JPH0774724B2 - Inspection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Outline of the Invention] The present invention relates to an inspection object inspection method for inspecting the position, inclination, and the like of an inspection object mounted on a substrate, and simply relates to the position and inclination of the inspection object. By not only acquiring, but also by setting the window for recognizing the symbol drawn on the surface of the object to be inspected from the information of its position and inclination,
Even if the insertion position of the object to be inspected is shifted or tilted, it is possible to accurately set the window at the position where the symbol is actually drawn, and as a result, the symbol recognition is performed. This is a high-speed inspection that minimizes the amount of data to be processed.

[Industrial application field]

The present invention relates to a method for inspecting an object to be inspected, which is mounted on a board and has a symbol drawn on its surface, and more specifically, an inspection of an object to inspect the position, inclination, the above-mentioned symbol, etc. of the object to be inspected with respect to the board. Regarding the method.

[Conventional technology]

In recent years, the automation of the manufacturing process has been greatly advanced due to the improvement in the production amount of electronic devices and the like, but in many cases, the inspection process, especially the visual inspection of electronic devices, is based on visual inspection. However, such a visual inspection has a problem that it cannot cope with a rapid increase in the production amount, and therefore an electric tester for solving such a problem has been developed.
As an inspection method using this electric tester, for example, a method is known in which an image of a component mounted on a printed circuit board is picked up from above via an image pickup means, and the presence or position of the component is inspected.

However, according to such an inspection method, the object to be inspected cannot be detected with good contrast, the position of the object to be inspected attached to the substrate cannot be detected, and the object to be inspected attached to the substrate is lifted (completely inserted). There are many problems such as not being able to detect the three-dimensional posture in the case of (not performed), and it was difficult to put it into practical use as an inspection method.

Therefore, 1 which was proposed to solve such a problem
As one inspection device, a light-section projector using a light-section method as shown in FIG. 23 is known. This principle is the 23rd
As shown in the figure, first, the light from the light source 2 is converted into the light on the elongated slit through the slit 3 and the slit lens 4, and the angle is inclined with respect to the upper surface of the object 1 to be inspected (for example, 45 °). Illuminate with a slit image (slit image) 5
To form. Then, the slit image 5 is projected onto the projection lens 6
By projecting onto the projection surface 7 by means of the above, an unevenness enlarged image 8 having a cross-sectional shape obtained by cutting the surface of the object 1 to be inspected by a plane including the slit and the optical axis can be obtained.

FIG. 24 shows the configuration of a light-section inspection device that can be considered using the principle of such a light-section method. In the figure, an object to be inspected 1 such as a capacitor mounted on a printed circuit board (not shown) is
The object 1 is placed on the Y stage 11, a He-Ne laser is used as the light source 2, and a cylindrical lens is used as the slit 3. Then, the object 1 to be inspected is imaged from above by the imaging means 9 via the lens system 10. The picked-up image signal obtained by the image pickup means 9 is converted into a digital signal by the analog-digital converter 12 and stored in the image memory 13. When the computer 14 arbitrarily accesses the stored image signal, the image pickup device is displayed on the display device 15 including the CRT connected through the interface. In this captured image, the x-axis direction corresponds to the position along the substrate surface, the y-axis direction corresponds to the height from the substrate, and the slit image 37 included in the captured image has a height h at a predetermined x-position, for example.
By knowing that it is (x), the position and height of the object 1 to be inspected with respect to the printed circuit board can be obtained.

In the above configuration example, the XY stage 11 is drive-controlled by the computer 14 and can be moved vertically and horizontally as needed.

[Problems to be solved by the invention]

According to the inspection device having the above-described configuration, the height of the object 1 to be inspected can be detected, but the following various problems have occurred.

(1) There is much data to be processed, and high-speed inspection cannot be expected.

(2) An object on which a symbol is drawn, such as a key on a keyboard, or a symbol such as a tantalum capacitor. In the case where a symbol showing polarity is drawn on the surface, the symbol cannot be detected. In particular, if the insertion position of the object to be inspected with respect to the substrate is displaced or tilted, the position of the symbol drawn on the object to be inspected is also displaced, so even if the image sensing means for detecting the symbol is used. Even if they are provided separately, the position of the symbol cannot be accurately grasped, and the symbol cannot be recognized at all.

(3) The brightness of the image remarkably changes due to the gloss, color, etc. of the surface of the object to be inspected and the inclination of mounting.

(5) The surface of the object to be inspected illuminates depending on the illumination, and it is difficult to read the symbol drawn on the object to be inspected.

The present invention has been made in view of the above problems, and an object thereof is only to position or tilt the object to be inspected, even if the position of the object to be inspected is shifted or tilted. It is another object of the present invention to provide a method for inspecting an object capable of inspecting a symbol drawn on the surface at an extremely high speed.

[Means for solving problems]

In order to achieve the above object, the present invention provides a method for inspecting an object, which is mounted on a substrate and inspects the position, inclination, and the symbol of a three-dimensional object to be inspected with a symbol drawn on the surface with respect to the substrate. The following steps are provided.

(1) A step of irradiating a slit-shaped laser beam having a first wavelength by the first illuminating means to the object to be measured on the substrate at an inclined incident angle.

(2) A step of irradiating the object to be inspected with light having a second wavelength different from the first wavelength by the second illuminating means.

(3) A step of separating the lights of the first and second wavelengths, which are respectively irradiated by the first and second illumination means and reflected from the object to be inspected, by a wavelength selection mirror.

(4) Detecting the slit-shaped light of the first wavelength obtained by separating with the wavelength selection mirror, obtaining a structural image including a slit image corresponding to the slit-shaped light, and Storing in image memory.

(5) The light of the second wavelength obtained by being separated by the wavelength selection mirror is detected to obtain a symbol image including the symbol drawn on the object to be inspected, and the symbol image is stored in an image memory. Step to store.

(6) A step of setting a window at a position where a normal object to be inspected with respect to the structure image stored in the image memory, and cutting out the slit image in the window.

(7) A step of obtaining the position and inclination of the object to be inspected from the position and inclination of the cut out slit image, and determining that there is no object to be inspected when the length of the slit image is equal to or less than a specified value.

(8) A step of setting a window at a position where the symbol is drawn on the object to be inspected, based on the obtained position and inclination of the object to be inspected.

