JPS61225604A - Dimension measurement apparatus - Google Patents

Abstract

PURPOSE:To enable dimension measurement without depending on a value of the threshold level, by arranging a dimension comparing master in the vicinity of a specimen and obtaining a bit number of the difference between them. CONSTITUTION:A dimensionally known dimension compomporing master 21 and a dimensionally unknown specimen 22 are positioned with a distance S apart. And, a beam of light from a source 20 is irradiated to the master 21 and the specimen 22 simultaneously. Next, optical noise included in transmitted beams 201 from the specimen 22 and the master 21 are silenced in an image convertor 23 for conversion of a magnifying ratio of the image. Then, a solid image sensor 24 which received beams of light 202 from the convertor 23 generates a video signal 203 by converting a light quantity value per each bit into an electric intensity value for calculation of the dimension of the specimen 22.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a measuring method for measuring the dimensions of minute members.

[Prior art and problems]

In recent years, precision processing technology and automatic assembly technology have advanced.

There is a growing demand for online measurement of the dimensions of minute members with high precision.

On the other hand, with recent advances in electronic technology, solid-state image sensors are increasingly being used in various fields of optical measurement. Particularly in dimensional measurement, the object to be measured is irradiated with an original beam, the image is magnified by a lens system, the image is formed on a solid-state image sensor, and the light intensity of each electron (bit) on the sensor is measured. A common method is to obtain the dimensions by performing binarization of brightness and darkness.

Here, the method used to convert an analog video signal obtained by a solid-state image sensor into a digital 21I signal has a great influence on the measurement accuracy of dimensions. IE2
Figures (1) to (3) show conventional dimension measurement methods.

Figure 2 (1) is an original schematic diagram showing the conventional principle of dimension measurement.
10 is a light source, 11 is an object to be measured whose dimensions such as diameter are measured, 12 is a lens system, and 13 is a solid-state image sensor.

Light from a light source 10 is irradiated onto an object 11 to be measured, the light is magnified by a lens system 12, and an image is formed on an image sensor 13. FIG. 2(2) shows a video signal photoelectrically converted by the image sensor 13, and FIG. 2(3) shows a waveform diagram of a binary signal. 100 is a video signal, and 103 is a binary signal. A portion where light is blocked by the object to be measured 11 is dark, and a portion through which light is transmitted is bright. The places 101 and 102 where the brightness changes generally change with a certain slope. After determining the threshold level of this video signal 100, it is converted into a 2@ signal 103 of 0 and 1 as shown in FIG. 2 (3). The number of bits included in the dark area up to 18 is calculated and converted into size information.

At this time, as shown in FIG. 2 (2), the video signal 100
Since the change in brightness has a certain slope, when the threshold level is 14 and 15, the number of bits included in the dark area is 16 and 17, and the dimensions change depending on the value of the threshold level. It has the disadvantage of being

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to overcome these drawbacks and provide a precision measurement method that can measure dimensions without depending on the threshold level value.

[Means for solving problems]

The present invention places another dimension comparison object whose dimensions are known near the object to be measured, and simultaneously irradiates both the object to be measured and the dimension comparison object with an original light. is received by an image sensor, the video signal output from the image sensor is binarized at a certain arbitrary threshold level, and the first signal from the falling edge of the binary signal generated by the object under test to the next rising edge is detected. The size of the dimension comparison object is determined by calculating the difference between the number of bits included in the period and the number of bits included in the second period from the falling edge to the next rising edge of the binarized signal by the dimension comparison object. and the number of bits of the above-mentioned difference.

[Embodiments of the invention]

The present invention will be explained in detail below using figures.

FIG. 1 is a block diagram showing the principle of dimension measurement according to the present invention.

20 in FIG. 1 is a light transmitting unit that converts the shape of the light source and the original beam from the light source. 21 is a dimension comparison object whose dimensions are known; 22 is a measured object whose dimensions are unknown; dimension comparison object 2;
1 and the object to be measured 22 are placed a certain distance S apart. A light beam 200 emitted from a light transmitting section 20 is irradiated onto the size comparison object 21 and the object to be measured 22 at the same time. Reference numeral 23 denotes an image conversion unit, which converts transmitted light 20 from the object to be measured 22 and the dimension comparison object 21.
1. It reduces the optical noise involved in the image processing and converts the magnification of the image. The original beam 202 from the image converter 26 is received by the solid-state image sensor 24. The solid-state image sensor 24 has a one-dimensional linear array structure of 4096 bits (pixels) composed of, for example, CCD elements. A video signal 203 is created by converting the light amount value for each bit of the solid-state image sensor 24 into a voltage value, and a data processing section 25 performs various calculations to calculate dimensions.

Next, the operation of the data processing section 25 will be explained in detail.

FIG. 3(1) is a waveform diagram showing an example of a video signal obtained by the measurement method shown in FIG.

