JPH11183136A - Sectional and 3-dimensional shape measuring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To measure the sectional shape of an object to be measured such as an etched product in a nondestructive way at high speed by extracting the oblique deformed shape of a plane of pattern light from images obtained by a plurality of image pickup means, subjected it to opposite oblique transformation to calculate the coordinates of the location of the plane of pattern light, processing a plurality of images to synthesize the coordinates of the location of the plane of pattern light. SOLUTION: Grid pattern light 22 and 24 are projected onto a sample 20 from a plurality of directions. Sectional images 26 and 28 are formed on the sample 20, and their images are picked up from a plurality of oblique directions by two cameras 30 and 32 with partially overlapped fields of view. From the images obtained by the cameras 30 and 32, the oblique deformed shape of a plane of pattern light indicated in the image picked up by the camera 30 obliquely above the sample 20 on the left side and the image picked up by the camera 32 obliquely above the sample 20 on the right side is extracted. This is subjected to opposite oblique transformation to be a normal sectional image perpendicular with respect to the axis of the sample 20. After calculating the coordinates of the location of the plane of pattern light, a sectional shape is obtained by synthesizing the coordinates of the edge of the light plane so that images 26A and 28B may be overlaid on each other.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a cross-section and three-dimensional shape measuring apparatus, and more particularly to a cross-section or three-dimensional shape of an etching product such as a lead frame on which a semiconductor integrated circuit is mounted, a shadow mask, an aperture grill, and the like. The present invention relates to a cross-section or three-dimensional shape measuring apparatus suitable for use in measuring a cross section or a three-dimensional shape measuring device capable of non-destructively and rapidly measuring a cross-section or three-dimensional shape.

[0002]

2. Description of the Related Art As shown in FIG. 17, the etched cross-sectional shape of a metal microfabricated product such as a lead frame, a shadow mask, an aperture grill, etc. manufactured by etching, for example, the outer lead of a lead frame, is as shown in FIG. Therefore, it deviates from a desired rectangular pattern. In FIG. 17, reference numeral 10 denotes an outer lead formed by an etchant sprayed from the spray nozzle 12, and reference numeral 14 denotes a resist for leaving the outer lead during etching.

Therefore, it is necessary to predict a deviation from a desired pattern in advance and form the resist 14 with a corrected pattern.

For this purpose, it is necessary to accurately know the actual shape of the outer lead 10. Conventionally, a sample obtained by solidifying a sample with a resin is ground by a polishing machine, and a cross section of a desired portion is taken out to obtain a scanning type. Take an enlarged photo with an electron microscope,
The measurement was taking place.

[0005]

However, since it is a destructive inspection, when measuring, not only must the sample be destroyed, but the measurement is very time-consuming.
Furthermore, it is difficult to accurately expose the surface of the portion to be measured by surface polishing by a polishing machine after cutting. Scanning electron microscope photographs are good for qualitative observation, but it is difficult to use them for quantitative measurement as they are, and there is a problem that it is necessary to trace a drawing for measurement. I was

The present invention has been made to solve the above-mentioned conventional problems, and has been made in consideration of a lead frame, a shadow mask,
A first object is to measure the cross-sectional shape of an etching product or the like containing an aperture glyph at high speed without destruction.

A second object of the present invention is to measure the three-dimensional shape of the product as described above.

[0008]

According to the present invention, there is provided a cross-sectional shape measuring apparatus comprising: a pattern light projecting means for projecting predetermined pattern light onto a measurement object from a plurality of directions; A plurality of photographing means having at least partially overlapping visual fields for photographing from an oblique direction, a means for extracting an obliquely deformed shape of a pattern light surface from an image obtained by the photographing means, Means for calculating the coordinates of the pattern light surface position by inversely obliquely transforming the deformed shape, and synthesizing the coordinates of the pattern light surface position obtained by processing the images of the plurality of photographing means, and The first problem has been solved by using a means for obtaining a cross-sectional shape of the above.

Further, the pattern light is a lattice stripe pattern light.

[0010] Further, the grating pattern light is interference pattern light generated by a diffraction grating.

According to the present invention, there is further provided a cross-sectional shape measuring apparatus as described above, further comprising means for relatively moving a projection position of the pattern light with respect to the measurement object, and a measurement object obtained at each pattern light projection position. Means for obtaining a three-dimensional shape by synthesizing the cross-sectional shapes of the above-mentioned items, thereby solving the second problem.