(9) A step of recognizing a symbol in the set window.

[Action]

In the present invention, two types of light are emitted to the object to be inspected on the substrate. That is, one light is a slit-shaped laser light having a first wavelength and is emitted at an inclined incident angle with respect to the object to be inspected, and the other light has a second wavelength different from the first wavelength. It is light, and the entire surface of the object to be inspected is illuminated. The lights of the first and second wavelengths are reflected by the object to be inspected, separated from each other by the wavelength selection mirror, detected by the image pickup means, etc., and stored in the image memory as a structural image and a symbol image, respectively. The structural image includes a slit image corresponding to the slit-shaped light, while the symbol image includes an image of a symbol drawn on the surface of the object to be inspected. Based on these structural images and symbol images, processing for recognizing the position, tilt, and symbol of the object to be inspected is performed as follows.

First, with respect to the structure image, a window is set at a position where a normal inspection object is present (that is, a position where the inspection object should be), and the slit image is cut out from the window. If the position of the object to be inspected is shifted or tilted, the state is reflected in the position and inclination of the slit image.By knowing the position and inclination of this slit image, the position and inclination of the object to be inspected can be determined. You can ask. At this time, when the length of the slit image is equal to or less than the specified value, the height of the object to be inspected does not reach the original height, that is, it is determined that the object to be inspected is not at that position. .

On the other hand, if the position of the object to be inspected is displaced or tilted, the position of the symbol drawn on the surface of the inspected object may also deviate from its original position, so in the present invention, in the previous step From the obtained position and inclination of the object to be inspected, a window is set at the position where the symbol is drawn on the object to be inspected (the position where the symbol is actually present), and the symbol in this window is recognized.

As described above, in the present invention, by using lights of two kinds of wavelengths, not only the position and the inclination of the object to be inspected are inspected by the light section method, but they are drawn on the surface of the object to be inspected. It is also possible to recognize symbols.

Further, since the window is always set for recognizing the position, the inclination, and the symbol of the object to be inspected, the amount of data to be subjected to the recognition processing is suppressed as much as possible, and as a result, high-speed inspection is realized.
Moreover, by using the information on the position and inclination of the object to be inspected, the window for symbol recognition can be accurately set at the position where the symbol is actually drawn, so even if the position of the object to be inspected is displaced, Even if it is tilted or inclined, accurate and high-speed inspection can be performed.

〔Example〕

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a configuration diagram of an optical system in a mounting component inspection apparatus to which an embodiment of the present invention is applied, FIG. 2 is a block diagram of the apparatus, and FIG. 3 is a diagram showing an irradiation position of slit light in the present embodiment. Is.

In FIG. 1, an object 1 to be inspected is a three-dimensional electric component having polarity such as a tantalum capacitor, and is mounted on a printed circuit board (not shown) (shown as a printed circuit board 31 in FIG. 2) to be X-shaped. -It is placed on the Y stage 11 and can be moved vertically and horizontally as appropriate under the control of a computer 14 (FIG. 2) described later. As shown in FIG. 3, on the surface of the object to be inspected 1, a symbol indicating the polarity (for example, the positive side) of the object to be inspected 1
18 is drawn.

As a first illuminating means, a light source 2 such as a He-Ne laser that outputs red (first wavelength) laser light, and a cylindrical lens 3 that converts the laser light from this light source 2 into slit-shaped light are provided. The slit-shaped light is applied to the object 1 to be inspected from obliquely above. By this light irradiation, a slit image 5a corresponding to the cut surface by the slit-shaped light is formed on the upper surface of the object to be inspected 1 along a direction orthogonal to the longitudinal direction of the object to be inspected 1. Further, the light source 2 and the cylindrical lens 3 are provided with another pair of the light source 16 and the cylindrical lens 17, which are formed in the same combination, and if necessary, another slit is formed in a direction orthogonal to the slit-shaped light. By irradiating the object-shaped object 1, another slit image 5b is formed on the upper surface of the object 1 in the direction orthogonal to the slit image 5a (longitudinal direction of the object 1). In this embodiment, the light source 2
And an illumination means including the cylindrical lens 3, and a light source
16 and the illumination means consisting of the cylindrical lens 17,
The slit image can be formed by switching appropriately according to the arrangement direction of the object to be inspected 1 and forming a slit image along a direction orthogonal to the longitudinal direction of the object to be inspected 1 like the slit image 5a shown in FIG. 1, for example. Use the other lighting method. Moreover, the irradiation position of the slit-shaped light (the position where the slit image is formed is the third
As shown in the figure, it is set at two positions (shown as first and second irradiation positions 32 and 33) in the front part and the rear part in the longitudinal direction of the test object 1. Such switching of the irradiation position is performed by the computer 14 described later by the drive motor 30 of the XY stage 11.
(Fig. 2) is driven and controlled to move the object 1 to be inspected to a desired position.

On the other hand, as a second illumination means, a fiber ring 22 is arranged above the object 1 to be inspected, the fiber ring 22 is connected to one end of a linear optical fiber 21, and the other end of the optical fiber 2 is a blue light. It is connected to the lamp house 19 via a filter 20 that transmits only light. This allows the lamp house 19
Of the light from the light source inside, only the blue (second wavelength) light is propagated through the filter 20 into the optical fiber 21, and the object to be inspected indicated by the symbol 18 is drawn from around the fiber ring 22. 1 The entire surface is irradiated onto the upper surface as indicated by an arrow. Although FIG. 1 schematically shows the optical fiber 21 and the fiber ring 22, to be precise, the optical fiber 21 is composed of a bundle of a plurality of optical fibers, one end of which is connected to the filter 20. The fiber ring 22 is formed by arranging them in contact with each other and arranging them in a ring shape with the other end facing the object to be inspected 1. The slit-shaped red laser light obtained from the first illumination means passes through the area surrounded by the fiber ring 22 and is irradiated on the object 1 to be inspected.

A red reflection mirror 23 that selectively reflects only red light and transmits light of other colors is arranged above the object to be inspected 1 as a wavelength selection mirror and is reflected from the object to be inspected 1. The red and blue lights are separated by the red reflecting mirror 23. Then, the red light reflected by the red reflection mirror 23 is imaged by the first imaging means such as a television camera through the imaging lens 24, while the blue light transmitted through the red reflection mirror 23 is imaged by the imaging lens 26. An image is taken by the second imaging means 27 such as a television camera via the. As a result, a “structure image” including the slit image 5a or 5b is obtained from the first image pickup means, while
A "symbol image" containing the signal 18 is obtained from the second image pickup means.

