The present invention relates to a shape measurement apparatus and a shape
10 measurement method.
Background Ali
[0002]
A slab, which is a steel semi-finished product, a thick sheet produced using
15 the slab, and the like are conveyed on a production line composed of a plurality of
rolls during the production process thereof. At this time, shape measurement using
what is called a light-section method is performed in order to measure the surface
height of a rigid body such as the slab or the thick sheet. However, when a rigid
body such as a slab or a thick sheet is conveyed on a production line, there has been a
20 problem that the fluctuation in the surface height derived from the vertical movement
and rotation of the rigid body (hereinafter, referred to as "disturbance") is
superimposed on the measured surface height, and the tme surface height cannot be
measured.
25
[0003]
To address the problem mentioned above, a teclmology shown in Patent
Literature 1 below proposes forming, in addition to a light-section line for the
original shape measurement formed in the width direction of a rigid body to be
measured, another light-section line in a direction oblique to the light-section line
mentioned above (directions that are not mutually parallel). In the technology, the
30 measurement of the same point of the rigid body to be measured, which is originally
supposed to have the same surface height, is performed twice for each of a plurality
PCT/JP2016/062801
2/76
of points of different longitudinal-direction positions and different width-direction
positions. After that, the magnitude of disturbance (vettical movement and
rotation) by which the surface heights of the plurality of points mentioned above
coincide most favorably is obtained by optimization calculation, and the effect of
5 disturbance is removed from the measurement result.
10
15
Citation List
Patent Literature
[0004]
Patent Literature I: JP 2013-221799A
Summmy of Invention
Technical Problem
[0005]
However, in the technology shown in Patent Literature 1 above, if the
measurement error is large in the surface height measurement of each measurement
point, the optimization calculation may not converge correctly. Furthermore, the
technology shown in Patent Literature 1 above has a problem that errors are
superimposed on the measurement result in the case where three types of disturbance
20 of vettical movement (translation in the height direction), rotation around the
longitudinal-direction axis, and rotation around the width-direction axis, which may
exist as disturbance, exist simultaneously.
[0006]
Thus, the present invention has been made in view of the problems
25 mentioned above, and an object of the present invention is to provide a shape
measurement apparatus and a shape measurement method that can measure the
smface height of a rigid body to be measured more accurately even if any one of
three types of disturbance of translation in the height direction, rotation around the
longitudinal-direction axis, and rotation around the width-direction axis has occurred
30 during conveyance.
PCT/JP2016/062801
3/76
Solution to Problem
[0007]
According to an aspect of the present invention in order to achieve the
above-mentioned object, there is provided a shape measurement apparatus that
5 measures a shape of a rigid body to be measured by means of a plurality of lightsection
lines based on a plurality of linear laser light beams applied to a surface of
the rigid body to be measured from a plurality of linear laser light sources moving
relative to the rigid body to be measured along a longitudinal direction of the rigid
body to be measured. The shape measurement apparatus includes: an imaging
10 apparatus that applies three beams of the linear laser light to the surface of the rigid
body to be measured moving relatively along the longitudinal direction and images
reflected light of the three beams of the linear laser light from the surface of the rigid
body to be measured at a prescribed longitudinal-direction interval; and an arithmetic
processing apparatus that performs image processing on captured images related to
15 the light-section lines imaged by the imaging apparatus and calculates a surface
shape of the rigid body to be measured. The imaging apparatus includes a first
linear laser light source that emits a shape-measuring light-section line that is the
light-section line extending in a width direction of the rigid body to be measured and
is used to calculate the surface shape of the rigid body to be measured, a second
20 linear laser light source that emits a first correcting light-section line that is parallel
to the longitudinal direction of the rigid body to be measured and crosses the shapemeasuring
light -section line, and is used to correct an effect of disturbance acting on
the rigid body to be measured, a third linear laser light source that emits a second
correcting light-section line that is parallel to the longitudinal direction of the rigid
25 body to be measured, crosses the shape-measuring light-section line, and exists in a
width-direction position of the rigid body to be measured different from the first
correcting light -section line, and is used to correct an effect of disturbance acting on
the rigid body to be measured, a first camera that images the shape-measuring lightsection
line at each time corresponding to a prescribed longitudinal-direction interval
30 and generates a captured image of the shape-measuring light-section line at each time,
and a second camera that images the correcting light-section lines at each time
PCT/JP2016/062801
4/76
corresponding to a prescribed longitudinal-direction interval and generates a captured
image of the correcting light-section lines at each time, and the arithmetic processing
apparatus includes a shape data calculation unit that, on the basis of the captured
image of the shape-measuring light-section line at each time generated by the first
5 camera, calculates shape data that show a three-dimensional shape of the smface of
the rigid body to be measured and in which a measurement error derived from the
disturbance is superimposed, a disturbance estimation unit that performs, on a
plurality of points of different longitudinal-direction positions of the first correcting
light-section line, height change value acquisition processing of acquiring, from
10 height measurement values related to a surface height of the rigid body to be
measured acquired at different two times for the same position of the rigid body to be
measured, a height change value derived from the disturbance at the position, using
captured images of the first correcting light -section line, performs the height change
value acquisition processing on a plurality of points of different longitudinal-
15 direction positions of the second cmTecting light-section line using captured images
of the second correcting light-section line, and estimates the amount of height
fluctuation derived 11-om the disturbance superimposed in the shape data, using a
plurality of height change values derived from the disturbance obtained from
captured images of the first correcting light-section line and a plurality of height
20 change values derived from the disturbance obtained from captured images of the
second correcting light-section line, and a correction unit that subtracts the amount of
height fluctuation from the shape data and thereby corrects the measurement error
derived from the disturbance.
