JP4690183B2 - Thickness measuring device and thickness measuring method - Google Patents

Description

  The present invention relates to a thickness measuring apparatus or a thickness measuring method for measuring the thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container.

  Containers such as bottles are used for various purposes. In recent years, with the main purpose of resource saving, the thickness of the container has been manufactured as thin as possible. The limit of the thickness of the container is a standard that does not break even if there is some vibration or contact during use or transportation.

  However, it is known that the breakage of containers during use or transport is not dependent solely on the wall thickness, but rather on the non-uniformity of the wall thickness. Usually, containers such as beer bottles and liquor bottles are manufactured by a blowing method, so that the thickness distribution of the containers tends to be uneven in the circumferential direction. For this reason, it is necessary to inspect the thickness of such a container over the entire circumference and to eliminate the defective thickness container as a defective product. Therefore, a technique for measuring the thickness of the container in a non-contact and instantaneously (real time) with high accuracy is eagerly desired.

  For example, Patent Documents 1 to 3 disclose a method for measuring the thickness of a container using laser light as a method for measuring the thickness of the container in a non-contact manner and in real time with high accuracy.

  The invention of Patent Document 1 separates reflected light from the bin surface and refracted light from the back surface with a lens, and measures the thickness from the size (index) of the separation. And in patent document 1, only one imaging lens is used for separation (imaging) of reflected light and refracted light.

  Further, the invention of Patent Document 2 irradiates laser light perpendicularly to the measurement glass surface, and captures the shape of the scattering surface on the front surface and the back surface (the interval between spots of scattered light on the front surface and the back surface) as an image screen. That's it. And in invention of patent document 2, thickness is measured from the space | interval of the spot by the scattered light of the surface and back surface caught as a video screen. That is, the invention of Patent Document 2 images scattering surfaces on both the outer wall and the inner wall with respect to incident light incident substantially perpendicularly on the glass surface, and measures the bin thickness from the positional relationship. In Patent Document 2, the measured thickness is compared with a predetermined inspection reference value, and the quality of the thickness value is determined.

Further, in Patent Document 3, an oblique irradiation cross-sectional image of a glass thick portion is captured by a CCD camera from a certain direction, and a thickness value is measured from the shape. That is, the thickness is obtained by imaging the trajectory of the thick portion of the light beam irradiated to the glass bottle. As a feature point, in order to remove the influence of disturbance light, a filter that transmits only the wavelength of the laser beam is used.
Japanese Patent Laid-Open No. 1-195305 (released August 7, 1991) JP-A-6-201336 (published July 19, 1994) JP 11-190614 A (published July 13, 1999)

  However, the above-described conventional configuration has the following problems.

  None of the conventional configurations described in Patent Documents 1 to 3 has a measure against the environment at the time of measurement, that is, the rotation of the center at the time of rotation, which is indispensable for measurement in the entire circumferential direction. Therefore, highly accurate measurement is possible if there is no center swing.

  However, it is technically difficult to limit the size of such swinging on the wall thickness measurement line. For this reason, the conventional configuration described above has a drawback that a measurement error occurs.

  The present invention has been made in view of the above-mentioned problems, and its purpose is to provide a container thickness measuring apparatus capable of highly accurate thickness measurement even when the center of the container is displaced by touching or the like. And it is providing the thickness measuring method of a container.

  In order to solve the above problems, the thickness measuring apparatus of the present invention is a thickness measuring apparatus that measures the thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container. A laser light source for irradiating the outer wall surface of the container with laser light, an optical path distribution means for distributing the scattered light to a first optical path and a second optical path different from the first optical path, and a first from the optical path distribution means Condensing means for condensing the scattered light via the optical path, a first imaging device for reading the scattered light collected by the condensing means as a scattering spot on the outer wall surface and the inner wall surface, and the optical path distribution A second imaging element that reads scattered light from the means through the second optical path as a specular reflection spot, a scattering spot that is read by the first imaging element, and a positive image that is read by the second imaging element. Based on reflection spot location information It is characterized in that a wall thickness calculating means for calculating a thickness for displacement and its displaced center position of the center of the container.

  In a conventional thickness measuring apparatus, when applied to the measurement of the thickness of a container such as a glass bottle, the center position of the container moves for each measurement, so that accurate thickness measurement becomes difficult.

  However, according to the above configuration, the thickness calculation means is based on the position information of the scattering spot read by the first image sensor and the specular reflection spot read by the second image sensor. Since the displacement of the center position of the container is calculated, it is possible to appropriately detect the center position of the container even when the center of the container is displaced by touching during actual thickness measurement. And according to said structure, since the thickness calculation means calculates the thickness with respect to the displaced center position, it becomes possible to reduce the measurement error which arises around the center position. For this reason, according to said structure, even when the center of a container displaces by touching etc., the thickness measuring apparatus of a container which can measure a highly accurate thickness can be provided.

In the thickness measuring apparatus of the present invention, the center O of the container is displaced to the center O ′, the distance between the center O and the center O ′ is ΔR ₀ , and the laser light irradiation position S ₁ and the center O are And the straight line connecting the center O and the center O ′ is θ _R0 , the thickness calculating means calculates the displacement of the center position of the container as coordinates (ΔR ₀ , θ _R0 ) Is preferably calculated.

With the above configuration, the center position of the container can be detected more accurately by calculating the displacement of the center position of the container as coordinates (ΔR ₀ , θ _R0 ).

Further, in the thickness measuring apparatus of the present invention, the radius of the circle formed by the outer wall surface and R _1, the radius of the circle formed by the inner wall surface and R _2, the incident angle of the laser beam and theta ₁ The laser beam is refracted at the outer wall surface when the distance between the first image sensor and the light collecting means is b, the refractive index of the container is n, and the center position of the container is the center O. The angle of refraction is r _1, and when the intersection point where the straight line connecting the center of the first image sensor and the center of the condensing means intersects the outer wall surface is D, the distance between the intersection D and the condensing means is is a, a distance between the origin position O _B of the intersection D and the second imaging element and O _{B D,} when the center O of the container is displaced to the center O ', read in the first image sensor scattered spots of the outer wall surface that is, the distance between the origin position O _a of the first imaging element l _A1 ' And, when the center O of the container is displaced to the center O ', the specular reflection spot of the outer wall surface read by the second image sensor, the distance between the origin position O _B of the second image sensor In the case of l _B1 ′, the thickness calculation means is represented by the following formulas (B-1) to (B-9):
φ ₁ '= tan ^-1 (l _A1 ' / b) (B-1)
sinr ₂ = (R ₁ / R ₂ ) sinr ₁ = (R ₁ / nR ₂ ) sinθ ₁ (B-2)
k ₁ = [R ₁ sin {(r ₂ −r ₁ ) / 2}] / sin {θ ₁ + (r ₂ −r ₁ ) / 2} (B-3)
k ₂ = {(a + k ₁ ) sinφ ₁ ′} / sin (φ ₁ ′ + 2θ ₁ + r ₂ −r ₁ ) (B-4)
_{k 3 = [{(O B} D) + k 1} 2 + l B1 '2] 1/2 (B-5)
α ₁ = sin ⁻¹ (l _B1 ′ / k ₃ ) (B-6)
k ₄ = {k ₂ ² + k ₃ ² −2 k ₂ k ₃ cos (2θ ₁ + r ₂ −r ₁ −α ₁ )} ^1/2 (B-7)
α ₂ = sin ⁻¹ {k ₂ sin (2θ ₁ + r ₂ −r ₁ −α ₁ ) / k ₄ } (B-8)
θ ₁ ′ = (α ₂ + 2θ ₁ + r ₂ −r ₁ −α ₁ ) / 2 (B-9)
Calculate
θ _R0 = sin ⁻¹ [{sin θ ₁ (k ₂ −k ₁ ) + R ₁ sin (θ ₁ ′ −θ ₁ )} / ΔR ₀ ]
+ (R ₂ −r ₁ ) (6-1)
ΔR ₀ = [R ₁ ² (sinθ ₁ '−sinθ ₁ ) ²
+ {(K ₂ −k ₁ ) −R ₁ (sin θ ₁ ′ −sin θ ₁ )} ² ] ^1/2 (6-2)
Preferably, the displacement of the center position of the container is calculated based on the above.

Θ ₁ ′, k ₁ , k ₂ are calculated by the above formulas (B-1) to (B-9), and the displacement of the center position of the container is obtained based on the above formula (6-2). As a result, the center position of the container can be detected more accurately.

Furthermore, in the thickness measurement apparatus of the present invention, when the center O of the container is displaced to the center O ′, the scattered spot on the inner wall surface read by the first image sensor, and the first image sensor When the distance from the origin position O _A is l _A2 ′ and the center O of the container is displaced to the center O ′, the regular reflection spot on the inner wall surface read by the second image sensor, the distance between the origin position O _B of the image sensor l _{B2 'and,} the radius of the circle the center position is formed by the inner wall surface of the container displaced R _2' in the case of the above thickness calculating unit represented by the following formula (C-1) to (C-22)
t ₁ = k ₂ sin (2θ ₁ + r ₂ −r ₁ ) / sinφ ₁ ′ (C−1)
φ ₂ '= tan ^-1 (l _A2 ' / b) (C-2)
β ₁ = 2θ ₁ + r ₂ −r ₁ −θ ₁ ′ −φ ₂ ′ (C-3)
t ₂ = t ₁ sin (φ ₁ ′ + φ ₂ ′) / sin β ₁ (C-4)
t ₃ = − (R ₁ −t ₂ ) cos β ₁ + {R ₁ ² − (R ₁ −t ₂ ) ² sin ² β ₁ } ^1/2 (C-5)
β ₂ = sin ⁻¹ (sin β ₁ t ₃ / R ₁ ) (C-6)
t ₄ = R ₁ {2 (1-cosβ ₂ )} ^1/2 (C-7)
t ₅ = (R ₁ ² + ΔR ₀ ² -2R ₁ ΔR ₀ cos θ _R0 ) ^1/2 (C-8)
β ₄ = sin ^-1 [{t ₁ sin (φ ₁ '+ 2θ ₁ + r ₂ -r ₁ -θ ₁ ')
−asin (2θ ₁ + r ₂ −r ₁ −θ ₁ ′)} / t ₅ ] (C-9)
t ₇ = {R ₁ ² + t ₅ ² −2R ₁ t ₅ cos (β ₂ −β ₄ )} ^1/2 (C-10)
β ₅ = sin ⁻¹ [{R ₁ sin (β ₂ −β ₄ )} / t ₇ ] (C-11)
t ₈ = {t ₅ sin θ ₁ + R ₁ sin (β ₂ −β ₄ −θ ₁ )} / sinφ ₂ ′ (C−12)
t ₉ = [(O _B D) ² + t ₇ ² +2 (O _B D) {t ₅ cos θ ₁ −R ₁ cos (β ₂ −β ₄ −θ ₁ )}] ^1/2
(C-13)
t ₁₀ = [l _B2 ' ² + t ₉ ² -2l _B2 ' {t ₅ sinθ ₁ + R ₁ sin (β ₂ -β ₄ -θ ₁ )}] ^1/2
(C-14)
t ₁₁ = [l _B2 ' ² + {(O _B D) -a} ² ] ^1/2 (C-15)
β ₇ = cos ⁻¹ [(t ₈ ² + t ₁₀ ² −t ₁₁ ² ) / 2t ₈ t ₁₀ ] (C-16)
θ ₁ ″ = β ₁ −β ₂ + β ₇ (C−17)
sinr ₁ ″ = sin θ ₁ ″ / n (C-18)
sinr ₁ ′ = sin θ ₁ ′ / n (C-19)
_{t 12 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '') (C-20)
_{t 13 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '+ r 1'') (C-21)
t ₁₄ = (t ₁₂ ² + t ₁₃ ² −2t ₁₂ t ₁₃ cosr ₁ ′) ^1/2 (C−22)
Calculate
R ₂ ′ = {(R ₁ −t ₁₂ ) ² + t ₁₄ ²
+ 2t ₁₄ (R ₁ −t ₁₂ ) cos (β ₂ + r ₁ ″)} ^1/2 (7)
Preferably, the radius R ₂ ′ is calculated based on

T ₁₂ , t ₁₄ , β ₂ , r ₁ ″ are calculated by the above formulas (C-1) to (C-22), and the inner wall surface of the container whose center position is displaced based on the above formula (7) The radius R ₂ ′ of the circle formed by is obtained. Thereby, it is possible to accurately detect the radius R ₂ ′ generated when the center of the container is displaced by touching or the like. Further, the thickness of the container can be calculated by the difference between the radius of the circle formed by the outer wall surface and the radius of the circle formed by the inner wall surface. Therefore, according to the above configuration, the thickness of the container can be detected with higher accuracy even when the center of the container is displaced by touching or the like.

Further, in the thickness measuring apparatus of the present invention, the radius of the circle formed by the outer wall surface and R _1, the radius of the circle the center position is formed by the inner wall surface of the container that is displaced when the R _{2 '} The wall thickness calculating means preferably calculates the wall thickness of the container as a difference between the radius R ₁ and the radius R ₂ ′.

Usually, a container such as a bottle is manufactured by a mold method (spraying). For this reason, even when the center of the container is displaced by touching or the like, the radius of the circle formed by the outer wall surface of the container can be regarded as constant. Therefore, by calculating the difference between the radius R ₁ and the radius R ₂ ′ according to the above configuration, the container thickness can be obtained more simply.

  In order to solve the above problems, the thickness measuring apparatus of the present invention is a thickness measuring apparatus that measures the thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container. A laser light source for irradiating the outer wall surface of the container with laser light, an optical path distribution means for distributing the scattered light to a first optical path and a second optical path different from the first optical path, and a first optical path from the optical path distribution means. Of the scattered light passing through one optical path, the first light collecting means for collecting scattered light scattered from the outer wall surface, and the scattered light collected by the first light collecting means A third imaging element that reads as a scattered spot; a second condensing unit that condenses the scattered light scattered from the inner wall surface out of the scattered light from the optical path distributing unit via the first optical path; and the second Scattered light collected by the light collecting means is used as a scattering spot on the inner wall surface. Read by the fourth imaging element, the second imaging element that reads the scattered light from the optical path distribution means via the second optical path as a regular reflection spot, and the third and fourth imaging elements. And a thickness calculation means for calculating the displacement of the center position of the container and the thickness of the displaced center position based on the position information of the scattered spot and the specular reflection spot read by the second image sensor. It is characterized by that.

  According to said structure, the 1st condensing means which condenses the scattered light scattered from an outer wall surface among the scattered lights via the 1st optical path from the said optical path distribution means, and the said 1st condensing means The third imaging device that reads the scattered light collected by the above as a scattering spot on the outer wall surface, and the scattered light scattered from the inner wall surface among the scattered light passing through the first optical path from the optical path distribution means is collected. Since it has the 2nd condensing means to light and the 4th image sensor which reads the scattered light condensed by the 2nd condensing means as a scattering spot of the inner wall surface, it is provided with the outer wall surface of the container Alternatively, there is an advantage that the sensitivity to a minute change of the inner wall surface becomes good and the measurement error due to the aberration of the light collecting means can be reduced.

