JPH07111333B2 - External shape measuring device for test objects such as tires - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an external shape measuring device for an object such as a tire.

[0002]

2. Description of the Related Art An outer shape measuring device used in a tire uniformity machine and used for measuring the outer shape of a tire is conventionally disclosed in Japanese Examined Patent Publication No. 62-232507. As shown in FIG. 6, this external shape measuring device is composed of an optical displacement meter 2 such as a laser for scanning a measurement site of a tire 1 having an uneven surface and a measurement site based on an output signal of the optical displacement meter 2. Signal correction means 3 that removes and corrects the sampling data corresponding to the uneven portion, measurement means 4 that performs a predetermined shape measurement based on the corrected signal, and output means 5 that outputs the measurement result. It

As shown in FIG. 7, the tire 1 is mounted on a tire rotating mechanism 6 and is rotationally driven about the vertical axis. The optical displacement meter 2 is supported by the tips of a pair of upper and lower support arms 7, and is arranged so that the sidewall portion of the tire 1 is sandwiched from both sides. Each support arm 7 is connected to a vertical movement mechanism 8 and a horizontal movement mechanism 9, and is driven by a vertical movement motor 10 so that it can move vertically symmetrically about the tire 1 and also by a horizontal movement motor 11. It is possible to move in the left and right directions at the same time.

The output signal of the optical displacement meter 2 is converted from an analog signal to a digital signal by the A / D converter 12,
It is read into the arithmetic processing unit 13 as sampling data. In this sampling data, when the sidewall part of the tire 1 is scanned, the uneven component corresponding to the decoration such as characters, and when the radial part is scanned,
Concavo-convex component corresponding to the tread groove is included. Therefore, after removing these unevenness components from the sampling data by the signal correcting means 3, the measuring means 4 measures the outer shape of the tire 1 such as runout or bumpy.

That is, as shown in the outline of the processing procedure in FIG. 8, first, in step S1, the data of the optical displacement meter 2 is transferred to the tire 1
Take in only one lap. In this case, if the optical displacement meter 2 scans the sidewall portion of the tire 1, the sampling data includes the uneven component due to the character.
Next, in step S2, data processing for removing the unevenness component from the sampling data is executed. Then, in step S3, the measurement mode is determined. When the runout measurement mode is selected, the process proceeds to step S4, and when the bumpy detection mode is selected, the process proceeds to step S5 to execute each process. In the runout measurement process in step S4, the runout amount is calculated from the corrected data, and in the bumpy detection process in step S5, the existence of bumpy is detected from the corrected data.

Character component (unevenness component) in step S2
Are grouped to reduce the data processing time by collecting a group of close points in the sampling data into one group, trapezoidal pattern component removal processing to remove the trapezoidal pattern component, and large mountain / small mountain pattern component. This is performed in the procedure of Oyama / Oyama pattern component removal processing. Then, the trapezoidal pattern and the large mountain / small mountain pattern to be removed are discriminated by comparing the level of the angle formed by three adjacent sampling data with a preset value.

For example, in FIG. 9A, M = tan
α (increment per unit length), for example

[0008]

[Equation 1]

At the time, the point (X _i , Y _i ) is removed. Further, in FIG. 9B, K = tan β (increment for unit length), and for example,

[0010]

[Equation 2]

At the time, the point (X _i , Y _i ) is removed.

[0012]

DISCLOSURE OF THE INVENTION Conventionally, as to the determination of the unevenness component to be removed, as shown in Expression 1 and Expression 2, the angle level formed by three adjacent sampling data is compared with a preset value. Done by However, due to the timing of the sampling data and the measurement error of the optical displacement meter 2, the reproducibility of detecting the uneven shape is not perfect, and the angle formed by the 3 data slightly changes.

Therefore, in the uneven component forming an angle close to a preset value, sometimes the repeated process of removing the uneven component is performed or sometimes the process is not activated, resulting in runout or bumpy. The reproducibility may decrease in the measurement result, and it is difficult to preset an appropriate set value. In view of such conventional problems, an object of the present invention is to enable removal of uneven components having good reproducibility without depending on subtle reproducibility of sampling data only by presetting a frequency to be removed. And

[0014]

The present invention provides a means for that purpose.
As a step , an optical displacement meter 2 for scanning a measurement site of a subject 1 such as a tire having an uneven portion on the surface, and a signal correcting means 3 for removing and correcting sampling data corresponding to the uneven portion 3
And an external shape measuring device for an object such as a tire provided with a measuring means 4 for performing a predetermined shape measurement on the basis of the corrected signal. Conversion function to convert to
It is provided with a removal function of removing a component having a frequency equal to or higher than a preset frequency and an inverse transform function of transforming the frequency domain into the time domain by inverse fast Fourier transform.

