JP2000258354A - Inspection method and device of projection part - Google Patents

Abstract

PROBLEM TO BE SOLVED: To rapidly inspect such defect as lack and position deviation for each projection by applying light to an object to be inspected from three directions and with three colors for picking up a color image, and obtaining brightness and hue images for creating a label data. SOLUTION: The illumination part of a lighting/input part 20 applies light to an object to be inspected from three different directions and with three different colors, the color image is picked up by a CCD camera 26. A brightness image creation part 32 of an image-processing part 30 creates a brightness image from a color image and performs labeling by a labeling part 34 for recording a result, and a hue image is created for labeling by a hue image creation part 36. A classification image creation part 40 refers to a classification database 38 for storing the range of defect and hue values in advance by classification and correspondence, and classifies a hue image and creates a classification image. A label creation part 42 creates a label data from them, a defect inspection part 44 inspects such defect as lack, position deviation, size, deformation, and height for each projection from them, and judges whether the object to be inspected is conforming or not.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for inspecting a projection, and more particularly, to a method of disposing solder pins on a ball grid array (BGA) board for directly mounting an electric component without a pin. It is suitable for use in the inspection of the solder ball that is placed, is arranged at a predetermined position on the surface to be inspected, the surface is mirror-like, and the outer shape is a spherical surface, or a missing or displaced projection of a part thereof, The present invention relates to a method and an apparatus for inspecting a protrusion for inspecting a defect such as a size, a deformation, and a height.

[0002]

2. Description of the Related Art There are several methods for mounting electric components on a printed circuit board. One of the methods is as shown in FIG. 1 (perspective view), FIG. 2 (top view), and FIG. 3 (side view). , B
There is a method in which a solder ball 12 is placed at a predetermined position on the GA substrate 10 where an electric component is to be mounted, the electric component is placed, and the solder ball is heated and melted to perform soldering.

In order to securely mount an electric component on such a BGA board, it is necessary that solder balls 12 of a predetermined size, shape, and height are arranged at predetermined positions on the BGA board 10. As shown in the top view of FIG. 4A, there is no part missing, or as shown in the top view of FIG. As shown in the top view of FIG. 4, the size of the solder ball 12 is different from a predetermined size, or FIG.
As shown in the top view and the side view of FIG. 4D, the shape is broken and deformed, or as shown in the side view of FIG.
If the height is higher or lower than the specified height,
Since electronic components are not securely mounted, it is necessary to inspect these defects.

For this reason, conventionally, for example, FIG. 5 (perspective view)
6 and FIG. 7 (side view), the slit light 1 irradiating the solder ball 12 from the slit light source 14 is shown.
5 illuminates from one direction, and images a slit light 15 of the slit light source 14 with a CCD camera 16 arranged in a different direction from the slit light source 14 to obtain a cross-sectional shape.
Inspection by the so-called light-section method or, as shown in FIG. 8 (perspective view) and FIG. 9 (side view), a ring-shaped light source 1
Inspection is performed by a so-called oblique illumination method in which the solder ball 12 is illuminated obliquely by the light 19 generated in 8 and an image is obtained by the CCD camera 16 arranged directly above the BGA substrate 10.

[0005]

However, in the former inspection using the light section method, the inspection of the height can be performed relatively accurately, but the inspection takes a long time because the area to be inspected is scanned. In addition, since the slit light source 14 and the CCD camera 16 must be arranged at a certain angle, the slit light source 14 is arranged directly above the substrate 10 and the camera 16 is arranged obliquely. 14 and camera 16
In both of the arrangements of FIG. 7 in which both are arranged obliquely, as shown in the lower part of FIGS. 6 and 7, there is a part which is always uninspected, and is not suitable for examination of size or deformation. For example, it is possible to reduce the uninspected portion by using both the arrangements of FIGS. 6 and 7 and using a plurality of scanning directions. However, since the number of scanning directions increases, the inspection takes more time.