In the present invention, as shown in FIG. 1, pattern lights 22 and 24 are projected from a plurality of directions, here two directions, onto a measurement object, here a sample 20 which is an inner lead of a lead frame. Thereby, the sample 20
Since the cross-sectional images 26 and 28 are formed on the upper surface, the cross-sectional images 26 and 28 are viewed from a plurality of oblique directions (two directions in FIG. 1) and two cameras 30 partially overlapping the field of view (here, the upper surface 20A of the sample 20). , 32. In the figure, reference numeral 31 denotes a camera 3
The shooting area 0 and the shooting area 33 are the shooting areas of the camera 32.

Next, from the images obtained by the cameras 30 and 32, FIG. 2 (image taken by the camera 30 obliquely above the left side of the sample 20) and FIG. 3 (image taken by the camera 32 obliquely above the right side of the sample 20) The oblique deformation shape of the pattern light surface as shown in FIG. 2 and 3
, 26A and 28A are images representing the same upper surface 20A of the sample 20.

Further, the obliquely deformed shape as shown in FIGS. 2 and 3 is inversely obliquely transformed according to the positions and orientations of the cameras 30 and 32 to form a regular cross-sectional image perpendicular to the axial direction of the sample 20. After calculating the coordinates of the pattern light surface position, for example, the image 26 with respect to the upper surface 20A of the sample 20 is obtained.
By synthesizing the coordinates of the light surface edges so that A and 28B just overlap and connecting the lower ends with straight lines, a complete cross-sectional shape of the sample 20 as shown in FIG. 4 can be obtained.

As shown in FIG. 5, when an adjacent sample disturbs and a complete side wall image cannot be obtained, an obliquely upper portion taken from the directions A and B in FIG. In addition to the images captured by the two cameras, a diagonally lower part (FIG. 5)
(C direction and D direction) can also be used together.

In the above description, a simple pattern light of several stripes is used as the pattern light.
The type of the pattern light is not limited to this, and an interference pattern light formed using a lattice fringe pattern or an interference fringe 39 generated by a diffraction grating 38 as shown in FIG. 6 can also be used. At this time, the spread Δy of the local maximum of the interference fringes 39 can be reduced by increasing the number of slits of the diffraction grating per unit length.

Further, during the projection of the pattern light onto the object to be measured, the object to be measured and the pattern light are relatively moved, and the cross-sectional shape of the object to be measured obtained at each pattern light projection position is synthesized to obtain a three-dimensional shape. It is also possible.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to the drawings, an embodiment of the present invention will be described in detail for the case of two-direction photographing for simplicity.

The first embodiment of the present invention relates to an apparatus for measuring a three-dimensional shape of a lead frame using a grid pattern, as shown in FIG. First and second pattern light generators 40 and 42 including light sources for projecting pattern light from two directions, and the pattern light generator 4
First and second cameras (image input units) 3 having a partially overlapped field of view for photographing the pattern light projected from 0 and 42 onto the sample from two oblique directions and inputting an image.
0, 32, an instruction input unit (for example, a keyboard or a mouse) 48 for giving various instructions, and moving mechanisms 50, 52, 54 for moving the pattern light creating units 40, 42 and the cameras 30, 32, respectively. 56, first and second image storage units (for example, frame memories) 60 and 62 for storing original images input from the first and second cameras 30 and 32, respectively, and the image storage units 60 and 62. And second for displaying the original image stored in the monitor for monitoring.
From the original images stored in the first and second image storage units 60 and 62, as shown in FIG. 2 and FIG. An oblique deformed shape extraction unit 68 for extracting a deformed shape;
A projection reverse conversion processing unit 70 that corrects the angle distortion due to oblique incidence by performing an inverse oblique transformation on the obliquely deformed shape, and, for example, a left oblique image obtained by the first camera 30 as shown in FIG. An image obtained by correcting the angle of the image viewed from above and an image obtained by correcting the angle of the image viewed obliquely from the upper right as shown in FIG. 3 are combined, and for example, the lower side is supplemented with a straight line to complete a regular cross-sectional shape. Cross-section synthesis section 72 and cross-section shape storage section 74 for storing the cross-section shape completed by cross-section synthesis section 72
A cross-sectional shape display unit 76 for displaying the cross-sectional shape stored in the cross-sectional shape storage unit 74; and a measurement result printing unit 78 for printing the cross-sectional shape stored in the cross-sectional shape storage unit 74.

A sample movement instruction given by the instruction input unit 48 is used to move a sample table 82 for fixing a sample in the vertical and horizontal directions and to rotate around a horizontal axis and a vertical axis. The sample is input to a sample moving mechanism 80 including a rotating stage, a vertical moving stage, and an XY stage, and the sample is moved. The amount of movement of the sample table 82 by the sample moving mechanism 80 is measured by the sample table coordinate measurement unit 84 and stored in the coordinate storage unit 86.

