JPS62232507A - Shape measuring device - Google Patents

Abstract

PURPOSE:To measure the outer shape of an object to be inspected without selecting a measuring line even if unnecessary unevenness such as a raised printed character is present, by removing the signal pattern component predetermined in correspondence to an uneven part from the output signal of an optical displacement meter. CONSTITUTION:The output signal of an optical displacement meter 7 utilizing laser or LED is inputted to an operational processing unit 14 through an A/D converter 13. The apparatus 14 has a correction part 15 an a measuring part 16, and the correction part 5 operates so as to remove the signal pattern component (trapezoid, large ridge, small ridge) predetermined in correspondence to the raised printed character on the side wall of a tire 3 from the sampling data of the displacement meter 7. The measuring part 16 performs measurement relating to the outer shape of the tire such as line-out or bumpy on the basis of the data corrected in the correction part 15 and a result is outputted to an output part 7 and set as the judge material in the selection of a tire rank in combination with other tire measuring parameter.

Description

Detailed Description of the Invention (Industrial Field of Application) This invention is used to measure the external shape of an object, such as a shape measuring device that is used in the Taikame 7 Coniformity Machine to measure the external shape of a tire. The present invention relates to a shape measuring device for measuring crystallization.

(Conventional technology and its problems) Tire uniformity is generally defined as rigidity 1.
It is classified into three categories: size (shape), weight uniformity. Tire uniformity machines have long been known as devices for examining these uniformities, and these machines carry out tests or measurements to inspect each of the above-mentioned uniformities on tires that are transported along a predetermined inspection route. executed sequentially.

Examples of uniformity in size (shape) include runout (hereinafter referred to as runout) and bumpy, and runout includes longitudinal runout (hereinafter referred to as radial runout) that represents the variation in tire radius at the tire tread (full surface position). , and lateral runout (hereinafter referred to as lateral runout), which represents a change in tire width at the tire sidewall (lateral position). Bumpy refers to minute irregularities that appear on the tire surface. These runouts and bumpies are mainly caused by the non-uniform rigidity of the rubber/cord composite due to the uneven distribution of the tread rubber, carcass cord, belt, and other components that make up the tire. This is due to variations in the tire manufacturing process.

FIGS. 25 and 26 are structural explanatory diagrams schematically showing a conventional shape measuring device for measuring the outer shape of both side surfaces of a tire for the purpose of inspecting the runout and bumpy. FIG. 25 shows a contact type using a potentiometer 1 as a displacement meter, and FIG. 26 shows a non-contact type using a capacitive displacement meter 2 as a displacement meter. In these devices, the tie V73 as the subject is attached to the tire rotation mechanism 4 and rotated at a predetermined number of rotations.

The potentiometer 1 and the electrostatic capacitance ω-type displacement 42 are supported at the tip of the support arm 5, are arranged to sandwich both sides of the tire 3, and scan both sides of the tire 3. In the case of the contact type shown in FIG.
- A) The shape of the side surface of the tire 3 is measured by capturing the vibration of 6 with the potentiometer 1, and in the case of the non-contact type shown in Figure 26, the change in capacitance due to distance variation is measured using the capacitance displacement By making two measurements in total, the shape of the side surface of the tire 3 is measured.

However, the conventional device shown in FIG. 25 using the potentiometer 1 has a problem in that when the rotational speed of the tire 3 is increased to perform measurements, the bumpy waveform cannot be followed due to the bounce of the roller 6. For example, the spring constant of a potentiometer is 2
4 K9 f / 20 pm is the limit, and not only does it take time to measure, but when data sampling is performed at the same time as the stiffness test,
Failure to comply with 6 Orpm of ASO standards. In addition, with the conventional device shown in Fig. 26 that uses the capacitive displacement meter 2, the measurement spot diameter is as large as approximately 10#I, making it difficult to accurately capture minute irregularities on the surface of the tie 17. Furthermore, the tire surface is curved. Therefore, errors are likely to occur in large measurement spots.