The structure image and the symbol image thus obtained are input to the image selection circuit 29 as shown in FIG.
The image is selected according to the command from 14. And
Image that has passed through this image selection circuit 29 (analog image)
Is converted into a digital image by the analog-to-digital converter 12 and then stored in the image memory 13. In this way, the computer 14 executes a predetermined process based on the structure image and the symbol image stored in the image memory 13 to determine the position, inclination, floating state, direction, etc. of the object 1 to be inspected with respect to the printed circuit board 31. inspect. Computer
The reference numeral 14 is preliminarily provided with data such as the type of the object to be inspected, the normal (original) position, and the direction, and these data are used as appropriate.

The processing steps performed by the computer 14 in this embodiment will be described below in more detail.

(A) First, by appropriately driving the XY stage 11, the slit light is irradiated to the first irradiation position 32 of the object 1 to be inspected, and the slit image thereof is picked up by the first image pickup means 25. Select the structural image obtained by the image selection circuit 29,
The digitally converted structure image is stored in the image memory 13.

(B) With respect to the structural image stored in the image memory 13, a window is set at a normal position of the object 1 to be inspected on the printed circuit board 31, and a slit image is cut out from this window. In FIG. 5B, the object 1 to be inspected is the printed circuit board 31.
FIG. 6 (B) shows an example of setting the window 35 when it is arranged in parallel with the printed circuit board 31.
7 is a setting example of the window 35 when the window 35 is inclined. In these examples, cutting is performed at both ends of the continuous slit image 37, and the position of the cutting is shown as the cutting portion 36.

(C) The actual position and inclination (inclination angle θ) of the object 1 to be inspected are obtained from the position and inclination of the cut-out slit image. Such a position and inclination can be obtained based on the principle of the conventional light cutting method. At the same time, the length of the slit is compared with a specified value (a value corresponding to the minimum height of the object to be inspected 1 or less), and if it is less than the specified value, it is placed on the printed circuit board. It is determined that there is no object to be inspected.

(D) Subsequently, as shown in FIG.
Whether the object 1 to be inspected is correctly inserted into the normal insertion position (normal insertion position) 33a based on the inclination (inclination angle θ) of
Alternatively, it is determined whether or not it is inserted into the next wrong insertion position 33b, 33c.

When the object 1 to be inspected is arranged parallel to the printed circuit board 31, the slit image included in the structural image becomes a parallel slit image 37 as shown in (i) of FIG. 4 (A), for example. , (I) of FIG. 4 (A), when the object 1 to be inspected is arranged at an angle of θ in the width direction as shown in FIG. 4 (b).
Since the slit image 37 is inclined as shown in i), the inclination angle θ of the object 1 to be inspected can be obtained from the inclination angle of the slit image 37. On the other hand, from the difference between the position of the slit image formed on the upper surface of the inspection object 1 and the position of the slit image formed on the printed board in the slit image, the height H of the inspection object 1 is calculated.
Can be asked.

When the tilt angle θ and the height H are obtained in this way, for example, as shown in FIG. 4 (B), the tilted state of the test object 1 is imaged by using a mathematical expression such as M ₁ = M + H sin θ. The actual insertion position M ₁ can be obtained by correcting the position M obtained in this way. Here, the above M and M ₁ indicate the position in the x-axis direction. By comparing the actual insertion position M ₁ obtained in this way with a previously known normal insertion position, it is possible to accurately detect the position of the test object 1.

(E) Next, by appropriately driving the XY stage 11, the slit light is irradiated onto the second irradiation position 33 of the object 1 to be inspected, and the slit image thereof is detected by the first image pickup means 25. Select the structural image obtained by imaging in the image selection circuit 29,
The digitally converted structure image is stored in the image memory 13.

(F) For the structure image stored in this image memory 13,
By repeating the processes similar to the above (B) to (D), the inspection of the position, inclination, presence or absence of the object 1 to be inspected is performed.

(G) Object to be inspected 1 obtained by the above processes (A) to (F)
The difference between the average height at the first irradiation position 32 and the average height at the second irradiation position 33 is calculated, and when the value is equal to or more than a predetermined specified value, the object 1 to be measured is printed on the printed circuit board 31. On the other hand, it is judged that it is a “floating defect” that has emerged.

If the object to be inspected 1 is a tantalum capacitor, the object to be inspected 1 inserted into the printed circuit board 31 has two lead electrodes. For example, one of the lead electrodes is deeply inserted and soldered. When the lead electrodes are shallowly inserted and soldered, the object 1 to be inspected is in a floating state. In such a state, as shown in FIG.
When the slit light is emitted to the second irradiation positions 32 and 33, the fourth irradiation is performed.
Since the heights h (x) and h ′ (x) in the y-axis direction in the two slit images are different from each other as shown in (iii) of FIG. (A), these h (x) and h ′ (x) are different. Corresponds to the size of the floating amount, and from this difference it can be determined whether or not it is a “floating defect”.

(H) Next, the symbol image obtained from the second image pickup means 27 is selected by the image selection circuit 29, and the digitally modulated symbol image is stored in the image memory 13.

(I) For the symbol image stored in the image memory 13, the symbol on the object 1 to be inspected is calculated based on the position, the inclination, and the lift amount of the object 1 to be inspected, which are obtained by the processes (A) to (G). Set the windows (forward direction window and reverse direction window) at the 18 positions to be drawn and the position on the opposite side. FIG. 5 (A) shows the forward and reverse windows 34 when the object 1 to be inspected is arranged parallel to the printed circuit board 31.
6A is an example of setting a and 34b, and FIG. 6A shows a positive value when the object 1 to be inspected is arranged to be inclined with respect to the printed circuit board 31,
It is a setting example of the reverse windows 34a and 34b.

