JPS5835406A - Configuration measuring method utilizing moire fringes - Google Patents

Abstract

PURPOSE:To make the measurement highly accurate, by moving a projecting grating, and detecting the variation in the amount of light at a point to be measured as the phase of the Moire fringes by using a specified expression. CONSTITUTION:The mesh shaped projecting grating 4 is moved in the surface direction at a constant speed, and light is irradiated from a light source 6 through a condenser lens 5. The transmitted light is projected on a body to be checked 1 through a projecting lens 2. The reflected light is focused on a secondary image pickup element 10 through a focusing lens 3 and a relay lens 9. Its output signal is supplied to an operating circuit 11. A triangular wave is taken out of each position of the mesh and approximated by a sine wave S. A phase standard sine wave S0 is generated through an input and output device 12. The difference phi between S and S0 is computed by using the specified expression. Said computation is executed for every address in the mesh, and the configuration of the object is determined. Therefore, the measurement of the configuration is performed highly accurately.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is adjusted to the engraving column j1 by *rvH4.

With this kind of device, Hiro, #l! Rotate the grating on the light receiving side and the light receiving side, and the resulting peak reed II! There are many ways to perform light reception and analysis wm return using makametsu etc. (6). Since the information on the west side of the Hiro object cannot be obtained with this type of device, we input that information with makumetsu at the time of analysis. In addition, in order to conduct information J & 11 with only the hatare stripes that were used for yj71 II, there were many 116 strings between Peng and stripes that were hidden and 1m long. Another drawback was that it was difficult to control, and it was impossible to control it if the reflectance of the object varied depending on the location.

Hiroshi Motoya's discussion of the drawbacks of the above-mentioned conventional example and analysis #
The purpose of the present invention is to provide a glue measuring method that simplifies the calculation process and enables high III-WIL meetings with constant accuracy when there is no meeting order.

The present invention will be explained in detail with reference to the illustrated embodiment KIM.

In No. **, 1 is the object to be tested, and the projection lens is facing the object to be tested! , place line image lens ink. A projection lens 4, a condenser lens 6, and a light source 6 are arranged in sequence behind the projection lens 2. l shadow grating 4 is drive circuit 7
Faith from #4 ##cll! The light source 6 illuminates the pupil position of the projection lens 2 by means of the lens 5.

On the other hand, behind the imaging lens ink, there are arranged a grid, an imaging grating, a Delley lens 9, and a secondary ytass lens 10. The signal from the image sensor 10 is input to the confidence calculation circuit 1111, and the output line from the calculation circuit 11 is input to the moving grating 40 drive circuit 7. The training input/output device 12 is then connected to the yakuzakaku 1111. The arithmetic and arithmetic circuit 11 has arithmetic functions that control the movement direction and estimated motion amount of the projection grid 4, and analyzes the shape of the object based on the $motion amount and input data from the image sensor 10. , to have the function of displaying the intense calculation results.

First, a state in which the projection grating 4 is stationary will be explained.

Considering a point M on the iml-th test object 1, the points where M is conjugate to the projection lens 2 and the imaging lens 8 are λ' and X, respectively.
shall be. As shown in the 211th image forming grating 8, #j! Metsu$/! Think about it separately and in the mesh! to meet
Let the ground be a (a, m) as shown by m8rllK. The amount of light incident on this aK shown by the shaded area differs depending on the state in which the image of 4 is incident on 1. 4a3) shows a state in which the black portion of the projection grating 4 that does not transmit light and the similar black portion of the imaging grating 8 are superimposed, and the light amount of a is at its maximum. Also, in the 411th Th), the black part of the grid 4 and the white part of the grid 8 that transmit light overlap, resulting in the lowest amount of light, and the 41st! ! (C) shows a state in which the black part of the grid 4 and the white part of the grid 8 overlap in half, and the value of the maple grid is an intermediate value between the above two cases.

Next, #l glue child 4 is ** lens! Let us consider the case of moving to the grid 4 at a constant S station without changing the angle to, that is, the air gap. In this case, the IIk lattice 40 distribution is, for example, a white part and a black lI part, and in the case of a rectangular chart, the quantity is as shown in Figure S! It comes out like a square liquid Φ surrounding steel koji. In the 111th step, set point 1 on the weight of the object to be examined in the same way as point M, and consider b (*'%I) that meets the conjugate point B' with consideration m on the imaging grid weight,
This b (m', d) is shown in Figures 4 and 6.
It is assumed that the material has the same elasticity as 1). In this state, if you move l1jI1111I child 4 S*, you can obtain a, b (1) where the point M and point 1 are located on the plane of the plane.
j! As shown in Ryohiro @ 4■, the eclipse (the amount of light on the
When point 1 and point 1 are not within the same range, a balance 11 occurs at a%b@light quantity as shown in the 7th column. Mmm, B
If &@ is a slight unevenness difference between the first orders of moiré fringes, the value of a%btD output mmm decreases ■ As shown in ■, the order of 4 are! It represents the amount of unevenness of the points and B steps with respect to IK. l! Therefore, by keeping the phase difference between the outputs of Hatake and b constant, it is possible to detect the amount of unevenness of points and B from k.