A more important problem is that, as shown in the schematic plan view of FIG. 27(b), most tires have floating letters on their sidewalls to indicate their type, etc. Therefore, when attempting to measure the center of the sidewall, which is the standard inspection position, as shown by the arrow mark A in the schematic cross-sectional view (a-a section) of Fig. 27(a), the In the conventional device, the roller 6 bounces significantly due to the repetition of characters, making measurement impossible. Furthermore, in the conventional device shown in Fig. 26,
Since the measurement spot of the capacitive displacement meter 2 is large, the data obtained at the center of the sidewall is the average of text and non-text parts, making it difficult to ensure accurate measurement. Even if you try to avoid floating letters, there are often no lines on the sidewall where floating letters do not exist, and even if there is, the width is very narrow, making it difficult to position the displacement meter.

For this reason, in the past, instead of the standard inspection position shown by the arrow mark in Figure 27(a), it was unavoidable to measure the part that avoided the floating letters, as shown by the arrow mark B. That was the reality.

(Objective of the Invention) The object of the present invention is to solve the problems of the prior art as described above, and to measure the test object without selecting a measurement line even if there are unnecessary irregularities such as floating letters on the surface of the test object such as a tire. It is an object of the present invention to provide a shape measuring device capable of measuring an external shape and also capable of obtaining accurate measurement data at high speed.

(Means for Achieving the Object) In order to achieve the above object, a shape measuring device according to the present invention includes an optical displacement meter that scans a predetermined measurement region of a subject having an uneven surface, and signal correction means for receiving an output signal from the meter and removing a predetermined signal pattern component corresponding to the uneven portion from the output signal;
and measuring means for performing predetermined shape measurement based on the signal corrected by the signal correcting means.

(Embodiment) Fig. 1 is an explanatory diagram showing a mechanical part of an embodiment in which a shape measuring device according to the present invention is applied to measuring the outer shape of a tire, and Fig. 2 is a block diagram showing a signal processing part thereof. .

As shown in FIG. 1, a tire 3, which is an object to be inspected, is mounted on a tire rotation mechanism 4 in the same manner as in the past, and
It is rotated at a standard rotation speed of 6Q rpm. Floating letters are formed on the sidewall of the tire 3, as shown in FIG. As the displacement t1, an optical displacement meter 7 using a laser, an LED, or the like is used because it is non-contact and has a small measurement spot. For example, in the case of a laser type, the measurement spot diameter is about 0.3#, and LED
In the case of the type, the diameter of the measurement spot is about 3 mm, which is much smaller than that of the electrostatic displacement meter.

This optical displacement meter 7 is supported by the tips of a pair of support arms 8, and is arranged in an oval shape with both sides of the tire 3 sandwiched therebetween. The support arm 8 is connected to a vertical movement mechanism 9 and a horizontal movement mechanism 10, and is driven by a vertical movement motor 11 to be able to move vertically symmetrically around the tire 3, and is also driven by a horizontal movement motor 12 to move left and right. It is configured to be able to move simultaneously in both directions. Now support arm 8
As shown in the figure, the optical displacement meter 7 is positioned at a position where it scans the center of both nightwalls of the tire 3.

The output signal of the optical displacement meter 7 is A as shown in FIG.
The analog signal is converted into a digital signal by the /D converter 13, and then taken into the arithmetic processing unit 14. The number of samples at this time is 36 per 1'71 rotation of the tie.
0 points (approximately 4 InIR intervals for regular passenger car tires)
Although more accurate measurement is possible if the number of points is more than 720, it is preferable to measure with a sampling number of 720 points. The arithmetic processing device 14 includes a correction section 15 and a measurement section 16. The correction unit 15 extracts predetermined signal pattern components from the sampling data of the optical displacement meter 7 corresponding to the floating characters on the sidewall of the tire 3 (in this embodiment, a trapezoid, a dog mountain, a small mountain, etc. 3 patterns). δ1 again? The I11 section 16 measures the outer shape of the tire, such as runout and bumpiness, based on the data corrected by the correction section 15. This measurement result is outputted to an output unit 17 having, for example, a display, and used together with other tire measurement parameters as a basis for determining tire rank selection.