25
[0008]
It is preferable that the disturbance estimation unit approximate, with a
straight line, height change values derived from the disturbance at a plurality of
points on the first correcting light-section line and estimates a height change value
derived from the disturbance at an intersection point of the straight line and the
shape-measuring light-section line, approximate, with a straight line, height change
30 values derived li'om the disturbance at a plurality of points on the second correcting
light-section line and estimates a height change value derived from the disturbance at
II
~~--·1
5
PCT/JP2016/062801
5176
an intersection point of the straight line and the shape-measuring light-section line,
and estimate the amount of height fluctuation by means of a straight line connecting
the height change values derived from the disturbance at the two intersection points.
[0009]
It is preferable that each of the first camera and the second camera perform
imaging at each time corresponding to a prescribed longitudinal-direction interval
and generates N (N being an integer of 2 or more) captured images, and that the
disturbance estimation unit calculate the amount of height fluctuation on the
assumption that the disturbance has not occurred in a 1st captured image.
10 [0010]
It is preferable that an imaging timing of the first camera and the second
camera be controlled so that a common irradiation region that is a portion of the rigid
body to be measured irradiated with the correcting light-section line in common
exists in captured images of the second camera captured at mutually adjacent
15 imaging times, and that the disturbance estimation unit calculate a height change
value derived from the disturbance for the plurality of points falling under the
common irradiation region of each of the first correcting light -section line and the
second correcting light-section line.
20
[00 II]
It is preferable that using an apparent surface height including the height
change value obtained from an i+l-th captured image (i = 1, 2, ... , N-1) ofthe second
camera and a surface height that is obtained Jiom an i-th captured image of the
second camera and that is after the height change value in the common irradiation
region of the i-th captured image is removed, the disturbance estimation unit
25 calculate the height change value in the i+ 1-th captured image and a surface height
after the height change value is removed.
[0012]
It is preferable that the disturbance estimation unit calculate the height
change value in an i-th captured image (i = 2, ... , N) of the second camera with a 1st
30 captured image of the second camera as a reference.
[0013]
PCT/JP2016/062801
6176
It is preferable that the first linear laser light source, the second linear laser
light source, and the third linear laser light source be provided such that an optical
axis of each light source is perpendicular to a plane defined by a longitudinal
direction and a width direction of the rigid body to be measured.
5 [0014]
It is preferable that an angle between an optical axis of the first camera and
an optical axis of the first linear laser light source, an angle between a line of sight of
the second camera and an optical axis of the second linear laser light source, and an
angle between the line of sight of the second camera and an optical axis of the third
10 linear laser light source be mutually independently not less than 30 degrees and not
more than 60 degrees.
