JP2006108563A - Component recognizing method and apparatus thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a component recognizing method and apparatus for recognizing components without any recognition error by executing image recognizing processes through extraction of images of the photographed electronic components, while moving the components attracted and held with nozzle without increase in luminance of the lighting with use of urea sensor. <P>SOLUTION: The component recognizing apparatus generates accurate image recognizing data by synthesizing a plurality of images photographed during chip components 18 attracted and held with the nozzle 16 are moving within the sight of a CCD camera 29, and then executing the image recognizing process through generation of images including clear difference in gradation. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

The present invention relates to a component recognition method and apparatus for recognizing the position of an electronic component from an image obtained by imaging the electronic component sucked and held by a nozzle from below.

In the field of electronic component mounting, a method of applying image recognition is widely used as a method for accurately positioning an electronic component. In this method, the position of an electrode, an edge portion or a center of gravity of an electronic component is detected by recognizing an image obtained by imaging the electronic component.

In the position detection of the electronic component by this image recognition, the electronic component is picked up and held by a nozzle and moved above the image pickup apparatus to image the electronic component. Conventionally, as this type of imaging apparatus, there has been one using a two-dimensional area sensor composed of a CCD camera, for example.

By the way, in the above-mentioned conventional one, in order to image the electronic component, the mounting head unit is moved so that the electronic component is positioned above the imaging device by sucking and holding the electronic component with the suction nozzle of the mounting head unit. It must be stopped once. Therefore, there is a problem that the time required for imaging the electronic component becomes long and the tact time becomes long.

Therefore, by adopting a line sensor as a one-dimensional sensor instead of the two-dimensional area sensor such as the CCD camera, the electronic component is moved across the line sensor while the electronic component is being transferred. An image of a component is captured at a fixed timing, and an imaging position of the electronic component is detected from the image data, thereby shortening the imaging time of the electronic component (see, for example, Patent Document 1).
JP-A-7-263897

  However, the area sensor is generally 512 pixels in the longitudinal direction, whereas the line sensor is formed with a high density of 1000 to 7500 pixels in a line in the longitudinal direction. By adopting the line sensor, although the imaging time of the electronic component is shortened, there is a problem that the number of data is large due to the large number of pixels, and it takes time for image processing. In addition, since the image is captured in a short time in order to capture the image while moving, it is necessary to increase the luminance of the illumination.

  Therefore, the present invention uses an area sensor to move the electronic component picked up and held by the nozzle without increasing the brightness of the illumination, captures the image of the captured electronic component, performs image recognition processing, and performs a recognition error. It is an object of the present invention to provide a component recognition method and apparatus that do not have any components.

According to the first aspect of the present invention, an electronic component is mounted by irradiating the electronic component with light while picking up and holding the electronic component on the suction surface of the nozzle and performing image recognition processing on the image of the electronic component. In the component recognition method of the mounting device, the electronic component is captured a plurality of times and image data is taken into the component recognition device while the electronic component crosses the field of view of the imaging unit of the component recognition device, and the plurality of images are captured. The above-described problem is solved by adopting a method of performing image recognition processing on the image data obtained by combining the data after correcting the position.

According to the second aspect of the present invention, the electronic component is picked up by irradiating the electronic component with light in a state where the electronic component is sucked and held on the suction surface of the nozzle, and an image of the electronic component is subjected to image recognition processing and mounted. In the component recognition device of the component mounting apparatus, an illumination unit that irradiates light to the electronic component, an imaging unit that images the electronic component, an image storage unit that stores a plurality of captured images, and image recognition from the image storage unit An extraction data storage unit that extracts and stores image data necessary for processing for each image; an image recognition calculation unit that performs image processing on the image data stored in the extraction data storage unit; An image synthesis unit that synthesizes the image synthesis data, and a control unit that transfers image recognition data obtained by processing the image data synthesized by the image synthesis unit by the image recognition calculation unit, or controls the imaging unit Constructed from, it solves the problems.