On the other hand, in the latter inspection by the oblique illumination method,
Since the imaging of the inspection object does not take much time, the inspection time can be shortened as a whole. However, only the bright portion 12A shown in FIG. 10 is illuminated in the input image, and the dark portion 12B is not illuminated. Therefore, as shown in FIG. 11, only the ring-shaped portion where the illumination light is specularly reflected with respect to the camera is inspected, not the entire protrusion. Therefore, FIG.
As shown in FIG. 2 and FIG. 13, abnormalities such as deformation outside the area cannot be inspected, and particularly, height defects that cause solder errors can hardly be inspected.

Further, as a problem common to the light cutting method and the oblique illumination method, there is a problem that all defects such as lack of a projection, displacement, size, deformation, and height cannot be inspected at the same time. Was.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and it is possible to simultaneously detect defects such as missing, displacement, size, deformation, and height of each projection from one piece of color image data. It is an object of the present invention to be able to perform an inspection, and to enable an inspection at a high inspection speed with the same or higher accuracy as the conventional method.

[0009]

SUMMARY OF THE INVENTION According to the present invention, there is provided a method for inspecting a projection which is disposed at a predetermined position on a surface to be inspected and has a mirror-like surface and a spherical outer shape or a part thereof. In the method for inspecting a projection part, an object to be inspected is illuminated with colors different from each other from three different directions, a color image of the illuminated object is captured, and a brightness image and a hue image are obtained from the obtained color image. Is calculated, and label data is created from the obtained lightness image and hue image. Using the created label data, a defect including a defect, a displacement, a size, a deformation, and a height is inspected for each projection. Thus, the above-mentioned problem has been solved.

In a similar projection inspection apparatus, 3
Illuminating means for illuminating the object with different colors from two different directions, imaging means for capturing a color image of the illuminated object, and calculating a brightness image and a hue image from the obtained color image Calculating means, label data creating means for creating label data from the obtained brightness image and hue image, and using the created label data, determine the missing, positional displacement, size, deformation, height for each projection. The above problem is also solved by providing a defect inspection means for inspecting a defect including the defect.

[0011]

Embodiments of the present invention will be described below in detail with reference to the drawings.

In this embodiment, as shown in FIG. 14, an illumination unit for illuminating an inspected object from three different directions with different colors, and a color image of the illuminated inspected object as a whole. An illumination / input unit 20 including a CCD camera for capturing an image of an image, calculating a brightness image and a hue image from the color image obtained by the illumination / input unit 20, and label data from the obtained brightness image and hue image. Using the created label data, an image processing unit 30 that inspects each projection for defects such as missing, misalignment, size, deformation, and height, and the quality obtained by the image processing unit 30 A display unit 50 such as a monitor for displaying the determination result
And a control unit 60 for controlling the components.

As shown in detail in FIG. 15, the illumination / input unit 20 outputs different colors (eg, θ1) from three different directions θ1, θ2, θ3 (eg, θ1 <θ2 <θ3).
= Blue B, θ 2 = green G, θ 3 = red R), for example, ring-shaped light sources 21, 22, 23 illuminating the BGA substrate 10, which is the object to be inspected, on the stage 8. 23
A lighting unit 24 for lighting a light source;
An area sensor type color CCD camera 26 is provided for photographing a color image of the BGA substrate 10 illuminated by the reflected light of the BGA substrate 10 from right above, for example.

Note that the light sources 21, 22, and 23
As in the modification shown in FIG. 6, the light sources need not be ring-shaped light sources.