  The wall thickness measuring method of the present invention is a wall thickness measuring method for measuring the wall thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container in order to solve the above problems. The scattered light scattered from the outer wall surface and the inner wall surface of the container is distributed to the first optical path and a second optical path different from the first optical path, and the scattered light via the first optical path is collected, The image pickup device 1 reads the scattered spot on the outer wall surface and the inner wall surface, and the scattered light passing through the second optical path is read as a specular reflection spot on the second image pickup device. Based on the read scattering spot and the position information of the specular reflection spot read by the second image sensor, the displacement of the center position of the container and the thickness with respect to the displaced center position are calculated. .

  According to said structure, based on the positional information on the scattering spot read by the said 1st image sensor, and the regular reflection spot read by the 2nd image sensor, the displacement of the center position of a container and its Since the thickness with respect to the displaced center position is calculated, the center position of the container can be appropriately detected even when the center of the container is displaced by touching or the like during actual thickness measurement. And according to said structure, since the thickness calculation means calculates the thickness with respect to the displaced center position, it becomes possible to reduce the measurement error which arises around the center position. For this reason, according to said structure, even when the center of a container displaces by touching etc., the thickness measuring method of a container which can measure a highly accurate thickness can be provided.

  The wall thickness measuring method of the present invention is a wall thickness measuring method for measuring the wall thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container in order to solve the above problems. The scattered light scattered from the outer wall surface and the inner wall surface of the container is distributed to the first optical path and the second optical path different from the first optical path, and the scattered light passing through the first optical path is separated from the outer wall surface. Condensed scattered light scattered, read as a scattering spot on the outer wall surface with a third image sensor, condensing scattered light scattered from the inner wall surface among the scattered light via the first optical path, The fourth image sensor reads as a scattering spot on the inner wall surface, and the scattered light passing through the second optical path is read as a regular reflection spot by the second image sensor, and the third and fourth image sensors are used. Read by the scattered spot and the second image sensor Based on the position information of the specularly reflected spot it is characterized by calculating the thickness for displacement and its displaced center position of the center of the container.

  According to said structure, the scattered light scattered from an outer wall surface among the scattered lights via a 1st optical path is condensed, and it reads as a scattered spot of the said outer wall surface with a 3rd image pick-up element, and the 1st optical path The scattered light scattered from the inner wall surface among the scattered light via the light is collected and read as a scattered spot on the inner wall surface by the fourth image sensor, so that it is sensitive to minute changes in the outer wall surface or inner wall surface of the container. And the measurement error due to the aberration of the condensing means can be reduced.

  As described above, the thickness measuring apparatus of the present invention distributes the scattered light to the first optical path and the second optical path different from the first optical path, by irradiating the outer wall surface of the container with the laser light. Optical path distributing means, condensing means for condensing scattered light from the optical path distributing means via the first optical path, and scattering light collected by the condensing means on the outer wall surface and the inner wall surface. A first imaging element that reads as a spot, a second imaging element that reads scattered light from the optical path distribution means via the second optical path as a regular reflection spot, and a scattering spot that is read by the first imaging element And a thickness calculating means for calculating the displacement of the center position of the container and the thickness of the displaced center position based on the position information of the specular reflection spot read by the second image sensor. is there.

  Further, as described above, the thickness measuring method of the present invention distributes the scattered light scattered from the outer wall surface and the inner wall surface of the container to the first optical path and the second optical path different from the first optical path, The scattered light passing through the first optical path is collected, read by the first image sensor as a scattering spot on the outer wall surface and the inner wall surface, and the scattered light passing through the second optical path is read by the second image sensor. Displacement of the center position of the container and its displacement based on the position information of the scatter spot read by the first image sensor and the specular spot read by the second image sensor as a specular spot. In this configuration, the wall thickness with respect to the center position is calculated.

  In addition, as described above, the thickness measuring apparatus of the present invention includes a laser light source that irradiates laser light on the outer wall surface of the container, and the scattered light that is separated from the first optical path and the second optical path different from the first optical path. An optical path distribution unit that distributes the light, a first condensing unit that collects the scattered light scattered from the outer wall surface among the scattered light from the optical path distribution unit via the first optical path, and the first condensing unit. A third imaging element that reads the scattered light collected by the means as a scattered spot on the outer wall surface, and scattered light scattered from the inner wall surface among the scattered light from the optical path distribution means via the first optical path. A second light collecting means for collecting light, a fourth imaging element for reading the scattered light collected by the second light collecting means as a scattered spot on the inner wall surface, and a second optical path from the optical path distributing means. 2nd image of reading scattered light through Based on the positional information of the element, the scattering spot read by the third and fourth imaging elements, and the specular reflection spot read by the second imaging element, the displacement of the center position of the container and its displacement And a thickness calculating means for calculating the thickness with respect to the center position.

  Further, as described above, the thickness measuring method of the present invention distributes the scattered light scattered from the outer wall surface and the inner wall surface of the container to the first optical path and the second optical path different from the first optical path, Of the scattered light that passes through the first optical path, the scattered light that is scattered from the outer wall surface is collected and read as a scattered spot on the outer wall surface by a third image sensor, and the scattered light that passes through the first optical path The scattered light scattered from the inner wall surface is collected and read as a scattered spot on the inner wall surface by the fourth image sensor, and the scattered light passing through the second optical path is specularly reflected by the second image sensor. The displacement of the center position of the container and its displacement based on the position information of the scattering spot read by the third and fourth image sensors and the specular spot read by the second image sensor In this configuration, the wall thickness with respect to the center position is calculated.

  This makes it possible to appropriately detect the center position of the container even when the center of the container is displaced by touching or the like during actual wall thickness measurement. And since the thickness calculation means calculates the thickness with respect to the displaced center position, it becomes possible to reduce the measurement error that occurs around the center position. Therefore, even when the center of the container is displaced by touching or the like, highly accurate wall thickness measurement can be realized.

  In the present invention, both the scattered light from the light (laser light) irradiation region and the regular reflection light on the front and back surfaces of the bin are used together to measure the thickness. In the present invention, first, the center position of the bin is measured from the specularly reflected light on the front and back surfaces of the bin (hereinafter referred to as a regular reflection method). Then, the scattered light with respect to the center position is imaged by the imaging device, and an accurate thickness value is calculated based on the imaging point (hereinafter referred to as an imaging method).

  The thickness of the container can be measured by either the imaging method or the regular reflection method. When the thickness of a container is measured by the specular reflection method, it is necessary that the measurement surface is substantially mirror-like due to the reflection characteristics of the container. However, in an actual bin, it may be a scattering surface locally, or the normal direction of the measurement surface may be greatly inclined with respect to the normal direction of other regions. In such a case, the reflected light intensity may be uniform over the entire surface of the light receiving surface of the image sensor, or the spot of the reflected light may deviate greatly from the light receiving surface. As a result, when measuring the thickness of the container by the specular reflection method, measurement may be impossible.

  On the other hand, considering the accuracy of measurement, the thickness measurement by the specular reflection method is highly accurate. Therefore, the combined use of the imaging method and the regular reflection method is considered the best. Conventional wall thickness measurement is performed on a container whose center position is fixed. However, in this case, the incident angle of the laser beam changes due to the displacement of the center position, and a measurement error corresponding to the change occurs.

  In the present invention, the two methods of the imaging method and the specular reflection method, particularly the specular reflection method, are used for detecting the displacement of the center position at the time of measurement, while the imaging method is particularly used for the original thickness measurement. The thickness of the container is measured by a so-called combination method. Thereby, it is possible to realize a highly accurate thickness measurement that does not cause a measurement error corresponding to the displacement of the center position.

  An embodiment of the present invention will be described below with reference to FIGS. FIG. 1 is an explanatory diagram illustrating a system example of the thickness measuring apparatus according to the present embodiment.

  As shown in FIG. 1, the wall thickness measuring device of the present embodiment (hereinafter referred to as the wall thickness measuring device) is mainly composed of a sensor unit 1 and a data processing unit 2. This wall thickness measuring device is a device for measuring the wall thickness of the container 3. In this embodiment, this wall thickness measuring apparatus measures the wall thickness of the container 3 that is dynamically displaced in real time.

  As shown in FIG. 1, the sensor unit 1 includes a measurement unit 4 and a CPU (computer) 5. The measurement unit 4 transmits data measured for the thickness of the container 3 to the CPU 5. The CPU 5 calculates the wall thickness of the container 3 based on the data input from the measurement unit 4. The calculation of the wall thickness by the CPU 5 is mainly performed by the data processing unit 2. The “thickness calculating means” here means a device including the CPU 5 and the data processing unit 2.

As shown in FIG. 1, the measurement unit 4 includes a control unit 7, a detection unit 8, an amplifier 9, an amplifier 9 ′, an A / D (Analog / Digital) converter 10, and an A / D converter 10 ′. Yes. The control unit 7 of the measurement unit 4 is for controlling the entire measurement unit 4. The detection unit 5 includes a laser light source 11, a half mirror (light distribution unit) 12, a lens (condensing unit) 13, a CCD _A (first image sensor), and a CCD _B (second image sensor). . The detection unit 5 outputs laser light from the laser light source 11 under the control of the control unit 7 and irradiates the irradiation position S ₁ on the outer wall surface M of the container 3 with laser light. The scattered light scattered from the outer wall surface M and the inner wall surface N of the container 3 is read by the CCD _A and the CCD _B through the half mirror 12 and the lens 13. The detection unit 5 outputs analog signals based on the position information read by the CCD _A and the CCD _B to the amplifier 9 and the amplifier 9 ′, respectively.

The amplifier 9 and the amplifier 9 ′ are devices for amplifying an analog signal. The amplifier 9 and the amplifier 9 ′ amplify the analog signal input from the CCD _A and the CCD _B of the detection unit 5 to amplify the analog signal. To output. The A / D converter 10 and the A / D converter 10 ′ convert an analog signal into a digital signal.

Here, a more detailed configuration of the detection unit 5 of the present embodiment will be described. As shown in FIG. 1, the detection unit 5 has a configuration in which a lens 13 is provided between the CCD _A and the container 3, and a half mirror 12 is provided between the lens 3 and the container 3.

In the detection unit 5, laser light is emitted from the laser light source 11 and enters the irradiation position S ₁ of the outer wall surface M of the container 3. The laser light incident on the irradiation position S ₁ becomes scattered light S _A3 scattered on the outer wall surface M of the container 3 and scattered light S _B3 scattered on the point S ₂ on the inner wall surface N of the container 3. Both scattered lights are guided to the half mirror 12.

The half mirror 12 has both the scattered light (scattered light S _A3 and scattered light S _B3 ) scattered to the CCD _A (scattered light via the first optical path) and the scattered light (second optical path) guided to the CCD _B. Functioning as light distribution means for distributing the light to (scattered light via Scattered light _{S A3} and scattered light _{S B3} guided to CCD _A is the lens 13, is focused on the CCD _A. The CCD _A captures (images) and reads the scattered light S _A3 and the scattered light S _B3 collected by the lens 13 as scattered spots on the outer wall surface M and the inner wall surface N.

On the other hand, the scattered light guided to the CCD _B is received by the CCD _B. The CCD _B receives the scattered light S _A3 and the scattered light S _B3 as regular reflection spots on the outer wall surface M and the inner wall surface N. Note that when the scattered light S _A3 and the scattered light S _B3 are received, the light receiving spots have a certain extent. The CCD _B reads the point having the largest light quantity among such light receiving spots as a regular reflection spot.

Thus, in this thickness measuring apparatus, CCD _A reads as scattered spots on outer wall surface M and inner wall surface N, while CCD _B reads scattered light S _A3 and scattered light S _B3 as regular reflection spots. ing. The scattered spot imaged by the CCD _A has a certain degree of contrast, and is reflected clearly and about the size of the irradiated surface. On the other hand, since the regular reflection spot received by the CCD _B is not a focused spot, the light intensity at the center is maximized, and the light intensity appears to decrease toward the periphery. Moreover, the tendency of the light intensity in a regular reflection spot becomes so strong that the scattering degree of an irradiation surface (outer wall surface M of the container 3) is large.

  In the present thickness measuring apparatus, the read information regarding the scattered light and the read information regarding the regular reflection light are distinguished from each other by utilizing the difference in properties between the scattering spot and the regular reflection spot.

In this thickness measuring apparatus, the signals of the reading position information imaged on the CCD _A and the CCD _B are amplified and discretized by the amplifiers 9 and 9 ′ and the A / D converters 10 and 10 ′ to the CPU 5. It is guided.

  The signal of the reading position information guided to the CPU 5 is transferred to the data processing unit 2. The data processing unit 2 calculates, determines, and monitors the thickness of the container 3 based on the signal of the reading position information. The CPU 5 and the data processing unit 2 are connected by a channel link driver 18 and a channel link receiver 19 so that a signal of reading position information is transferred to the data processing unit 2 at high speed.

  The channel link driver 18 and the channel link receiver 19 are specifications of a high-speed communication interface using LVDS (Low Voltage Differential Signaling). This interface achieves small wiring and high speed by using four 7-bit serials and a clock line. In addition, a driver (transmission) IC and a receiver (reception) IC are used as a pair for this interface.

  As shown in FIG. 1, the data processing unit 2 includes an FPGA unit 14, a DSP unit 15, a determination unit 16, and a display unit 17.

  The FPGA (Field Programmable Gate Array) unit 14 is a programmable IC circuit. The FPGA unit 14 configures an IC circuit by sending a design drawing of a microprocessor or ASIC in addition to a logic circuit (logic circuit). The FPGA unit 14 is reconfigurable and has a circuit configuration that can be rewritten many times.

The DSP (Digital Signal Processor) 15 calculates the thickness of the container 3 based on the signal of the reading position information imaged on the CCD _A and the CCD _B. The DSP unit 15 is a processor for causing digital calculation at high speed. In this wall thickness measuring apparatus, high-speed calculation and processing of a large amount of digital data such as images (images read by CCD _A and CCD _B ) is a major bottleneck for the CPU 5 that manages the system. For this reason, in the present thickness measuring apparatus, the DSP unit 15 performs such calculation / processing. Therefore, the internal structure of the DSP unit 15 is configured to perform such an operation at high speed.

  Then, the determination unit 16 determines whether the thickness of the container 3 calculated by the DSP unit 15 is good or bad. The display unit 17 displays the thickness of the container 3 calculated by the DSP unit and the determination result of the thickness by the determination unit 16.

Specifically, the DSP unit 15 uses the reading position information imaged on the CCD _A and the CCD _{B to} determine the center position of the container 3 that moves to the center position for each thickness measurement by the specular reflection method described later. While calculating, the thickness of the container 3 with respect to the center position is calculated. Thereby, in this thickness measuring apparatus, even when the center of the container 3 is displaced by touching or the like, it is possible to realize thickness measurement with higher accuracy than in the past.

  Moreover, the determination part 16 determines whether the thickness of the container 3 calculated in this way is appropriate, and is a defective product based on the set thickness value of the manufactured container 3. It is determined whether or not.

  The reason why the container breaks may be that the wall thickness is large in addition to the thin wall thickness. Even when the thickness of the container is locally thick, the container may be damaged at the boundary portion (the boundary portion between the thick wall portion and the thin wall portion). The determination unit 16 determines that the product is a normal product if the measured thickness value is within a predetermined range with reference to the set thickness value. And when the measured thickness value is outside a predetermined range, it is determined that the product is defective. For example, when the container 3 is a beer bottle, the range of thickness determination by the determination unit 16 is 3.5 mm ± 0, 1 mm.