When a reference position (for example, the tip position O of the inner lead shown in FIG. 1) in the image display unit 64 or 66 is designated by the instruction input unit 48 using, for example, a pointer, the pointer coordinate calculation unit 90. The coordinates of the reference position are calculated and stored in the coordinate storage unit 86.

As shown in FIG. 8, the coordinate storage unit 86 stores coordinates x, corresponding to the three-dimensional position of the sample table 82,
y and z, and two-dimensional designated coordinates X and Y corresponding to the position of the pointer designated in the image by the designated input unit 48 are stored.

The coordinates stored in the coordinate storage unit 86 are input to the pattern light surface calculation unit 92, and are used for calculating the pattern light surface position together with the information of the pattern moving mechanism, and the position of the image input camera. A parameter storage unit 96 is input to a camera parameter calculation unit 94 that calculates information (referred to as a camera parameter) such as a direction and a direction, and the pattern light surface position calculated by the pattern light surface calculation unit 92.
And is used for the inverse projection conversion processing in the inverse projection conversion processing unit 70.

The sample table 82, the pattern light generating unit 4
An example of the positional relationship between 0 or 42 and the camera 30 or 32 is shown in FIG. In the figure, 100 is a laser light source, 102
Is a spatial filter composed of a condensing lens 104 and a pinhole 106; 108 is a collimator for collimating light passing through the spatial filter 102; 112 is for projecting a grid pattern onto the surface of the sample 20 Reference numeral 118 denotes a video camera constituting the camera 30 or 32, and reference numeral 120 denotes a lens barrel including a long working lens having a large depth of focus and a wide focusing range.

The laser light source 100 has a short wavelength, for example, 488.
An argon laser of 89 to 514.5 nm can be used.

The spatial filter 102 removes noise by condensing only desired light by a condensing lens 104 and passing it through a pinhole 106.

An aperture is provided in the long working lens 120 so that the numerical aperture NA can be adjusted.

Next, referring to FIG. 10, a processing procedure in the present embodiment will be described.

First, at step 1000, the sample 20 is set on the sample table 82.

Next, in step 1010, a camera and a pattern light are set. Specifically, as shown in FIG.
So that both side walls of the sample 20 are completely visible,
The positions and directions of the first and second cameras 30 and 32 are determined, and the first and second pattern lights 22 and 24
Simultaneously hit the sample 20 from the direction, and
In A, the pattern is positioned so that the two pattern lights 26A and 28A completely match.

Next, in step 1020, coordinates for obtaining camera parameters indicating the position and direction of the camera are input, and the position of the camera is calibrated. Specifically, as shown in FIG. 11, an S-system which is a coordinate system of a camera imaging surface (for example, a CCD), an O-system which is a coordinate system of a surface of the sample table 82, and a projection coordinate system onto the S-system O'-system (coordinate system with the origins O'x, O'y, O'z at the center of the enlarged projection when it is considered that the sample on the sample table is enlarged and photographed on the imaging surface) Are set, and the coordinates of the O-system and the corresponding S-system coordinates on the imaging surface are determined.
The camera parameters are calculated by the least square method.

More specifically, as shown in FIG. 12, the point of interest P 1 is placed on the sample table 82 irrespective of the sample 20.
(P1x, P1y, P1z) o is determined, and as shown in FIG. 13, the coordinates P1 '(P1'x, P1'y) s on the corresponding imaging surface are obtained. Hereinafter, the sample table 82 is sequentially moved up, down, left and right, and Pi (Pix, Piy, Piz) and Pi ′ (Pi′x
, Pi′y) s are sequentially obtained, and the camera parameters are estimated by the least squares method based on these data.
For example, if the number of camera parameters is 8, 8
Enter coordinates greater than the point. This calibration only needs to be performed once unless the camera is moved.

Next, the routine proceeds to step 1030, where the projection position of the pattern is determined. If the projection pattern consists of multiple parallel slit light planes, first determine a specific position on the sample, consider it as the origin, move the specific position, and bring it to a certain slit light plane The coordinates are obtained from the moving amount of the stage. In this way, the slit light surface can be determined by obtaining at least three points on the slit light surface so as to be linearly independent. Furthermore, for the slit light plane parallel to it,
By shifting the specific position again so as to be on the slit light surface, it can be determined by calculating the shift amount from the slit light surface obtained first. Hereinafter, the number of the slit light surfaces may be obtained.