(J) In the two windows set for the symbol image, the process for symbol recognition is performed, and when the symbol 18 is recognized in the forward direction window, the polarities of the object 1 to be measured are correct. If the symbol 18 is recognized in the reverse direction window, it is determined that the polarity of the object 1 to be inspected is opposite (direction is opposite).
Specifically, in the above two windows, for example, the number of bits whose brightness is equal to or larger than a predetermined value is obtained, and if the number of bits in the forward direction window is smaller than the number of bits in the backward direction window, symbol 18 indicates the backward direction window. Since it exists inside, it is determined that the direction of the object to be inspected 1 is opposite. In FIG. 5 (A) and FIG. 6 (A), the symbol 18 indicates the forward direction window 34.
Since it exists in a, it is judged to be normal. Even if the symbol 18 is a simple mark as shown in FIG. 3 or a character as shown in FIG. 6 (A), accurate direction determination is possible.

Through the series of processes (A) to (J) described above, the inspection of one inspected object such as a polar tantalum capacitor is completed. When there are a plurality of parts or the like to be inspected on the printed circuit board, the above processing is repeated for each object to be inspected.

Therefore, according to the present embodiment, the position, inclination, and
Not only is it possible to perform an accurate inspection of the floating state, etc., but since the window for symbol recognition can be set accurately from the information of the position and inclination, even if the insertion position of the object 1 to be inspected is shifted or tilted. Even if it is, the symbol drawn on the test object can be recognized, so that the direction of the test object can be accurately inspected. Moreover, not only the window for inspecting the position and tilt but also the window for symbol recognition can be set accurately, so the amount of data to be processed can be suppressed as much as possible, and the inspection time can be greatly shortened. it can.

The second lighting means adopted in the above embodiment is
As shown in FIG. 1, the object 1 to be inspected is illuminated with blue light from above by the fiber ring 22. A schematic side view and a plan view of this illumination means are shown in FIGS. 7 (A) and (B), respectively. When the object 1 to be inspected illuminated by the fiber ring 22 is attached to the printed circuit board 31 while being inclined as shown in FIG.
Illumination light 38 from 22 becomes specular reflection light, and second imaging means 27
In some cases, the upper surface of the object 1 to be inspected shines as a glossy surface 39 as shown in FIG. 7 (B), and is drawn on the surface of the object 1 to be inspected. It becomes impossible to read the symbol 18. In the symbol image obtained from the second image pickup means 27 in such a state, for example, as in the symbol image 40 shown in FIG. Not only does it become a pure white reflecting surface 41 due to the reflected light, but also the different-colored main body portion not marked is also reflected in white. In such a case, the generation of specular reflection light can be suppressed by bringing the fiber ring 22 close to the object 1 to be inspected, but the object 1 to be inspected becomes a shadow of a component arranged adjacent thereto. .

Therefore, it is desirable to configure, for example, as shown in FIG. 8 as a means for reading the symbol 18 drawn on the object 1 to be inspected with high contrast. This configuration is made in view of the fact that the object to be inspected 1 hardly tilts in the longitudinal direction with respect to the printed circuit board 31 and the object to be inspected 1 often tilts in the width direction. As shown in FIG. 8 (A), the inclination angle θ ₁ of the object 1 to be inspected is about ± 20 ° at the maximum in the width direction. The mounting direction of the object to be inspected 1 may be along the longitudinal direction of the object to be inspected 1 in the x-axis direction of the printed circuit board 31 or in the direction of y-axis. Inclination in the front-back direction (x-axis direction) or left-right direction (y-axis direction) of 31. So, in FIG.
Light-shielding plates 43 that shield the front and rear and right and left regions of the fiber ring 22 are provided so that light is not emitted from the direction in which the object 1 to be inspected is likely to tilt, that is, light is emitted from the direction in which regular reflection light is generated. It is constructed so as not to be broken.

With such a configuration, the symbol 18 drawn on the object 1 to be inspected
Can be clearly imaged by the second imaging means 27.
That is, as in the symbol image 40 shown in FIG. 9 (B), the portion where the symbol 18 is drawn on the upper surface of the object 1 to be inspected is pure white by the specular reflection light as compared with the case of FIG. 9 (A). And the symbol 18 can be read with good contrast.

In addition, the slit image is cut out from the structural image in the above-described embodiment as shown in FIG.
The left and right ends of the continuous portion of the slit image 37 are cut out as the cutout portion 36. In general, the slit image 37 is composed of the slit image 37a on the flat surface and the slit image 37b on the side surface. Therefore, in the method of cutting out at the left and right ends of the continuous portion as described above, If there is an inclination and a slit image is formed on the side surface portion, or if it is in contact with another adjacent component, the unnecessary slit image may be cut out. Therefore, there may be a problem that the position of the object to be inspected 1 is erroneously detected. Therefore, a clipping method for eliminating such an adverse effect will be described below.

In this cutting method, first cutting is performed first, and then second cutting is performed. First, the primary cutout will be described. Now, as shown in FIG. 10, x in the window 35
A position in the y-axis direction of the slit image 37 at an arbitrary position x in the axial direction will be represented as a height h (x). When there is no slit image in the window, h (x) = 0.
FIG. 11 shows the relationship between the window and the cutout position of the slit image, and FIG. 12 shows a flowchart showing the cutout processing. In Fig. 11, the origin 0 at the left end of the window 35
To the right end x _{end in the} x-axis direction to obtain the height h (x). Then, cutout (primary cutout) is performed at a position where at least one of the following conditions (i) to (iv) is satisfied, that is, at a position where there is a large difference in height or a position where the inclination changes significantly. In the example of FIG. 11, the condition (ii)
The cutouts are set at the position C ₁ that satisfies the condition and the position C ₂ that satisfies the condition (iv).

(I) | s (x) | ≧ h _diff (ii) h (x)> 0 and h (x−1) = 0 (iii) h (x) = 0 and h (x−1)> 0 (iv ) | S (x + 1) −s (x−1) | ≧ s _diff where s (x) = h (x + 1) −h (x−1), where s (x) is the inclination of the slit image. Correspond. Also h
_diff and s _diff are the height break allowance and the slope break allowance, respectively.

Such a primary cutting process will be described in detail below with reference to the flowchart of FIG.

First, in step 48 of FIG. 12, a position x in the x-axis direction, a slit flag s _flg that is a flag that is set to “1” when a slit image exists at this position x, and the left and right cutout positions. Both C ₁ and C ₂ are set to the initial value “0”. Next, in step 49, the height h (x) of the slit image at the position x is read out, and in step 50, the height h (x) is read.
It is determined whether or not (x) = 0, that is, whether or not there is a slit image at the position x.

