JPH09106459A - Position deviation inspecting method for electronic parts - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve inspection accuracy by extracting a shape and a color related to inspection from the color picked-up images of electronic parts, discriminating the reliability of the measured data to normal data by fuzzy inference and inspecting its position deviation. SOLUTION: A CCD camera 10 the image of chip parts 3 positioned and fixed on a conductive pattern 2 of a wring board 1 just from the upside in color and sets an inspection window 14 slightly bigger than the chip parts 3 on the image pickup picture. Besides, an image processor 12 processes the color picked-up image of the chip parts 3 and extracts the shape and color of a main body section related to the inspection. Then, black-and-white inversion is performed to that color extracted image between the color of the chip parts 3 and the surrounding color, and a white image 3a of the chip parts 3 is extracted. Further, image feature amounts such as position and angle are measured from that white image 3a. Then, a host computer 13 discriminates the normality of the positions are angles of the chip parts 3 from the measured data of image feature amounts of position and angle, etc., by fuzzy inference to inspect its position deviation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting a positional deviation of an electronic component, particularly a chip component, which is positioned and fixed on a wiring board.

[0002]

2. Description of the Related Art When a chip component (electronic component) is positioned on a wiring board and fixed by soldering, it is necessary to inspect the positional deviation of the component.
The following is given with reference to (g). First, the first means is based on an image processing method. As shown in FIG. 3A, the chip component (3) which is positioned and soldered and fixed on the conductive pattern (2) of the wiring board (1) is fixed. A CCD camera (4) that images from above and an illumination light source (5) that irradiates the chip component (3) on the wiring board (1) with specular reflection illumination light from diagonally above.
And (5).

In the above structure, first, the light source (5) is attached to the chip part (3) which is positioned and fixed on the wiring board (1).
CCD camera (4) irradiating regular reflection illumination light from (5)
The chip component (3) is imaged by the method of FIG.
As shown in (c), the inspection window (6) is set at the regular mounting position of the chip component (3) on the imaging screen. Then, since the main body portion of the chip part (3) shown by the shaded portion in the drawing is reflected by the illumination light and is bright, the number of pixels (bright portion area) of the bright portion in the inspection window (6) is counted.
Then, a threshold value is set in consideration of an error due to illumination conditions and the like, and as shown in FIG. 3B, if the number of pixels of the bright part in the inspection window is equal to or larger than the threshold value, the entire bright part is determined. Is in the inspection window (6) and the chip part (3)
Indicates that there is no misregistration at the regular position. On the other hand, if the number of pixels in the bright portion within the inspection window is less than or equal to the threshold value, as shown in FIG. 3C, part of the bright portion protrudes outside the inspection window (6). At this time, it is determined that the chip component (3) is misaligned due to an abnormal position such as inclination or parallel misalignment.

Next, the second inspection means is based on a three-dimensional measurement method, and has a laser light source (7) as shown in FIG. 3) is scanned and its position is measured, and the positional deviation is inspected by comparing with the reference position.

The third inspection means is based on the triangulation distance measuring method, and as shown in FIG. 3 (e), it has an optical displacement sensor (9), and the displacement sensor (9) emits a laser beam. Light and reflected light are received, and the height (Ho) from the surface of the wiring board (1) to the upper surface of the chip component (3) is measured. Then, as shown in FIG. 3F, when the height displacement (H) at the XhYh coordinate position on the substrate is measured, the substrate level (H
There is a sharp change between a) and the chip level (Hb). Therefore, as shown in FIG. 3 (g), the displacement sensor (9) is scanned to draw a spiral trajectory (J), and the chip component (3) is moved from the position of the sharp change point (E) of the displacement (H). ) Edge (E)
Is detected to inspect whether or not there is a positional deviation.

