JP2004031727A - Electronic-component recognizing apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an electronic-component apparatus and method whereby the accuracy of an image can be secured. <P>SOLUTION: In an electronic-component mounting apparatus, the electronic-component recognizing apparatus captures an image of an electronic component P held by a transferring head by a camera 9 to recognize the image. In this electronic-component recognizing apparatus, the picture elements falling in a picture-element range PX1 (1-n) which are included in a plurality of picture elements of the line sensor of the camera 9 are set as photometric picture elements. The picture elements falling in a picture-element range PX2 other than the one PX1 are set as photographinic picture elements for acquiring the image signal of the electronic component P. When acquiring an image, an image signal 22 outputted from the picture-element range PX2 is corrected based on a photometric reference signal 21 obtained by guiding the illumination light of an image-signal illuminating portion 10 by a reflecting plate 20. Therefore, the variation of an image which is caused by the variation of the relative moving speed of the camera 9 to the moving speed of the electronic component P can be corrected properly. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an electronic component recognizing device and an electronic component recognizing method for recognizing a position and a shape of an electronic component used in an electronic component mounting apparatus.
[0002]
[Prior art]
2. Description of the Related Art A line camera is known as an imaging unit for performing position recognition of an electronic component by image processing in an electronic component mounting apparatus or the like. In this line camera, pixels that store charges according to the amount of received light are arranged in a line. When an image of an object to be imaged is formed on a line camera by an optical system, charges corresponding to image data of the object are stored in each pixel. By sequentially outputting the electric charge of each pixel as an electric signal, a one-dimensional image in the pixel arrangement direction, that is, the main scanning direction can be obtained. Then, desired two-dimensional image data is acquired by moving the electronic components in a sub-scanning direction orthogonal to the main scanning direction and arranging a plurality of one-dimensional images obtained in parallel. In the electronic component mounting apparatus, the transfer head holding the electronic component to be recognized is horizontally moved above the line camera, thereby performing the movement in the sub-scanning direction.
[0003]
[Problems to be solved by the invention]
In order to obtain an appropriate image with such a line camera, it is necessary to keep the relative speed between the recognition target and the line camera in the sub-scanning as constant as possible. However, in the electronic component mounting apparatus, the moving speed of the transfer head is not always strictly constant due to various factors such as an error of a feed mechanism such as a ball screw and a control error. This variation in the moving speed in the sub-scanning direction causes variation in brightness for each one-dimensional image, and the variation reduces image accuracy. As described above, the conventional electronic component recognition device using a one-dimensional camera such as a CCD line camera has a problem that it is difficult to ensure image accuracy.
[0004]
Therefore, an object of the present invention is to provide an electronic component recognition device and an electronic component recognition method that can ensure image accuracy.
[0005]
[Means for Solving the Problems]
An electronic component recognizing device according to claim 1, wherein the electronic component recognizing device is an electronic component recognizing device that captures and recognizes an image of an electronic component held by a transfer head in the electronic component mounting apparatus, and includes a nozzle that sucks and holds the electronic component. A transfer head, a one-dimensional camera having a plurality of pixels arranged in a row, and moving means for moving the transfer head relative to the one-dimensional camera in a direction orthogonal to the arrangement direction of the pixels And an image recognition unit that recognizes the electronic component based on a plurality of one-dimensional images obtained by imaging the electronic component relatively moved by the moving unit with the one-dimensional camera, and an imaging area of the one-dimensional camera. An illumination unit for illuminating the passing electronic component; a light guiding unit for guiding illumination light of the illumination unit to a pixel set as a photometric reference pixel among the plurality of pixels; And a picture signal correcting means for correcting the image signal output from the pixel on the basis of the signal.
[0006]
3. An electronic component recognition method according to claim 2, wherein the electronic component mounting apparatus recognizes an image of the electronic component held by the transfer head by capturing the image with a one-dimensional camera having a plurality of pixels arranged in a row. A relative movement step of moving a transfer head, which holds the electronic component by suction, relative to a one-dimensional camera in a direction orthogonal to the arrangement direction of the pixels; and An image recognition step of recognizing the electronic component based on a plurality of one-dimensional images obtained by imaging with the one-dimensional camera. In this image recognition step, the electronic component passing through the imaging area of the one-dimensional camera The illumination light of the illumination unit for illuminating is guided to a pixel set as a photometric reference pixel among the plurality of pixels to receive light, and an image correction electric signal output from the photometric reference pixel is output. Zui by correcting the image signal output from the pixel.
