JPH0486562A - Speckle speed detector - Google Patents

Abstract

PURPOSE:To enable the detection of a speed at a sufficiently high accuracy even for an object to be inspected with a limited diffusing property of light such as transparent film by providing a high pass filter between two photo detecting sensors and a mutual correlation processor. CONSTITUTION:Received light signal from photo detecting sensors 4A and 4B are converted into voltages with i/v converters 7A and 7B. Then, the signals are made to pass through high pass filters 8A and 8B to remove a low frequency component and a high frequency component alone is inputted into a mutual correlation processor 5. Therefore, a high frequency component alone is subjected to a mutual correlation processing computation, which allows increase in the number of waves entering a data area where the computation processing is performed to achieve a higher detection accuracy. Moreover, a low frequency noise component as caused by an interference in a web 1 is cut hence achieving a higher accuracy in this respect too. Transmission light passing through the web 1 is transmitted through narrow slits 10A and 10B formed in a direction of a speckle translation to be guided to the sensors 4A and 4B through a prism 12 thereby enabling sharp detection of changes in quantity of light with respect to a translation speed pertaining to the direction of the speckle translation.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a speckle speed detection device for measuring the moving speed of a web or the like.

[Conventional technology]

The laser speckle velocity detection method is known, for example, from a report published in "Laser Research J, Vol. 8, No. 2, pp. 379 onwards. The basic idea is that the object to be detected has some kind of unevenness. The moving speed of the object to be detected is measured by detecting the movement of the uneven portion to be detected at a position separated by a predetermined distance.

To summarize this with reference to FIG. 1, when an object to be detected such as a web is moving to the right, a laser beam from a laser light source 2 such as He-Ne is applied to the surface of the object through a projection lens 3. irradiate. The transmitted light or reflected light of the laser light on this web 1 is detected by the light receiving sensors 4A and 4B arranged in parallel in the direction of translation of the speckles, and the amount of time shift of the detected optical signals is detected by the cross-correlation processor 5. The speed of the web 1 is detected by cross-correlation processing.

Now, the moving speed of the web 1 is V, the translational magnification is σ, the moving speed of speckles on the light receiving sensor surface is V, the separation distance between the light receiving sensors 4A and 4B is X, the light emitting lens 3 and the web 1
The distance between the beam waist 6 and the nib 1 is 2.
1 The distance between web 1 and light receiving sensors 4A and 4B is R.
Then, the time deviation amount τd of the signals detected by the light receiving sensors 4A and 4B is given by equation (3).

■−σV ・・・・・・ (1) σ=1+R/Z ・・・・・・ (2) r d =
X/v=X/aV...(3)
Here, X and σ are known.

In addition, as shown in FIG. 2, the light receiving sensor 4B shows a similar light receiving signal waveform with a time difference τd relative to the light receiving signal waveform of the light receiving sensor 4A, so if the time difference τd is measured, , (3), the target moving speed V of the web 1 can be determined.

[Problem to be solved by the invention]

Such a laser speckle speed detection method can be effectively applied to a transparent web moving at high speed in that it can be detected without contact.

However, transparent films with high smoothness such as PET (polyethylene terephthalate) and TAC (triacetate) have low diffusivity, especially in the uncoated state. Therefore, it was difficult to achieve high speed detection accuracy.

Therefore, an object of the present invention is to obtain sufficiently high speed detection accuracy even for objects to be inspected that have low light diffusivity, such as transparent films.

[Means to solve the problem]

The above-mentioned problem is to irradiate a laser beam onto an object to be detected, and to receive the light intensity fluctuation due to the parallel movement of the reflected or transmitted speckle pattern by at least two light receiving sensors spaced apart in the direction of movement of the speckles. This problem is solved by installing a high-pass filter between each of the light receiving sensors and the cross-correlation processor in a device that detects the moving speed of an object by determining the amount of time shift of the signals from the light receiving sensors through cross-correlation processing. can.