And in invention of Claim 3, in the components recognition apparatus of the electronic component mounting apparatus of Claim 2, in addition to a coaxial illumination part, the said illumination means is from at least one of a downward illumination part and a side illumination part. And having an illumination switching unit that switches driving of the illumination unit based on the control of the control unit.

  According to the component recognition method and apparatus in the electronic component mounting apparatus of the present invention, the electronic component (for example, a chip component) sucked and held by the nozzle in the upper visual field of the component recognition device is imaged while moving without stopping. Time can be reduced. Then, the position of the image of the electronic component is corrected so that the image of the electronic component is located at the center of the image frame for each image by moving the image a plurality of times. Further, the position-corrected image is synthesized. That is, one image is generated by summing up the pixel values of each image. In order to capture images while moving electronic parts, images are captured in a short time so that there is no blurring, resulting in a dark image with little gradation difference, but by generating multiple images, an image with a clear gradation difference is generated. Since the image recognition process is performed, the image recognition can be performed with high accuracy. In other words, it is possible to generate image recognition data while greatly reducing image recognition errors while reducing imaging time.

  Furthermore, in addition to the coaxial illumination unit, a plurality of illumination units such as a lower illumination unit and a side illumination unit are provided, so that when the electronic component is located near the center of the field of view of the imaging unit (e.g., CCD camera), the coaxial illumination unit and the lower part It is suitable to illuminate with the illumination unit and illuminate with the coaxial illumination unit and the side illumination unit or the lower illumination unit and the side illumination unit when the electronic component is located outside the center of the field of view of the imaging unit Image recognition mistakes can be reduced because images can be taken with illumination under conditions.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view of an electronic component mounting apparatus in which a component recognition device according to an embodiment of the present invention is incorporated. 2 is a diagram showing the positional relationship between the electronic component and the imaging unit at the time of imaging, FIG. 3 is an explanatory diagram for synthesizing a plurality of images of the electronic component, and FIG. 4 is a diagram showing correction of nozzle movement with respect to the imaging unit. FIG. 5 is a diagram of illumination means of the component recognition device, FIG. 6 is a block diagram showing the system configuration of the component recognition device, and FIG. 7 is a flowchart showing the operation of the component recognition device using image processing.

  First, the entire structure of the electronic component mounting apparatus 10 will be described with reference to FIG. In FIG. 1, reference numeral 13 denotes a substrate transfer device. A substrate 19 transported by the substrate transport device 13 is held near the center of the substrate transport device 13. The mounting head unit 11 is supported by the XY transfer unit 12 and can move horizontally. The mounting head portion 11 is provided with a plurality of nozzles 16 for sucking and holding electronic components (chip components) 18. The mounting head portion 11 vacuum-sucks the chip component 18 supplied to the electronic component supply position of the electronic component supply device (feeder) 14 to the lower end portion of the nozzle 16 and moves it XY to the upper portion of the component recognition device 15. The component recognizing device 15 images the lower surface of the chip component 18 moving on the upper surface a plurality of times, calculates the suction displacement amount (XY error, angle error) of the chip component 18 sucked and held by the image recognition process, and the board 19 The position is corrected with respect to the regular mounting position and mounted on the substrate 19.

  With reference to FIG. 2 and FIG. 3, the positional relationship of imaging and the captured image will be described. In FIG. 2, 16 is a nozzle, 18 is a chip component, 29 is a CCD camera as an imaging unit, and CL is the center of the visual field of the CCD camera. The chip components 18 are imaged in the order of FIG. 2A, FIG. 2B, and FIG. That is, the nozzle component 16 picks up and holds the chip component 18 while moving while picking up an image at a position in front of the visual field center CL of the CCD camera 29 by a length L1, and then picking up an image at the position of the visual field center CL. An image is taken at a position that is a length L2 past from the position. In the embodiment, L1 = L2 = 10 mm.