  Hereinafter, the measurement of the container thickness by the imaging method and the measurement of the center position of the container by the regular reflection method will be described in more detail.

[1. (Measurement of container wall thickness by imaging method)
[1.1 Measurement principle (imaging method)]
The imaging method utilizes the fact that both scattered lights from the outer wall surface (front surface) and the inner wall surface (back surface) of a container such as a bottle are imaged on the image plane (on the CCD camera). When these scattering surfaces are displaced up and down due to the displacement of the bin thickness or the bin center position, both imaging points on the CCD move. Then, by measuring the amount of movement of both imaging points due to this displacement, the bin thickness can be measured. A container such as a bottle has a predetermined curvature. For this reason, the theoretical analysis for measuring the thickness by the imaging method should of course be performed for a predetermined curvature.

  However, first, in 1.2, considering the ease of theoretical analysis, a transparent parallel plate whose front and back surfaces are parallel to each other (this corresponds to the case where the curvature of the bin is assumed to be infinite). ) Will be examined. This analysis can be used for the validity of the analysis result of the bin having the curvature and the estimation of the simulation result in the present invention. Next, in 1.3, the thickness measurement in a container such as a glass bottle, which is the object of the present invention, will be examined.

[1.2 Thickness measurement method of transparent parallel plate]
The optical system using the imaging method includes two optical systems: a one-lens method that forms an image on both the front and back surfaces with a single lens, and a two-lens method that forms an image with two lenses. Can be considered. In the present invention, either the single lens method or the double lens method may be applied.

  As a matter of course, the single lens method has an advantage that the configuration of the optical system is simple. However, the single lens method has a drawback that a measurement error due to lens aberration or the like becomes large. On the other hand, the two-lens method has an advantage that the configuration of the optical system becomes complicated. However, the two-lens method has a drawback that measurement errors due to lens aberration and the like are reduced. Therefore, in the present invention, which of the single lens method and the double lens method is applied depends on the tolerance of the measurement error. Hereinafter, the single lens method and the double lens method will be described.

[1.2.1 Single lens method]
FIG. 2 is an explanatory diagram for explaining the principle of thickness measurement by the single lens method (imaging method). In the single lens method, a transparent parallel plate having a thickness X is irradiated with laser light G ₁ at an incident angle θ. Then, the scattered light S scattered from the front surface E and the back surface F of the transparent parallel plate is collected by a lens L having a focal length f and imaged on a CCD camera (hereinafter simply referred to as a CCD). Here, it is assumed that the optical axis of the scattered light S is DD ′.

In FIG. 2, the laser beam G ₁ is irradiated to the object point A ₀ on the surface E of the transparent parallel plate. Then, the laser beam G ₁ includes a scattered light S _A scattered in various directions around the angle θ at the surface E, a refractive light reaching the object point C ₀ of the back surface F refracted at refraction angle r Divided. Then, this refracted light is scattered in various directions centered on the angle r at the object point C ₀ on the back surface F, and after reaching the object point B ₀ on the surface E, in various directions centered on the angle θ. Emitted. In FIG. 2, the scattered light that is scattered at the object point C ₀ of the back surface F and the scattered light S _B. The scattered light S _A passes through the lens L and is condensed on the image forming point A ₀ ′ of the CCD. On the other hand, the scattered light S _B scattered at an angle r at the object point C ₀ on the back surface F passes through the lens L and is condensed on the image forming point B ₀ ′ of the CCD. Thereby, the object point A ₀ on the front surface E is imaged at the CCD image forming point A ₀ ′, while the object point C ₀ on the back surface F is imaged at the CCD image forming point B ₀ ′. Is done.

At that time, the imaging points A ₀ ′ and B ₀ ′ depend on the thickness X of the transparent parallel plate as shown in the figure. That is, the distance (L _A + L _B ) between the imaging point A ₀ ′ and the imaging point B ₀ ′ shown in FIG. 2 depends on the thickness X of the transparent parallel plate. Therefore, in the one-lens method, the thickness X can be measured based on the imaging points (A ₀ ′ and B ₀ ′) on the CCD. In the one-lens method, the measurement principle is that the thickness X is measured from the image formation points (A ₀ ′ and B ₀ ′) on the CCD.

Note that the position D in the optical system shown in FIG. 2 is the center position between A _{0 and} B ₀ . However, in the optical system using this one-lens method, the position D does not have to be a center position between A _{0 and} B ₀ . The position D is an intersection where a straight line connecting the center D ′ of the CCD and the center O of the lens L intersects the transparent parallel plate installed as a reference, and is a predetermined position.

Thus, in the single lens method, as shown in FIG. 2, the image formation point A ₀ ′ where the scattered light S _A from the front surface E is imaged by the CCD and the scattered light S _B from the back surface F are Both of the image formation points B ₀ ′ imaged by the CCD are imaged at a position considerably distant from the center point D ′ on the CCD camera. For this reason, in the one-lens method, the thickness detection sensitivity decreases and the thickness measurement error increases.

The standard thickness X of the transparent parallel plate can be obtained by a simple geometric technique. More specifically, if the refraction angle of the laser beam G ₁ r, the refractive index of the transparent parallel plate is n, the relationship of the following equation (1) holds. From the relationship of this formula (1), it is possible to obtain the standard thickness X of the transparent parallel plate.
X = {acos θsin (φ _A + φ _B )} / {2 tanr cos (θ + φ _A ) cos (θ−φ _B )} (1)
here,
sinθ / sinr = n, (1-1)
1 / a + 1 / b = 1 / f, (1-2)
_{tanφ A = L A / b,} tanφ B = L B / b. (1-3)
In these equations, a is the object distance (distance between the transparent parallel plate and the lens L), b is the image distance (distance between the CCD and the lens L), and f is the focal length of the lens L. Φ is a solid angle viewed from the lens. That is, φ _A is an angle formed by a straight line connecting the origin position D ′ of the CCD and the center O of the lens L and a straight line connecting the imaging point A ₀ on the CCD and the center O of the lens L. Φ _B represents an angle formed by a straight line connecting the origin position D ′ of the CCD and the center O of the lens L, and a straight line connecting the imaging point B ₀ on the CCD and the center O of the lens L. Show.

  The above-described thickness measurement method is based on the premise that the surface of the transparent parallel plate to be measured does not change up and down for each measurement. However, in actual measurement, that is, real-time measurement, the front and back surfaces of the transparent parallel plate to be measured may be displaced up and down every time the thickness is measured (in addition, such displacement is, for example, a container having a predetermined curvature). Corresponds to a displacement in which the center position is deviated).

As shown in FIG. 2, when the surface E is displaced by a displacement Δy ₁ due to the vertical displacement, the surface E ′ is the surface E ′, and when the back surface F is displaced by the displacement Δy ₂ is the back surface F ′, the laser beam G _{1 is used.} The optical path is as shown by a broken line. In other words, the irradiation position of the laser beam _{G 1} is composed of the point _{A 0} to point _{A 1.} The laser light G ₁ is scattered light S _A ′ scattered in various directions around the angle θ on the surface E ′, and refracted light refracted at the refraction angle r and reaches the point C ₁ on the back surface F ′. And divided. The refracted light is scattered in various directions centered on the angle r at the point C ₁ on the back surface F ′, and after reaching the point B ₁ on the front surface E ′, in various directions centered on the angle θ. Emitted. In FIG. 2, the scattered light scattered at the point C ₁ on the back surface F ′ is defined as scattered light S _B ′. The scattered light S _A ′ passes through the lens L and is condensed on the CCD image formation point A ₁ ′. On the other hand, the scattered light S _B ′ scattered at an angle r at the point C ₁ on the back surface F ′ passes through the lens L and is condensed on the image forming point B ₁ ′ of the CCD.

In this case, as shown in the figure, the imaging points A _{1 'is} the displacement [Delta] y ₁ surface E, imaging points A _0' has a position displaced by [Delta] L _A from. Further, the image forming point _{B 1} 'is the displacement [Delta] y ₂ of the rear surface F, the image forming point _{B 0'} has a position displaced by [Delta] L _B from. That is, when the illustrated, imaging points _{A 1} 'is the origin position D on the CCD' while a position apart _{L A} + [Delta] L _A from the image forming point _{B 1} 'is the origin position D on the CCD 'is a position apart _L B -Delta L _B from.

The displacement Δy _{1 of the} surface E can be calculated based on the displacement ΔL _A of the imaging point A ₀ ′ read on the CCD. The displacement [Delta] y ₂ of the rear surface F can be calculated based on the displacement [Delta] L _B of imaging points _{A 0} 'displacement [Delta] L _A and the imaging point _{B 0'} of. The displacement of the wall thickness X is obtained by calculating the difference between the displacement Ey ₁ on the front surface E and the displacement Δy ₂ on the back surface F.

Note that if the reference thickness is X and the thickness of the transparent parallel plate to be measured is X ′, the thickness X ′ can be obtained by a simple geometric technique. More specifically, if the refraction angle of the laser beam G ₁ r, the refractive index of the transparent parallel plate is n, the relationship of the following equation (2) holds.
X '= {(acosθ-Δy 1) sin (φ A' + φ B ')} /
_{{2tanrcos (θ + φ A '} ) cos (θ-φ B')}, (2)
here,
tanφ _A ′ = {(L _A + ΔL _A ) / L _A } tanφ _A , (2-1)
tan φ _B ′ = {(L _B + ΔL _B ) / L _B } tan φ _B , (2-2)
Δy ₁ = {acos θsin (φ _A + φ _B )} /
{Cos (θ + φ _A ) sin (2θ + φ _A ′)} (2-3)
Δy ₂ = X−X ′ + Δy ₁ , (2-4)
In the equation, φ _A ′ is formed by a straight line connecting the origin position D ′ of the CCD and the center O of the lens L and a straight line connecting the imaging point A ₁ ′ on the CCD and the center O of the lens L. Φ _B ′ is formed by a straight line connecting the origin position D ′ of the CCD and the center O of the lens L and a straight line connecting the imaging point B ₁ ′ on the CCD and the center O of the lens L. Indicates the angle to be played.

As shown in the above formula, the wall thickness X ′ can be obtained directly from the formula (2). That is, slight displacements (Δy ₁ and Δy ₂ ) on the front surface and the back surface are not necessary for calculating the wall thickness X ′. However, these values are necessary for the bin thickness calculation as discussed in 1.3 below.

[1.2.2 Two-lens method]
FIG. 3 is an explanatory diagram for explaining the principle of thickness measurement by the two-lens method (imaging method). As shown in the figure, in the second lens method, by irradiating a laser beam G ₂ at an incident angle θ at a point A on the surface E of the transparent parallel plate having a thickness d.

The laser beam G ₂ irradiated to the point A on the surface E is scattered light S _A1 scattered in various directions around the angle θ from the point A on the surface E, and the point C on the back surface F refracted at a refraction angle r. It is divided into refracted light that reaches. The refracted light is scattered at various points around the angle r at the point C on the back surface F, and after reaching the point B on the surface E, is emitted in various directions around the angle θ. In FIG. 3, the scattered light scattered at the point C on the back surface F is the scattered light S _B1 .

Of the laser light G ₂ , the scattered light S _A1 scattered from the point A on the surface E is applied to the imaging point O _A ′ on the CCD _A1 by the lens L _A (first condensing means) having the focal length f _A. Focused. On the other hand, the scattered light _{S B1} scattered at an angle r at point C of the back surface F is the lens _{L B} of the focal length _{f B,} it is focused on the imaging point _{O B} 'on the _{CCD B1.} Thus, an object point A on the surface E is 'one imaged at object point C of the back surface F is CCD imaging point _{O B'} imaging point _{O A} of the _{CCD A1} is imaged by The Here, the optical axis of the scattered light S _A1 is O _A ′ -A, and the optical axis of the scattered light S _B1 is OB′- _B .

In two lenses method, the thickness measurement greatly differs according to one lens method described above, with respect to both the scattered light S _B1 scattered from the scattered light S _A1 and the back F scattered from the surface E, respectively CCD ( CCD _A1 and CCD _B1 ) are prepared. Then, at the position of the transparent parallel plate serving as a reference, the imaging point of the scattered light S _A1 and the scattered light S _B1 can be set at the center of the CCD (CCD _A1 and CCD _B1 ).

  Therefore, the two-lens method has an advantage that the sensitivity to a minute change in the reference transparent parallel plate position is better than that of the single-lens method, and the measurement error due to lens aberration can be reduced. is there.

Here, as shown in FIG. 3, the case where the front surface E is displaced by the displacement Δy _{1 to} become the front surface E ′ and the back surface F is displaced by the displacement Δy ₂ to be the rear surface F ′ due to the vertical displacement will be described. At this time, the irradiation position of the laser beam G ₂ is, becomes the object point A 'from the object point A. Scattered light S _A1 ′ scattered in various directions around the angle θ at the object point A ′ on the surface E ′ is condensed by the lens L _A on the image point X _A on the CCD _A1 . As a result, the object point A ′ is imaged at the image point X _A on the CCD _A1 .

Also, of the laser beam G _2, point A 'refracted light refracted by the refraction angle r at the rear surface F' reaches the point C 'of. Then, the light is scattered in various directions centered on the angle r at the point C ′ on the back surface F, reaches the point B ′ on the front surface E ′, and is emitted in various directions centered on the angle θ. In FIG. 3, the scattered light scattered at the point C ′ on the back surface F is defined as scattered light S _B1 ′. Scattered light _{S B1} 'is the lens _{L B} of the focal length _{f B,} it is focused on the imaging point _{X B} in the _{CCD B1.} Thereby, the point C ′ on the back surface F is imaged at the image point X _B on the CCD _B1 .

Here, as shown in FIG. 3, when the distance between the point A and the point B is L, and the distance between the point A ′ and the point B ′ is L ′, the thickness d of the reference transparent parallel plate and the measurement The thickness d ′ of the transparent parallel plate to be obtained is as shown in the following formula (3-1).
d = L / (2 tanr), d ′ = L ′ / (2 tanr), (3-1)
Therefore, the difference dd ′ between the wall thickness d and the wall thickness d ′ is obtained as in the following formula (3).
_{d-d '= {- a} A sinφ A1 sin (2θ + φ B) + sinφ B1 sin (2θ + φ A1)} /
_{{2tanrsin (2θ + φ A1)} cos (θ + φ B1)}, (3)
here,
_{tanφ A1 = l A / b A} , tanφ B1 = l B / b B, (3-2)
1 / a _A + 1 / b _A = 1 / f _A , 1 / a _B + 1 / b _B = 1 / f _B (3-3)
In these formulas, L is an irradiation position A of the laser beam G ₂ at the surface, the scattered light S _B1 of the laser beam G ₂ is scattered by the rear surface refracted is the distance between the position B to reach the surface, L 'is, after the displacement, the irradiation position a of the laser beam G ₂ at the surface' distance between, and the laser beam G ₂ is refracted scattered light S _B1 scattered at the back side _'position B which reaches the surface' It is. a is the object distance (the distance between the lens and the transparent parallel plate), where a _A is the distance between the lens L _A and the irradiation position A, and a _B is the distance between the lens L _B and the position B. Distance. b is the image distance (the distance between the CCD and the lens L), the distance between the center _{O A1} and the imaging point _{O A} 'of the formula in _{b A} lens _{L A,} the center of the _{b B} lens _{L B} O _B1 and the imaging point is the distance between _{O B} '. Also, phi is a solid angle viewed from the lens, is wherein phi _A1, and the straight line connecting the center _{O A1} of the imaging point _{O A} 'and the lens _{L A} of _{CCD A1,} an image on _{CCD A1} by displacement It has been shown the angle formed by the straight line connecting the center _{O A1} of the imaging points _{X a} and the lens _{L a,} phi _B1 is the imaging point of the _{CCD B1} and _{O B} 'and the center _{O B1} of the lens _{L B} a straight line connecting the shows the angle formed by the straight line connecting the center O _B1 of the imaged imaging point X _B and the lens L _B on CCD _B1 by displacement.