If h (x) = 0 is determined in step 50, it means that there is no slit image at the position x.
From now on, it is determined in step 52 whether or not the slit flag s _flg = 0. If s _flg = 0, it means that the slit image has not been detected yet. Therefore, in step 53, 1 is added to the position x. Set the next position. And step 54
Then, it is determined whether or not the position x is equal to or less than the end position x _end which is the right end of the window 35. If x ≦ x _end , the process returns to step 49.

When it is determined in step 50 that h (x) = 0 is not satisfied, that is, when it is found that the slit image exists at the position x, the slit flag s _flg =
It is determined whether or not 0. If s _flg = 0, it means that the left edge of the slit image is detected, and the above condition (i
Since i) holds, proceed to step 56 and cut out the left end C
It sets the value of the position x at that time to _1, set to "1" to the slit flag s _flg, proceeds to step 53.

On the other hand, if it is determined in step 51 that s _flg = 0 is not established, that is, if the left end of the slit image has already been detected, the process proceeds to step 57, where x and x
Difference in height at −2 s (x−1) = h (x) −h (x
-2) is determined, and it is determined in step 58 whether this difference is equal to or greater than the height disconnection allowable value h _diff , that is, whether the above condition (i) is satisfied. If the insulation value of the height difference s (x-1) is smaller than the height disconnection allowable value h _diff , the process further proceeds to step 59, and the height difference s (x-1) at the position x-1.
Value (slope difference) ss = s (x-1) -s (x-3) obtained by subtracting the height difference s (x-3) at the position x-3 from the absolute value of the slope difference ss. Value slope disconnection tolerance s _diff
It is determined in step 60 whether or not the above is satisfied, that is, whether or not the above condition (iv) is satisfied. If the absolute value of the slope difference ss is smaller than the slope disconnection allowable value s _diff, it is determined that the position x at that time should not be clipped, and the process proceeds to step 53.

When the absolute value of the height difference s (x-1) is equal to or _larger than the height disconnection allowable value _{hdiff in} step 58 (that is, when the above condition (i) is satisfied), or the above step
If the absolute value of the slope difference ss is greater than or equal to the slope disconnection allowable value s _diff at 60 (that is, the above condition (iv) is satisfied), or the slit flag s _flg = 0 at step 52 above.
If it is not (that is, if the above condition (iii) is satisfied), it is the right end position where the cutout is to be performed, so the process proceeds to step 61, and the cutout right end C ₂ is one position more than the position x at that time. Set the value of the previous position x-1.
Then, in step 62, the cutout right end C ₂ and the cutout left end C ₁
X of the slit image cut out by taking the difference with
The size S in the axial direction is calculated. If the cutout position is correct, the size S should be approximately equal to the lateral width of the object to be inspected (it becomes slightly smaller as it is inclined). Therefore, in order to determine whether or not the cutout position is correct, the process proceeds to step 63, and the size S is a value (Sn × MIN) or larger, and S ≧ Sn × M
If it is IN, the two cutout positions C ₁ and C ₂ are regarded as correct, and the processing ends.

On the other hand, if S ≧ Sn × MIN in step 63, the size of the slit image cut out is abnormally small. Therefore, the position x at that time is not set as the cutout right end C ₂ but the first cutout position. Is the cutout left edge C ₁
Is correct, and it can be judged that the position already set at the left end C ₁ of the cutout is incorrect (that is, the left end position of the unnecessary slit image formed on the side surface of the object to be inspected, for example). . Therefore, in such a case, the routine proceeds to step 56, where the position x at that time is newly set to the cutout left end C ₁ and ₁ is set to the slit flag s _flg . Then, the same processing is continued until the real cutout right end C ₂ is found.

Here, the primary cutting process consisting of the above steps,
A concrete explanation will be given with reference to FIG. When the object 1 to be inspected is mounted at an inclination, as shown in FIG. 13 (A),
There will be a slit image 37b of the side surface that is not necessary for detecting the position and inclination in the window 35, but this slit image 37b of the side surface is continuous with one end of the slit image 37b of the flat surface and is a series of images. The slit image 37 is formed, and the other end of the slit image 37a on the plane portion is not continuous with any slit image. Therefore, in order to cut out only the slit image 37a of the flat portion from the inside of the window 35, first, as shown in FIG. 13 (B), the height in the y-axis direction is increased from the position of x = 0 in the x-axis direction. The height h (x) is calculated, and the position 45 at which the height h (x) changes abruptly is set as the left cutout position (cutout left end) C ₁ as shown in FIG. 13 (D). When the cutout left end C ₁ is found in this way, this time, as shown in FIG. 13 (C), the slope s (x) is obtained from the values of the heights h (x) at positions adjacent to each other. The position 46 at which the inclination s (x) greatly changes is defined as the cutout position at the right end (cutout right end) C ₂ as shown in FIG. 13 (D).

The object 1 to be inspected may be tilted in the opposite direction to that in the case of FIG. 13 (A), but in such a case, the slit image 37b on the side surface portion is located at the left end of the slit image 37a on the flat surface portion. It becomes a connected state. For such slit image, first, the left end C ₁ cut the right edge of the slit image 37b of the side portion, but results in the right end C ₂ cut the right edge of the slit image 37b, the slit image which is cut out In this manner Size S (= C ₂ −C ₁ ) is the minimum allowable size (= Sn × MI)
It will be below N), and it can be seen that it is not the correct cutting position. Therefore, in such a case, the right end of the slit image 37b on the side surface (that is, the left end of the slit image 37a on the flat surface) is cut out again as the left end C _1, and then the position where the height changes abruptly is found. , The right end of the slit image 37a of the plane portion can be cut out and set as the right end C ₂ .

Next, the secondary cutout performed after the above-mentioned primary cutout will be described with reference to the flowchart of FIG. First, in step 65, the size S (= C ₂ −C ₁ ) of the slit image cut out by the above-described primary cutting processing is obtained, and then the process proceeds to the next step 66, in which the size S is the size of the object 1 to be inspected. It is determined whether or not it is larger than a value (Sn × max) obtained by multiplying the normal width Sn of the width by an allowable maximum value max (for example, about 1.2), and if S> Sn × max, two cutting positions C ₁ , C ₂ is correct, and the whole process is terminated.