  In FIG. 3, 30 is an image frame indicating the range of the screen imaged by the CCD camera 29 and corresponds to the field of view of the CCD camera 29. Reference numeral 30a denotes a detection frame indicating the range of image processing within the image frame 30. The image captured in FIG. 2 (a) is FIG. 3 (a) and 18a shows an image of the chip component 18, and the image captured in FIG. 2 (b) is FIG. 3 (b) and 18b is the chip component. FIG. 3C shows an image of 18, and FIG. 3C shows an image taken in FIG. 2C, and 18 c shows an image of the chip component 18.

  Further, image composition will be described with reference to FIG. In FIG. 4, 30 is an image frame, 16a, 16b, and 16c are images of the suction surface of the nozzle 16. The left-right direction of the paper surface is the X axis and the front-back direction of the paper surface is the Y axis. First, an image of the suction surface of the nozzle 16 is captured in advance in order to capture the movement locus of the nozzle 16. The suction surface of the nozzle 16 is imaged at each position in FIGS. 2A, 2 </ b> B, and 2 </ b> C without sucking the chip component 18. 4a, 16b, and 16c correspond to images of the suction surface of the nozzle 16 at each position, and are images displayed in the individual image frames 30, but are represented by one image frame 30 for convenience. When imaging the suction surface of the nozzle 16, it is desirable to stop and image at each position in order to obtain a good image, and a white jig may be attached to the suction surface of the nozzle 16.

Next, centroid positions 16a (Xa, Ya), 16b (Xb, Yb), 16c (Xc, Yc) of 16a, 16b, and 16c are calculated by image recognition processing, and an approximate straight line L that passes through these three centroids is obtained. Thus, the inclination θ in the X-axis direction is calculated. Here, the composition of the image, which will be described later, will be described with reference to the composition of the center of gravity. The following relational expression is used to align the center of gravity 16a with the center of gravity 16b and the center of gravity 16c with the center of gravity 16b.
16b (Xb, Yb)
= 16a (Xa + L1, Ya + L1 * tan θ)
= 16c (Xc-L2, Yc-L2 * tan θ)
In the image processing, it is necessary to obtain in advance the image size (mm / pixel) per pixel in the set visual field size. Thereby, when converting the position and angle of an image, it is generally performed by coordinate conversion in units of pixels.

  Here, the images are synthesized. However, if the image frames 30 are simply overlapped, the state shown in FIG. In FIG. 3A, the detection frame 30a is set so that the center of the detection frame 30a becomes the center of gravity 16a of FIG. 4, and the center of the detection frame 30a and the center of the image frame 30 are aligned. Perform coordinate transformation on pixel data. Similarly, in FIG. 3B, the detection frame 30a is set so that the center of the detection frame 30a becomes the center of gravity 16b of FIG. 4, and the detection frame 30a and the center of the image frame 30 coincide with each other. Further, coordinate conversion is performed on the pixel data in the pixel 30a, and the detection frame 30a is set so that the center of the detection frame 30a is the center of gravity 16c in FIG. Coordinate conversion is performed on the pixel data in the detection frame 30a so that the centers of the image frames 30 match.

  From the above processing, three pixel values can be acquired at one coordinate after coordinate conversion. Therefore, the calculation of adding the three pixel values at each coordinate to obtain the pixel value of the coordinate is performed on all the coordinates in the detection frame 30a. To generate one synthesized image. FIG. 3E shows a synthesized image. In the figure, 18d is an image of the chip component 18 after synthesis. Thereby, image data having a large gradation difference can be obtained. For example, the gradation difference between the low pixel value 35 and the high pixel value 60 is 25, and by combining these three, image data with a gradation difference of 75 can be obtained. Since the pixel value is represented by 8 bits, if the pixel value is from 0 to 255 but exceeds 255 in the image composition, 255 is subtracted from the maximum pixel value after composition, and 1 / The image may be synthesized after the value of 3 is uniformly subtracted from the pixel value of the image data before synthesis.

  Image recognition processing is performed using the image data having a large gradation difference, and image recognition data is generated. That is, the center of gravity of the chip component 18 and the inclination with respect to the XY direction are obtained. Since the center of the nozzle 16 is the center of the detection frame 30a, the amount of suction deviation (XY error, angle error) of the chip component 18 is calculated.