Then, the correction means 3 compares the original sampling data with the primary processing result data which has been subjected to the primary processing of the original sampling data by the converting function, the removing function and the inverse converting function, and compares the primary sampling data with the primary processing result data. With a replacement function that replaces the original sampling data that exceeds the average value of the preset data immediately before its proximity, and a function that processes the data after replacement with the conversion function, the removal function, and the inverse conversion function. Is.

[0016]

After the sampling data from the optical displacement meter is transformed into the frequency domain by the fast Fourier transform, the component having a frequency higher than the preset frequency is removed. Next, the frequency domain is transformed into the time domain by the inverse fast Fourier transform, the primary processing result data subjected to the primary processing is compared with the original sampling data, and the original sampling data exceeding the primary processing result data is close to its proximity. Replace with the average value of the preset data immediately before. Then, the data after the replacement is subjected to the primary processing again.

In this way, it is possible to remove the uneven component having good reproducibility without depending on the subtle reproducibility of the sampling data by simply setting the frequency to be removed in advance, and at the practical level of time. It is possible to remove irregularities inside. In addition, the data after the replacement with the average value is set to 1 again.
Since the process is repeated after the next process, it is possible to obtain a result that is asymptotic to the envelope of the peaks or valleys of the irregularities. Therefore, as a result of the specific component removal processing, it is possible to prevent extreme variations in processing results that are sometimes left behind or removed.

[0018]

Embodiments of the present invention will now be described in detail with reference to the drawings. The support mechanism of the optical displacement meter 2 is substantially the same as that shown in FIG. 7, and the outline of the signal processing system is substantially the same as that shown in FIG. The arithmetic processing unit 13 has an architecture incorporating a DSP or other high-speed signal processing chip, and has a conversion function of converting sampling data into a frequency domain by a fast Fourier transform and a component having a frequency higher than a preset frequency. Removal function, an inverse transform function for transforming the frequency domain into the time domain by an inverse fast Fourier transform, primary processing result data obtained by primary processing of the original sampling data by the transform function, the removal function and the inverse transform function, and A replacement function of comparing the original sampling data and replacing the original sampling data exceeding the primary processing result data with an average value of preset data immediately before the proximity,
It is possible to instantly execute the function of processing the data after the replacement by the conversion function, the removal function, and the inverse conversion function.

Next, the calculation process for measuring the outer shape of the tire 1 will be described with reference to the flowchart shown in FIG. The number of repetitions is two, the first cutoff frequency is 10 Hz, and the second cutoff frequency is 32 Hz. First, in step S6, the sampling data of the optical displacement meter 2 is fetched for one rotation of the tire 1. Next step
Fast Fourier transform (FFT) of sampling data in S7
After the conversion into the frequency domain in step S8, in step S8, a preset frequency, that is, a component of 10 Hz or higher is removed.
Then, in step S9, the data from which the frequency component of 10 Hz or higher is removed is processed by the inverse fast Fourier transform (inverse FFT) to transform the frequency domain into the time domain.

As described above, the sampling data is subjected to the fast Fourier transform, the high frequency component removal and the inverse fast Fourier transform.
When the next process is performed, the data of the primary process result becomes as shown in FIG. Therefore, the number of times of processing is determined in step S10, and if it is the first time, in step 11, the data of the primary processing result and the original sampling data are compared. Then, if the original sampling data exceeds the data of the primary processing result, the cutoff frequency is changed to 32 Hz in step S12, and then the process returns to step S7 to perform fast Fourier transform, removal of high frequency components, and inverse fast Fourier transform. Each process is repeated to perform the secondary process. Then, in step 11, the data of the secondary processing result and the original sampling data are compared and the same processing as described above is performed. At this time, if the original sampling data is smaller than the data of the secondary processing result, in step 13, as shown in FIG. 2, it is replaced with the average value of the preset data immediately before the proximity and the data is filled.

The cutoff frequency in each iterative process can be set in advance for each iterative step, and by setting an appropriate cutoff frequency in view of the characteristic frequency characteristic of the measurement target, a small N number is used for convergence. It can be done. Looking at the characteristic frequency characteristics, it is possible to converge with a small number of processings by gradually setting the cutoff frequency close to the frequency of the envelope to be obtained.