As shown in detail in FIG. 17, the image processing unit 30 creates a brightness image of a color image obtained by the CCD camera 26, and a brightness image creation unit 32 that creates the brightness image of the color image. A labeling unit 34 for performing labeling for each brightness image and recording the result.
A hue image creating unit 3 that creates a hue image of a color image obtained by the CCD camera 26 and performs labeling using the labeling result by the labeling unit 34
6, a classification database 38 in which the ranges of the defect and the hue value are stored in advance in association with the classification, and by referring to the classification database 38, the hue image created by the hue image creation unit 36 A classification image creating unit 40 that creates a classification image by assigning a classification corresponding to the hue value in units; and the number of objects, the center of gravity, and the distance between the centers of gravity obtained from the brightness image created by the brightness image creating unit 32. distance,
Label data generation by calculating label data using the area of the target, the (horizontal and vertical) fillet diameter obtained from the classified image generated by the classified image generating unit 40, and the labeling result by the labeling unit 34, and recording the label data Part 42
And a defect inspection unit 44 that performs a defect inspection, a positional deviation inspection, a size / deformation / height inspection using the label data created by the label data creation unit 42, and a defect detected by the defect inspection unit 44. And a pass / fail judgment unit 46 that rejects the whole when any one of the displacement, the size, the deformation, and the height is rejected.

In the brightness image creating section 32, as shown in FIG. 18, an R (red) image 27R, a G (green) image 27G, and a B (blue) image 27B input from the CCD camera 26 are synthesized. A portion where the sum of the R component, the G component, and the B component is equal to or larger than the threshold value is set as an inspection target corresponding to a solder ball, and a monochrome image 33 with the rest as a background.
A is created, and then the object is expanded and contracted with respect to the background, and the monochrome lightness image 3 in which the solder ball portion is clarified
Create 3B.

The labeling section 34 labels an object using the brightness image 33B created by the brightness image creating section 32, and records the result.

As shown in FIG. 19, the hue image creating section 36 combines the R image 27R, the G image 27G, and the B image 27B of the input image obtained by the CCD camera 16 and performs hue conversion in pixel units. To convert the image into a monochrome hue image 37.

More specifically, the hue value H of each pixel of the input image is obtained by using, for example, the following Haydn's hue conversion formula.

I) When R, G ≧ B, H = (GB) / (R + G−2B) (1) ii) When G, B ≧ R, H = ｛(BR) / (G + B−) 2R)｝ + 1 (2) iii) When R, B ≧ G H = {(RG) / (B + R−2G)} + 2 (3)

As a result, the hue value H is obtained as a real value between 0 and 3 and becomes as shown in FIG. Normally, in order to form an image based on the hue value, from the viewpoint of processing, a hue value newly taking an integer value between 0 and 255 from the obtained hue value H using the following equation (4). A value Himg is obtained (hereinafter, Himg is treated as a hue value).

Himg = (255/3) H (4) (Truncation after decimal point)

That is, since the light intensity of the image data in each irradiation direction obtained by photographing differs depending on the surface state (degree of inclination) of the sample, the R and G of the RGB color image synthesized based on the surface state of the sample surface. Based on the principle that the color of B becomes different, the synthesized color image corresponding to the surface state of the sample is color-coded for each pixel, and the hue of each color-coded pixel is accurately represented by a numerical value. (Referred to as hue value), and the hue value of each pixel is classified and classified with reference to a database of the correspondence between the surface state of the sample and the range of the hue value obtained in advance. And non-fatal surface flaws (referred to as pseudo defects). Therefore, the defect and the pseudo defect can be automatically classified based on the hue value, and the classification can be easily performed.

The classification image creating section 40 refers to a classification database 38 as shown in FIG. 21 in which the ranges of the defect and the hue value are associated with the classification in advance, and converts the hue-converted image data into pixel data. A classification corresponding to the hue value is given in units to create a classification image 41. In the case of FIG. 2, the hue values Himg = 0 to 41, 21 corresponding to red (R)
Hue values 42 corresponding to A and green (G) are classified into 4 to 255 categories.
Hull values 128 corresponding to B and blue (B)
213 are set as C. For convenience, A ~
Each classification of C is identified by assigning a numerical value representing the hue of each classification. In other words, each pixel has a numerical value representing the hue of each classification.