[0006]

The problem to be solved is that, in the case of the image processing system, when a wide range of threshold error is considered, a slight positional deviation (for example, vertical and horizontal parallel 0.3 mm or less,
If the inclination is 5 ° or less), it cannot be judged because it falls within the error of the threshold value, and if the error of the threshold value is narrowed, the influence of lighting conditions etc. will be added and the judgment will be erroneous. There is a limit. Also, in the case of the three-dimensional measurement method, the scanning mechanism is complicated, so the cost is high (about 10 million yen higher than the image processing method), and the scanning time is long and the inspection time is image processing. There is a problem that it takes 5 to 10 times as much as the method. Further, in the case of the triangulation method, the determination can be made if the conductive pattern (2) of the wiring board (1) is a copper foil, but if it is solder plating, the laser light will be diffusely reflected and the chip component ( There is a problem that the edge portion of 3) is erroneously detected and erroneously determined.

[0007]

SUMMARY OF THE INVENTION According to the present invention, a step of color-imaging an electronic component positioned and fixed on a wiring board and extracting a shape and a color related to an inspection from a color picked-up image, and image-processing the color-extracted image. Then, the image feature amount is measured, the certainty factor of the measured data with respect to the normal data is determined by fuzzy inference, and the positional deviation of the electronic component is inspected.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of a position deviation inspection method for electronic parts according to the present invention will be described below with reference to FIGS. First, FIG. 1 (a) is an apparatus configuration diagram for carrying out the method of the present invention. In the figure, (10) is a CCD camera, and (1)
1) is a light source for illumination, (12) is an image processing device, and (13) is a host computer. The CCD camera (10) picks up a color image of the chip component (3) positioned and fixed on the conductive pattern (2) of the wiring board (1) from directly above,
As shown in FIGS. 1B and 1C, an inspection window (14) larger than the chip component (3) is set on the imaging screen. The illumination light source (11) irradiates the chip component (3) with specular reflection illumination light obliquely from above. The image processing device (12) image-processes the color picked-up image of the chip part (3), and the shape and color of the main body part related to the inspection (however, the extraction color is set to a part color such as brown or black). To extract. And FIG.
As shown in (d) and (e), the color-extracted image is inverted in black and white by the color of the chip component (3) and its surrounding color (for example, the substrate color is green), and the white image of the chip component (3) ( Take out 3a). Then, the image features such as position and angle are measured from the white image (3a), and the measurement data is sent to the host computer (13). The host computer (13) is located,
The position of the chip part (3) and the normality of the angle are determined by fuzzy inference from the measurement data of the image feature amount such as the angle, and the positional deviation is inspected.

The operation of the present invention based on the above configuration will be described below. First, the chip component (3) is positioned on the wiring board (1) and fixed by soldering. Then, the chip component (3) is irradiated with regular reflection illumination light from the light sources (11) and (11), and the CCD camera (10) is used. The chip component (3) is imaged from directly above. Then, as shown in FIGS. 1 (b) and 1 (c), the inspection window (14) is provided at the regular mounting position of the chip component (3) on the imaging screen.
Set. Next, when the shape and color of the main body portion related to the inspection of the chip part (3) shown in the shaded area in the figure are extracted by the image processing device (12), as shown in FIGS. A white image (3a) of the chip part (3) is taken out by reversing the surrounding color in black and white. And the white image (3a)
The image feature amount of the position and the angle is measured, and the measurement data is sent to the host computer (13). Therefore, from the measurement data of the above position and angle, the host computer (13)
In FIG. 1, fuzzy inference is performed to determine the normality of the position and angle of the chip component (3), and for example, the presence or absence of positional deviation of each white image (3a) (3a) of FIGS.

