 reads the component data stored in the memory, drives and controls the head unit drive source, and moves the head unit 12 onto the nozzle changer 16. Then, the designated nozzle is attached to each head 12 a of the head unit 12 from the nozzle changer 16. Next, the controller 53 drives and controls the head unit drive source, moves the head unit 12 to the component supply position of the electronic component supply device 15, and sucks the electronic component 3.

  When the electronic component 3 is picked up, the main controller 50 reads the height of the electronic component 3 currently picked up from the component data stored in the memory. Then, the suction head 12a is raised until the lower surface of the sucked electronic component 3 reaches a predetermined component recognition position (step S1 in FIG. 6). In this state, the lower surface of the electronic component 3 is irradiated with light from the two illuminations 23, and the two CCD cameras 22 simultaneously image the lower surface of the electronic component 3 from obliquely below.

  The two CCD acquired images captured by the CCD camera 22 are input to the image calculation device 51 (step S2). When the image of the lower surface of the component captured obliquely from the lower side is captured by the image calculation device 51 from directly below. A trapezoidal correction calculation is performed so as to obtain the same shape (step S3). That is, if two CCD acquired images taken obliquely from below are shown in FIG. 22A, the image shown in FIG. 22B can be obtained by trapezoidal correction.

  Next, the keystone correction image is rotationally corrected to a specific reference angle (step S4). As shown in FIG. 23, it is assumed that the imaging angles of the electronic component 3 by the two CCD cameras 22a and 22b are deviated from each other by 45 ° with respect to the specific reference angle. In this case, the trapezoidally corrected image of the first CCD acquired image captured by the CCD camera 22a is rotated 45 ° counterclockwise around the CCD center position to obtain an acquisition posture at a specific reference angle. Rotation correction is performed as follows. On the other hand, the trapezoidally corrected image of the second CCD acquired image captured by the CCD camera 22b is rotated 45 ° clockwise around the center position of the CCD so that the acquired posture is obtained at a specific reference angle. Rotation correction is performed. The two images after the rotation correction are as shown in FIG.

  The specific reference angle is not limited to the angle shown in FIG. For example, the imaging angle of the electronic component 3 by one CCD camera 22a can be set as the specific reference angle. In this case, it is not necessary to perform rotation correction on the image after the trapezoid correction of the first CCD acquired image captured by the CCD camera 22a, and after the keystone correction of the second CCD acquired image captured by the CCD camera 22b. Only the image may be rotated 90 ° clockwise around the CCD center position.

The image after the rotation correction is multiplied by (LLy / Ly) so as to be enlarged to an actual ratio (step S5). The image obtained in this way is sent to the image recognition device 52 and is handled as an image obtained by imaging the lower surface of the component from directly below.
The image recognition device 52 uses one of the two corrected images, recognizes the shape of the electronic component 3 in the image by image processing such as edge detection, and recognizes the center position of the recognized electronic component 3. And the amount of deviation between the CCD center position is recognized. The amount of deviation of the suction position recognized here is added to the mounting coordinates of the electronic component 3 when the component is mounted on the circuit board 5 later. Thereby, the electronic component 3 can be correctly mounted at a desired position on the circuit board 5.

  Next, the two correction images are overlapped to recognize a suction angle shift or a terminal error of the electronic component 3. First, the type of the electronic component 3 currently picked up is read from the component data stored in the memory (step S8). Here, if the electronic component 3 currently attracted is a lead terminal component as shown in FIG. 22 (Yes in step S9), the lead terminal comparison is performed based on the two superimposed correction images. Then, it is determined whether or not a coplanarity error and a coreity error have occurred in the lead terminal (step S10).