If the thickness of the slit light surface at the time of projecting the pattern cannot be ignored, it is assumed that a portion having a large light amount becomes the slit light surface. That is, in the vicinity of the specific position, the point where the amount of light becomes maximum is considered to be the slit light surface. That is, the specific position described above is moved, and at the same time, the light amount is tracked on the screen to find a point where the light amount peak is formed. This can also be extended to the case where a plurality of parallel slit light surfaces are formed by repeating the same number of times as the number of slit light surfaces in the same manner as described above.

Further, even in the case where the projection pattern has a lattice shape, the position of the slit light surface can be determined by applying the above procedure to each of the vertical and horizontal slit light surfaces. .

Next, proceeding to step 1040, the sample table 82 is moved, and the sample position is adjusted so that the pattern light hits the position to be measured.

Next, the process proceeds to steps 1050 and 1052, in which the first and second cameras 30 and 32 perform section photographing in parallel.

Next, the process proceeds to steps 1060 and 1062, and an obliquely deformed shape is extracted from the images obtained by the cameras 30 and 32.

Next, the process proceeds to steps 1070 and 1072, and the actual three-dimensional coordinates of the cross-sectional position where the pattern light is applied are calculated from the camera parameters and the pattern light surface position by inverse oblique transformation.

Next, proceeding to step 1080, the three-dimensional coordinates of the cross-sectional shapes obtained in steps 1070 and 1072 are combined, and the cross-sectional shape at that position is synthesized by synthesizing the sample cross-sectional shape with the lower surface shape as a straight line. obtain.

Next, the process proceeds to step 1090, and returns to step 1050 when it is necessary to move to the next position.

On the other hand, in the judgment at step 1090,
When it is determined that the measurement at all positions has been completed, the measurement is completed.

In the present embodiment, since the pattern light formed by using the laser light source is used as the cross-section forming light to be projected on the measuring object, high-precision measurement is possible.

Next, a second embodiment of the present invention will be described in detail.

In this embodiment, as shown in FIG. 14, the same pattern plate 112 and sample table 8 as those in the first embodiment are used.
2. In the three-dimensional apparatus having the sample moving mechanism 80, the video camera 118, and the lens barrel 120, instead of the laser light source, a lamp house 130 including a highly parallel normal light source such as a xenon lamp or a mercury lamp may be used. It was made.

In the figure, 132 is an optical fiber for guiding the light generated in the lamp house 130 to the pattern plate 112.

The other points are the same as in the first embodiment, and the description is omitted.

Next, a third embodiment of the present invention in which pattern light is formed by using interference fringes 39 generated by the diffraction grating 38 as shown in FIG. 6 will be described in detail.

In this embodiment, in the same three-dimensional shape measuring apparatus as in the first embodiment, a diffraction grating plate is arranged instead of the pattern plate 112, and the interference pattern light as shown in FIG. Is projected on the screen.

In the present embodiment, since the light quantity distribution takes a very regular repetitive distribution, a plurality of light quantity distribution data are taken in while shifting the phase of the projection pattern based on the regularity, and the light quantity peak is not necessarily captured by the light quantity peak. Instead, the shape measurement can be performed based on the fluctuation of the light amount distribution.

For example, when the position of the point B in FIG. 15 is to be obtained, since the peak of the light amount distribution is not formed unlike the point A, the position of the point B is determined by detecting the peak of the light amount. I can't. However, as shown in FIG. 16, the phase difference between points A and B is estimated by calculating the light amount distribution at point B in several pattern projections shifted in phase as shown in FIG. Point A and B
The distance between points can be determined. Therefore, if the position of the point A is determined from the light amount peak, the position can be determined even for a point having an intermediate light amount between the peaks. Therefore, by this method, the number of measurement target points can be increased, and the shape of each part can be determined more precisely.

In particular, when the light source is a laser, an appropriate filter is applied by utilizing the difference between the frequency of the speckle and the frequency component of the light quantity distribution fluctuation due to the periodicity due to the other interference patterns. It also has the advantage that it can be removed.

Further, for example, by making the projection pattern a grid pattern, it is also possible to perform stereo calculation based on photographing from a plurality of cameras.

In each of the above embodiments, the shape of the tip of the lead frame was measured. However, the present invention is not limited to this, and other etching products such as a shadow mask and an aperture grill can be used. Alternatively, it is apparent that the present invention can be similarly applied to measurement of a cross-sectional shape or a three-dimensional shape of a general product other than the etching product.