[0007]
According to the present invention, in an image recognition step of recognizing an electronic component based on a plurality of one-dimensional images obtained by imaging a moving electronic component with a one-dimensional camera, an electronic component passing through an imaging area of the one-dimensional camera is used. The illumination light of the illumination unit that illuminates the component is guided to the pixel set as the photometric reference pixel among the plurality of pixels and received, and is output from the pixel based on the image correction electric signal output from the photometric reference pixel. By correcting the obtained image signal, it is possible to appropriately correct the variation in the image caused by the variation in the relative moving speed.
[0008]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view of an electronic component mounting apparatus according to one embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of an electronic component recognition apparatus according to one embodiment of the present invention, and FIG. FIG. 3B is an explanatory diagram of a configuration of a camera of the electronic component recognition device according to one embodiment, FIG. 3B is an explanatory diagram of an output signal of a line sensor of the electronic component recognition device of one embodiment of the present invention, and FIG. FIG. 5 is a diagram showing a signal output timing of a line sensor of the electronic component recognition device of one embodiment. FIG. 5 is an explanatory diagram of a signal output of the line sensor of the electronic component recognition device of one embodiment of the present invention.
[0009]
First, an electronic component mounting apparatus will be described with reference to FIG. In FIG. 1, a transport path 2 is provided in the center of a base 1 in the X direction. The transport path 2 transports and positions the substrate 3. Component supply units 4 for supplying electronic components are arranged on both sides of the transport path 2, and a large number of tape feeders 5 are arranged in each component supply unit 4. The tape feeder 5 stores the electronic components held on the tape, and feeds the electronic components to the pickup position by feeding the tape at a pitch.
[0010]
Two Y-axis tables 6A and 6B are arranged in parallel on both side ends of the upper surface of the base 1, and an X-axis table 7 on which a transfer head 8 is mounted is mounted on the Y-axis tables 6A and 6B. Have been. The transfer head 8 includes a plurality of suction nozzles 8a (FIG. 2) for sucking electronic components at a lower end. By driving the Y-axis tables 6A and 6B and the X-axis table, the transfer head 8 moves horizontally, picks up electronic components from the pickup position of the tape feeder 5, and transfers and mounts them on the substrate 3.
[0011]
Next, with reference to FIGS. 2 and 3, an electronic component recognizing device that captures and recognizes an image of an electronic component held by the transfer head 8 in the electronic component mounting apparatus will be described. A camera 9 that reads an image of an electronic component is disposed on a movement path of the transfer head 8 between the transport path 2 and the component supply unit 4. The camera 9 is a line camera (one-dimensional camera), and includes a line sensor 11 in which a plurality of pixels having photoelectric conversion elements are arranged in a row in the Y direction. On both sides of the camera 9, illumination units 10 for illuminating electronic components passing through the imaging area of the camera 9 are arranged.
[0012]
With the electronic component P held by the suction nozzle 8a of the transfer head 8 positioned on the camera 9 as shown in FIG. 2, the illumination unit 10 is turned on to illuminate the electronic component P from below. An optical image of the electronic component P is formed on each pixel of the line sensor 11. Then, an image signal obtained by converting the optical image into an electric signal is constantly output to the signal capturing unit 12.
[0013]
The X-axis table 7 for moving the transfer head 8 includes a motor 7a and a feed screw 7c, and the motor 7a is controlled by a motor control unit 13. The motor control unit 13 controls the motor 7a in accordance with a speed command and a position command transmitted via the CPU 19, whereby the transfer head 8 moves in the X direction with respect to the camera 9 in a predetermined movement pattern.