[1.3 Measuring method of container thickness with respect to center position of fixed container]
Hereinafter, thickness measurement in a container such as a glass bottle, which is an object of the present invention, will be described with reference to FIG.

  When the above-described imaging method is actually applied to the measurement of the wall thickness of a container such as a glass bottle, the center position of the container moves for each measurement, which makes analysis difficult and accurate measurement difficult. Here, the thickness measuring method by the two-lens method for the center position of the fixed container will be described first.

  FIG. 4 is an explanatory diagram for explaining the principle of the container thickness measuring method with respect to the center position of the fixed container. Here, an example in which the above-described two-lens method is applied to the measurement of the container thickness will be described. However, in the present invention, the above-described single lens method may be applied to the measurement of the container thickness.

In the thickness measuring method, and irradiated with a laser beam G ₃ at an incident angle theta ₁ to the point S ₁ of the outer wall surface M of the container having a wall thickness W. The laser beam G ₃ applied to the point S ₁ on the outer wall surface M is refracted at a refraction angle r ₁ with scattered light S _A2 scattered in various directions around the angle θ ₁ from the point S _{1 on} the outer wall surface M. It is divided into refracted light that reaches the point S ₂ on the inner wall surface N. Then, this refracted light is the inner wall surface and scattered in various directions around the angle r ₂ in S ₂ points N, after reaching the point S ₃ of the outer wall surface M, a variety around the angle theta ₁ Emitted in the direction. In Figure 4, and the scattered light that is scattered in various directions around the angle r ₂ at the point S ₂ of the inner wall surface N and the scattered light S _B2.

The scattered light S _A2 is condensed at the origin position O ₁ on the CCD ₁ by the lens L ₁ having the focal length f ₁ . On the other hand, the scattered light S _B2 is condensed at the origin position O ₂ on the CCD ₂ by the lens L ₂ having the focal length f ₂ . Thus, the point _{S 1} of the outer wall surface M, while being focused on the origin position _{O 1} on the CCD _1, the point _{S 2} on the inner wall surface N is imaged origin position _{O 2} on the CCD _2.

In the thickness measurement of the container to which the two-lens method is applied, the point S ₁ that is the irradiation position of the laser beam G ₃ with respect to the container having the standard thickness W (R ₁ -R ₂ ), The point S ₃ of the outer wall surface M (corresponding to the point S ₂ ) is adjusted so as to be imaged at the origin positions O ₁ and O ₂ on the CCD ₁ and CCD ₂ , respectively. That is, the scattered light S _A2 and the scattered light S _B2 are adjusted so as to be condensed at the origin positions O ₁ and O ₂ on the CCD ₁ and the CCD ₂ , respectively.

When the outer wall surface M and the inner wall surface N are slightly displaced due to the change in thickness, these two image formation points move from the origin positions O ₁ and O ₂ . Hereinafter, the optical path when displaced due to a change in thickness will be described.

As shown in FIG. 4, the case where the outer wall surface M is displaced by a displacement ΔR _{1 to} become an outer wall surface M ′ and the inner wall surface N is displaced by a displacement ΔR ₂ to become an inner wall surface N ′ due to the displacement of the container will be described. .

At this time, the laser beam G ₃ are, is irradiated at an incident angle theta _{1 'to'} S ₁ point _'outer wall surface M. Then, the laser beam G ₃ are, the inner wall surface of the refracted at refraction angle r _{1 'and} the angle theta _1' center and the scattered light S _A2 scattered in various directions in which the _'at' point S ₁ of the _'outer wall surface M N It is divided into refracted light that reaches 'point S ₂ '. The scattered light S _A2 ′ is condensed at an image forming point O ₁ ′ on the CCD ₁ by the lens L ₁ . As a result, the point S ₁ ′ on the outer wall surface M ′ is imaged at the imaging point O ₁ ′ on the CCD ₁ .

Further, the refracted light refracted at the point S ₁ ′ is scattered in various directions centered on the angle r ₂ ′ at the point S ₂ ′ on the inner wall surface N ′, and reaches the point S ₃ ′ on the outer wall surface M ′. It reaches and exits in various directions around the angle θ ₁ ′. In FIG. 4, the scattered light scattered at the point S ₂ ′ on the inner wall surface N ′ is taken as scattered light S _B2 ′. The scattered light S _B2 ′ is condensed at the image forming point O ₂ ′ on the CCD ₂ by the lens L ₂ . As a result, the point S ₂ ′ on the inner wall surface N ′ is imaged at the imaging point O ₂ ′ on the CCD ₂ .

The wall thickness W ′ of the container to be measured is basically the radius of a circle formed by the outer wall surface M ′ after displacement (hereinafter referred to as the outer diameter) R ₁ + ΔR ₁ and the inner wall surface N after displacement. It can be obtained from the difference between the radius of the circle formed by '(hereinafter referred to as the inner diameter) R ₂ + ΔR ₂ .

The displacement ΔR ₁ is obtained from the displacement amount P ₁ from the imaging point O ₁ to the imaging point O ₁ ′ on the CCD ₁ . On the other hand, the displacement [Delta] R ₂ is the above displacement amount _{P 1,} obtained from displacement of _{P 2} Metropolitan from the imaging point _{O 2} on the CCD ₂ to _{O 2} '.

The displacement ΔR ₁ and the displacement ΔR ₂ are obtained as follows.
ΔR ₂ = {R ₁ + ΔR ₁ } sinr ₁ ′ / sinr ₂ ′ −R ₂ , (4)
Here, δ = a ₁ P ₁ / (P ₁ cos 2θ ₁ + b ₁ sin 2θ ₁ ) (4-1)
ΔR ₁ = −R ₁ + {δ ² + R ₁ ² + 2δR ₁ cos θ ₁ } ^1/2 (4-2)
sinθ ₁ ′ = R ₁ sinθ ₁ / {δ ² + R ₁ ² + 2δR ₁ cosθ ₁ } ^1/2 , (4-3)
sinβ = {R ₁ sin (θ ₁ + Δφ ₂ ) + a ₂ sinΔφ ₂ } / (R ₁ + ΔR ₁ ), (4-4)
tanΔφ ₂ = p ₂ / b ₂ , (4-5)
r ₂ ′ = −π / 2 + r ₂ −r ₁ + r ₁ ′ + θ ₁ + (− θ ₁ ′ + Δφ ₂ + β) / 2, (4-6)
sinr ₁ ′ = sin θ ₁ ′ / n, (4-7)
A container such as a bottle is usually manufactured by a mold method (spraying). For this reason, the outer diameter of the container is constant. Therefore, the change in the thickness of the container for each measurement is mainly caused by the change in the inner diameter when the center position of the container for each measurement is not changed. In this case, it can be considered that ΔR ₁ = 0 in the above formula.

Therefore, the thickness W ′ of the container to be measured can be obtained by substituting the above formula (4) into the following formula (5).
W ′ = (R ₁ + ΔR ₁ ) − (R ₂ + ΔR ₂ )
= R ₁ − (R ₂ + ΔR ₂ ) (5)
Hereinafter, the derivation method of said Formula (4) and Formula (4-1)-(4-7) is demonstrated based on FIGS. 5 to 7 are explanatory views for explaining a method of calculating the displacement [Delta] R ₁ and the displacement [Delta] R _2. 5 to 7 are based on FIG.

Here, the case where the center position of the container is fixed will be described. For this reason, the displacements ΔR ₁ and ΔR ₂ of the outer diameter R ₁ and inner diameter R _{2 of} the container are obtained with respect to the fixed center position O.

First, the displacement ΔR ₁ of the outer diameter R ₁ due to the change in the wall thickness of the container and the change from the incident angle θ ₁ to θ ₁ ′ (θ ₁ −θ ₁ ′) are the triangle ΔS ₁ OS shown in FIG. Applying the sine and cosine theorems for ₁ ', it is obtained as follows.
(R ₁ + ΔR ₁ ) ² = δ ² + R ₁ ² -2δR ₁ cos (π−θ ₁ ), (A-1)
Here, δ = a ₁ P ₁ / (P ₁ cos 2θ ₁ + b ₁ sin 2θ ₁ ), (4-1)
When the distance between the point S ₁ and the point S ₁ ′ is δ, the equation (A-1) is obtained from the cosine theorem. Here, δ is given by the following equation (A-2) by applying the sine theorem to the triangle ΔC ₁ S ₁ S ₁ ′. Accordingly, the displacement [Delta] R ₁ of the outer diameter _{R 1,} using the above equation (A-1), obtained as follows.
ΔR ₁ = −R ₁ + {δ ² + R ₁ ² + 2δR ₁ cos θ ₁ } ^1/2 . (4-2)
Further, by using the sine theorem for the triangle ΔC ₁ S ₁ S ₁ ′, the following equation (A-2) is obtained for θ ₁ ′.
R ₁ / sin θ ₁ ′ = (R ₁ + ΔR ₁ ) / sin (π−θ ₁ ), (A-2)
From the above formula (A-2), the following formula (4-3) is obtained.
sinθ ₁ ′ = R ₁ sinθ ₁ / {δ ² + R ₁ ² + 2δR ₁ cosθ ₁ } ^1/2 (4-3)
Next, sin β is obtained by applying the cosine theorem and the sine theorem to the triangle ΔC ₂ OS ₃ shown in FIG.
δ ₁ ² = R ₁ ² + a ₂ ² -2R ₁ a ₂ cos (π-θ ₁ ), (A-3)
δ ₁ / sin (π−θ ₁ ) = R ₁ / sin α, (A-4)
Here, δ ₁ indicates a distance between the point S ₁ and the center S _2, and α is an angle OC ₂ S ₃ (an angle formed by the straight line OC ₂ and the straight line C ₂ S ₃ ).

Next, the following formula (A-5) is obtained for the angle β between the straight line C ₂ S ₃ ′ and the straight line OS ₃ ′.
(R ₁ + ΔR ₁ ) / sin (α + Δφ ₂ ) = δ ₁ / sin β, (A-5)
Thereby, it becomes the following formula (4-4).
sinβ = {R ₁ sin (θ ₁ + Δφ ₂ ) + a ₂ sinΔφ ₂ } / (R ₁ + ΔR ₁ ), (4-4)
Here, tanΔφ ₂ = p ₂ / b ₂ (4-5)
Finally, regarding the displacement ΔR ₂ of the inner diameter R ₂ , the relationship of the following equation is obtained by applying the sine theorem to the triangle ΔOS ₁ 'S ₂ ' shown in FIG.
(R ₁ + ΔR ₁ ) / sin (π−r ₂ ′) = (R ₂ + ΔR ₂ ) / sinr ₁ ′, (A-6)
Thus, the following formula (A-7) is further obtained.
ΔR ₂ = (R ₁ + ΔR ₁ ) sinr ₁ ′ / sinr ₂ ′ −R ₂ , (A-7)
here,
r ₂ ′ = −π / 2 + r ₂ −r ₁ + r ₁ ′ + θ ₁ + (− θ ₁ ′ + Δφ ₂ + β) / 2, (4-6)
sinr ₁ ′ = sin θ ₁ ′ / n, (4-7)
[2. (Measurement method of container thickness with respect to the center position of the displaced container)
[2.1 Regular reflection method]
In the specular reflection method, a CCD that receives specular reflection light from the outer wall surface and the inner wall surface of a container such as a bottle is provided, and the positional relationship between the light reception positions (regular reflection light spots) of the specular reflection light and the above-described 1. 2.2 The center position of the container is obtained from both the image formation positional relationship on the CCD _A1 and the CCD _B1 obtained by the two-lens method. In the present invention, the thickness with respect to the center position obtained by the regular reflection method is obtained by the above-described imaging method.

  The wall thickness measurement method described in 1.3 above assumes a state in which the center position of the container does not change for each measurement. However, in actuality, when measuring the thickness of a container such as a glass bottle in real time, stop the movement of the container conveyed by the belt conveyor and rotate it at least once at that position to increase the circumferential thickness of the container. The method of measuring is taken. Then, the center position of the container moves (displaces) due to the swinging that occurs when it rotates at least once for each measurement. For this reason, the thickness measurement method described in 1.3 above is not practically used. In the present invention, the thickness measurement by the imaging method can be accurately performed under the condition that the center position of the container moves. In the present invention, the center position of the container can be measured by the specular reflection method. The thickness measurement method described in 1.3 can be applied to the center position of the container measured by the regular reflection method. Hereinafter, the thickness measurement principle in the present invention will be described in more detail.

[2.2 Measurement principle]
The above-described imaging method itself guarantees a certain degree of accuracy. However, when higher measurement accuracy is required, it is necessary to suppress as much as possible the measurement error caused by the wobbling of the center position of the container during the thickness measurement, which is considered to be the largest cause of error.

  However, it is technically difficult to limit the size of such swinging on the wall thickness measurement line. Even in such a case, a method capable of measuring with high accuracy is required. In the present invention, by using the regular reflection method and the imaging method in combination, high-accuracy measurement is realized in consideration of an error due to the swing of the center position of the container.

  As described below, this method basically measures the displacement due to the swing of the center position of the container (glass bottle) by the specular reflection method, and the outer wall surface position and the inner wall position relative to the displaced center position by the imaging method. The amount of displacement with respect to the wall surface position is accurately measured.

FIG. 8 is an explanatory diagram for explaining the principle of the thickness measuring method of the present invention. Here, the measurement method using the single lens method considering practicality is described. As shown in the figure, the CCD _A receives the scattered light from the outer wall surface and the inner wall surface, and the CCD _B receives the regular reflection light from the outer wall surface and the inner wall surface. FIG. 8 shows a configuration in which CCD _B is provided on the back surface of CCD _A. However, in an actual thickness measuring apparatus, for example, as shown in FIG. Separation means (half mirror) for separation is provided, and CCD _B is not provided on the back side of CCD _A. In FIG. 8, in order to simplify the optical path of the scattered light from the wall surface and the inner wall surface, the CCD _B is provided on the back surface of the CCD _A.

Further, “scattered light” refers to light diffused from the irradiation position irradiated with laser light. On the other hand, “regularly reflected light” refers to light that is reflected at the same angle as the incident angle of laser light among the scattered light. In the configuration of FIG. 8, scattered light from the outer wall surface and the inner wall surface is condensed on the CCD _A by a lens. A part of the scattered light from the outer wall surface and the inner wall surface is received by the CCD _B. Generally, the light in the regular reflection direction has the highest light intensity among the scattered light. For this reason, in the CCD _B , spots of specularly reflected light from the outer wall surface and the inner wall surface are formed. In FIG. 8, regarding the optical path from the outer wall surface to the lens, in order to simplify the optical path of the scattered light, the scattered light received by the CCD _A (scattered light S _A3 and scattered light S _B3 ) and the CCD _B The regular reflection light received is shown as the same straight line.