When a coplanarity error has occurred at the position E1 in FIG. 22C, when two corrected images are superimposed, a part of the outer edge portion of the lead terminal is shifted as shown in FIG. 15B. . In addition, when a reality error has occurred at the position E2 in FIG. 22 (c), when the two images are superimposed, the outer edge of the lead terminal is gradually changed as shown in FIG. 15 (c). Shift back and forth or left and right. In this way, by detecting the deviation of the outer edge portion of the lead terminal, it is possible to recognize whether or not a terminal error has occurred, and to specify the type of error when a terminal error has occurred. .
Even when the electronic component 3 that is currently attracted is a ball terminal component or a land terminal component, as in the case of the lead terminal component, the terminal feature point (in the case of the ball terminal component) between the two corrected images Indicates the presence or absence of a terminal error by comparing the center point of the terminal and the center or corner of the terminal in the case of a land terminal component.

  In addition, when the electronic component 3 currently sucked is a chip component (No in steps S9, S11, and S13), the deviation of the outer shape of the chip component is recognized using two superimposed correction images. When the chip part is sucked and sucked in the height direction, as shown in FIG. 21C, each correction image is different from the normal image, and the two correction images are superimposed. However, the outer shape of the chip parts does not match. Thus, by detecting the deviation of the outer shape of the chip component, it is possible to recognize the adsorption angle deviation of the chip component.

As described above, when the displacement angle of the electronic component 3 or a terminal error is detected by comparing the two correction images, the mounting of the electronic component 3 currently attracted to the circuit board 5 is stopped. .
On the other hand, if it is determined that the suction angle shift or the terminal error of the electronic component 3 is not detected by comparing the two corrected images and the currently sucked electronic component 3 is normal, the controller 53 12 is moved to a predetermined component mounting position on the circuit board 5. The electronic component 3 is mounted on the circuit board 5 by lowering the suction nozzle 12b at this component mounting position.

  As described above, in the present embodiment, two images obtained by imaging the electronic component 3 from obliquely below in two directions having different angles around the nozzle axis are captured from directly below the component in the specific reference angle direction by geometric transformation. Convert as a mapping to a dimensional image. Specifically, when the height of the suction head 12b is adjusted so that the center of the lower surface of the component is positioned at the focal position of the two CCD cameras 22, the positional relationship of the lower surface of the component, the CCD camera 22 and the CCD in the same. The trapezoidally deformed part lower surface image projected onto the CCD is geometrically converted from the distance, tilt angle, azimuth angle, etc. between them into the original rectangular image. Then, the two two-dimensional images that have undergone mapping conversion are compared, and the positional deviation between the feature points of the component image is detected to recognize the height deviation of the electronic component 3 and its terminals.

Therefore, an image obtained by imaging the lower surface of the component from directly below can be acquired without placing a camera directly below the electronic component 3. Therefore, it is possible to increase the degree of freedom of camera arrangement. In addition, since there is little deterioration of the image on the lower surface of the component due to mapping conversion (decrease in resolution in the tilt direction is about 1 / √2≈0.7 times even if taken at a tilt angle of 45 degrees), accurate image recognition is performed. be able to.
Furthermore, the image of the bottom surface of the component, which is the most important for recognizing coplanarity errors, is that when there is no coplanarity error, the images of the bottom surface of the component overlap accurately. For example, interpolation between two acquired images is the value of each pixel. It is also possible to average only, and more accurate image recognition is also possible.

(effect)
As described above, in the above embodiment, the CCD camera is arranged directly under or near the electronic component in order to perform the component recognition by correcting the two-dimensional image obtained by capturing the electronic component from diagonally below to the two-dimensional image captured from directly below. There is no need to do. Therefore, the CCD camera can be arranged at a position where interference between the CCD cameras and interference between the CCD camera and the suction head can be prevented. By preventing interference between the CCD camera and the suction head, safety and work efficiency can be improved.

Moreover, since a CCD camera is used as a means for imaging electronic components, a captured image can be acquired in an instant. That is, since it is not necessary to scan and acquire a component bottom surface image unlike a line sensor, it is fast and does not require a scanning mechanism, thereby suppressing control complexity and device enlargement. Can do.
Furthermore, since the CCD camera is mounted on the head unit, component recognition can be performed while moving the head unit from the component supply position of the electronic component supply device to the component mounting position on the circuit board. Therefore, there is no restriction on the mounting locus of the head unit, and the component mounting efficiency can be improved. Further, since the positional deviation between the CCD camera and the suction head tip can be reduced, component recognition can be performed with high accuracy.