The functions of the arithmetic processing unit 14 described above can be realized using, for example, a microcomputer, and FIG. 3 is a flowchart showing an overview of the processing procedure at that time. First, in step S1, the output data of the optical displacement meter 7 is collected for one rotation of the tire 3 (Odog ~
360d (! (3)) Since the optical displacement meter 7 is now positioned at the center of the sidewall of the tire 3, it scans over the floating letters formed on the sidewall as shown in Figure 4. Therefore, the sampling data obtained at this time contains noise components (character components) due to the unevenness of the floating characters, as shown in FIG. 5(a).

Next, in step S2, data processing is performed to remove the character component from the captured sampling data. This data processing will be described in detail later, but the correction data obtained thereby represents only the original shape variation of the surface of the tire 3, as shown in FIG. 5(b). 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 the respective processes. In the runout measurement process in step S4, runout IRQ (lateral runout in this case) is calculated from the corrected data as shown in FIG. ), the presence of bumpy BP is detected from the corrected data. In addition, Figure 5 (C
), the data curve is relatively smooth because the same effect as applying moving average processing to the data and passing it through a low-pass filter is obtained.

FIG. 6 is a flowchart schematically showing an example of an algorithm for character component removal processing in step S2 in FIG. In the character component removal process in this example,
Data patterns corresponding to floating characters are defined as three types of data patterns: trapezoid, inuyama, and koyama, as will be described in detail later, and among these, the data component of the trapezoid pattern is removed in step $9.
In addition, the data components of the Inuyama and Oyama patterns are removed in step S10. Further, by performing these processes, unnecessary sampling data other than characters (noise, etc.) can also be removed.

In order to further enable high-speed processing, the concept of grooving is introduced, and grooving processing based on this concept is performed in step S7.

In the first step S6, the steps 87°S9. S
10, namely, grooving processing, trapezoidal pattern component removal processing, and Inuyama.

Various parameters necessary for the hill pattern component removal process are set, and the number of repetitions N is set to determine how many times the process is to be repeated.

Next, in step S7, grooving processing (the details will be described later)
is executed, and after setting the counter to 1-1 in the subsequent step S8, step S9. In S10, trapezoidal pattern component removal processing and Inuyama and Oyama component removal processing are respectively executed. Then, in step S11, it is determined whether i≧N, that is, whether the processing number 1 has reached the set repetition number N. If "No", the counter is updated by 1 in step 812, and then steps 89 and 810 are performed again. The process is repeatedly executed. Then, when i≧N,
The character component removal process shown in FIG. 6 ends.

FIG. 7 J3 and FIG. 8 represent step S of FIG. 6 described above.
7 is an explanatory diagram showing the concept of grooving processing in FIG. In the grooving process, a collection of very close points (P) in sampling data is treated as one group (G). For groups of two or more points, only the first and last points are left as the representative points of the group, and for groups of only one point, that point is left as the representative point of the group. Subsequent data processing is performed with reference to only the representative point (that is, points other than the representative point are ignored).

A parameter called bound is introduced as a criterion for the above grouping. Then, a collection of points falling within the range of the appropriately set grooving parameter bound is set as one group. FIG. 7 shows 19 sampling data from point 21 to point P19, and these sampling data belong to group G.
~Grooved into 8 groups of grooving. That is, in group G, five points P2 to P6 that fall within the bound range with point P2 as the reference point are grouped into one group, and in group G4, point P2 is grouped into one group.
Three points P that fall within the bound range with 8 as the reference point
8 to P1o are grouped into one group. Also, in group G5, boun is set using point P11 as the reference point.
Five points P11 to P15 falling within the range of d are grouped into one group, and in group G6, point P16
62 points P within the range of bound using as the reference point
, P are grouped into one group. Other points P, P7. p18. P19 is group G1. with only one point. G3. G7. Each groove is grooved as G8.