On the other hand, if S> Sn × max in step 66, the size of the cut slit image is abnormally large, so it can be determined that either of the _two cut positions C ₁ and C ₂ is incorrect. Therefore, we have to find a new cutting position. Therefore, first in step 68, a value obtained by dividing the end position x _end of the window into two is obtained, and this value is set as the center position x _{CENT of the} window. Then, step 69
, _Find the absolute value of the difference between the left edge C ₁ of the cutout and the center position x _CENT of the window, and use this value as the distance Lw from the center position x _CENT of the window to the left edge C ₁ of the cutout, and the right edge C _{2 of} the cutout and the center of the window. _Obtain the absolute value of the difference from the position x _CENT, and cut this value from the window center position x _CENT to the left end C
The distance to ₂ is Rw.

Then, in step 70, the above distances Lw and Rw are compared with each other, and if Lw <Rw, it can be determined that the position of the cutout left end C ₁ is correct, but the position of the cutout right end C ₂ is too close to the right direction. In step 72, a value obtained by adding the normal size Sn of the lateral width of the object _{1 to} the value of the cutout left end C ₁ is set as a new cutout right end C ₂ , and then the entire process is ended. On the other hand, unless Lw <Rw, since the position of the cut-out right end C ₂ is correct the position of the cut-out left C ₁ can be determined to have too close to the left, at step 71, a test from the value of the cut-out right end C ₂ A value obtained by subtracting the normal size Sn of the width of the object 1 is set as a new cutout left end C ₁ , and then the entire process is terminated. That is, two cutting positions C ₁ and C ₂
The cutout position closer to the center position x _CENT of the window is judged to be correct, and another cutout position is newly set at a position separated by Sn from that position.

15 and 17 show cutout examples obtained by the above-described primary and secondary cutout processing. FIG. 15 shows the case where the object 1 to be inspected is mounted at an angle to the substrate and is not in contact with the adjacent parts. In such a case, the cutting position obtained by the primary cutting process is performed.
C ₁ and C ₂ are the correct cutout positions, so that only the slit image 37 on the flat surface can be cut out accurately.
On the other hand, FIG. 17 shows a slit image obtained when the object to be inspected 1a adjacent to the object to be inspected 1a is in contact with the object to be inspected 1 as shown in FIG. Cutout left end C _{1 of the} cutout positions C ₁ and C ₂ obtained by
Although but shifted to the left edge of the object to be inspected 1a, it is possible to set the cut left C ₁ correct position by performing the secondary cut, thus extracting only the slit image of the flat portion of the test object 1 properly be able to.

By the way, the surface condition of the object 1 to be inspected, such as color, gloss,
If the degree of inclination is different, the brightness when capturing an image may change significantly. Therefore, there may be a problem that the stability and the inspection accuracy with respect to the fluctuation of the light amount are deteriorated. Therefore, as one means for solving such a problem, an inspection method will be described below in which an image is detected in multiple gradations and an image with the maximum brightness is extracted from these.

First, as shown in FIG. 19, the window 35 is set in the area to be inspected, and the height h (x) at each position in the x-axis direction is detected. At that time, The image is scanned in the y-axis direction at each position x, and the bit having the highest brightness is detected as a part of the slit image 37. Since the slit image 37 thus obtained has a width of several bits in the y-axis direction, the height h
(X). In addition, when the image obtained by detecting the maximum brightness is a slit image, this slit image may be detected at two or more locations. In such a case, the one with the larger image width is the real slit. Make it a statue. When the detected maximum brightness is lower than the minimum required brightness, it is assumed that the slit image 37 does not exist.

Such processing will be described in detail below with reference to FIGS. 18 and 20. Note that this process is performed for each position x in the x-axis direction.

In FIG. 18, first in step 75, the slice level SL for determining the brightness level is set to the basic slice level BSL.
(Brightness slightly higher than the background brightness of the slit image:
(See FIG. 20), the position y (y address) in the y-axis direction, the slit flag S that is set to “1” when the luminance exceeds the slice level SL, and the slit image in the y-axis direction. The slit width buffer LB indicating the width is set to the initial value 0 in both cases.

Then, in the next step 76, the luminance at the position y (y = 0 at the beginning) is measured, and the value is set to the measured luminance I. In step 77, the measured luminance I is the slice level SL (first SL =
BSL) or above. If not I ≧ SL, it is determined in step 48 whether the slit flag S is 0 or not. If S = 0, it means that the brightness equal to or higher than the slice level SL has not yet been obtained up to the current position y, so that the next position is set by adding 1 to the position y in step 79. Then, in step 80, the end position y _EN (measurement end y address) where the position y is the upper end of the window 35
If y ≦ y _EN , it returns to step 76.

Repeat this process (steps 76-80),
If it is determined in step 77 that I ≧ SL, it can be determined that the position y has reached the position of the slit image. Therefore, in step 84, the luminance I is larger than the slice level SL, or It is determined whether it is equal to the slice level SL. If I> SL, in step 85, the slice level SL is reset to be high so as to be equal to the brightness I at that time, and the slit width buffer LB is set to 0. Then, in step 89, the lower end position H of the slit image
The value of the position y at that time is set in the slit lower end buffer H ₁ B indicating ₁ (see FIG. 19), and 1 is set in the slit flag S as shown in FIG. After that, the processing shifts to the above step 79 and the same processing is repeated. That is, each time the brightness I at the new position y exceeds the slice level SL, the slice level SL is newly set high and the slit lower end buffer H ₁ B is updated with the value of the position y at that time. It will be.

If I> SL in the above step 84, that is, if I = SL, then in step 84 the slit flag S is still 0, that is, whether the position y has not yet reached the position of the slit image. , S =
If it is 0, it means that the position of the slit image is reached for the first time at the current position y, so the process of step 89 is performed. On the other hand, if S = 0 in step 86, that is, if S = 1, it means that the position y has already reached the position of the slit image, and the brightness I at the current position y is before that. Since it is equal to the brightness I at the position y−1 of the (steps 84 and 85), step 89
Is performed, that is, without updating the slit lower end buffer H ₁ B, the process directly proceeds to step 79.