300 is a video signal, where the dark area 30 is a shadow caused by the size comparison object 21, the dark area 62 is a shadow caused by the object to be measured 22, and the bright area 61 is a shadow caused by the size comparison object 21. This is a light transmitting portion due to the gap SK' between the object 22 and the object 22 to be measured. This video signal 300 is binarized by setting a threshold level 33, but as explained in the example of FIG.
Since the dark to bright change curves 303 and 304 of 1% 302 change with a certain slope, the intersection point A-B-C-D of the threshold level 33 and the video signal depends on the value of the threshold level. It will be different.

The reason why there is a slope with a change in brightness and darkness is thought to be due to the original diffraction, aberration of the lens system, positional fluctuation of the object to be measured, etc.

FIG. 3+21 is a waveform diagram showing a binary signal.

It is clear that an accurate dimension cannot be obtained even if the number of bits X included between the CDs of the dark area 32 due to the object to be measured 22 is determined independently.

According to the experimental results, the threshold level is 10% to 3.
It has been confirmed that in the region down to approximately 0%, the slope of the change in brightness and darkness is the same at each location. Therefore, the pitch P1 between A and 0 that changes from bright to dark is constant depending on the value of the threshold level, and the pitch P2 between B and 0 that changes from dark to bright is also at the threshold level. −j” Constant Tee.

Here, the number of bits included in the first period from the falling edge of the binary signal 305 produced by the object under test 22 to the next rising edge is calculated from the falling edge of the binary signal 305 produced by the X1 size comparison object 21 to the next rising edge. Let Y be the number of bits included in the second period, and let Z be the number of bits included in the period from the rising edge of the binary signal to the next falling edge due to the gap between the dimension comparison object 21 and the measured object 22. Then, P
Since I=Y+Z and P2=Z+X, P2-P1=X-
Y, and the difference in the number of bits included in the first period and the second period does not depend on the value of the threshold level.

The dimension of the object to be measured 220 is Lx, and the dimension of the dimension comparison object 21 is L
When y is the conversion coefficient from the number of bits to the dimension determined by the constants of the optical system, etc., L x −L y = K
(X − Y), so the number of bits of the difference mentioned above and t,
By adding the values y and y using the conversion coefficient K, it is possible to obtain the value Lx that does not depend on the threshold level.

Here, the distance between the object to be measured 22 and the size comparison object 21 is an amount irrelevant to the calculation. Furthermore, the video signal 300
If there is some noise, the number of bits of the above-mentioned difference can be calculated analytically by removing the noise and approximating the measured data using the least squares method or the like.

How to perform stable measurements in the above explanation.

It is necessary to obtain a stable video signal.

FIG. 4 is an optical path diagram showing an embodiment of the light transmitting section 20 and the image converting section 23 shown in FIG.

40 is a He-N e laser, 41 is a cylindrical lens, and 42 is a plano-convex lens, which constitute the light transmitting section 20.

An elongated elliptical beam with a nearly one-dimensional spread is formed from the plano-convex lens 42, and is directed to the object to be measured 22 and the object to be compared in size 2.
1 is irradiated at the same time.

Next, 43 and 45 are plano-convex lenses, and 44 is a spatial filter, which constitute the image conversion section 23.

If the dynamic range of brightness and darkness of the video signal is widened, 2
In order to stabilize the value conversion level, a spatial filter 44 is installed to convert the beam into the elliptical beam described above and to cut harmonic noise of the light due to diffraction or the like. Furthermore, in order to improve the precision of dimension measurement, the image is magnified by a magnification determined by the ratio of the focal lengths of the plano-convex lenses 43 and 45.

Furthermore, since the original beam emitted from the light transmitting section 20 is a parallel light beam, even if the position of the object to be measured changes, it hardly affects the magnification of the optical system, and therefore does not cause measurement errors.

〔Effect of the invention〕

As is clear from the above embodiments, according to the present invention, by placing a dimension comparison object close to the object to be measured and determining the number of bits of the difference between the two, high accuracy can be achieved without depending on the threshold level. The dimensions of the object to be measured can be calculated.

[Brief explanation of drawings]

Fig. 1 is a block diagram of dimension measurement according to the present invention, Fig. 2 is an optical path diagram showing the principle of conventional dimension measurement and a waveform diagram showing two sides of a video signal, and Fig. 3 is a video signal and binary value diagram according to the present invention. FIG. 4 is an optical path diagram showing an embodiment of the optical system. 20...Light sending unit, 21...Dimension comparison object, 22...Measurement object, 23...Image conversion unit, 24...・Image sensor-1 Figure 181

Claims (1)

[Claims] The object to be measured whose dimensions are to be measured is irradiated with light, an optical image from the object to be measured is received by an image sensor, a threshold level of a video signal output from the image sensor is determined, and the threshold level of the video signal output from the image sensor is determined. In the dimension measurement method of binarizing and measuring the dimensions, a dimension comparison object whose dimensions are known is placed near the object to be measured, and both the measurement object and the dimension comparison object are At the same time, light is irradiated, and the number of bits included in the first period from the falling edge of the binary signal from the object to be measured to the next rising edge, and the number of bits included in the first period from the falling edge of the binary signal from the dimension comparison object to the next rising edge. A dimension measuring method characterized in that the dimension of the object to be measured is determined by calculating the difference between the number of bits and the number of bits included in the second period up to the rise and the dimension of the dimension comparison object.