The above explanation is based on the two addresses a%b on the Kofu image grid 6, but the same can be said about Metsu V Kokin IIK. The shape of the object can be determined by doing the same thing for other addresses for jIgtl and analyzing it taking into account connectivity. k for actually performing the above analysis is
Returning a large amount of data requires an arithmetic circuit 11 for analysis. The light intensity of each mesh with respect to the position of the projection grating 4 is output 11-by a two-dimensional imaging element 10 such as a TV camera, the change in the light intensity of each mesh with respect to the movement of the grating 4 is detected, and the number is adjusted to a predetermined address. The shape of the three-dimensional object can be calculated by detecting the photoresistance.In this direction, the reflectance of the object is different as shown in jI9■, and a shape with 44 stripes is obtained, resulting in 4111 points. In other words, by simply placing *, you can find the power that is difficult to test using relatively simple calculations.

This operation is carried out as follows with an operation path of 11 km.
First of all, the triangular wave-like output of the Hataare edition at the place where the mesh is formed lol! This is approximated to IwL liquid 1.

Here, the fluid length that is 1 falcon is the leap f: Calculate the cerebrospinal fluid -■Route 1-
2, the s11φ of the JIIIL strong fluid is detected from this one wave and the peaks of each mesh.゛Divide one period of I joy n & which is the standard into N (
N is a number of 2IllIDII (Otori>1), and the -1 axis is divided into N output sounds of the synchronized S and Yugo liquid fields. The average l11 degree of each section of the mo was divided into ro.

Pl, Pxs...PH-1. Recommended 1011m
The above method is explained, and the case of N-4 is shown. In this case, it can be generally expressed as the phase ill-ray method.

That is, in the case of N proverb 4, the monkey, φwaa m-' ((Pg-Pl) / (Po-I
's)) becomes.

This calculation is performed for each metshu's honor turtle, and the occurrence is
4. Calculate the positional deviation of each writing place from the jE value calculation. Then, it can be expressed like an object shape line method using peak-area fringes.

4=d(d-f)-11-(1/(f-I,-(1-
f)-N-p) Here, as shown in the 11th cabinet, to K is the length of each order mlL of the basket array fringes from the reference position, 4 subtraction projection lens! and the chief judge from the imaging lens i to the reference position, L
is the distance between the projection lens 2 and the imaging lens, fm*shadow lens 2. Let the focal length of the imaging lens ink, t, represent the grating pitch, and φ, the phase of the reference position.
It is possible to calculate the shape of the object by considering the irregularity of C7. At this time, the phase of the scale of Hataare Il Hiroboku 40
'.

The size 1F of the mesh of the imaging grating 6 is determined by the image sensor 10 or the arithmetic circuit 11, and regarding the direction perpendicular to the naruko, it is sufficient that the length S of one period of the grating is 1 yen, and the mesh size 1F of the lattice 6 is determined by the size 1F of the mesh. Regarding the direction, #I41II Hiro of the Chosen, being careful of the above-mentioned III Koji example, Hiro set the moving grid to *S 4, but fixed all the lattices 4 and fixed the imaging grid 8. A total of 4 meals (obtains the same effect. Projection grid 4
The same effect can be obtained even if the and imaging gratings are moved in the opposite direction, and if the movements of the respective gratings 4 and 8 are different.

There are no restrictions on the size of Metsu V3-, and K says it's okay to have a small feast. The same effect can be obtained even if an optical grating is not used for moving the imaging grating 6, and the grating is generated by an electric transmitter.

As explained above, the method of shape adjustment using the A vlaK according to the present author is based on the method of adjusting the shape using the 4-array version. It is possible to obtain a high precision IIL"?, which was a detection error, and it is also possible to obtain a track record for the object. Also, since the reflectance varies among objects, it is accurate even if the location is Kj1. The first step is to obtain such information, and the direction of the unevenness of the object can be automatically determined based on the moving direction of the grid.

[Brief explanation of drawings]

Part 1II is an example of the configuration of a device for carrying out the shape measurement method using peak array fringes according to the present invention, and Fig. 10 of the Iris 2rlA~ edition explains the origin of the Zero Derivation. Fig. 11 is a layout diagram of the 11th line optical system. Mark 4 #1 & test object, 2 is projection lens v a imaging lens, 4 is projection grating, ink is Tsundenna lens, 6 is light source,
7 is Houdou Omoji, 8 is an imaging grating, 9 is a relay lens, 1G
is the image sensor, 1N! The arithmetic circuit 12 is an input/output device. Patent Applicant: Canon Co., Ltd. Yusha No. 16 1113 Kaku G (n, m) 1411 1s5 ■ Horsetail η- 1111 Kaku

Claims (1)

[Claims] By moving the t aS grating and detecting the change in light intensity at the point to be measured as the phase of the lattice array, the grating is used to detect the shape of the object to be inspected. Shimaran Denkata Shibu 11cjl
and converting the light amount change into a phase. However, the equal fraction of the amount of behavior for one period of the N-ray projection grating is the amount of light modulated by the fringe at the point e47 at the position MKmll corresponding to the NK* of the r-th projection mm. Glue-like 11jI by Kago Are IIK, which is characterized by positioning 1 # and layering 4 martial arts! Method.