Hereinafter, the optical path of the scattered light from the outer wall surface and the inner wall surface will be described in detail. As shown in FIG. 8, in this thickness measurement method, a laser beam G ₄ is irradiated at an incident angle θ ₁ to a point S ₁ on the outer wall surface M of a container having a thickness W. The laser beam G ₄ applied to the point S ₁ on the outer wall surface M is refracted at a refraction angle r ₁ with scattered light S _A3 scattered in various directions around the angle θ ₁ from the point S _{1 on} the outer wall surface M. It is divided into refracted light that reaches the point S ₂ on the inner wall surface N. Then, this refracted light is the inner wall surface and scattered in various directions around the angle r ₂ in S ₂ points N, after reaching the point S ₃ of the outer wall surface M, a variety around the angle theta ₁ Emitted in the direction. In Figure 8, and the scattered light that is scattered in various directions around the angle r ₂ at the point S ₂ of the inner wall surface N and the scattered light S _B3.

The scattered light S _A3 and the scattered light S _B3 are condensed on the image forming point P _A1 and the image forming point P _A2 on the CCD _A by the lens L ₃ having the focal length f ₃ . As a result, the point S _{1 on} the outer wall surface M and the point S _{2 on the} inner wall surface N are imaged on the image forming point P _A1 and the image forming point P _A2 on the CCD _A , respectively.

In the present invention, the scattered light S _A3 and the scattered light S _B3 are received by the CCD _B. On the CCD _B , light receiving points (reflected light irradiation positions) P _B1 and P _B2 are imaged as spots of specularly reflected light S _A3 and scattered light S _B3 .

The imaging points P _A1 and P _A2 are read as distances l _A1 and l _A2 from the origin position O _A of the CCD _A. The light receiving points P _B1 and P _B2 are read as distances l _B1 and l _B2 from the origin position O _B of the CCD _B.

Here, as shown in FIG. 8, when the center O of the container is displaced to the center O ′ due to the swing, this displacement can be defined by the displacement amount ΔR ₀ and the displacement angle θ _R0 . The displacement amount ΔR ₀ represents the distance between the center O and the center O ′. Further, the displacement angle θ _R0 represents an angle formed by a straight line S ₁ O connecting the irradiation position S ₁ and the center O and a straight line OO ′ connecting the center O and the center O ′.

  A case where the outer wall surface M becomes the outer wall surface M ′ and the inner wall surface N becomes the inner wall surface N ′ due to such displacement will be described.

The laser beam G ₄ is irradiated to the point S ₁ ′ on the outer wall surface M ′ at an incident angle θ ₁ ′. Then, the laser beam G ₄ are outer wall surface M 'of the points S _1' at an angle theta _{1 'scattered} in various directions around the scattered light S _A3' and the refractive angle r ₁ refracted inner wall surface N by _' It is divided into refracted light that reaches 'point S ₂ '. Then, this refracted light is scattered in various directions around the _'angle r _{2 at'} S ₂ points of the inner wall surface N, after reaching the 'point S _3' of the outer wall surface M, an angle theta _{1 '} The light is emitted in various directions around the center. In Figure 8, and the scattered light that is scattered in various directions around the angle r ₂ at the point S ₂ of the inner wall surface N and the scattered light S _{B3 '.}

Lens _{L 3} is condensed by, CCD _A scattered light on the _{S A3} 'and the scattered light _{S B3'} image point of _{P A1} 'and the imaging point _{P A2'} is the distance from the center position _{O A} of the CCD _A Read as l _A1 ′ and l _A2 ′. On the other hand, the specular reflection light from the upper outer wall surface M 'and on the inner wall surface N' is the center _{O B} 'point at a distance of _{P B1' l B1} from the CCD _B, and a point at a distance of _{l B2} 'P Irradiate _B2 '.

When the center O of the container is displaced to the center O ′ due to swinging, the outer diameter R _{1 of the} container is changed to the outer diameter R ₁ ′. On the other hand, it is assumed that the inner diameter R ₂ of the container changes to the inner diameter R ₂ ′ at that time. Therefore, the wall thickness W ′ of the container to be measured can be obtained from the difference between the outer diameter R ₁ ′ of the container after displacement and the inner diameter R ₂ ′ of the container after displacement.

The displacement (displacement amount ΔR ₀ and displacement angle θ _R0 ) of the center position of the container is obtained as in the following formulas (6-1) and (6-2).
θ _R0 = sin ⁻¹ [{sin θ ₁ (k ₂ −k ₁ ) + R ₁ sin (θ ₁ ′ −θ ₁ )} / ΔR ₀ ]
+ (R ₂ −r ₁ ) (6-1)
ΔR ₀ = [R ₁ ² (sinθ ₁ '−sinθ ₁ ) ²
+ {(K ₂ −k ₁ ) −R ₁ (sin θ ₁ ′ −sin θ ₁ )} ² ] ^1/2 (6-2)
here,
φ ₁ '= tan ^-1 (l _A1 ' / b) (B-1)
sinr ₂ = (R ₁ / R ₂ ) sinr ₁ = (R ₁ / nR ₂ ) sinθ ₁ (B-2)
k ₁ = [R ₁ sin {(r ₂ −r ₁ ) / 2}] / sin {θ ₁ + (r ₂ −r ₁ ) / 2} (B-3)
k ₂ = {(a + k ₁ ) sinφ ₁ ′} / sin (φ ₁ ′ + 2θ ₁ + r ₂ −r ₁ ) (B-4)
_{k 3 = [{(O B} D) + k 1} 2 + l B1 '2] 1/2 (B-5)
α ₁ = sin ⁻¹ (l _B1 ′ / k ₃ ) (B-6)
k ₄ = {k ₂ ² + k ₃ ² −2 k ₂ k ₃ cos (2θ ₁ + r ₂ −r ₁ −α ₁ )} ^1/2 (B-7)
α ₂ = sin ⁻¹ {k ₂ sin (2θ ₁ + r ₂ −r ₁ −α ₁ ) / k ₄ } (B-8)
θ ₁ ′ = (α ₂ + 2θ ₁ + r ₂ −r ₁ −α ₁ ) / 2 (B-9)
In the above formula (B-1) ~ (B -9), b represents the distance between the CCD _A and the lens _{L 3,} n is the refractive index of the container, _{r 1} is the center position of the vessel center when that is a O, a refractive angle of the laser beam G ₄ is refracted by the outer wall surface M. And a is the intersection point D and the origin position O _{A of} CCD _A , where D is the intersection point where the straight line O _A C connecting the origin position O _A of the CCD _A and the center C of the lens L ₃ intersects the outer wall surface M. It indicates a distance, (O _B D) shows the distance between the origin position _{O B} of the intersection D and CCD _B. L _A1 ′ is a scattering spot (imaging point P _A1 ) of the outer wall M read by the CCD _{A and} the origin position O _A of the CCD _A when the center O of the container is displaced to the center O ′. L _B1 ′ is a specular reflection spot (light receiving point P _B1 ) of the outer wall M read by the CCD _B when the center O of the container is displaced to the center O ′, and the CCD _B It indicates the distance between the origin position O _B.

Hereinafter, the derivation method of the said Formula (6-1), Formula (6-2), and Formula (B-1)-(B-9) is demonstrated based on FIG. FIG. 9 is an explanatory diagram for explaining a method of deriving the displacement (displacement amount ΔR ₀ and displacement angle θ _R0 ) of the center position of the container. In addition, each code | symbol shown by FIG. 9 is based on FIG. In FIG. 8, a point D is an intersection where the straight line O _B C and the outer wall surface M intersect. The point D ′ is an intersection where the straight line O _B C and the straight line S ₁ S ₁ ′ intersect. Then, the distance between the point D and the point D ′ = the distance between the point D ′ and the irradiation position S ₁ = k _1, and the distance between the point D ′ and the irradiation position S ₁ ′ is k _2. Yes. Furthermore, the angle formed by the straight line DD ′ and the straight line S ₁ D ′ is 2θ ₁ + r ₂ −r ₁ . Further, if the angle formed between the straight line S ₁ ′ S ₁ and the straight line S ₁ O ′ is α ₃ , the angle formed between the straight line OS ₁ and the straight line S ₁ O ′ is π−θ ₁ −α ₃ .

The angle φ ₁ ′ shown in FIG. 9 is experimentally obtained from the following formula (B-1).
φ ₁ '= tan ^-1 (l _A1 ' / b). (B-1)
Further, by applying the sine theorem to the triangular OS ₃ S ₂ , the following formula (B-2) is established with respect to the angle r ₂ .
sinr ₂ = (R ₁ / R ₂ ) sinr ₁ = (R ₁ / nR ₂ ) sinθ _1. (B-2)
The distance k ₁ between the point D and the point D ′ is expressed by the following formula (B-3) by applying the sine theorem for the triangle DOD ′.
k ₁ = [R ₁ sin {(r ₂ −r ₁ ) / 2}] / sin {θ ₁ + (r ₂ −r ₁ ) / 2}. (B-3)
Similarly, the distance k ₂ between the point D ′ and the irradiation position S ₁ ′ is expressed by the following formula (B-4) by applying the sine theorem for the triangle CD ′S ₁ ′.
k ₂ = {(a + k ₁ ) sinφ ₁ ′} / sin (φ ₁ ′ + 2θ ₁ + r ₂ −r ₁ ). (B-4)
Further, the distance k ₃ between the imaging point P _B1 ′ and the point D ′ is expressed by the following equation (B-5) by applying the cosine theorem to the triangle ΔO _B D′ P _B1 ′. Become.
_{k 3 = [{(O B} D) + k 1} 2 + l B1 '2] 1/2. (B-5)
In the formula _{(B-5), O B} D indicates the distance between the origin position _{O B} and the point D.

Further, the angle α ₁ shown in FIG. 9 is experimentally expressed by the following equation (B-6) using the value k ₃ and the imaging point l _B1 ′.
α ₁ = sin ⁻¹ (l _B1 ′ / k ₃ ). (B-6)
Further, the distance k ₄ between the imaging point P _B1 ′ and the irradiation position S ₁ ′ can be expressed by the following equation (B-7) by applying the cosine theorem to the triangle P _B1 ′ D′ S ₁ ′. It becomes like this.
k ₄ = {k ₂ ² + k ₃ ² −2k ₂ k ₃ cos (2θ ₁ + r ₂ −r ₁ −α ₁ )} ^1/2 . (B-7)
Further, the angle α ₂ shown in FIG. 9 is obtained as shown in the following formula (B-8) by applying the sine theorem to this triangle P _B1 'D'S ₁ '.
α ₂ = sin ⁻¹ {k ₂ sin (2θ ₁ + r ₂ −r ₁ −α ₁ ) / k ₄ }, (B-8)
Further, the incident angle θ ₁ ′ is expressed as in the following formula (B-9).
θ ₁ ′ = (α ₂ + 2θ ₁ + r ₂ −r ₁ −α ₁ ) / 2. (B-9)
When these equations (B-1) to (B-9) are applied to the equation of the cosine theorem for the triangle S ₁ OO ′, the displacements (ΔR ₀ and θ _R0 ) of the bin center position coordinates are expressed by the following equation (6) -1) and (6-2).
θ _R0 = sin ⁻¹ [{sin θ ₁ (k ₂ −k ₁ ) + R ₁ sin (θ ₁ ′ −θ ₁ )} / ΔR ₀ ]
+ (R ₂ −r ₁ ) (6-1)
ΔR ₀ = [R ₁ ² (sinθ ₁ '−sinθ ₁ ) ²
+ {(K ₂ −k ₁ ) −R ₁ (sin θ ₁ ′ −sin θ ₁ )} ² ] ^1/2 (6-2)
Further, as shown in FIG. 8, the radius (hereinafter referred to as the inner diameter) R ₂ of the circle formed by the inner wall surface N changes to the inner diameter R ₂ ′ due to the displacement caused by the swinging. The inner diameter R ₂ ′ is obtained as shown in the following formula (7). The following equation (7) makes it possible to accurately measure the inner diameter R ₂ ′ even when the center position of the container is shifted for each measurement.
R ₂ ′ = {(R ₁ −t ₁₂ ) ² + t ₁₄ ²
+ 2t ₁₄ (R ₁ −t ₁₂ ) cos (β ₂ + r ₁ ″)} ^1/2 (7)
here,
t ₁ = k ₂ sin (2θ ₁ + r ₂ −r ₁ ) / sinφ ₁ ′ (C−1)
φ ₂ '= tan ^-1 (l _A2 ' / b) (C-2)
β ₁ = 2θ ₁ + r ₂ −r ₁ −θ ₁ ′ −φ ₂ ′ (C-3)
t ₂ = t ₁ sin (φ ₁ ′ + φ ₂ ′) / sin β ₁ (C-4)
t ₃ = − (R ₁ −t ₂ ) cos β ₁ + {R ₁ ² − (R ₁ −t ₂ ) ² sin ² β ₁ } ^1/2 (C-5)
β ₂ = sin ⁻¹ (sin β ₁ t ₃ / R ₁ ) (C-6)
t ₄ = R ₁ {2 (1-cosβ ₂ )} ^1/2 (C-7)
t ₅ = (R ₁ ² + ΔR ₀ ² -2R ₁ ΔR ₀ cos θ _R0 ) ^1/2 (C-8)
β ₄ = sin ^-1 [{t ₁ sin (φ ₁ '+ 2θ ₁ + r ₂ -r ₁ -θ ₁ ')
−asin (2θ ₁ + r ₂ −r ₁ −θ ₁ ′)} / t ₅ ] (C-9)
t ₇ = {R ₁ ² + t ₅ ² −2R ₁ t ₅ cos (β ₂ −β ₄ )} ^1/2 (C-10)
β ₅ = sin ⁻¹ [{R ₁ sin (β ₂ −β ₄ )} / t ₇ ] (C-11)
t ₈ = {t ₅ sin θ ₁ + R ₁ sin (β ₂ −β ₄ −θ ₁ )} / sinφ ₂ ′ (C−12)
t ₉ = [(O _B D) ² + t ₇ ² +2 (O _B D) {t ₅ cos θ ₁ −R ₁ cos (β ₂ −β ₄ −θ ₁ )}] ^1/2
(C-13)
t ₁₀ = [l _B2 ' ² + t ₉ ² -2l _B2 ' {t ₅ sinθ ₁ + R ₁ sin (β ₂ -β ₄ -θ ₁ )}] ^1/2
(C-14)
t ₁₁ = [l _B2 ' ² + {(O _B D) -a} ² ] ^1/2 (C-15)
β ₇ = cos ⁻¹ [(t ₈ ² + t ₁₀ ² −t ₁₁ ² ) / 2t ₈ t ₁₀ ] (C-16)
θ ₁ ″ = β ₁ −β ₂ + β ₇ (C−17)
sinr ₁ ″ = sin θ ₁ ″ / n (C-18)
sinr ₁ ′ = sin θ ₁ ′ / n (C-19)
_{t 12 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '') (C-20)
_{t 13 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '+ r 1'') (C-21)
t ₁₄ = (t ₁₂ ² + t ₁₃ ² −2t ₁₂ t ₁₃ cosr ₁ ′) ^1/2 (C−22)
In the above formulas (C-1) to (C-22), l _A2 ′ is the scattering spot (in the inner wall surface N read by the CCD _A ) when the center O of the container is displaced to the center O ′ ( The distance between the imaging point P _A2 ) and the origin position O _A of the CCD _A is shown.