If I ≧ SL is not satisfied in step 77, that is, if the brightness I is lower than the slice level SL, it is checked in step 78 whether the slit flag S is still 0, and if S = 0. , Position y has not yet reached the position of the slit image, so step 79
Go to and repeat the same process. On the other hand, if S = 1 in step 78 (see step 89), it means that the position y has already reached the position of the slit image, and then the brightness I becomes lower than the slice level SL (that is, the slit image SL). Therefore, in such a case, the process proceeds to step 81, the slit flag S is set to 0, and the slit upper end buffer H indicating the upper end position H ₂ (see FIG. 19) of the slit image is displayed. ₂ B then position y
The value of the position y−1, which is one before, is set, and the slit image is obtained by subtracting the value of the slit lower end buffer H ₁ B set in step 89 from the value of the slit upper end buffer H ₂ B. The temporary width (temporary slit width) L of

Then, in step 82, it is judged whether or not the temporary slit width L is smaller than the value of the slit width buffer LB (initially LB = 0), and if L <LB, the temporary slit width L
Is not the true width of the slit image, and the process proceeds directly to step 79. On the other hand, in step 82 above, L <LB
If not, the process proceeds to step 83 to set the value of the provisional slit width L at that time in the slit width buffer LB, and
The value of the slit lower end buffer H ₁ B at that time is the lower end position H ₁ of the slit image, and the value of the slit upper end buffer H ₂ B is the upper end position H ₂ of the slit image. After that, the routine proceeds to step 79, and the same processing is repeated.

As a result of the above processing, as shown in FIG. 20, a pixel group having a maximum luminance in the y-axis direction (a group of consecutive pixels), of which the widest pixel group is the slit image (lower end position H ₁ , The upper end position H ₂ ) is detected. And the 18th
Although not shown in the flowchart of the figure, the lower end position H ₁
And the center position of the upper end position H ₂
(X).

By such processing, the effects as shown in FIGS. 21 and 22 can be obtained. FIG. 22 (A) shows the object 1 to be inspected by irradiating the inner surface of the circle indicated by W diagonally with the laser beam by the light cutting method. The measurement example at this time is as shown in FIG. 21 (A). Is large, the slit image 37 will spread, but
Since the peak detection 89 is performed like the light intensity shown in FIG. 21B for processing, the slit image 37 becomes extremely clear as shown in FIG. 21B and is not affected by the disturbance of the image noise due to the change in the light amount. High-precision inspection is possible.

In each of the embodiments described above, only mounted components having a polarity such as a tantalum capacitor are shown as the object to be inspected, but the object to be inspected of the present invention does not necessarily have such a polarity. It does not need to be a part that has, for example, each key of a keyboard is also an inspection target. In other words, it is a shape (three-dimensional shape) that can be detected by the light-section method, and also has some symbol on its surface (for example, a symbol indicating polarity in the case of a tantalum capacitor), and in the case of each key of the keyboard. Can be inspected by the method of the present invention as long as a symbol (character, etc.) drawn on the key top is drawn. The shape that can be detected by the light section method is not limited to a simple cube or a rectangular parallelepiped, and it may be a shape such as a triangular prism, and it can be detected if there is a flat portion on the surface, that is, The shape of the object to be inspected includes a wide range that the person skilled in the art can commonly think of. Further, the position where the symbol is drawn may be drawn on any part of the surface of the object to be inspected, as long as it can be illuminated by the second illuminating means. Does not have to be drawn on a flat surface.

Further, in the above-mentioned embodiment, since the component having the polarity is used as the object to be inspected, the work for detecting the direction (polarity) is also performed, but the inspection is performed simply for the purpose of symbol recognition such as the key arrangement detection of the keyboard. In the case of, it is not necessary to detect the direction. In particular, when the direction (polarity) is detected as in the above embodiment, two windows, normal and reverse, are set in the symbol image, but only one window is set for simply performing symbol recognition. Just enough.

Further, in FIG. 1, a set of slit light illuminating means (2, 3) is provided.
Besides, one set of slit light illuminating means (16, 17) is provided, but in the present invention, it is not always necessary to provide two sets of slit light illuminating means. According to the principle of the light section method shown in FIGS. 3 and 4, it is possible to detect the position and the inclination of the object to be inspected by only one slit light illuminating means. Of course, it is useful to improve the inspection function by providing two sets of slit light illuminating means as shown in FIG. 1 so that slit images can be formed in directions intersecting with each other.

In addition, in the above embodiment, two slit images are formed on the object to be inspected as shown in FIG. 3, but this is for inspecting the floating state of the object to be inspected. If only position and tilt inspections are performed without performing, only one slit image is sufficient.

〔The invention's effect〕

According to the present invention, not only the window for slit image recognition is set at the position where the object to be inspected in the structural image, but also the position and inclination information of the object to be inspected obtained based on the slit image. By using, the window for symbol recognition can be set at the position where the symbol is actually drawn on the surface of the object to be inspected in the symbol image, so even if the insertion position of the object to be inspected is shifted or tilted. Even if it is, it is possible not only to accurately inspect the position and inclination of the object to be inspected, but also to accurately recognize the symbol on the surface. Moreover, as described above, not only the window for slit image recognition but also the window for symbol recognition can be set, so that the amount of data to be recognized can be significantly reduced. High-speed inspection can be realized.

Further, when the slit image is cut out from the window of the structural image, the cutout is performed at two positions where the slit image greatly changes (primary cutout), and the size of the object to be inspected obtained based on this is larger than the specified value. At this time, one of the above two cutting positions, which is closer to the center of the window, is set as a correct cutting position, and another position separated by the above specified value from this position to the other cutting position is set to another position.
If one correct cutout position is set (secondary cutout), only the slit image of the plane portion of the object to be inspected can be cut out accurately even when other adjacent parts are in contact with each other.