[Effect]

In detecting the moving speed of uncoated PET and TAC, the present inventors discovered that since these have low laser light diffusivity,
It was discovered that the speckle size increased and the proportion of high frequency components in the total signal was therefore low, resulting in poor detection accuracy.

However, according to the present invention, by providing a high-pass filter between each light-receiving sensor and the cross-correlation processor, low frequencies can be removed and cross-correlation processing can be performed only on high-frequency components. Detection accuracy increases as the number of waves entering the data area to be processed increases.

Moreover, since low frequency noise components caused by interference in the web are cut, accuracy can be improved from this aspect as well.

[Specific structure of the invention]

The present invention will be explained in more detail below.

FIG. 3 shows an example of the device of the present invention, in which the light receiving sensor 4
The received light signals from A and 4B are first converted into voltage by an IV converter (current-voltage converter), then passed through high-pass filters 8A and 8B to remove low frequency components, and only high frequency components are processed by a cross-correlation processor. 5 is input.

On the other hand, in this embodiment, the transmitted light of the web 1 passes through the cylindrical lens 11, the slit plate 10, and the prism 12, and is guided to each of the light receiving sensors 4A and 4B.

As can be seen from FIG. 4, the slit plate 10 includes:
Narrow slits IOA and IOB are formed in the speckle translation direction. Due to the presence of the slits 10A and IOB, as shown in Fig. 5, if the light receiving areas of the light receiving sensors 4A and 4B are 4A* and 4B* when these are not provided, it can be easily inferred from the figure. In addition, it becomes possible to sharply detect changes in the amount of light with respect to the translation speed in the speckle translation direction.

Further, a cylindrical lens 11 is disposed in front of the slit plate 10 to condense laser light in a direction perpendicular to the direction of speckle translation. Further, a prism 12 is arranged behind the slit plate 10, and
The structure is such that the laser beams transmitted through the IOA and IOB are separated and guided to the corresponding light receiving sensors 4A and 4B, respectively.

Each of the light reception signals from the light reception sensors 4A and 4B is subjected to cross-correlation processing by the correlation processor 5 as in the conventional example, and the speed of the web 1 is detected.

On the other hand, as in the above example, if a narrowing means such as the pick-up filter 10A or IOB is provided to narrow the translation direction of speckles with respect to the speckle component of the optical signal taken into the light-receiving sensor 4A14B, the light-receiving sensor 4A and 4B are
Since light is detected only in a narrow region in the translational direction, steep changes in light amount can be detected, thereby increasing velocity resolution. As a result, detection accuracy increases.

On the other hand, for the direction orthogonal to the speckle translation direction,
If a magnifying means such as the cylindrical lens 11 is provided, the target signal relative to noise can be detected with high accuracy, that is, the S/N ratio can be increased.

In addition, by using this enlarging means, it is possible to receive light over a wider range than the light receiving area defined by the light receiving area of the light receiving sensor itself, so it is easy to set and position the light receiving sensor in a direction perpendicular to the speckle translation direction. Become.

The optimum range of the cutoff frequency 50 of the high-pass filter used in the present invention was experimentally derived. As a result, it was found that the highest detection accuracy can be obtained by defining the magnification ratio, v: sample speed, d slit width).

As the light-receiving sensor in the present invention, any appropriate sensor can be used, such as a photodiode FD or an avalanche photodiode APD.

Examples include a photodiode array PDA and a solid-state image sensor CCD. Further, the enlarging means and the narrowing means in the above embodiments may be arranged in either order.

〔Example〕

Next, examples will clarify the effects of the present invention.

Speckle speed detection was carried out using the apparatus shown in FIG. 3 and under the following measurement conditions.

■Distance between beam waist and detected object; 50mm ■Distance between detected object and slit; 55mm ■Irradiation beam diameter on detected object surface; 3.6mm ■Slit length; 7III!
l ■Slit width: 40μm ■Slit interval: 800μm ■Detected object speed: 70m/mm ■Constant number of bypass filters: 10kHz
PET) bases with a thickness of 180 μm and 100 μm, and TAC (TAC) bases with a thickness of 120 μm were used.