The thickness W ′ of the container to be measured is obtained by substituting the above formula (7) into the following formula (8).
W ′ = R ₁ ′ −R ₂ ′ (8)
Here, as explained in 1.3, a container such as a bottle is usually manufactured by a mold method (spraying). For this reason, the outer diameter of the container is constant. For this reason, in FIG. 8, it can be considered that outer diameter R ₁ = outer diameter R ₁ ′.

That is, in the optical system shown in FIG. 8, the thickness W ′ of the container to be measured is expressed by the following equation.
W ′ = R ₁ −R ₂ ′ (9)
Hereinafter, the derivation method of the said Formula (7) and Formula (C-1)-(C-22) is demonstrated based on FIG.10 and FIG.11. 10 and 11 are explanatory diagrams for explaining a method of deriving the inner diameter R ₂ ′. Note that the reference numerals shown in FIGS. 10 and 11 are based on FIG. In FIG. 10, a point S ₁ ″ is an intersection where the straight line S ₁ ′ O ′ and the straight line S ₃ ′ C intersect.

The distance t ₁ between the center C and the irradiation position S ₁ ′ shown in FIG. 10 is expressed by the following formula (C-1) by applying the sine theorem to the triangle CD ′S ₁ ′. become.
t ₁ = k ₂ sin (2θ ₁ + r ₂ −r ₁ ) / sinφ ₁ ′, (C−1)
Here, the angle φ ₂ ′ shown in FIG. 10 is given by the following equation (C-2).
φ ₂ '= tan ^-1 (l _A2 ' / b). (C-2)
Therefore, the angle beta ₁ shown in FIG. 10 is given as the following formula (C-3).
β ₁ = 2θ ₁ + r ₂ −r ₁ −θ ₁ ′ −φ ₂ ′, (C-3)
The distance t ₂ between the irradiation position S ₁ ′ and the point S ₁ ″ is expressed by the following equation (C-4) by applying the cosine theorem to the triangle CS ₁ ″ S ₁ ′. Given to.
t ₂ = t ₁ sin (φ ₁ ′ + φ ₂ ′) / sin β ₁ (C-4)
Similarly, the distance t ₃ between the point S ₃ ′ and the point S ₁ ″ is expressed by the following formula (C-5) by applying the cosine theorem to the triangle S ₃ ′ O ′S ₁ ″. Is given as follows.
t ₃ = − (R ₁ −t ₂ ) cos β ₁ + {R ₁ ² − (R ₁ −t ₂ ) ² sin ² β ₁ } ^1/2 , (C-5)
Further, the angle β ₂ shown in FIG. 10 is obtained as shown in the following formula (C-5) by applying the sine theorem to the triangle S ₃ ′ O ′S ₁ ″.
β ₂ = sin ⁻¹ (sin β ₁ t ₃ / R ₁ ). (C-6)
The distance t ₄ between the point S ₃ ′ and the irradiation position S ₁ ′ is obtained by applying the cosine theorem to the triangle S ₃ ′ O ′S ₁ ′ as shown in the following formula (C-7). It is done.
t ₄ = R ₁ {2 (1-cosβ ₂ )} ^1/2 , (C-7)
And, 'the distance t ₅ between, triangular DOO' point D and the center O by applying the cosine theorem to be obtained as the following formula (C-8).
t ₅ = (R ₁ ² + ΔR ₀ ² -2R ₁ ΔR ₀ cos θ _R0 ) ^1/2 , (C-8)
Further, when these values and the sine theorem for the triangle DO ′S ₁ ′ are used, the angle β ₄ is obtained as in the following formula (C-9).
β ₄ = sin ^-1 [{t ₁ sin (φ ₁ '+ 2θ ₁ + r ₂ -r ₁ -θ ₁ ')
−asin (2θ ₁ + r ₂ −r ₁ −θ ₁ ′)} / t ₅ ] (C-9)
Therefore, the distance t ₇ between the point D and the point S ₃ ′ is obtained from the cosine theorem for the triangle S ₃ ′ O′D as shown in the following formula (C-10).
t ₇ = {R ₁ ² + t ₅ ² −2R ₁ t ₅ cos (β ₂ −β ₄ )} ^1/2 (C-10)
Further, the angle β ₅ is expressed by the following formula (C-11) from the sine theorem for the triangle S ₃ 'O'D. Note that the angle β ₅ is an angle formed by the straight line S ₃ ′ D and the straight line O′D.
β ₅ = sin ⁻¹ [{R ₁ sin (β ₂ −β ₄ )} / t ₇ ] (C-11)
A distance t ₈ between the center C and the point S ₃ ′ is obtained by applying the sine theorem to the triangle CS ₃ ′ D as shown in the following formula (C-12).
t ₈ = {t ₅ sin θ ₁ + R ₁ sin (β ₂ −β ₄ −θ ₁ )} / sinφ ₂ ′ (C−12)
The center _{O B} and the point _{S 3} 'distance _{t 9} between the triangle _O B _{S 3'} by applying the cosine theorem with respect to D, is determined as the following equation (C-13) . In the formula, O _B D indicates the distance between the center O _B and the point D.
t ₉ = [(O _B D) ² + t ₇ ² +2 (O _B D) {t ₅ cos θ ₁ −R ₁ cos (β ₂ −β ₄ −θ ₁ )}] ^1/2
(C-13)
The distance t ₁₀ between the image point P _B2 ′ and the point S ₃ ′ is expressed by the following equation (C-14) by applying the cosine theorem to the triangle O _B P _B2 ′ S ₃ . Desired.
t ₁₀ = [l _B2 ' ² + t ₉ ² -2l _B2 ' {t ₅ sinθ ₁ + R ₁ sin (β ₂ -β ₄ -θ ₁ )}] ^1/2
(C-14)
Further, the distance t ₁₁ between the image point P _B2 ′ and the center C is obtained by applying the cosine theorem to the triangle O _B P _B2 ′ C as shown in the following formula (C-15). It is done.
t ₁₁ = [l _B2 ' ² + {(O _B D) -a} ² ] ^1/2 (C-15)
The angle β ₇ is obtained by applying the cosine theorem to the triangle P _B2 'S ₃ ' C as shown in the following formula (C-16).
β ₇ = cos ⁻¹ [(t ₈ ² + t ₁₀ ² −t ₁₁ ² ) / 2t ₈ t ₁₀ ] (C-16)
here,
θ ₁ ″ = β ₁ −β ₂ + β ₇ (C−17)
sinr ₁ ″ = sin θ ₁ ″ / n (C-18)
sinr ₁ ′ = sin θ ₁ ′ / n (C-19) Next, using FIG. 11, various amounts necessary for calculating the inner diameter R ₂ ′ are obtained.

The distance t ₁₂ between the irradiation position S ₁ ′ and the point S ₂ ″ is expressed by the following formula (C-20) by using the sine theorem for the triangle S ₃ ′ S ₂ ″ S ₁ ′. Desired.
_{t 12 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '') (C-20)
The distance _{t 13} between the 'point _{S 2} and' illumination position _{S 1,} the use of the triangle _{S 3 'S 2'} sine theorem in _{S 1} ', the following formulas (C-21).
_{t 13 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '+ r 1'') (C-21)
Further, the distance _{t 12} between 'the point S _2' point _{S 2} 'and, by using a triangle _{S 3' S 2} 'cosine theorem S _2' ', the following formulas (C-22).
t ₁₄ = (t ₁₂ ² + t ₁₃ ² −2t ₁₂ t ₁₃ cosr ₁ ′) ^1/2 (C−22)
OS 2 _'' and so far obtained equation and triangle [Delta] S ₂ 'OS _2' 'inside diameter R ₂ of the cosine theorem _for' is given by the following equation.
R ₂ ′ = {(R ₁ −t ₁₂ ) ² + t ₁₄ ²
+ 2t ₁₄ (R ₁ −t ₁₂ ) cos (β ₂ + r ₁ ″)} ^1/2 (7)
As described above, in the present invention, the instantaneous instantaneous inner diameter R ₂ ′ in the circumferential direction (rotation direction) is calculated based on the above equation (7). Even if the center position of the container is shifted for each measurement due to this rotation, the inner diameter R ₂ ′ can be accurately measured. Then, by calculating the difference between the outer diameter R ₁ ′ (= R ₁ ) and the measured inner diameter R ₂ ′ for each measurement, it is possible to accurately measure the wall thickness W ′ of the container to be measured. become.

Further, in the present invention, in order to measure the thickness of a container such as a bottle, the thickness of the bottle is measured by advancing sequentially while refracting the laser beam at the boundary (outer wall surface). Such a method is referred to as a “ray tracing method”. Since the present invention uses the ray tracing method for the center position of the container that changes at each measurement, there is no measurement error except for lens aberrations, particularly spherical aberration.
[3. Results and examination)
[3.1 Resolution (resolution) of wall thickness measurement]
In order to estimate the resolution of the thickness measurement by the imaging method described above, a simulation experiment was performed using the optical design software “Zemax”. This software is basically for simulation based on the light tracking method by refraction at each boundary. FIG. 12 is a diagram showing an optical arrangement for this simulation.

This simulation is based on the optical system shown in FIG. That is, this simulation is a simulation of how the imaging point P _A1 and the imaging point P _A2 imaged on the CCD _A move when the center O of the container is displaced to the center O ′. It is.

As shown in FIG. 12, in this simulation, the incident angle θ ₁ is determined as an angle with respect to the center (maximum) position of the container. Further, the simulation conditions are such that the incident angle θ ₁ is 45 °, the outer diameter R ₁ is 55 mm, and the wall thickness W (outer diameter R ₁ -inner diameter R ₂ ) is 4 mm. The lens _{L 3,} the focal length _{f 3} and 13 mm, and the diameter and 12.7 mm, has a lens thickness and 5.5 mm. Further, the distance between the center O _{A of the} CCD _A and the center C of the lens L ₃ is 23.558441 mm, and the distance between the center C of the lens and the point D is 25 mm.

Based on the above conditions, the amount of movement of the imaging point P _A1 and the imaging point P _A2 when the center O of the container is displaced to the center O ′ was simulated. The movement amount of the imaging point _{P A1} and the imaging point _{P A2} is the imaging point _{P A1} and the imaging point _{P A2,} is the distance _{l A1} and _{l A2} between the center _{O A.} Further, as shown in FIG. 12, the displacement of the center O 'from the center O is defined by the displacement amount [Delta] R ₀ and the displacement angle theta _R0.

The simulation results are shown in Table 1. Table 1 shows the movement of the imaging point P _A1 and the imaging point P _A2 imaged on the CCD _A in mm units when the container center O is displaced to the center O ′. 1 (a) shows the amount of movement (l _A1 ) of the image point P _A1 , and Table 1 (b) shows the amount of movement (l _A2 ) of the image point P _A2 . Note that “−” in the table indicates the moving direction of the image forming point. When the moving directions of the image forming point P _A1 and the image forming point P _A2 shown in FIG. Indicates the opposite direction.

When the change in thickness of the container is ΔR, the change in the image point on CCD _A is Δd, and the resolution (resolution) of CCD _A is ΔD _res , the detection resolution (resolution) of this method, that is, detectable The minimum wall thickness ΔR _res is obtained as in the following formula (10).
ΔR _res = (ΔR / Δd) · ΔD _res . (10)
The detection resolution ΔR _res is determined by the optical arrangement of the members, such as the incident angle θ ₁ , the positional relationship between the lens L ₃ and the CCD _A , the displacement of the center O of the container, the thickness W of the container, or the like. Depending on the dimensions of the Table 2 shows the detection resolution ΔR _res calculated from the above equation (10) based on the data of Table 1 (a) and Table 1 (b) obtained by the simulation. In Table 2, the detection resolution ΔR _res is calculated assuming that the resolution (resolution) ΔD _res of CCD _A is 10 μm.

Table 2 (a) shows that in the optical system shown in FIG. 12, the radius of curvature (inner diameter R ₂ ) of the inner wall surface N is changed from 51 mm to 51.1 mm, and the wall thickness W of the container is reduced by 0.1 mm. cases, to simulate the variation Δd of the imaging point _{P A2} on CCD _a, shows a result of calculating a detection resolution [Delta] R _res the above equation (10), Table. 2 (b), the inner wall surface N curvature radius When the (inner diameter R ₂ ) is changed from 51 mm to 50.9 mm, and the wall thickness W of the container is increased by 0.1 mm, the amount of change Δd of the imaging point P _A2 on the CCD _A is simulated, and the above formula (10) shows the result of calculating the detection resolution ΔR _res , and Table 2 (c) shows the change in the radius of curvature (outer diameter R ₁ ) of the outer wall surface M from 55 mm to 54.9 mm, and the wall thickness W of the container the case was thin as 0.1mm, CCD _a Simulating the variation Δd of the imaging point _{P A1,} it shows a result of calculating a detection resolution [Delta] R _res the above equation (10), Table. 2 (d), the radius of curvature of the outer wall surface M (outer diameter _R 1) Is changed from 55 mm to 54.9 mm, and the amount of change Δd of the imaging point P _A1 on the CCD _A when the wall thickness W of the container is reduced by 0.1 mm is simulated and detected from the above equation (10). Table 2 (e) shows the result of calculating the resolution ΔR _res , and Table 2 (e) changes the curvature radius (outer diameter R ₁ ) of the outer wall surface M from 55 mm to 55.1 mm, and the container thickness W is 0.1 mm. When the thickness is increased, the amount of change Δd of the imaging point P _A1 on the CCD _A is simulated, and the detection resolution ΔR _res is calculated from the above equation (10). Table 2 (f) shows the outer wall M Curvature radius (outer diameter R ₁ ) from 55 mm to 5 The amount of change Δd of the imaging point P _A1 on the CCD _A when the thickness W of the container is increased by 0.1 mm is simulated, and the detection resolution ΔR _res is calculated from the above equation (10). The result of calculating is shown.

As shown in Table 2 (a) to Table 2 (f), the range of the detection resolution ΔR _res for the optical system in FIG. 12 is 10 to 50 μm. The measurement error required for practical use with this thickness measurement method is 0.1 mm. Therefore, this detection resolution ΔR _res can withstand practical use.

  For reference, an example of the result of the center position obtained by the regular reflection method will be described.

Table 4 shows the result of simulating the amount of specular reflection light receiving position shift due to the center position shift of the container using the optical design software Zemax. This simulation is based on the optical system shown in FIG. 8 as described above. That is, this simulation is a simulation of how the imaging point P _B1 and the imaging point P _B2 imaged on the CCD _B move when the container center O is displaced to the center O ′. It is.

As shown in FIG. 13, the conditions of this simulation are that the incident angle θ ₁ is 45 °, the outer diameter R ₁ is 40 mm, and the wall thickness W (outer diameter R ₁ -inner diameter R ₂ ) is 3 mm. Yes. The distance between the center _{O B} and vessel CCD _B is set to 76.4mm. Further, as shown in FIG. 7, the displacement from the center O of the container to the center O ′ due to the swinging is defined by the displacement angle θ _R0 and the displacement distance ΔR ₀ .