In addition, since the imaging surface of the suction component is simultaneously imaged from two different directions by the two CCD cameras, the two images are images of the suction component under exactly the same conditions. Accordingly, it is possible to eliminate the influence of the suction position shift and suction angle shift of the suction component accompanying the movement of the head unit, and it is possible to suppress the component recognition error.
Furthermore, since component recognition is performed using a two-dimensional image, the processing can be simplified and the processing time can be shortened compared to the case where a three-dimensional image is used as in the conventional apparatus.
In addition, since various types of recognition (coplanarity error, collinearity error, adsorption position deviation, adsorption angle deviation, component size, etc.) can be acquired all at once, there is no need to provide multiple units and the configuration is relatively inexpensive. Can do.

(Modification)
In the above embodiment, as shown in FIG. 24, a CCD camera in which the CCD is arranged in parallel with the lower surface of the component can be used. As a result, since the CCD acquired image is not distorted as in the case where the CCD is inclined with respect to the lower surface of the component, the above-mentioned trapezoidal correction can be omitted, and high-speed calculation is possible. However, since the CCD is inclined with respect to the lens (or the lens itself is also inclined), the usable range of the CCD is limited due to the depth of focus, lens aberration, and the like, and it is difficult to obtain a large image. Therefore, in applications where only a small image is used, it is possible to obtain the same effect as in the above embodiment, which is useful.

  In the above-described embodiment, the case where the electronic component 3 is imaged from different directions using the two CCD cameras 22 has been described. However, the electronic component 3 may be imaged from different directions using the single CCD camera 22. it can. In this case, as shown in FIG. 25, one set of CCD camera 22 and illumination 23 are arranged to face each suction head 12a. Then, after the electronic component 3 is imaged by the CCD camera 22 at the first rotation angle (for example, 0 °), the suction head 12a is rotated around the nozzle axis, and again at the second rotation angle (for example, 90 °). The electronic component 3 is imaged by the CCD camera 22. In this way, the electronic component 3 is imaged from two different directions.

  However, in this case, as shown in FIG. 26, when the center of rotation of the head and the center of the tip of the head are different, an error due to the difference between the amount of eccentricity of the head and its direction due to component rotation occurs. That is, an error occurs when the CCD center positions (centers of the head tips) are overlapped after performing trapezoidal correction and rotation correction on the two CCD acquired images shown in FIGS. Will occur. Therefore, the eccentric amount and direction of the head tip at the time of component imaging are stored in the memory in advance, and the keystone correction and the rotation correction are performed on the two CCD acquired images, and then stored in the memory. By using the eccentric amount and direction of the tip of the head, the image after correction is translated and then the images are superimposed. The eccentric amount of the head tip and the direction thereof are acquired by imaging a calibration graphic in which each dimension and angle of the graphic can be accurately grasped in advance.

In this way, by imaging the electronic component 3 from different directions using one CCD camera 22, it is possible to realize a reduction in the number of components and cost and a reduction in the size of the apparatus.
Furthermore, in the above embodiment, a large high-precision camera 25 shown in FIG. 28 can be used instead of the CCD camera 22. In this case, the electronic component 3 is imaged using the reflecting plate 26 fixed to the column 21b. Thereby, component recognition with higher accuracy can be performed.

  In the above embodiment, it may be difficult to measure the positional relationship between the position of the lower surface of the component and the CCD in the CCD camera with high accuracy in performing geometric calculation. Therefore, as shown in FIG. 29A, sample parts (calibration parts) having a known pattern and shape, such as shapes, dimensions, and angles, are prepared, and the calibration parts are imaged in advance. The CCD acquired image as shown in FIG. 29B and the image after mapping conversion are stored in the memory, and the CCD acquired image and the image after mapping conversion are correctly stored in the memory during the component recognition processing. Thus, linear interpolation such as movement, rotation, enlargement, reduction, etc., or proportional interpolation may be performed. As a result, the accuracy can be easily increased, so that inexpensive and small cameras and lenses can be used.