As a result of the above-described grooving, the amount of sampling data to be referred to in subsequent data processing is reduced as shown in FIG. 8, thereby making it possible to shorten the data processing time.

FIGS. 9 and 10 show step S in FIG. 6 described above.
FIG. 11 is an explanatory diagram showing the concept of the trapezoidal pattern component removal process in FIG. 9, and FIG. 11 is a flowchart showing the processing procedure at that time. To perform this process, d
, d, and two +18X l1In parameters are prepared. d is a parameter used to recognize the rise of the trapezoid aX, and d ・
is a parameter used to evaluate the end of the trapezoid head. Note that the trapezoid here refers to one in which there are two or more samples at the head of the trapezoid.

In FIG. 9, 19 sampling points from point 21 to point P19 are shown, and these data are processed to remove trapezoidal pattern components according to the algorithm shown in FIG. That is, first, in step S13, it is determined whether or not the data group ax rises with a displacement of d or more, and the subsequent displacement remains within the range of d, 11n. In the example of Figure 9, point P
50 of the displacement exceeding d from to point P
Since there is a max rise, point P5 is first recognized as the rise point of the trapezoid. Next, using point P6 as a reference point, points Pn+In to P whose displacement is within the range d are recognized as trapezoidal heads, and the range G where the displacement is d is recognized.
12 The point P13 that first deviates from the n++n circle is recognized as the falling point of the trapezoid.

In such a case, the determination in step 813 is "Yes" and the process proceeds to step S14.

On the other hand, if the determination in step 313 is NO, there is no trapezoidal pattern component, and the process ends. In step S14, the process of removing the trapezoidal pattern component recognized as described above is executed.

That is, as shown in FIG. 10, from point 26 to point P of the trapezoid head
12 is discarded, and instead, linear interpolation is performed between the rising point P5 and the falling point P13 of the trapezoid to create new points P6' to P13.
12' becomes data that replaces the original sampling data P6 to P12. The sampling data thus obtained as shown in FIG. 9 is corrected as shown in FIG. 10 by trapezoidal pattern component removal processing.

FIGS. 12 to 15 are explanatory diagrams showing the concept of the Inuyama and Oyama pattern component removal processing in step S10 of FIG. , FIG. 14, and FIG. 15 relate to Inuyama pattern component removal.

The basic idea of this Inuyama/Koyama pattern component removal processing is to remove portions with relatively large gradients that appear in the sampling data by regarding them as characters or noise. In order to evaluate the point to be removed at this time, it is possible to evaluate the point to be removed using, for example, three points including the point in front of the point. And these 3
There are four possible patterns of points as shown in FIGS. 16(a) to 16(d).

First, of the pattern (a), tanθ ・tanθ2
is less than a certain negative value (i.e. points A, B.

C) with sharp corners are removed. As a parameter for evaluating this, M is used as described later. Then (b
) patterns are not removed. Also (C).

Among the patterns in (d), those in which tanθ and tanθ2 are larger than a certain positive value (that is, those in which points A, B, and C continue to rise or fall at a steep slope) are removed.

As a parameter for evaluating this, K is used as described below.

The above parameters M and K can be given as angles (dea), but since this makes it difficult to understand the process, it is preferable to give them as unit lengths, as shown in Figure 17.
Parameters are defined by increments (how many layers rises or falls when moving forward by 1 m) with respect to 1 m). In FIG. 17(a), M=tanα (increment with respect to unit length), and for example, and y・−yH−1>O...(1) One point at the throat (
×・, y・) are removed.

Further, in FIG. 17(b), when K = tanβ (increment with respect to unit length), and for example (2), the point (×·, y·) is removed.

Now, if the scanning radius of the optical displacement meter 7 is r, the data numbering interval ΔX can be expressed as Δx=2πr/n (3). However, n is the number of samples per rotation of the tire. Here, if we set x-=n・ΔX (n,=1~n)...(4), then
The above equations (1) and (2) are respectively y・−yi−i
yi+1-! i/1−(M×Δx)>
X n , -n.