When the processing is performed in this way, as is clear from FIGS. 3A and 3B, as the number of times of processing increases, it gradually approaches a concave or convex envelope. Then, it converges substantially when N = 2, and by changing the value of N, the frequency of the unevenness component to be finally removed can be set. Moreover, 1024
Point, N = 2 processing can be executed in about 150 msec, so the processing time is negligible even in online measurement.

In the example of FIG. 3, the surface of the tire having a tread pattern consisting of the projections 14 and the grooves 15 as shown in FIG. 4 is scanned and sampled in the radial direction at the scanning position 16. The horizontal axis shows the data obtained by scanning the entire circumference from 0 to 360 °, and the vertical axis shows the scale from 0 to 10 mm, and the groove depth of the tread groove 15 is about 5 mm. It should be noted that in this example, N = 2 gives a sufficient asymptote to the envelope of the sampling data.

FIG. 5 shows a pattern when the floating character 1a on the surface of the tire 1 is scanned. The cross-sectional shape of the float 1a is shown in Fig. 5.
Although it can be represented by the five types shown in (A), the sampling data corresponding to them are various as shown in (B) of FIG. However, even with these sampling data, the uneven component can be removed by repeating the above-described processing.

In the example, the outer shape of the sidewall of the tire 1 is measured. However, if the optical displacement meter 2 is arranged on the radial side, the radial side can be measured by the same procedure. It is also possible to measure the outer shape of the. However, when removing characters on the sidewall side, the valley side (concave portion) data is acquired as the hole filling data in FIG. 2, whereas the peak side (convex portion) data is acquired on the radial side.

The subject may be something other than the tire 1.

[0027]

According to the present invention, the signal correction means 13 has a conversion function of converting sampling data into a frequency domain by a fast Fourier transform, a removal function of removing a component having a frequency higher than a preset frequency, and a frequency Since it has an inverse transform function that transforms the domain into the time domain by the inverse fast Fourier transform, it does not depend on the subtle reproducibility of the sampling data just by setting the removal frequency in advance, and the uneven component with good reproducibility is obtained. Can be removed and the signal correction means 13
In addition, the original sampling data is compared with the primary processing result data obtained by first processing the original sampling data by the conversion function, the removal function, and the inverse conversion function, and the original sampling data exceeding the primary processing result data are brought close to each other. By providing a replacement function that replaces the average value of the immediately preceding preset data, and a function that applies the converted data to the processing by the conversion function, the removal function, and the inverse conversion function, the extreme processing result by repeating the processing Can be prevented from occurring.

[Brief description of drawings]

FIG. 1 is a flowchart of a processing procedure showing an embodiment of the present invention.

FIG. 2 is an explanatory diagram of hole filling processing according to an embodiment of the present invention.

FIG. 3 is a waveform chart of a processing result showing an embodiment of the present invention.

FIG. 4 is a diagram of a tread pattern showing an embodiment of the present invention.

FIG. 5 is an explanatory diagram of floating characters and sampling data according to an embodiment of the present invention.

FIG. 6 is a block diagram showing a conventional measuring device.

FIG. 7 is a configuration diagram showing a conventional measuring device.

FIG. 8 is a flowchart showing a conventional processing procedure.

FIG. 9 is an explanatory diagram of a conventional pattern removal process.

[Explanation of symbols]

 1 subject (tire) 2 optical displacement meter 3 signal correcting means 4 measuring means 13 arithmetic processing unit

Continuation of the front page (56) Reference JP-A-2-42306 (JP, A) JP-A-3-54407 (JP, A) JP-A-3-179206 (JP, A) JP-B-1-51122 (JP , B2)

Claims (1)

[Claims] 1. An object to be inspected, such as a tire, having an uneven surface
An optical displacement meter (2) that scans the measurement site of (1), a signal correction means (3) that removes and corrects the sampling data corresponding to the uneven portion, and performs a predetermined shape measurement based on the corrected signal. In the external shape measuring device for an object such as a tire provided with the measuring means (4), the signal correcting means (3) has a conversion function of converting the sampling data into the frequency domain by the fast Fourier transform, and is preset. a removal function of removing frequency or more components, and an inverse transform function to convert the time domain by inverse fast Fourier transform from the frequency domain, the variable original sampling data to said signal correcting means (3)
Primary processing by conversion function, removal function and inverse conversion function 1
Compare the next processing result data with the original sampling data
The original sampling data that exceeds the primary processing result data
And the average value of the preset data immediately before its proximity
The exchange function to exchange and the data after the exchange are
Apply processing by conversion function, removal function and inverse conversion function
An outer shape measuring apparatus for a subject such as a tire, which is provided with a function .