In the above embodiment, a process for correcting the acquired image is added in consideration of the case where the component height registered in the component data stored in the memory is different from the component height of the actually sucked electronic component 3. May be.
In this method, the height of the suction nozzle 12b is adjusted so that the center position of the lower surface of the component comes to the component recognition position. Therefore, for example, when the actual height of the suction component is higher than the component height registered in the component data (when the component is thick), as shown in FIG. It will be located below the part lower surface position. Then, a shift occurs in the focal positions of the two CCD cameras 22 (the point where the lower surface of the component and the recognition center line of the CCD camera 22 intersect). Therefore, when parts images are acquired by the two CCD cameras 22 and the two images are superimposed at the time of parts recognition, as shown in FIG. 30B, all the terminals are translated in a certain direction by a certain amount. It is mistakenly judged as a terminal error.

  However, as long as the bottom surface of the component is flat, the component height hardly affects the mounting quality unless it is extremely extreme. Therefore, when one component image is translated with respect to the other component image in a certain direction by a certain amount within an allowable range, each of the two images shows the same location on the lower surface of the component. Based on the feature points, the terminal error is recognized after correcting the one component image to be translated by the predetermined amount in the direction opposite to the predetermined direction. At this time, even if the correction by the parallel movement is performed, if the terminal is displaced, the mounting of this component is stopped as a terminal error. Even when the parallel movement amount is out of the allowable range, a component height error is assumed as the component height affects the mounting quality, and the mounting of the component is stopped.

(Application examples)
In the above-described embodiment, the case where the number of captured images of the electronic component 3 is two has been described. However, the number may be three or more.
In the above embodiment, the case where the CCD camera is applied as the means for imaging the electronic component 3 has been described. However, a CMOS camera can also be applied.
Furthermore, in the above-described embodiment, the case where one electronic component 3 is imaged by one CCD camera 22 has been described. However, the positional relationship between the lower surface of the electronic component 3 and the CCD camera 22 is known, and the camera resolution is sufficient. If so, a single CCD camera 22 may image a plurality of parts at once.
In the above embodiment, the case where the image after the trapezoid correction is subjected to the rotation correction and the ratio expansion is performed in the component recognition process of FIG. 6 has been described. However, the order of the rotation correction process and the ratio expansion process is reversed. May be.

Furthermore, in the above-described embodiment, the case where the CCD camera 22 is configured to receive the reflected light from the illumination light irradiated to the electronic component 3 from the illumination 23 and image the lower surface of the electronic component 3 has been described. It is good also as a structure which receives the transmitted light of the illumination light irradiated to the electronic component 3 from 23, and images the lower surface of the electronic component 3. FIG. In this case, as shown in FIG. 31, the illumination 23 is provided at a position that is point-symmetric with respect to the CCD camera 22 around the tip position of the suction nozzle 12b when the lower surface of the electronic component 3 is the component recognition position. What is necessary is just to arrange. Thereby, the lower surface of the electronic component 3 can be imaged as a shadow image.
In the above embodiment, the case where the CCD camera 22 is fixed to the head unit 12 has been described. However, the CCD camera 22 may be fixed to the base of the electronic component mounting apparatus 1. Thereby, the increase in the weight of the head unit 12 can be suppressed.

  DESCRIPTION OF SYMBOLS 1 ... Electronic component mounting apparatus, 3 ... Electronic component, 5 ... Circuit board, 11 ... Conveying rail, 12 ... Head unit, 12a ... Suction head, 12b ... Suction nozzle, 13 ... X-axis gantry, 14 ... Y-axis gantry, 15 DESCRIPTION OF SYMBOLS ... Electronic component supply device, 16 ... Nozzle changer, 20 ... Component recognition mechanism, 21a, 21b ... Strut, 22 ... CCD camera, 23 ... Illumination, 50 ... Main control device, 51 ... Image calculation device, 52 ... Image recognition device, 53 ... Controller

Claims (14)