Based on the above conditions, the amount of movement of the imaging point P _B1 and the imaging point P _B2 when the center O of the container is displaced to the center O ′ was simulated. The movement amount of the imaging point _{P B1} and the imaging point _{P B2} is the imaging point _{P B1} and the imaging point _{P B2,} and a distance _{l B1} and _{l B2} between the origin position _{O B.}

Table 4 preliminarily determines the shift of the simulation center position, and obtains the light reception position of the regular reflection light with respect to the determined center position shift. That is, Table 4 shows the results of simulations of distances l _B1 and l _B2 with respect to a predetermined displacement angle θ _R0 and displacement distance ΔR ₀ .

However, the regular reflection method according to the present invention obtains the deviation of the center position from the light receiving position of the regular reflection light based on the above formulas (6-1) and (6-2).
[3.2 Thickness measurement error]
Here, the error of the wall thickness measurement according to the present invention will be examined. As the main causes of this measurement error, the following three causes (a) to (c) are considered.
(A) Installation position error of containers such as transparent parallel plates and bottles during measurement.
(B) Spherical aberration of the lens (imaging lens).
(C) CCD reading error.

  Among these causes, (a) is an error in actual wall thickness measurement. On the other hand, (b) and (c) are errors due to the optical system itself that performs wall thickness measurement.

  The present invention is the development of a bin thickness measurement method and system. For this purpose, in the present invention, the thickness with respect to the center position obtained by both the regular reflection method in addition to the imaging method is obtained by the above-described imaging method. Further, this imaging method is measured by a ray tracing method. In such a method, first, the center of the container is accurately obtained by the specular reflection method, and the wall thickness is obtained with respect to this center position, so basically no error occurs. However, this method cannot easily estimate the measurement error due to the displacement of the center of the container.

  In this section, in order to estimate the measurement error due to the displacement of the center of the container, the measurement error is roughly estimated from the analysis of the center of the fixed container. When the center position of the container changes, the incident angle changes as a result. This change in the incident angle causes an error in the thickness measurement as shown in equations (1) and (3). Therefore, the change in the incident angle is first calculated, and then this change Δθ is applied to the expression (1) or (3) for the transparent parallel plate instead of the expression (4) for a container such as a bottle. Approximate error in thickness measurement.

[Incident angle error due to container swinging]
When measuring the thickness of a container such as a glass bottle in real time, a method is adopted in which the movement of the container conveyed by the belt conveyor is stopped, and the thickness of the container in the circumferential direction is measured by at least one rotation at that position. . The greatest error factor in measuring the thickness of the container is an error caused by the swing of the center position of the container that occurs when the container is rotated at least once. This error can be examined by incorporating the incident angle change into the error. Hereinafter, the incident angle error due to the swinging of a container such as a glass bottle will be described. In addition, a mode that an incident angle changes with the change of the center position of a container is as showing in FIG.

That is, as shown in FIG. 8, 'when displaced, by displacement distance [Delta] R ₀ between the displacement angle theta _R0, the incident angle of the laser beam from theta ₁ theta _1' center O of the container center O changes. The incident angle θ ₁ ′ is obtained as in the following formula (11). Equation (11) is calculated when the incident angle θ ₁ = 45 °.
θ ₁ ′ = sin ⁻¹ [{R ₁ + ΔR ₀ 2 ^1/2 cos (θ _R0 + π / 4)} / R ₁ 2 ^1/2 ]. (11)
Hereinafter, a method for deriving the above equation (11) will be described with reference to FIG. FIG. 14 is an explanatory diagram for explaining a method of deriving an incident angle error due to the shake of the container. In addition, each code | symbol shown by 14 is based on FIG. In FIG. 14, the incident angle θ ₁ = 45 °. Further, considering an xy coordinate system in which the center O of the fixed (reference) container is the origin (0, 0), the center O ′ after displacement is (Δx ₀ , Δy ₀ ).

As shown in FIG. 14, by the movement of the center position of the container, the incident angle is changed in theta _{1 'from} theta _1. The bin center position O ′ (Δx ₀ , Δy ₀ ) on the radius ΔR ₀ is given by the following equation (D-1).
Δx ₀ = ΔR ₀ sinθ _R0 , Δy ₀ = ΔR ₀ cosθ _R0 . (D-1)
Therefore, an arbitrary point coordinate on the circle S ₂ whose center is O ′ is expressed as the following formula (D-2).
x = R ₁ cosβ + ΔR ₀ sinθ _R0 , y = R ₁ sinβ + ΔR ₀ cosθ _R0 . (D-2)
A straight line L with an inclination of 45 ° is given by the following formula (D-3).
y = x + R ₀ . (D-3)
From simple geometric considerations, the following relationship is obtained.
_{R 0 sinβ + ΔR 0 cosθ R0} = R 0 cosβ + ΔR 0 sinθ R0 + R 0, (D-4)
cos (β + π / 4) = {R ₀ −ΔR ₀ √2 cos (θ _R0 + π / 4)} / R ₀ √2. (D-5)
Therefore, from the relationship of β−π / 4 = θ ₁ ′, θ ₁ ′ finally becomes the following equation.
θ ₁ ′ = sin ⁻¹ [{R ₁ −ΔR ₀ √2 cos (θ _R0 + π / 4)} / R ₁ √2]. (11)
In the above equation (11), the minimum value θ ′ _m and the maximum value Θ ′ _M of the incident angle θ ₁ ′ are theoretically calculated from simple geometric considerations. That is, the minimum value θ ′ _m and the maximum value Θ ′ _M are values when cos (θ _R0 + π / 4) = 1 and cos (θ _R0 + π / 4) = − 1, respectively. That is, in the above formula (11), when θ _R0 = −π / 4, θ ₁ ′ becomes the minimum value θ ′ _m . On the other hand, when θ _R0 = π / 4, θ ₁ ′ becomes the maximum value Θ ′ _M. Table 5 shows the maximum value θ ′ _max and the minimum value θ ′ _max for several values of ΔR ₀ / R ₀ .

For example, when the incident angle θ = 45 °, the minimum value θ ′ _m and the maximum value Θ ′ _M are as shown in Table 5. Table 5 is a table showing the minimum value θ ′ _m and the maximum value Θ ′ _M when the ratio between the outer diameter R ₁ and ΔR ₀ is changed.

As shown in Table 5, the difference between the minimum value θ ′ _m and the maximum value Θ ′ _M increases as the ratio between the outer diameter R ₁ and ΔR ₀ increases (that is, the outer diameter decreases). . This indicates that as the ratio between the outer diameter R ₁ and ΔR ₀ increases, the error of the incident angle θ ₁ ′ greatly affects.
[Thickness measurement error in the present invention]
When the incident angle changes from θ ₁ to θ ₁ ′ as described above, the error (θ ₁ ′ −θ ₁ ) of the incident angle is Δθ. The wall thickness measurement error ΔX resulting from the incident angle error Δθ can be obtained by substituting θ + Δθ into θ in the equations (1) and (3) related to the transparent parallel plate.

  When the single lens method is used as the imaging method, the measurement error ΔX caused by Δθ is expressed by the following equation (12-1).

Therefore, the relative error ΔX / X estimated by the first-order approximation of the thickness is given by the following equation (12).
ΔX / X≈Δθ {tan (θ + φ _A ) + tan (θ−φ _B ) −cot θ−tan θ−
sinθcosθ / (n ² −sin ² θ)}, (12)
When the two-lens method is used as the imaging method, the measurement error ΔX caused by Δθ is expressed by the following equation (13-1).

Therefore, the relative error ΔX ′ / X estimated by the first-order approximation of the thickness is given by the following equation (13).
ΔX ′ / X = Δθ [{− cot θ + tan (θ + φ _B ) −sin θ cos θ / (n ² −sin ² θ)}
+ 2a _A sin φ _A sin (2θ + φ _B ) {cot (2θ + φ _B ) −cot (2θ + φ _A )} /
_{{A A sinφ A sin (2θ} + φ B) -sinφ B sin (2θ + φ A)}], (13)
For _{example, θ = 45 °, tanφ A} ≒ l A / b ≒ tanφ B ≒ l B / b = 5mm / 50mm, if X = 5 mm, when using a single lens method, measurement errors ΔX caused by Δθ is The following formula (12 ′) is obtained.
ΔX / X = −0.30Δθ, (12 ′)
When the two-lens method is used, the measurement error ΔX caused by Δθ is expressed by the following equation (13 ′).
ΔX ′ / X = −0.12Δθ. (13 ′)
Assuming that, for example, R: ΔR ₀ = 15: 1 as the maximum value of Δθ, from Table 5, Δθ _max ≈ ± 5 °. When this angle is converted into radians, Δθ _max ≈ 0.087. Substituting this into equation (12 ′) gives
ΔX / X = −0.30 Δθ = 0.026
It becomes. For example, when the thickness X of the container is 5 mm (X = 5 mm),
ΔX = 0.026 × 5.0 ≒ 0.13mm
It becomes.

Also, if Δθ _max ≈0.087 is substituted into equation (13 ′),
ΔX ′ / X = −0.12 Δθ = 0.010
It becomes. And when the wall thickness X ′ of the container is 5 mm (X ′ = 5 mm),
ΔX ′ = 0.010 × 5.0≈0.05 mm
It becomes.

  Thus, the measurement error for both lens systems was 0.13 mm or less, which was found to be a practically allowable range.

  Hereinafter, for reference, a study on the cause of the above (b) or (c) when measuring the thickness of the transparent parallel plate will be described.

[Error by single lens method]
First, the measurement error ΔX when the thickness of the transparent parallel plate is measured by the single lens method will be examined. In the single-lens method shown in FIG. 2, the errors corresponding to the causes (b) and (c) are as follows.
(B) Spherical aberration of the lens L (imaging lens).
(C) Image point reading error on the CCD (Δl _A and Δl _B ).
Among these, errors (Δl _A and Δl _B ) corresponding to the cause of (c) can be examined based on the above equation (1). In particular, errors Δl _A and Δl _B corresponding to the cause of (c) can be handled as minute angle changes Δφ _A and Δφ _B on the CCD.
Hereinafter, the error corresponding to these two causes will be examined.

First, the measurement error ΔX caused by the reading errors Δφ _A and Δφ _B corresponding to the cause of (C) is obtained as the following equation (14-1).

Therefore, the relative error ΔX / X is given by the following equation (14).

Here, the magnitude of the actual error is approximated. The errors of Δφ _A and Δφ _B correspond to the reading error of the image point on the CCD. Therefore, the error of Δφ _A and Δφ _B can be considerably reduced.

Next, a measurement error caused by the spherical aberration of the lens L corresponding to the cause of (B) is simulated by the optical design software Zemax. This simulation is based on the optical system shown in FIG. That is, this simulation shows how the imaging point A ₀ ′ and the imaging point B ₀ ′ imaged on the CCD move when the thickness of the container is 3 mm and the back surface or the front surface is displaced. Is simulated.

  As shown in FIG. 15, the conditions for this simulation are that the incident angle θ is 45 °. When the intersection point between the center O of the lens and the straight line passing through the origin position D ′ of the light receiving surface of the CCD and the surface E of the sample (transparent parallel plate) is D, the distance between the point D and the center O (see FIG. (Corresponding to the distance a in 2) is 30 mm. The distance between the center O of the lens and the center D 'of the light receiving surface of the CCD (corresponding to the distance b in FIG. 2) is 46.5561406 mm. The focal length (F) of the lens (EDOMUNDO) is 18 mm, and the lens diameter (D) is 6 mm. Further, the diameter of the lens aperture is 2 mm.

Based on the above conditions, the amount of movement of the imaging point A ₀ ′ and the imaging point B ₀ ′ when the back surface or the front surface is displaced was simulated. The movement amount of the imaging points _{A 0} 'and the imaging point _{B 0'} is the imaging points _{A 0} 'and the imaging point _{B 0',} the distance _{l A} and _{l B} between the origin position D ' Yes. The examination results are shown in Table 6. X in Table 6 is a value obtained by substituting the obtained imaging points l _A and l _B into the equations (1-1) and (1).

  In this case, as shown in Table 6, a maximum measurement error of 40 μm occurs in X of the simulation result. This is because the spherical aberration is increased because the scattered light scattered on the front surface and the scattered light scattered on the back surface are imaged by one lens. On the other hand, in the two-lens method described later, the influence of the spherical aberration of the lens is almost eliminated.

[Errors due to the two-lens method]
Next, the measurement error ΔX when the thickness of the transparent parallel plate is measured by the two-lens method will be examined. Similarly to the above, the error corresponding to the cause of the above (b) becomes the reading error (Δl _A and Δl _B ) of the imaging point on the CCD. The errors Δl _A and Δl _B can be handled as minute angle changes Δφ _A and Δφ _B on the CCD. Hereinafter, the errors corresponding to these three causes will be examined.

The measurement error ΔX ′ caused by the reading errors Δφ _A and Δφ _B corresponding to the cause of (C) is obtained as follows.

In the above formula, when φ _A → φ _A + Δφ _A and φ _B → φ _B + Δφ _B are calculated, Δd with respect to these changes is calculated as the following formula (15).

Here, the magnitude of the actual error is approximated. The errors of Δφ _A and Δφ _B correspond to the reading error of the image point on the CCD. Therefore, the error of Δφ _A and Δφ _B can be considerably reduced.

Finally, the measurement error caused by the spherical aberration of the lens corresponding to the cause of (B) is simulated by the optical design software Zemax. This simulation is based on the optical system shown in FIG. That is, this simulation is a simulation of how the imaging point O _A ′ and the imaging point O _B ′ imaged on the CCD _A and the CCD _B move when the back surface or the front surface is displaced. It is.

As shown in FIG. 16, the conditions for this simulation are an incident angle θ of 45 °. When the intersection of the straight line passing through the center O _{A of} the lens and the imaging point (origin position) O _A ′ of the light receiving surface of the CCD and the surface E of the sample (transparent parallel plate) is A, The distance from the center O _A (corresponding to the distance a _A in FIG. 3) is 30 mm. Further, the distance between the lens center O _A and the image forming point O _A ′ on the light receiving surface of the CCD (corresponding to the distance b _A in FIG. 3) is 46.5561406 mm. The focal length (F) of the lens (EDOMUNDO) is 18 mm, and the lens diameter (D) is 6 mm. Further, the diameter of the lens aperture is 2 mm.

Based on the above conditions, the amount of movement of the imaging point O _A ′ and the imaging point O _B ′ when the back surface or the front surface is displaced was simulated. The moving amounts of the image forming point O _A ′ and the image forming point O _B ′ are distances l _A and l _B. The examination results are shown in Table 7. DD ′ in Table 7 is a value obtained by substituting the obtained distances l _A and l _B into equation (3).

  In this case, as shown in Table 7, it can be seen that there is no error in the simulation result D-D ′ according to the equation (3). This is because two different imaging lenses are used for the front and back surfaces, and spherical aberration hardly occurs.

〔in conclusion〕
The present invention irradiates laser light on the outer wall surface of the container from an oblique direction, and accurately determines the thickness of the bottle in real time by using a combination of the following two methods.
(1) The center position of the container is obtained from the image formation (light reception) position where the regular reflection light from the outer wall surface and the inner wall surface of the container forms an image on the CCD ₂ .
(2) Further, the scattered light from both the outer wall surface and the inner wall surface is imaged on the CCD ₁ by a lens, and the thickness of the container is obtained from the imaged point.