The label data generator 42 calculates the number of objects, the center of gravity, and the distance between the centers of gravity from the brightness image, and calculates label data corresponding to the area of the object and the (horizontal and vertical) fillet diameter from the classified image. And record.

If the number of objects matches the number of solder balls based on the label data created by the label data creating unit 42, the defect inspection unit 44 passes the missing inspection result. If so, the result of the missing inspection is rejected.

As shown in FIG. 22, the distances A, B, C, and D between the upper, lower, left, and right centers of gravity are obtained for each object.
As shown in FIG. 22, when the distances A, B, C, and D between the centers of gravity are all equal to or less than the upper limit and equal to or greater than the lower limit, the result of the displacement inspection is regarded as pass. As shown in the figure, if any one of the distances between the centers of gravity is outside the upper limit or the lower limit, the position shift inspection result is rejected.

Further, the area of interest and (vertical, horizontal)
The fillet diameter is compared with the upper and lower limits, respectively, and if both are within the range of the upper and lower limits, the result of the size / deformation / height inspection is considered to be acceptable, while either the area or the fillet diameter is either the upper limit or If the value is lower than the lower limit, the result of the size / deformation / height inspection is rejected. This is because, as shown in FIG. 23, the presence of a defect affects the area, fillet diameter, or the like of any of the classifications A, B, and C.

The pass / fail judgment unit 46 is provided with the defect inspection unit 4.
4 based on the defect inspection result,
If any one of the deformation and height fails,
Fail as a whole.

The display unit 50 is provided with the CCD camera 26.
And an input image input from the server, a quality determination result by the quality determination unit 46, and the like.

Hereinafter, a processing example according to the present invention will be described.

For example, as shown in FIG. 24, when the blue light B is irradiated at the lowest irradiation angle θ1, the green light G is irradiated at the next irradiation angle θ2, and the red light R is irradiated at the largest irradiation angle θ3,
As shown in FIG. 25, the input image of the CCD camera 26 viewed from the top has red at the vertex, green around it, and blue at the outermost, and the background is gray.

When the R image, the G image, and the B image of the input image shown in FIG. 25 are combined, as shown in FIG. 26, the object becomes white at a luminance of 255, for example, and the background becomes dark at a luminance value of 0, for example. An image with a certain brightness is obtained.

Using this brightness image, as shown in FIG. 27, for example, objects 1 to 4 are labeled, and the number of objects, the center of gravity of each object, and the distance between the centers of gravity of each object are obtained as label data. .

Next, as shown in FIG. 28, based on the labeling result, the hue is converted by the hue image creating section 36 only for the target portion to obtain a hue image.

Next, as shown in FIG. 29, the hue image is converted into a classified image using the data of the classification database 38. From this classification image, the area and fillet diameter of the target are obtained as label data.

FIG. 30 shows an example of label data.

In the above-described embodiment, the present invention is applied to the inspection of solder balls on a BGA substrate. However, the present invention is not limited to this, and the present invention is not limited to this. It is apparent that the present invention can be similarly applied to the inspection of a defect in a projection which is arranged at a predetermined position, has a mirror-like surface, and has a spherical outer shape, or a part thereof. Further, the color of the illumination light is not limited to a combination of red, blue, and green.

[0039]

According to the present invention, it is possible to simultaneously inspect a single piece of color image data for defects such as missing, positional deviation, size, deformation, and height of each projection. Also, since the time for inputting an image is short, the inspection speed is high. Further, it has an excellent effect that the inspection can be performed with the same or higher accuracy as the conventional method.

[Brief description of the drawings]

FIG. 1 is a perspective view showing a solder ball arranged on a BGA substrate to be inspected according to the present invention.

FIG. 2 is also a top view

FIG. 3 is a side view of the same.

FIG. 4 is a top view and a side view showing types of defects of a solder ball.

FIG. 5 is a perspective view for explaining an inspection state by a conventional light-section method.