[0014]
By taking an image with the camera 9 while moving the transfer head 8, the camera 9 acquires a scan image of the electronic component P. The X-axis table 7 moves the transfer head 8 holding the electronic component P, which is an imaging target, relative to the line sensor 11 of the camera 9 in a direction (X direction) orthogonal to the pixel arrangement direction (Y direction). It is a moving means for moving. Then, the electronic component P is identified and its position is detected by recognizing the obtained scanned image as an image. When the electronic component P is mounted on the board 3, the displacement is corrected based on the position detection result. .
[0015]
The motor 7a includes an encoder 7b, and a pulse signal output from the encoder 7b is sent to the motor control unit 13. The pulse detection unit 14 that has received the pulse signal from the motor control unit 13 detects the position of the transfer head 8 based on the pulse signal, and outputs the current position to the image capture command unit 15.
[0016]
Next, the camera 9 will be described with reference to FIG. The camera 9 has a configuration in which line sensors 11 are arranged below the optical system 9a in the Y direction. Here, of the plurality of pixels 11a (1 to m) of the line sensor 11, the pixel 11a located in the pixel range PX1 (1 to n) for n pixels from the first pixel is set as a photometric reference pixel. A light-shielding partition 11b is provided at a boundary with the pixel range PX2 (n + 1 to m) excluding the pixel range PX1.
[0017]
The pixels 11a in the pixel range PX1 receive only the reflected light of the illumination light emitted from the illumination unit 10 that is reflected on the optical system 9a by the reflector 20 as the photometric reference light. The reflection plate 20 is a light guiding unit that guides the illumination light of the illumination unit 10 to a pixel in the pixel range PX1 set as a photometric reference pixel among the plurality of pixels 11a. Then, as shown in FIG. 3B, the pixel 11a in the pixel range PX1 outputs the charge accumulated by receiving the light as the photometric reference signal 21. VA indicates a voltage value of the photometric reference signal 21 output as an analog signal.
[0018]
The pixel range PX2 excluding the pixel range PX1 is an image signal capturing range for capturing an image of the electronic component P, and these pixels 11a receive the reflected light from the electronic component P, as shown in FIG. As shown, the charge accumulated by the light reception is output as an image signal 22 including the image of the imaging target. VB indicates a voltage value of the image signal 22 output as an analog signal. Note that the pixel range PX1 as the photometric reference pixel is set to a so-called unnecessary pixel range excluding a range required for imaging the electronic component P.
[0019]
In FIG. 2, the signal capturing unit 12 connected to the line sensor 11 includes an image signal capturing unit 12a and a reference signal capturing unit 12b. The image signal capturing unit 12a captures the image signal output from the pixel 11a in the pixel range PX2 among the pixels of the line sensor 11, and the reference signal capturing unit 12b measures the photometry output from the pixel 11a in the pixel range PX1. Capture the reference signal.
[0020]
The signal capture from the line sensor 11 is performed when the signal capture unit 12 receives an image capture command signal output from the image capture command unit 15. That is, the pulse signal transmitted from the encoder 7b of the motor 7a that drives the transfer head 8 in the X direction is received and counted by the pulse detection unit 14, and the count value becomes a count value corresponding to the specific position of the transfer head 8. At the arrival timing, an image capture command signal is output to the signal capture unit 12.
[0021]
Then, with the image capture command signal as a trigger, the electric charges accumulated in the pixels are output as electric signals in each scanning operation in order from the top pixel of the line sensor 11. At this time, as described above, the signals output from the n first pixels (pixel range PX1) are captured by the reference signal capturing unit 12b, and the signals output from the subsequent pixels (pixel range PX2). Is captured by the image signal capturing unit 12a.
[0022]
The reference signal capturing unit 12b and the image signal capturing unit 12a are connected to an image correction reference value setting unit 16 and an image correction unit 17, respectively. The photometric reference signal 21 (voltage value VA) and the image signal 22 (voltage value VB) shown in FIG.
[0023]
Upon receiving the photometric reference signal 21, the image correction reference value setting unit 16 samples and holds a representative voltage value [VA] that is close to the average value of the voltage values VA, and stores this voltage value [VA] in the image correction unit 17. Communicate to The voltage value [VA] is used as a reference value for image signal correction in the image correction unit 17 as described later. As the voltage value [VA], a voltage value corresponding to a charge amount accumulated in a pixel located substantially at the center of the pixel range PX1 may be sampled and held, or the charge amount of all pixels in the pixel range PX1 may be sampled and held. An average value may be used.