  The measurement principles of the inventions in Patent Documents 1 to 3 described above all measure the wall thickness by an imaging method. However, all of the inventions of Patent Documents 1 to 3 use only one lens for imaging. In contrast, in the present invention, two different imaging lenses are used for imaging the outer wall surface and the inner wall surface of the container. For this reason, in this invention, compared with the conventional thickness measuring method, there exists an effect that the sensitivity and precision of a measurement go up.

  Furthermore, in the present invention, the imaging method and the regular reflection method are used in combination in order to avoid measurement errors caused by the shake of the container that always occur during actual real-time measurement. That is, in the present invention, the center position of the container is obtained from the image formation point on the CCD obtained by the image formation method and the regular reflection method, and the accurate thickness with respect to the center position is measured.

  Specifically, two CCD cameras are used for the regular reflection method and the imaging method. Then, an accurate thickness is measured with respect to the center position of the container obtained from the image forming points on the two CCD cameras.

  As described above, the present invention uses two lenses in the imaging method and further uses it in combination with the regular reflection method in order to reduce a measurement error caused by a wobbling or the like that occurs during actual measurement. As a result, it is possible to dramatically improve measurement sensitivity and reduce measurement errors.

  In order to measure the bottle thickness in real time, an imaging method using specular reflection method was developed. This system consists of two main subsystems. One is a sensor head including a laser as a light source, two CCD cameras for receiving both regular reflection light and scattered imaging light, and a lens for imaging the scattered light on the CCD camera. The other is a data processing subsystem composed of a personal computer with a built-in A / D converter and an amplifier and a display for displaying the thickness. According to a simulation experiment using the optical design software “Zemax”, it was found that the resolution, that is, the minimum measurable thickness was 10 to 50 μm when a CCD camera with a resolution of 10 μm was used. This value is sufficient for practical use.

  The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope shown in the claims. That is, embodiments obtained by combining technical means appropriately modified within the scope of the claims are also included in the technical scope of the present invention.

  The present invention can be applied to the use of measuring the thickness of a container such as a glass bottle.

BRIEF DESCRIPTION OF THE DRAWINGS It is explanatory drawing which shows embodiment of this invention and shows the system example of a thickness measuring apparatus. It is explanatory drawing for demonstrating the principle of the thickness measurement by the 1 sheet lens method (imaging method). It is explanatory drawing for demonstrating the principle of thickness measurement by the 2 lens method (imaging method). It is explanatory drawing for demonstrating the principle of the thickness measurement method of the container with respect to the center position of the fixed container. The method of deriving the displacement [Delta] R ₁ and the displacement [Delta] R ₂ is an explanatory view for explaining the. The method of deriving the displacement [Delta] R ₁ and the displacement [Delta] R ₂ is an explanatory view for explaining the. The method of deriving the displacement [Delta] R ₁ and the displacement [Delta] R ₂ is an explanatory view for explaining the. It is explanatory drawing for demonstrating the principle of the thickness measurement method of the container in which the center position has displaced by the whirling. It is explanatory drawing for demonstrating the derivation | leading-out method of the displacement of the center position of a container. It is explanatory drawing for demonstrating the derivation | leading-out method of the internal diameter in the container with respect to the displaced center position. It is explanatory drawing for demonstrating the derivation | leading-out method of the internal diameter in the container with respect to the displaced center position. It is a figure which shows the optical arrangement | positioning for simulation which approximates the resolution of thickness measurement. It is a figure which shows the optical arrangement | positioning for simulation of the amount of regular reflection light light reception position shift | offset | difference by the center position shift of a container. It is explanatory drawing for demonstrating the derivation | leading-out method of the incident angle of the laser beam which changed when the center of the container displaced. It is a figure which shows the optical arrangement | positioning for simulating the measurement error produced by the spherical aberration of a lens in the case of applying the single lens method. It is a figure which shows the optical arrangement | positioning for simulating the measurement error which arises by the spherical aberration of a lens at the time of applying the 2 lens method.

Explanation of symbols

1 Sensor part 2 Data processing part (Thickness calculation means)
3 Container 4 Measuring part 5 CPU (Thickness calculation means)
6 Amplifier 7 Control unit 8 Detection unit 9, 9 ′ Amplifier 10, 10 ′ A / D converter 11 Laser light source 12 Half mirror (optical path distribution means)
13 Lens (Condensing means)
14 FPGA part (thickness calculation means)
15 DSP part (thickness calculation means)
16 Judgment part (thickness calculation means)
17 Display unit 18 Channel link driver 19 Channel link driver

Claims (8)

A wall thickness measuring device for measuring the wall thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container,
A laser light source for irradiating the outer wall surface of the container with laser light;
Optical path distribution means for distributing the scattered light to a first optical path and a second optical path different from the first optical path;
Condensing means for condensing scattered light from the optical path distribution means via the first optical path;
A first imaging device that reads scattered light collected by the light collecting means as a scattered spot on the outer wall surface and the inner wall surface;
A second imaging element that reads scattered light from the optical path distribution means via the second optical path as a regular reflection spot;
Based on the position information of the scattering spot read by the first image sensor and the specular reflection spot read by the second image sensor, the displacement of the center position of the container and the thickness with respect to the displaced center position. And a thickness calculating means for calculating the thickness. The center O of the container is displaced to the center O ′, the distance between the center O and the center O ′ is ΔR ₀ , a straight line connecting the irradiation position S _{1 of the} laser beam and the center O, and the center O and the center O When the angle between the straight line connecting 'and θ _R0 is
2. The thickness measuring apparatus according to claim 1, wherein the thickness calculating means calculates the displacement of the center position of the container as coordinates (ΔR ₀ , θ _R0 ). The radius of the circle formed by the outer wall surface and R _1, the radius of the circle formed by the inner wall surface and R _2, the incident angle of the laser beam and theta _1, the first imaging device and the condenser The distance from the means is b, the refractive index of the container is n,
When the center position of the container is the center O, the refraction angle at which the laser beam is refracted on the outer wall surface is r ₁ ,
When the intersection of the straight line connecting the center of the origin position O _A and the focusing means of the first image sensor intersects with the outer wall surface is D, the distance between the intersection D and the condensing means is a, the intersection D the distance between the origin position O _B of the second imaging element and O _{B D} and,
When the center O of the container is displaced to the center O ′, the distance between the scattered spot on the outer wall surface read by the first image sensor and the origin position O _A of the first image sensor is defined as l _A1 ′. age,
When the center O of the container is displaced to the center O ', the specular reflection spot of the outer wall surface read by the second image sensor, the distance between the origin position O _B of the second image sensor l _B1 'If
The wall thickness calculating means is represented by the following formulas (B-1) to (B-9).
φ ₁ '= tan ^-1 (l _A1 ' / b) (B-1)
sinr ₂ = (R ₁ / R ₂ ) sinr ₁ = (R ₁ / nR ₂ ) sinθ ₁ (B-2)
k ₁ = [R ₁ sin {(r ₂ −r ₁ ) / 2}] / sin {θ ₁ + (r ₂ −r ₁ ) / 2} (B-3)
k ₂ = {(a + k ₁ ) sinφ ₁ ′} / sin (φ ₁ ′ + 2θ ₁ + r ₂ −r ₁ ) (B-4)
_{k 3 = [{(O B} D) + k 1} 2 + l B1 '2] 1/2 (B-5)
α ₁ = sin ⁻¹ (l _B1 ′ / k ₃ ) (B-6)
k ₄ = {k ₂ ² + k ₃ ² −2 k ₂ k ₃ cos (2θ ₁ + r ₂ −r ₁ −α ₁ )} ^1/2 (B-7)
α ₂ = sin ⁻¹ {k ₂ sin (2θ ₁ + r ₂ −r ₁ −α ₁ ) / k ₄ } (B-8)
θ ₁ ′ = (α ₂ + 2θ ₁ + r ₂ −r ₁ −α ₁ ) / 2 (B-9)
Calculate
θ _R0 = sin ⁻¹ [{sin θ ₁ (k ₂ −k ₁ ) + R ₁ sin (θ ₁ ′ −θ ₁ )} / ΔR ₀ ]
+ (R ₂ −r ₁ ) (6-1)
ΔR ₀ = [R ₁ ² (sinθ ₁ '−sinθ ₁ ) ²
+ {(K ₂ −k ₁ ) −R ₁ (sin θ ₁ ′ −sin θ ₁ )} ² ] ^1/2 (6-2)
The wall thickness measuring device according to claim 2, wherein the displacement of the center position of the container is calculated based on the above. When the center O of the container is displaced to the center O ′, the distance between the scattered spot on the inner wall surface read by the first image sensor and the origin position O _A of the first image sensor is defined as l _A2 ′. age,
When the center O of the container is displaced to the center O ', the specular reflection spot of the inner wall surface read by the second image sensor, the distance between the origin position O _B of the second image sensor l _B2 'age,
When the radius of the circle formed by the inner wall surface of the container whose center position is displaced is R ₂ ′,
The thickness calculation means is represented by the following formulas (C-1) to (C-22).
t ₁ = k ₂ sin (2θ ₁ + r ₂ −r ₁ ) / sinφ ₁ ′ (C−1)
φ ₂ '= tan ^-1 (l _A2 ' / b) (C-2)
β ₁ = 2θ ₁ + r ₂ −r ₁ −θ ₁ ′ −φ ₂ ′ (C-3)
t ₂ = t ₁ sin (φ ₁ ′ + φ ₂ ′) / sin β ₁ (C-4)
t ₃ = − (R ₁ −t ₂ ) cos β ₁ + {R ₁ ² − (R ₁ −t ₂ ) ² sin ² β ₁ } ^1/2 (C-5)
β ₂ = sin ⁻¹ (sin β ₁ t ₃ / R ₁ ) (C-6)
t ₄ = R ₁ {2 (1-cosβ ₂ )} ^1/2 (C-7)
t ₅ = (R ₁ ² + ΔR ₀ ² -2R ₁ ΔR ₀ cos θ _R0 ) ^1/2 (C-8)
β ₄ = sin ^-1 [{t ₁ sin (φ ₁ '+ 2θ ₁ + r ₂ -r ₁ -θ ₁ ')
−asin (2θ ₁ + r ₂ −r ₁ −θ ₁ ′)} / t ₅ ] (C-9)
t ₇ = {R ₁ ² + t ₅ ² −2R ₁ t ₅ cos (β ₂ −β ₄ )} ^1/2 (C-10)
β ₅ = sin ⁻¹ [{R ₁ sin (β ₂ −β ₄ )} / t ₇ ] (C-11)
t ₈ = {t ₅ sin θ ₁ + R ₁ sin (β ₂ −β ₄ −θ ₁ )} / sinφ ₂ ′ (C−12)
t ₉ = [(O _B D) ² + t ₇ ² +2 (O _B D) {t ₅ cos θ ₁ −R ₁ cos (β ₂ −β ₄ −θ ₁ )}] ^1/2
(C-13)
t ₁₀ = [l _B2 ' ² + t ₉ ² -2l _B2 ' {t ₅ sinθ ₁ + R ₁ sin (β ₂ -β ₄ -θ ₁ )}] ^1/2
(C-14)
t ₁₁ = [l _B2 ' ² + {(O _B D) -a} ² ] ^1/2 (C-15)
β ₇ = cos ⁻¹ [(t ₈ ² + t ₁₀ ² −t ₁₁ ² ) / 2t ₈ t ₁₀ ] (C-16)
θ ₁ ″ = β ₁ −β ₂ + β ₇ (C−17)
sinr ₁ ″ = sin θ ₁ ″ / n (C-18)
sinr ₁ ′ = sin θ ₁ ′ / n (C-19)
_{t 12 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '') (C-20)
_{t 13 = {t 4 cos (} β 2/2 + r 1 '')} / sin (β 2 + r 1 '+ r 1'') (C-21)
t ₁₄ = (t ₁₂ ² + t ₁₃ ² −2t ₁₂ t ₁₃ cosr ₁ ′) ^1/2 (C−22)
Calculate
R ₂ ′ = {(R ₁ −t ₁₂ ) ² + t ₁₄ ²
+ 2t ₁₄ (R ₁ −t ₁₂ ) cos (β ₂ + r ₁ ″)} ^1/2 (7)
The wall thickness measuring device according to claim 3, wherein the radius R ₂ 'is calculated based on the equation (1). Let R ₁ be the radius of the circle formed by the outer wall surface,
When the radius of the circle formed by the inner wall surface of the container whose center position is displaced is R ₂ ′,
The thickness computing means, the wall thickness of the container, the radius R ₁ and radius R ₂ of the preceding claims, characterized in that is adapted to calculate a difference between _'according to any one Thickness measuring device. A wall thickness measuring device for measuring the wall thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container,
A laser light source for irradiating the outer wall surface of the container with laser light;
Optical path distribution means for distributing the scattered light to a first optical path and a second optical path different from the first optical path;
A first condensing means for condensing the scattered light scattered from the outer wall surface out of the scattered light from the optical path distribution means via the first optical path;
A third imaging device that reads scattered light collected by the first light collecting means as a scattered spot on the outer wall surface;
A second condensing means for condensing the scattered light scattered from the inner wall surface out of the scattered light from the optical path distribution means via the first optical path;
A fourth imaging device that reads scattered light collected by the second light collecting means as a scattered spot on the inner wall surface;
A second imaging element that reads scattered light from the optical path distribution means via the second optical path as a regular reflection spot;
Based on the position information of the scattering spot read by the third and fourth image sensors and the position of the specular reflection spot read by the second image sensor, the displacement of the center position of the container and the displaced center position A wall thickness measuring device comprising a wall thickness calculating means for calculating the wall thickness with respect to. A wall thickness measuring method for measuring the wall thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container,
Distributing scattered light scattered from the outer wall surface and the inner wall surface of the container to the first optical path and a second optical path different from the first optical path;
The scattered light passing through the first optical path is collected and read as a scattering spot on the outer wall surface and the inner wall surface by the first image sensor,
Read the scattered light via the second optical path as a regular reflection spot on the second image sensor,
Based on the position information of the scattering spot read by the first image sensor and the specular reflection spot read by the second image sensor, the displacement of the center position of the container and the thickness with respect to the displaced center position. Thickness measurement method characterized by calculating A wall thickness measuring method for measuring the wall thickness of a container by detecting scattered light scattered from the outer wall surface and the inner wall surface of the container,
Distributing scattered light scattered from the outer wall surface and the inner wall surface of the container to the first optical path and a second optical path different from the first optical path;
Of the scattered light that passes through the first optical path, the scattered light scattered from the outer wall surface is collected, and is read as a scattered spot on the outer wall surface by a third image sensor,
Of the scattered light that passes through the first optical path, the scattered light scattered from the inner wall surface is collected and read as a scattered spot on the inner wall surface by a fourth image sensor,
Read the scattered light via the second optical path as a regular reflection spot on the second image sensor,
Based on the position information of the scattering spot read by the third and fourth image sensors and the position of the specular reflection spot read by the second image sensor, the displacement of the center position of the container and the displaced center position A method for measuring a wall thickness, comprising calculating a wall thickness for

Images