An electronic component mounting apparatus for sucking an electronic component by a suction nozzle in a head unit and mounting the electronic component at a predetermined position on a substrate,
Illuminating means for irradiating light onto the electronic component sucked by the suction nozzle;
A suction surface of the electronic component by the suction nozzle is received by receiving light emitted from the illumination unit in a state where the electronic component is sucked from above by the suction nozzle. Is an imaging means for imaging the opposite imaging surface from obliquely below;
A plurality of two-dimensional images obtained by imaging the imaging surface from a plurality of directions with different angles about the suction nozzle axis by the imaging means are converted into two-dimensional images obtained by viewing the imaging surface from directly below by geometric transformation. An electronic component mounting apparatus, comprising: a recognizing unit that recognizes the electronic component by comparing a plurality of two-dimensional images after geometric conversion.   The electronic component mounting apparatus according to claim 1, wherein the imaging unit includes a plurality of cameras, and images the imaging surface of the electronic component simultaneously from the plurality of directions.   2. The electronic device according to claim 1, wherein the imaging unit includes a single camera, and images the imaging surface of the electronic component a plurality of times by rotating the suction nozzle about the axis of the suction nozzle. Component mounting equipment.   4. The image pickup device according to claim 1, wherein the image pickup unit picks up the image pickup surface by receiving reflected light from light emitted from the illumination unit to the electronic component. 5. Electronic component mounting equipment.   The said illumination means and the said imaging means are arrange | positioned in the position which becomes line symmetrical with respect to the axis | shaft of the said suction nozzle at the time of the imaging of the said electronic component by the said imaging means. Electronic component mounting equipment.   The said imaging means images the said imaging surface by receiving the transmitted light of the light irradiated to the said electronic component from the said illumination means, The any one of Claims 1-3 characterized by the above-mentioned. Electronic component mounting equipment.   The said illumination means and the said imaging means are arrange | positioned in the position which becomes point-symmetrical with respect to the front-end | tip position of the said suction nozzle at the time of the imaging of the said electronic component by the said imaging means. Electronic component mounting equipment.   The electronic component mounting apparatus according to claim 1, wherein the imaging unit and the illumination unit are fixed to the head unit.   The electronic component mounting apparatus according to claim 1, wherein the imaging unit and the illumination unit are fixed to a base of the electronic component mounting apparatus. The imaging means is constituted by a CCD camera, and the CCD of the CCD camera is arranged to be inclined with respect to the imaging surface of the electronic component,
The recognizing means includes, as the geometric transformation, elevation angle correction for correcting distortion of a captured image with respect to the actual shape of the electronic component generated by the inclination of the CCD, and a predetermined specific reference angle around the suction nozzle axis. The rotation correction is performed so that the image obtained by correcting the elevation angle is rotated in the reverse direction around the center position of the image by the difference in the imaging angle of the imaging means around the suction nozzle axis. The electronic component mounting apparatus according to any one of? The imaging means is constituted by a CCD camera, and the CCD of the CCD camera is arranged in parallel to the imaging surface of the electronic component,
The recognizing means, as the geometric transformation, captures an image picked up by the image pickup means by a difference of an image pickup angle of the image pickup means around the suction nozzle axis from a predetermined specific reference angle around the suction nozzle axis. The electronic component mounting apparatus according to claim 1, wherein rotation correction is performed to rotate in a reverse direction around the center position of the image.   When the distance between the feature points of the electronic component when the plurality of two-dimensional images after geometric transformation are superimposed is outside the allowable range, the recognition means indicates that an abnormality has occurred in the electronic component. It recognizes, The electronic component mounting apparatus of any one of Claims 1-11 characterized by the above-mentioned.   The recognizing means detects an abnormality in the electronic component when a non-overlapping pixel area out of the pixel area forming the image of the electronic component when the plurality of two-dimensional images after geometric conversion are superimposed is outside an allowable range. The electronic component mounting apparatus according to claim 1, wherein the electronic component mounting apparatus recognizes the occurrence of the problem.   The said recognition means recognizes at least one of the shift | offset | difference of the adsorption angle from the horizontal surface of the said electronic component, and the deformation | transformation of the terminal of the said electronic component as an abnormality of the said electronic component. Electronic component mounting equipment.

Images