Also, when recognizing the slit image from the data in the window, scan the window in the height direction and set the widest part of the pixel group with the highest brightness as the center image of the slit image. With this, it is possible to ensure the stability of the inspection accuracy with respect to the fluctuation of the light amount.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a detection system of a mounted component inspection device to which an embodiment of the present invention is applied, FIG. 2 is a block diagram of a mounted component inspection device to which an embodiment of the present invention is applied, and FIG. FIG. 4A and FIG. 4B are perspective views of a test object showing slit light illumination positions on the test object in a detection system to which an embodiment of the present invention is applied. FIG. 6A is a diagram for explaining the determination of the normal insertion position in the case of being performed, where FIG. 6A is a plan view showing a slit image on the object to be inspected, and FIG. A side view of the object to be inspected, FIGS. 5A and 5B show examples of the symbol image and the structural image, respectively, and FIGS. 6A and 6B show other images of the symbol image and the structural image, respectively. The figure which shows an example, FIG. 7 (A), (B) is a side view and a top view which show an example of the illumination means used for the said mounting component inspection apparatus, 8 (A) and 8 (B) are a side view and a plan view showing another example of the illuminating means used in the mounted component inspection apparatus, and FIGS. 9 (A) and 9 (B) are FIGS. 7 and 8, respectively. FIG. 10 is a diagram showing an example of a symbol image obtained by using the illumination means of FIG. 10, FIG. 10 is a diagram for explaining the height h (x) of the slit image in the window, and FIG. 11 is a diagram for explaining the cutout portion in the window. FIG. 12 is a flow chart showing a primary cutting process in an embodiment of the present invention, and FIGS. 13 (A), (B), (C) and (D) are for explaining the primary cutting process of FIG. FIG. 14, FIG. 14 is a flow chart showing a secondary cutting process in one embodiment of the present invention, FIG. 15 is a plan view of a test object pattern showing a cutting method of a slit image in a window, and FIG. Fig. 17 is a diagram showing the imaging pattern when the mounted parts are in contact with each other. Fig. 17 shows the pattern of Fig. 16. FIG. 18 is a pattern diagram when the image is actually cut out, FIG. 18 is a flow chart for detecting the maximum luminance in one embodiment of the present invention, and FIG. 19 is a method for scanning the in-window slit for detecting the maximum luminance. Fig. 20 is a plane pattern diagram of the object to be measured, Fig. 20 is a diagram showing a luminance pattern along the y-axis for detecting the maximum luminance, Fig. 21 (A) is an enlarged view shown by W in Fig. 22 (A). FIG. 21 (B) is an imaging pattern diagram of a portion, FIG. 21 (B) is a slit pattern diagram in a window obtained by extracting the highest brightness from FIG. 21 (A), and FIG. 22 (A) is an oblique view of an object to be inspected. It is a perspective view of a laser beam irradiation using a light cutting method,
FIG. 22 (B) is a diagram showing the light intensity three-dimensionally, FIG. 23 is a principle diagram of a conventional light cutting method, and FIG. 24 is a configuration diagram of an inspection apparatus using the conventional light cutting method. 1 ... Object to be inspected, 2 ... Light source, 3 ... Cylindrical lens, 5a, 5b ... Slit image, 11 ... XY stage, 12 ... Analog-digital converter, 13 ... Image memory, 14 Computer, 16 Light source, 17 Cylindrical lens, 18 Symbol, 19 Lamphouse, 20 Blue transmission filter, 21 Optical fiber, 22 Fiber ring, 23 Red reflection Mirror, 24 ... Imaging lens, 25 ... First imaging means, 26 ... Imaging lens, 27 ... Second imaging means, 29 ... Image selection circuit, 30 ... Drive motor, 31 ... Printed circuit board, 34a …… forward window, 34b …… reverse window, 35 …… window, 36 …… cutout, 37 …… slit image.

Claims (7)

[Claims] 1. A method of inspecting an object for inspecting the position, inclination, and the symbol of a three-dimensional object to be inspected, which is mounted on a substrate and has a symbol drawn on the surface thereof, comprising: The step of irradiating the slit-shaped laser beam of the first wavelength with respect to the object to be measured on the substrate at an inclined incident angle, and the step of irradiating the second wavelength of the second wavelength different from the first wavelength by the second illumination means. Irradiating the entire surface of the object to be inspected with light, and mutually irradiating the light of the first and second wavelengths, which are respectively irradiated by the first and second illumination means and reflected from the object to be inspected, by a wavelength selection mirror. Separating step, and slit-shaped first obtained by separating with the wavelength selection mirror
Of light having a wavelength of, a structure image including a slit image corresponding to the slit-shaped light is obtained, and the structure image is stored in an image memory; and the structure image is separated by the wavelength selection mirror. Detecting light having a second wavelength to obtain a symbol image including the symbol drawn on the object to be inspected, and storing the symbol image in an image memory; and a structure stored in the image memory. With respect to the image, a step of setting a window at a position where a normal object to be inspected and cutting out the slit image in the window, and the position and inclination of the object to be inspected from the position and inclination of the cut out slit image And a step of determining that there is no object to be inspected when the length of the slit image is equal to or less than a specified value, from the obtained position and inclination of the object to be inspected, the symbol on the object to be inspected. Is drawn A method of inspecting an object, comprising: a step of setting a window at a position; and a step of recognizing a symbol in the set window. 2. The first illuminating means irradiates the slit-shaped laser light to two portions, namely, a front portion and a rear portion of a position where the object to be inspected is to be mounted on the substrate, and thereby the slit-shaped laser light is emitted. A structure image including two slit images corresponding to two locations is obtained, and the position, the inclination, and the floating state of the object to be inspected are detected from the positions and the inclinations of the two slit images. The method for inspecting an object according to item 1. 3. The symbol drawn on the surface of the object to be inspected is a symbol indicating the polarity of the object to be inspected, and the direction of the object to be inspected is obtained by recognizing the symbol. The method for inspecting an object according to claim 1 or 2. 4. A second window is set at a position opposite to the window set at the position where the symbol is drawn, and based on which of these two windows the symbol is recognized, The inspection method according to claim 3, wherein the direction of the inspection object is obtained. 5. When the slit image is cut out from the window set in the structural image, the cutting is performed at two positions where the slit image in the window greatly changes. The method for inspecting an object according to any one of items 4 to 4. 6. When the size of the object to be inspected obtained based on the two cutout positions is larger than a specified value, the cutout position closer to the center of the window among the two cutout positions is the correct cutout position. 6. The method for inspecting an object according to claim 5, wherein a position distant from the correct cutting position to the other cutting position by the specified value is set as another correct cutting position. . 7. When obtaining the position and inclination of the slit image cut out from the inside of the window set in the structural image, the inside of the window is scanned in the height direction at each position in the width direction, and the maximum is obtained. The test object inspection according to any one of claims 1 to 6, wherein the center position of the widest one of the pixel groups having brightness is set as the center image of the slit image. Method.