Here, the structure of the illumination means 20 of the component recognition apparatus 15 is demonstrated with reference to FIG. FIG. 5 shows the illumination means 20 including the coaxial illumination unit 23, the lower illumination unit 25, and the side illumination unit 27. A plurality of LEDs 21 are two-dimensionally (planar) arranged in the coaxial illumination unit 23 and attached to a frame 22 that is a vertical flat plate so that the arrangement surface of the LEDs 21 is vertical. In the figure, the frame body 22 is shown on the left and right, but is also arranged on the front and rear sides to surround the illumination unit. A diffusion plate 24 for forming uniform illumination light is disposed on the light projecting surface of the coaxial illumination unit 23. A half mirror 28 is disposed at an intersection of the vertical center line of the space surrounded by the frame 22 and the horizontal center line of the coaxial illumination unit 23 so as to form an angle of 45 degrees with respect to the coaxial illumination unit 23. The illumination light that has passed through the diffusion plate 24 is reflected upward by the half mirror 28 and irradiates the chip component 18 adsorbed and held by the nozzle 16 from directly below. The light reflected downward by the chip component 18 passes through the half mirror 28 and is imaged by the CCD camera 29.

A plurality of LEDs 21 are two-dimensionally (planar) arranged in the lower illumination unit 25, and are attached to the frame body 22 so that the arrangement surface of the LEDs 21 is horizontal above the coaxial illumination unit 23. In the drawing, the lower illumination unit 25 is shown on the left and right, but is also arranged on the front and rear sides so as to surround the illumination unit. It should be noted that the inner edge of each lower illumination section 25 (on the opposite side to the edge close to the frame 22) has a size that does not enter the visual field of the CCD camera 29. A diffusion plate 26 for forming uniform illumination light is disposed on the upper surface of the lower illumination unit 25, and the illumination light that has passed through the diffusion plate 2 irradiates the chip component 18 held by suction with the nozzle 16 obliquely from below. To do. The light reflected downward by the chip component 18 passes through the half mirror 28 and is imaged by the CCD camera 29.

A plurality of LEDs 21 are arrayed two-dimensionally (planar) in the side illumination section 27 and are attached to the frame body 22 so that the array plane of the LEDs 21 is vertical above the lower illumination section 25. Although the side illumination unit 27 is shown on the left and right in the figure, it is also arranged on the front and back sides to surround the illumination unit. In addition, the side illumination part 27 is arrange | positioned on the outer side (side near the frame 22) of the arrangement | sequence of LED21 of the downward illumination 25 so that the illumination of the downward illumination 25 may not be prevented. The illumination light irradiates the chip component 18 held by suction with the nozzle 16 from obliquely below. The light reflected downward by the chip component 18 passes through the half mirror 28 and is imaged by the CCD camera 29.

Next, the configuration of the control system of the component recognition device 15 will be described with reference to FIG. In FIG. 6, the CCD camera 29 images the chip component 18 and sends the image data to the image storage unit 32 via the A / D conversion unit 31. The image storage unit 32 stores the A / D converted image data. Then, in order to shorten the image recognition processing time, the data of the portion to be subjected to image processing (data in the detection frame 30a: see FIG. 3) is extracted from the image storage unit 32 and stored in the extracted data storage unit 33.

  The image recognition calculation unit 34 performs coordinate conversion on the image data stored in the extracted data storage unit 33 and transfers the image data to the image composition unit 35. Also, the image composition unit 35 is used as a work area, and the plurality of coordinate-converted image data are synthesized to generate a composite image in the image composition unit 35. Further, the composite image is received from the image composition unit 35, and image recognition processing is performed to detect the position and angle of the electrode, edge portion, or center of gravity of the chip component, and calculate the suction error. . The image recognition process is performed by a method specified in advance according to the type and size of the component, such as a pattern matching method or a binarization method.