[0055]

According to the present invention, it becomes possible to perform nondestructive and high-speed measurement of the cross-sectional shape of an etching product or the like.

[Brief description of the drawings]

FIG. 1 is a perspective view showing a state in which the tip of an inner lead is measured, for explaining the principle of the present invention.

FIG. 2 is a diagram showing an example of a pattern light cross-sectional shape obtained by a camera on the left side of FIG. 1;

FIG. 3 is a diagram showing an example of a pattern light cross-sectional shape obtained by a camera on the right side of FIG. 1;

FIG. 4 is a diagram showing an example of an inner lead cross-sectional shape obtained by performing an inverse oblique transformation on the shapes of FIGS. 2 and 3 and then combining the shapes;

FIG. 5 is a cross-sectional view showing a state where an image of a measurement target is photographed from four directions from above and below.

FIG. 6 is a sectional view showing a state where interference fringe pattern light is projected by a diffraction grating according to the present invention.

FIG. 7 is a block diagram showing an overall configuration according to the first embodiment of the present invention.

FIG. 8 is a table illustrating an example of coordinates stored in a coordinate storage unit according to the first embodiment;

FIG. 9 is a front view showing the overall arrangement of the first embodiment.

FIG. 10 is a flowchart showing a procedure of three-dimensional measurement according to the first embodiment;

FIG. 11 is a perspective view showing an example of a coordinate system used in the first embodiment.

FIG. 12 is a perspective view for explaining a method for determining camera parameters.

FIG. 13 is a front view of the same.

FIG. 14 is a front view showing the overall arrangement of the second embodiment of the present invention.

FIG. 15 is a perspective view for explaining a shape extraction method using a light quantity distribution.

FIG. 16 is a diagram showing a state in which the phase of the projection pattern is shifted.

FIG. 17 is a sectional view showing a state during etching for explaining the necessity of the present invention;

[Explanation of symbols]

 20 sample 22, 24 pattern light 26, 28 cross-sectional image 30, 32 camera 31, 33 photographing area 38 diffraction grating 39 interference fringe 40, 42 pattern light generator 48 instruction input unit 50, 52 , 54, 56 moving mechanism 60, 62 image storage unit 64, 66 image display unit 68 oblique transformation shape extraction unit 70 projective inverse transformation processing unit 72 cross-section synthesis unit 74 cross-sectional shape storage unit 80 sample Moving mechanism 82: sample table 84: sample table coordinate measurement unit 88: position designation input unit in image 90: pointer coordinate calculation unit 92: pattern light plane calculation unit 94: camera parameter calculation unit 100: laser light source 112: pattern plate 118 ... Video camera 130 ... Lamp house

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Teruaki Iinuma 1-1-1 Ichigaya-Kagacho, Shinjuku-ku, Tokyo Inside Dai Nippon Printing Co., Ltd. (72) Inventor Masataka Yamachi 1-chome, Ichigaya-cho, Shinjuku-ku, Tokyo No. 1 Dai Nippon Printing Co., Ltd. (72) Inventor Satoshi Watanabe 1-1-1, Ichigaya Kagacho, Shinjuku-ku, Tokyo Dai Nippon Printing Co., Ltd.

Claims (5)

[Claims] 1. A pattern light projecting means for projecting predetermined pattern light onto a measurement object from a plurality of directions, and at least a partial field of view for photographing the pattern light projected on the measurement object from a plurality of oblique directions. Duplicate,
A plurality of photographing means; a means for extracting an obliquely deformed shape of the pattern light surface from an image obtained by the photographing means; and an inverse oblique transformation of the obliquely deformed shape to obtain coordinates of the pattern light surface position. Means for calculating, and means for synthesizing the coordinates of the position of the pattern light surface obtained by processing the images of the plurality of photographing means to obtain a cross-sectional shape of the object to be measured. Shape measuring device. 2. A cross-sectional shape measuring apparatus according to claim 1, wherein said pattern light is a lattice stripe pattern light. 3. A cross-sectional shape measuring apparatus according to claim 2, wherein said lattice fringe pattern light is interference pattern light generated by a diffraction grating. 4. The cross-sectional shape measuring apparatus according to claim 1, further comprising: means for relatively moving a projection position of the pattern light with respect to the measurement target; and a measurement target obtained at each pattern light projection position. Means for obtaining a three-dimensional shape by synthesizing a cross-sectional shape of the three-dimensional shape. 5. An apparatus for measuring a cross-sectional shape and a three-dimensional shape, having a function of improving the accuracy of shape measurement using a pattern light quantity distribution.