0° 'i-1+・11...-(1') V゛-! i'H-1'17÷1-yi I-(KX Δ x)
×. , −. .

”-ni-1++1 + ...(2'), and further n・-n・=n・-n・=1 ...(5)+
+-1++1 +, so in the end -(M×Δx)2 〉 (y・-”i-1) x (yH+1-yH)...(
1") (KXΔX) < (y.-y.)x(y.-y.) + +-1++1 + (2") is obtained. Therefore, in simple terms, the above (1”).

The components of Inuyama and Oyama patterns can be determined using equation (2″).

FIG. 18 is a 70-chart showing in detail the procedure for removing Inuyama and Koyama pattern components in FIG. 6 above.

First, in step 315, three adjacent points are compared to determine whether the convexity is more than a certain level, that is, whether it is a hill pattern.If "yes", the process goes to step 316; if "no", the process goes to step S17. Each process will proceed.

For example, in the case of the sampling data in Figure 12, point P
4. The portions P5 and P6 are determined to be hill patterns.

This determination can be made using, for example, the above equation (1'').

In step S16, the hill pattern component determined in step 815 is removed. FIG. 13 is an explanatory diagram when removing the hill pattern component from the sampling data of FIG. New data P5' is inserted in place of the original sampling data P5.

In the following step 817, it is determined whether the data group is continuously rising and falling with a slope greater than a certain value, that is, whether it is a human mountain pattern. If "yes", the process advances to step 818. If the answer is "'No", the process ends. For example, in the case of the sampling data in FIG. 14, the portion from point 83 to point P8 is determined to be a crowd pattern.

This determination can be made using, for example, the above equation (2'').

In step 818, the human mountain pattern component determined in step 817 is removed. FIG. 15 is an explanatory diagram when removing the human mountain pattern component from the sampling data of FIG. 14. As shown in the figure, steep slopes P4 to P7 of the human mountain are removed, and points P
The original sampling data P is linearly interpolated between
- New data P' to P7' are inserted in place of P7. Note that when using the algorithm shown in FIG. 17, the peak P6 of the mountain in FIG. 14 is removed as corresponding to the mound pattern in FIG. 17(a).

FIG. 19 is an explanatory diagram when the above-described character component removal process is applied to various types of floating characters. The cross-sectional shapes of the floating letters on the tire surface can be represented by the five types shown in Figure 19 (a), and the corresponding gunpling data is shown in (b), depending on the sampling timing and the shape on the letter surface. It varies depending on the minute unevenness. However, those sampling data are
Crowd of people.

It is a combination of trapezoidal patterns, (2), (6), (
Even complex character sampling data patterns such as 8) can be completely removed by repeating the process.

For example, the data pattern (2) can be removed by repeating the small mountain pattern removal process (first time) and the mountain pattern removal process (second time), and the data patterns (6) and (8) can be removed using the human mountain pattern. It can be removed by repeating the removal process (first time) and the trapezoidal pattern removal process (second time). Other data patterns are removed by one-time processing; in the case of the data pattern (1), the small mountain pattern removal process, in the case of the data patterns (3) and (5), the human mountain pattern removal process, and (4) In the case of the data pattern, each data pattern is removed by trapezoidal pattern removal processing. Further, in the case of the data pattern (7), it is sequentially removed by the small mountain pattern removal process and the human mountain pattern removal process in one process.

FIG. 20 to FIG. 22 are diagrams showing actual processing results according to the above embodiment over the range of tire Odea to 180 degrees. In Fig. 20, (a) is the sampling data, (b) is the data after performing only the hill pattern removal process, and (C) is the data after performing the hill pattern removal process and the human mountain pattern removal process. are shown respectively. The configuration parameters are bound = 0.05
, M=0.2. K=4. It can be clearly seen that the sharp data fluctuations have been effectively corrected.