The results are shown in FIGS. 6 and 7.

Column (P) in FIG. 6 is a raw waveform diagram from each light receiving sensor without using a high-pass filter, column (Q) is a frequency component analysis diagram thereof, and (R) is a diagram of the raw waveform from each light receiving sensor when a high-pass filter is not used. The raw waveform diagram from each light-receiving sensor is shown in the case where the sensor is used, and the column (S) is a frequency component analysis diagram thereof.

According to FIG. 6, it can be seen that by using a high-pass filter, high frequency components are dominant in the signal given to the cross-correlation process.

Column (M) in FIG. 7 shows the variation in the amount of time shift after cross-correlation processing for the same unit when no high-pass filter is used, and column (N) shows the variation in the amount of time shift after cross-correlation processing for the same unit when a high-pass filter is used according to the present invention. This figure shows the variation in the amount of time shift after cross-correlation processing for the web.

As is clear from FIG. 7, according to the present invention, the variation in the amount of time shift can be reduced in all cases, whereas in the conventional example, the variation in the amount of time shift is large, especially in the transparent film base. The reduction rate is particularly remarkable when applied to a transparent film base.

FIGS. 8 and 9 show the speed detection accuracy when the cutoff frequency fc of the high-pass filter is changed using an optical system.

A PET base (18 μm) was used as the sample.

FIG. 8 shows an example where the sample speed is 33 to 38 m/nin, and FIG. 9 shows an example where the sample speed is 65 to 70 m/win.

Table 1 shows the light reception signal frequency f ap and the high pass filter cutoff frequency f that provides the highest speed detection accuracy in each case.

Table 1 From this, fc=-fS.

Correlation signal waveform diagram, FIG. 3 is a basic configuration diagram of the device of the present invention,
Figure 4 is a cross-sectional view of the main part taken 90 degrees apart, Figure 5 is an explanatory diagram of the slit, Figure 6 is a signal waveform diagram and frequency component analysis diagram from the light receiving sensor, and Figure 7 is a conventional example. 8 and 9 are correlation diagrams between the cutoff constant of the high-pass filter and the speed detection accuracy.

■...Web (object to be inspected), 2...Laser light source, 4
A, 4B... Light receiving sensor 5... Correlation processor, 8
A, 8B...High-pass filter As described above, according to the present invention, measurement accuracy is increased and it can be particularly effectively applied to speed detection using a transparent film base.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of the principle of speckle speed detection according to the present invention, Fig. 2 is a waveform at the light receiving sensor and accuracy [%] 1 Fig. (α) (b) 1 CHA door - 2 CI - t - ~ W) to ~ ψ J A Td - j Shoulder - IL number 9's 5i Kin Hansuke 1st peso) 1 mouth to one person (100μm) Gu Tsuku upper 1ro base (120μm) Pe Sotozu 15 month base (100μm) Yu Rikue-kake 1
Brother be'-hisa (120μm) to you, cut off the mistake, (180μm) Be-a-tX Bemis (ioOBml evening, nog#::n base (120μm) pe1 Soto 1 Rohesu, +180Bml Pe~nodo!t November to Ikkyu (100μm)...Tsukutsu 1st to Moon to Miss (120μm)... ``-'''''' -1 Procedural amendment (voluntary) 7゜The column of the detailed description of the invention in the description of the amendments is amended as follows on December 17, 1990.

Claims (1)

[Claims] (1) Laser light is irradiated onto the object to be detected, and the light intensity fluctuations due to the parallel movement of the reflected or transmitted speckle pattern are received by at least two light-receiving sensors spaced apart in the direction of movement of the speckles. A device for detecting the moving speed of an object by determining the amount of time shift of signals from the sensors by cross-correlation processing, characterized in that a high-pass filter is provided between each of the light receiving sensors and the cross-correlation processor. speckle speed detection device.