  The image composition unit 35 receives the plurality of image data after the coordinate conversion calculated by the image recognition calculation unit 34 as described above, and the image recognition calculation unit 34 internally generates a composite image of the plurality of image data after the coordinate conversion. The When the pixel value of the composite image exceeds 255, the composite image is discarded, and a composite image is generated inside again using the image data with the pixel value lowered.

The control unit 36 is a part that controls the entire component recognition apparatus 15, and the points related to the present invention will be described. The controller 36 performs imaging by sending a control signal for switching illumination and a trigger T during imaging. That is, the illumination switching control signal is transferred to the illumination switching unit 37 corresponding to each imaging position in FIGS. 2A, 2B, and 2C to irradiate the chip component 18 with illumination light and trigger Ta. , Tb, Tc are sent to the CCD camera 29, and the shutter is opened to capture an image. Further, the image recognition data such as the error between the suction position and the angle is acquired from the image recognition calculation unit 34 and transferred to the main control unit 38 of the electronic component mounting apparatus 10.

  Based on the control signal of the control unit 36, the illumination switching unit 37 operates any one of the coaxial illumination unit 23, the lower illumination unit 25, and the side illumination unit 27 to perform illumination, or performs illumination by combining the illuminations. Lower the to activate and illuminate.

  The configuration of the component recognition device 15 is as described above, and the operation thereof will be described below. In FIG. 7, from the start, the mounting head unit 11 sucks and holds the chip component 18 on the nozzle 16 and moves XY so that the chip component 18 crosses over the component recognition device 15 (field of view of the CCD camera 29) (ST1). .

  The control unit 36 sends a signal to the illumination switching unit 37 to operate the lower illumination unit 25 to irradiate illumination light, and the position of the chip component 18 is the center of the visual field of the CCD camera 29 as shown in FIG. When the length L1 before CL is reached, a trigger Ta is created in the CCD camera 29 to image the chip component 18, and the image data is taken into the image storage unit 32 via the A / D conversion unit 31 (ST2). ).

  The control unit 36 sends a signal to the illumination switching unit 37 to operate the coaxial illumination unit 23 to irradiate the illumination light, and the position of the chip component 18 is the visual field center CL of the CCD camera 29 as shown in FIG. Then, the trigger Tb is created in the CCD camera 29 to image the chip component 18, and the image data is taken into the image storage unit 32 via the A / D conversion unit 31 (ST3).

  The control unit 36 sends a signal to the illumination switching unit 37 to operate the lower illumination unit 25 to irradiate illumination light, and the position of the chip component 18 is the visual field center CL of the CCD camera 29 as shown in FIG. When the length L2 is passed, the trigger Tc is created in the CCD camera 29 to image the chip part 18, and the image data is taken into the image storage unit 32 via the A / D conversion unit 31 (ST4). ).

  The detection frame 30 a is applied to the image captured by the trigger Ta, and the image data in the detection frame 30 a is captured in the extracted data storage unit 33. The image recognition calculation unit 34 performs coordinate conversion on the image data captured in the extracted data storage unit 33 so that the center of the detection frame 30a and the center of the image frame 30 coincide with each other, and the image data after the coordinate conversion is combined with the image. The data is transferred to the unit 35 (ST5).

  The detection frame 30 a is applied to the image captured by the trigger Tb, and the image data in the detection frame 30 a is captured in the extracted data storage unit 33. The image recognition calculation unit 34 performs coordinate conversion on the image data captured in the extracted data storage unit 33 so that the center of the detection frame 30a and the center of the image frame 30 coincide with each other, and the image data after the coordinate conversion is combined with the image. The data is transferred to the unit 35 (ST6).

  The detection frame 30 a is applied to the image captured by the trigger Tc, and the image data in the detection frame 30 a is captured in the extracted data storage unit 33. The image recognition calculation unit 34 performs coordinate conversion on the image data captured in the extracted data storage unit 33 so that the center of the detection frame 30a and the center of the image frame 30 coincide with each other, and the image data after the coordinate conversion is combined with the image. The data is transferred to the unit 35 (ST7).