FIG. 6 is a side view when the slit light is irradiated from directly above.

FIG. 7 is a side view showing a case where the slit light is similarly applied from obliquely above.

FIG. 8 is a perspective view for explaining an inspection state by a conventional oblique illumination method.

FIG. 9 is a side view of the same.

FIG. 10 is a top view showing an example of the input image.

FIG. 11 is a side view showing the inspection unit.

FIG. 12 is a side view showing an example of an abnormality other than the inspection unit.

FIG. 13 is a side view showing an example of a height defect.

FIG. 14 is a block diagram showing an overall configuration of an embodiment of a projection inspection apparatus according to the present invention.

FIG. 15 is a side view showing a configuration example of a lighting unit of the embodiment.

FIG. 16 is a side view showing a modification of the illumination unit.

FIG. 17 is a block diagram showing a detailed configuration of the embodiment.

FIG. 18 is a block diagram showing processing contents of a brightness image creating unit.

FIG. 19 is a block diagram showing processing contents of a hue image creation unit.

FIG. 20 is a diagram showing the hue values calculated by the hue image creating unit in association with the hue circle;

FIG. 21 is a table showing an example of a classification database;

FIG. 22 is a top view showing a comparison between a pass example and a fail example in the displacement inspection of the defect inspection unit.

FIG. 23 is a plan view for explaining the principle of defect detection in the same size / deformation / height inspection.

FIG. 24 is a side view showing the state of light irradiation on the solder balls in a specific processing example of the embodiment.

FIG. 25 is a top view showing an example of the input image.

FIG. 26 is a diagram showing an example of a brightness image.

FIG. 27 is a diagram showing a state in which a brightness image is also labeled.

FIG. 28 is a diagram showing an example of a hue image.

FIG. 29 is a diagram showing an example of a classified image.

FIG. 30 is a table showing an example of label data.

[Explanation of symbols]

 Reference Signs List 8 Stage 10 BGA board 12 Solder ball 20 Illumination / input unit 21, 22, 23 Light source 26 Color CCD camera 27R, 27G, 27B Input image 30 Image processing unit 32 Brightness image creation unit 33B Brightness image 34 Labeling unit 36 Hue image creation unit 37 Hue image 38 Classification database 40 Classification image creation unit 41 Classification image 42 Label data creation unit 44 Defect inspection unit 46 Pass / fail judgment unit 50 Display unit 60 ... Control unit

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Kenta Hayashi 1-1-1 Ichigaya Kaga-cho, Shinjuku-ku, Tokyo Dai Nippon Printing Co., Ltd. F-term (reference) 2G051 AA90 AB02 AB20 BA01 BA08 CA03 CA04 EA11 EA16 EA17 EB01 ED07 ED23 5E319 AC01 CD51 CD53 GG09

Claims (2)

[Claims] 1. A method of inspecting a projection, which is arranged at a predetermined position on a surface to be inspected, for inspecting a defect of a projection having a mirror-like surface and a spherical outer shape or a part thereof, The object to be inspected is illuminated with different colors from three different directions, a color image of the illuminated object is captured, a brightness image and a hue image are calculated from the obtained color image, and the obtained brightness is obtained. Label data is created from an image and a hue image, and the generated label data is used to inspect each projection for defects including missing, misaligned, large, small, deformed, and height defects. Inspection methods. 2. A projection inspection apparatus for inspecting a defect of a projection disposed at a predetermined position on a surface to be inspected and having a mirror-like surface and a spherical outer shape or a part thereof, Illuminating means for illuminating the test object with different colors from three different directions; imaging means for capturing a color image of the illuminated test object; calculating a brightness image and a hue image from the obtained color image Calculation means, label data creation means for creating label data from the obtained brightness image and hue image, and using the created label data, omission, displacement, size, deformation, height for each projection. And a defect inspection means for inspecting a defect including: a projection inspection device.