[0024]
The image correcting unit 17 outputs a digitized one-dimensional image signal to the image recognizing unit 18 by A / D converting the image signal 22 which is an analog electric signal received from the image signal capturing unit 12a. I do. Then, the image recognition unit 18 obtains a two-dimensional image of the electronic component P to be recognized from a plurality of one-dimensional images obtained by sequentially capturing the electronic components P to be recognized that are relatively moved with respect to the camera 9, The electronic component P is recognized based on the two-dimensional image.
[0025]
Next, an image signal correction process performed by the image correction unit 17 at the time of A / D conversion of an image signal will be described with reference to FIGS. FIG. 4A shows a pulse p <b> 1 which is detected by the pulse detection unit 14 as the timing of signal capture from the line sensor 11 and output to the image capture command unit 15 among pulse signals transmitted from the encoder 7 b of the motor 7 a. , P2, p3,...
[0026]
FIG. 4B shows image data D1, D2, D3,... Output from the line sensor 11 at each scanning time in synchronization with these pulse signals. That is, the image data D1 obtained by exposing the line sensor 11 between the pulses p1 and p2 is output using the pulse P2 as a trigger, and the image data D2 obtained by exposing the line sensor 11 between the pulses p2 and p3. Is output with the pulse P3 as a trigger.
[0027]
At this time, as shown in FIG. 4A, the intervals T1, T2,... Indicating the intervals between the pulses are not always constant, and errors and control errors of the feed screw 7c constituting the moving means of the transfer head 8 are obtained. It often varies due to various factors. When the pulse interval varies, the exposure times t1 and t2 of the line sensor 11 become longer and shorter, and the amount of charge accumulated in each pixel during this time varies, so that the output image signals D1 and D2 have different intensities. That is, even when the light receiving amounts should be equal to each other, the brightness of the one-dimensional image acquired differs due to the variation in the relative moving speed of the transfer head 8, and the image accuracy decreases.
[0028]
In order to prevent such a decrease in image accuracy, in the electronic component recognition device of the present embodiment, when the image signal is A / D-converted by the image correction unit 17, a representative voltage value of the photometric reference signal [ [VA], the voltage value VB of the image signal 22 is corrected. That is, the voltage value VC of the digital image signal output from the image correction unit 17 to the image recognition unit 18 is not the voltage value VB transmitted from the image signal capturing unit 12a, but VC = VB / [VA]. The voltage value converted by the conversion formula is output. Therefore, the image correction reference value setting unit 16 and the image correction unit 17 serve as an image signal correction unit that corrects the image signal output from the pixel of the line sensor 11 based on the electric signal output from the photometric reference pixel. ing.
[0029]
This conversion process will be described with reference to FIG. 5 for an example of the image data D1 and D2 shown in FIG. FIGS. 5A and 5C show the photometric reference signal 21 and the image signal 22 captured by the signal capturing unit 12 for each of the image signals D1 and D2. 5B and 5D are image signals output based on the photometric reference signal 21 and the image signal 22, that is, VC1 = VB1 / [VA1] and VC2 = VB2 / [VA2]. The figure shows an image signal that is converted and output.
[0030]
Here, since both the photometric reference signal 21 and the image signal 22 are electrical signals based on the amount of charge exposed during a common exposure time, the voltage value VB and the voltage value [VA] are in a proportional relationship. is there. Therefore, by setting the value obtained by dividing VB by [VA] as the image signal VC, it is possible to obtain an image signal having the voltage value VC in which the influence of the variation in the exposure time is eliminated.
[0031]
That is, in the image recognition method by the electronic component recognition device, the transfer head 8 holding the electronic component P by suction is moved relatively to the camera 9 in a direction orthogonal to the arrangement direction of the pixels of the line sensor 11 ( Relative moving step), the electronic component P is recognized based on a plurality of one-dimensional images obtained by imaging the electronic component P during the relative movement with the camera 9 (image recognition step). Then, in this image recognition step, the illumination light of the illumination unit 10 that illuminates the electronic component P passing through the imaging area of the camera 9 is guided to the pixel 11a set as the photometric reference pixel among the plurality of pixels 11a, and received. The image signal output from the pixel is corrected based on the image correction electric signal output from the pixel 11a.