  The image recognition calculation unit 34 generates composite image data in the image composition unit 35 using the three image data after the coordinate conversion of the image composition unit 35 (ST8). Next, the process proceeds to a branch on whether the pixel value of the composite image data is 255 or less. If the pixel value of the composite image data exceeds 255, the pixel value is uniformly subtracted from the three image data before the composite (ST12). ) Three images are synthesized again.

  If the pixel value of the synthesized image data is 255 or less, the image recognition calculation unit 34 performs image recognition processing using the synthesized image data, calculates the center of gravity of the chip component 18 and the angle with respect to the XY direction, and Image recognition data such as a suction position and angle error is calculated (ST10). Then, the control unit 38 sends image recognition data to the main control unit 38 of the electronic component mounting apparatus 10 (ST11), and the process ends.

  In the above operation, the switching between the coaxial illumination unit 23 and the lower illumination unit 25 has been described. However, the side illumination unit 27 may be added to emphasize the edge portion of the chip component 18. Since the illumination method varies depending on the type of electronic component and the position where the image is taken, it is preferable to apply the method after selecting an optimal illumination unit and combining the illumination units in advance.

  In the embodiment, the chip component 18 is imaged three times while moving, and three images are combined. However, the CCD camera 29 is configured such that two images are combined and two images are combined and four images are combined and four images are combined. It is preferable to select in accordance with how the image is reflected.

It is a perspective view of the electronic component mounting apparatus with which the component recognition apparatus was integrated. It is a figure which shows the positional relationship of the electronic component and imaging part at the time of imaging. It is explanatory drawing for synthesize | combining this with several images of an electronic component. It is a figure which shows correction | amendment of the nozzle movement with respect to an imaging part. It is a figure of the illumination means of a components recognition apparatus. It is a block diagram which shows the system configuration | structure of a components recognition apparatus. It is a flowchart which shows operation | movement of the components recognition apparatus using an image process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Electronic component mounting apparatus 11 Mounting head part 12 XY transfer part 13 Board | substrate conveyance apparatus 14 Electronic component supply apparatus 15 Component recognition apparatus 16 Nozzle 18 Electronic component (chip component)
19 Substrate 20 Illumination means 23 Coaxial illumination unit 25 Lower illumination unit 27 Side illumination unit 29 Imaging unit (CCD camera)
30 Image frame 30a Detection frame 31 A / D conversion unit 32 Image storage unit 33 Extracted data storage unit 34 Image recognition calculation unit 35 Image composition unit 36 Control unit 37 Illumination switching unit 38 Main control unit

Claims (3)

In the component recognition method of the electronic component mounting apparatus in which the electronic component is sucked and held on the suction surface of the nozzle, the electronic component is irradiated with light and imaged, and the image of the electronic component is subjected to image recognition processing and mounted.
While the electronic component crosses the field of view of the imaging unit of the component recognition device, the electronic component is imaged a plurality of times, image data is taken into the component recognition device, and the plurality of image data are corrected and combined. A component recognition method for an electronic component mounting apparatus, wherein image recognition processing is performed on image data obtained by the synthesis. In the component recognition device of the electronic component mounting apparatus in which the electronic component is sucked and held on the suction surface of the nozzle, the electronic component is irradiated with light and imaged, and the image of the electronic component is subjected to image recognition processing and mounted.
Illumination means for irradiating light to the electronic component, an imaging unit for imaging the electronic component, an image storage unit for storing a plurality of captured images, and image data necessary for image recognition processing from the image storage unit for each image An extraction data storage unit that extracts and stores the image data, an image recognition calculation unit that performs image processing on the image data stored in the extraction data storage unit to generate image synthesis data, and an image that combines the image synthesis data Electronic component mounting comprising: a combining unit; and a control unit that transfers image recognition data obtained by processing the combined image data of the image combining unit by the image recognition calculation unit or controls the imaging unit Device recognition device. In addition to the coaxial illumination unit, the illumination unit includes at least one of a lower illumination unit and a side illumination unit, and includes an illumination switching unit that switches driving of the illumination unit based on control of the control unit. The component recognition apparatus for an electronic component mounting apparatus according to claim 2.

Images