Here, when performing the fuzzy inference, the image feature amount (position, angle) is discriminated by the probabilistic determination means. For example, position (X) (Y) and angle (Z),
The fuzzy membership functions (Ma) (Mb) (Mc) and judgment probabilities shown in FIGS. 2 (a) (b) (c) are combined with the results of judgments made by humans (normal and misregistration data). (C) is created respectively. However, (ZRa) (ZRb)
(ZRc) is a fuzzy set of normal data of position (X) (Y) and angle (Z), and (PSa) (PSb) (PSc) is misalignment of position (X) (Y) and angle (Z). It is a fuzzy set of data. Therefore, for example, if the position (X) (Y) and the angle (Z) are measured, the membership function (Ma)
At, the certainty factors (Aa) and (Ba) for the normal and misaligned data of the measurement data at the position (X) are detected. Similarly, in the membership functions (Mb) and (Mc), the confidence levels (Ab), (Bb), and (Ac) of the measurement data of the position (Y) and the angle (Z) for the normal and misaligned data
(Bc) is detected respectively. Then the certainty factor (Aa)
The larger (Ab) (Ac), the more certainty factor (Ba) (Bb) (B
The smaller c) is, the closer the measured data is to the normal data. Therefore, the total probability of normality (A = Aa) multiplied by the certainty factors (Aa) (Ab) (Ac) of normality at the position (X) (Y) and the angle (Z)
XAbxAc) multiplied by each confidence (Ba) (Bb) (Bc) of position (X) (Y) and angle (Z) position deviation, the total probability of position deviation (B = Ba x Bb x And Bc) and compare them with the decision probability (C). Then, when A>C> B, it is determined that the position of the chip component (3) is normal, and in other cases,
It is determined to be abnormal, and the chip component (3) is determined to be displaced.

[0011]

According to the present invention, when inspecting a positional deviation of an electronic component positioned and fixed on a wiring board, a shape and a color relating to the inspection are extracted from a color picked-up image of the electronic component, and the color extracted image is imaged. Image accuracy is processed and the confidence of the measured data with respect to normal data is determined by fuzzy inference to check the positional deviation of electronic parts, which greatly improves the inspection accuracy and makes it easier for conventional devices. Since it can be added to and incorporated in, the cost increase can be suppressed to a low level.

[Brief description of the drawings]

FIG. 1A is an apparatus configuration diagram for carrying out an electronic component position shift inspection method according to the present invention. (B) is a color captured image of the electronic component according to the present invention, showing the electronic component in a normal position. (C) is a color picked-up image of the electronic component according to the present invention, showing the electronic component in an abnormal position. (D) is a black and white inverted image of the electronic component according to the present invention, showing the electronic component in a normal position.
(E) is a black and white inverted image of the electronic component according to the present invention,
It is a figure which shows the electronic component in an abnormal position.

FIG. 2A is a membership function of a position X for performing fuzzy inference in the method for inspecting the positional deviation of an electronic component according to the present invention. (B) is a membership function of the position Y for performing fuzzy inference in the method for inspecting the displacement of the electronic component according to the present invention. (C) is an angular membership function for performing fuzzy inference in the electronic component position shift inspection method according to the present invention.

FIG. 3A is a device configuration diagram of an image processing system showing an example of a conventional method of inspecting a positional deviation of an electronic component. FIG. 3B is a captured image of an electronic component in the electronic component position shift inspection device shown in FIG. 3A, showing the electronic component in a normal position. FIG. 3C is a captured image of the electronic component in the method for inspecting the positional deviation of the electronic component shown in FIG. 3A, showing the electronic component at an abnormal position. (D) is a three-dimensional measurement system configuration diagram showing another example of the conventional displacement inspection method for electronic components. (E) is a triangulation type device configuration diagram showing another example of the conventional method for inspecting positional deviation of electronic parts. FIG. 3F is a graph showing the displacement of the ordinate-horizontal coordinate positions in the positional deviation inspection device for electronic components of the triangulation type shown in FIG. FIG. 3G is a plan view showing a sensor trajectory in the position shift inspection apparatus for electronic components of the triangulation distance measuring method shown in FIG.

[Explanation of symbols]

 1 wiring board 3 electronic parts (chip parts) 10 CCD camera 12 image processing device 13 host computer

Claims (1)

[Claims] 1. A step of color-imaging an electronic component positioned and fixed on a wiring board and extracting a shape and a color related to an inspection from the color imaged image, and image-processing the color extracted image to measure an image feature amount. And a step of inspecting the positional deviation of the electronic component by determining the certainty factor of the measured data with respect to the normal data by fuzzy inference, and inspecting the positional deviation of the electronic component.