In FIG. 21, (a) is sampling data, (b)
) is the data after performing the small mountain pattern removal process and the human mountain pattern removal process, and (C) is the data after performing the trapezoidal pattern removal process in addition to the small mountain pattern removal process and the Inuyama pattern removal process. It shows. The setting parameters are bound = 0.05, M-0,2
, K=4. And d = 0.3°aX d , = 0.1. It can be clearly seen that the trapezoidal pattern is effectively removed in the area indicated by the arrow by carrying out the trapezoidal pattern removal process.

Figure 22 shows the difference in the obtained data depending on the number of processing times, (a) is sampling data, (b)
(C) shows the data after one process of removing the small mountain pattern and the removal of the human mountain pattern, and (C) shows the data after the same removal process twice. All setting parameters are bou
nd = 0.05, M = 0.2°K = 7. It can be clearly seen that the circled area is effectively corrected by the two-time processing.

FIG. 23 is a mechanical explanatory diagram showing another embodiment of the present invention, in which the present invention is applied to measurement of the radial runout (the external shape of the tread portion) of the tire 3. The optical displacement meter 7 is supported at the tip of the support arm 18 and travels facing the center of the tread portion of the tire 3. Support arm 18
is connected to a horizontal movement mechanism 19 and driven by a motor 20 to be movable in the left and right direction. In the case of this embodiment, an algorithm for removing groove components (data components of the tread pattern) is required instead of removing character components in the sampling data. Since only the changing direction of the signal is reversed, the groove component can be removed sufficiently effectively using an algorithm similar to that in the above embodiment (that is, only the changing direction of the signal is considered reversed). 24th
The figure shows the results of actually removing groove components.
(a) shows sampling data, and (b) shows data after groove component removal processing. It is clearly seen that the groove components are effectively removed.

In the above embodiment, the case where the present invention is applied to measuring the outer shape of a tire is described in detail, but the present invention is not limited to tires, but can be applied to general measuring of the outer shape of a subject having an uneven surface. In this case as well, the same effects as in the above embodiment can be achieved.

(Effects of the Invention) As explained above, according to the present invention, even if there are unnecessary irregularities such as floating letters on the surface of the test object such as tie 17, the external shape of the test object can be measured without selecting a measurement line. In addition, it is possible to realize a shape measuring device that can measure and obtain accurate measurement data at high speed.

[Brief explanation of drawings]

FIGS. 1 and 2 are diagrams showing an embodiment of the present invention, FIG. 3 is a flowchart showing a processing procedure for measuring the outer shape of a tire to which this invention is applied, and FIG. 4 is an explanation showing scanning on floating letters. , FIG. 5 is an explanatory diagram showing signal changes, FIG. 6 is a flowchart showing the procedure of character component removal processing, FIGS. 7 and 8 are explanatory diagrams of grooving processing, and FIGS. 9 and 10 are An explanatory diagram of trapezoidal pattern removal processing, FIG. 11 is a flowchart showing the processing procedure, FIGS. 12 to 17 are explanatory diagrams of human mountain and small mountain pattern removal processing, and FIG.
Figure 19 is a flowchart showing the processing procedure, Figure 19 is an explanatory diagram when character component removal processing is applied to various types of floating characters, and Figures 20 to 22 are diagrams showing actual processing results according to the present invention. , FIG. 23 is a diagram showing another embodiment of the present invention, FIG. 24 is a diagram showing actual processing results according to another embodiment, and FIGS. 25 and 26 do not show the conventional shape measuring device 5. FIG. 27 is an explanatory diagram of floating letters and measurement parts on a tire. 3... Tire 7... Optical displacement meter 14... Arithmetic processing unit

Claims (1)

[Claims] (1) A device for measuring the external shape of a subject having an uneven surface, which includes an optical displacement meter that scans a predetermined measurement part of the subject, and an output signal of the optical displacement meter. signal correction means for receiving the signal and removing a predetermined signal pattern component corresponding to the uneven portion from the output signal;
A shape measuring device comprising: measuring means for performing predetermined shape measurement based on the signal corrected by the signal correcting means.