[0032]
In a two-dimensional image obtained by arranging such one-dimensional images in parallel, there is no variation in brightness due to variation in the relative movement speed between the camera 9 and the imaging target. , And high-precision image recognition can be performed. Since the image signal 22 is corrected by using the photometric reference signal 21 obtained from the same illumination, the image variation caused by the temporal change of the brightness of the illumination unit 10 is appropriately corrected, and the image signal 22 is corrected. Accurate image recognition can be performed.
[0033]
【The invention's effect】
According to the present invention, in an image recognition step of recognizing an electronic component based on a plurality of one-dimensional images obtained by imaging a moving electronic component with a one-dimensional camera, an electronic component passing through an imaging area of the one-dimensional camera is used. The illumination light of the illumination unit that illuminates the component is guided to the pixel set as the photometric reference pixel among the plurality of pixels and received, and is output from the pixel based on the image correction electric signal output from the photometric reference pixel. Since the corrected image signal is corrected, it is possible to appropriately correct the variation in the image caused by the variation in the relative moving speed, and to perform the high-accuracy image recognition by eliminating the error on the image which does not exist originally. .
[Brief description of the drawings]
FIG. 1 is a perspective view of an electronic component mounting apparatus according to an embodiment of the present invention; FIG. 2 is a block diagram illustrating a configuration of an electronic component recognition apparatus according to an embodiment of the present invention; FIG. FIG. 4B is a diagram illustrating the configuration of a camera of the electronic component recognition device according to one embodiment of the present invention. FIG. 4B is a diagram illustrating the output signal of a line sensor of the electronic component recognition device according to one embodiment of the present invention. FIG. 5 is a diagram showing a signal output timing of a line sensor of the electronic component recognition device according to the embodiment. FIG. 5 is an explanatory diagram of a signal output of the line sensor of the electronic component recognition device according to one embodiment of the present invention.
7 X-axis table 8 Transfer head 9 Camera 10 Illumination unit 11 Line sensor 11a Pixel 12 Signal acquisition unit 16 Image correction reference value setting unit 17 Image correction unit 18 Image recognition unit

Claims (2)

An electronic component recognizing device that captures and recognizes an image of an electronic component held by a transfer head in an electronic component mounting apparatus, wherein the transfer head includes a nozzle that sucks and holds the electronic component, and is arranged in a row. A one-dimensional camera having a plurality of pixels, moving means for moving the transfer head relative to the one-dimensional camera in a direction orthogonal to the arrangement direction of the pixels, and an electronic component relatively moved by the moving means An image recognition unit that recognizes the electronic component based on a plurality of one-dimensional images obtained by imaging the one-dimensional camera, and an illumination unit that illuminates an electronic component that passes through an imaging area of the one-dimensional camera. A light guiding unit for guiding the illumination light of the illumination unit to a pixel set as a photometric reference pixel among the plurality of pixels; and a light output unit that outputs the illumination light from the pixel based on an electric signal output from the photometric reference pixel. Electronic component recognition apparatus characterized by comprising an image signal correcting means for correcting the image signal. An electronic component recognition method in which an image of an electronic component held by a transfer head is captured and recognized by a one-dimensional camera having a plurality of pixels arranged in a row in an electronic component mounting apparatus, wherein the electronic component is suction-held. A relative moving step of relatively moving the mounted transfer head relative to the one-dimensional camera in a direction orthogonal to the arrangement direction of the pixels, and obtaining the electronic component during the relative movement by the one-dimensional camera. An image recognizing step of recognizing the electronic component based on a plurality of one-dimensional images. In the image recognizing step, the illumination light of an illumination unit that illuminates the electronic component passing through an imaging area of the one-dimensional camera is used. Out of the pixels, the image is output from the pixel based on the electrical signal for image correction output from the photometric reference pixel and guided to the pixel set as the photometric reference pixel to receive light. Electronic component recognizing method and correcting the issue.

Images