[0015]
According to an aspect of the present invention in order to achieve a shape
measurement method that measures a shape of a rigid body to be measured by means
15 of a plurality of light-section lines based on a plurality of linear laser light beams
applied to a surface of the rigid body to be measured fi"om a plurality of linear laser
light sources moving relative to the rigid body to be measured along a longitudinal
direction of the rigid body to be measured, there is provided the shape measurement
method including: an imaging step of imaging reflected light of three light-section
20 lines from the surface of the rigid body to be measured at a prescribed longitudinaldirection
interval by applying the three light-section lines to the surface of the rigid
body to be measured moving relatively along the longitudinal direction from an
imaging apparatus including a first linear laser light source that emits a shapemeasuring
light -section line that is the light -section line extending in a width
25 direction of the rigid body to be measured and is used to calculate a surface shape of
the rigid body to be measured, a second linear laser light source that emits a first
correcting light-section line that is parallel to the longitudinal direction of the rigid
body to be measured and crosses the shape-measuring light-section line, and is used
to correct an effect of disturbance acting on the rigid body to be measured, a third
30 linear laser light source that emits a second correcting light-section line that is
parallel to the longitudinal direction of the rigid body to be measured, crosses the
PCT/JP2016/062801
7176
shape-measuring light-section line, and exists in a width-direction position of the
rigid body to be measured different from the first correcting light-section line, and is
used to correct an effect of disturbance acting on the rigid body to be measured, a
first camera that images the shape-measuring light-section line at each time
5 corresponding to a prescribed longitudinal-direction interval and generates a captured
image of the shape-measuring light-section line at each time, and a second camera
that images the correcting light -section lines at each time corresponding to a
prescribed longitudinal-direction interval and generates a captured image of the
correcting light -section lines at each time; a shape data calculation step of, on the
10 basis of the captured image of the shape-measuring light-section line at each time
generated by the first camera, calculating shape data that show a three-dimensional
shape of the surface of the rigid body to be measured and in which a measurement
error derived Jiom the disturbance is superimposed; a disturbance estimation step of
performing, on a plurality of points of different longitudinal-direction positions of
15 the first correcting light-section line, height change value acquisition processing of
acquiring, fi·om height measurement values related to a surface height of the rigid
body to be measured acquired at different two times for the same position of the rigid
body to be measured, a height change value derived from the disturbance at the
position, using captured images of the first correcting light-section line, performing
20 the height change value acquisition processing on a plurality of points of different
longitudinal-direction positions of the second correcting light-section line using
captured images of the second correcting light-section line, and estimating the
amount of height fluctuation derived from the disturbance superimposed in the shape
data, using a plurality of height change values derived from the disturbance obtained
25 fi·om captured images of the first correcting light-section line and a plurality of
height change values derived from the disturbance obtained from captured images of
the second correcting light-section line; and a correction step of subtracting the
amount of height fluctuation from the shape data and thereby correcting the
measurement error derived from the disturbance.
30 [0016)
It is preferable that in the disturbance estimation step, height change values
PCT/JP2016/062801
8/76
derived from the disturbance at a plurality of points on the first correcting lightsection
line be approximated with a straight line and thereby a height change value
derived from the disturbance at an intersection point of the straight line and the
shape-measuring light-section line be estimated, height change values derived from
5 the disturbance at a plurality of points on the second correcting light-section line be
approximated with a straight line and thereby a height change value derived from the
disturbance at an intersection point of the straight line and the shape-measuring lightsection
line be estimated, and the amount of height fluctuation be estimated by
means of a straight line connecting the height change valnes derived from the
10 disturbance at the two intersection points.
[0017]
It is preferable that each of the first camera and the second camera perform
imaging at each time corresponding to a prescribed longitudinal-direction interval
and generate N (N being an integer of 2 or more) captured images, and that in the
15 disturbance estimation step, the amount of height fluctuation be calculated on the
assumption that the disturbance has not occurred in a I st captured image.
[00 I8]
It is preferable that an imaging timing of the first camera and the second
camera be controlled so that a common irradiation region that is a portion of the rigid
20 body to be measured irradiated with the correcting light-section line in conm1on
exists in captured images of the second camera imaged at mutually adjacent imaging
times, and in the disturbance estimation step, a height change value derived fi·om the
disturbance be calculated for the plurality of points falling under the connnon
irradiation region of each of the first correcting light-section line and the second
25 correcting light-section line.
[OOI9]
It is preferable that in the disturbance estimation step, using an apparent
surface height including the height change value obtained from an i+ I-th captured
image (i = I, 2, · ··, N -I) of the second camera and a surface height that is obtained
30 fi·om an i-th captured image of the second camera and that is after the height change
value in the common irradiation region of the i-th captured image is removed, the
PCT/JP2016/062801
9/76
height change value in the i+ 1-th captured image and a surface height after the height
change value is removed be calculated.
[0020]
It is preferable that in the disturbance estimation step, the height change
5 value in an i-th captured image (i = 2, ... , N) of the second camera be calculated with
a 1st captured image of the second camera as a reference.
[0021]
It is preferable that the first linear laser light source, the second linear laser
light source, and the third linear laser light source be provided such that an optical
10 axis of each light source is perpendicular to a plane defined by a longitudinal
direction and a width direction of the rigid body to be measured.
[0022]
It is preferable that an angle between an optical axis of the first camera and
an optical axis of the first linear laser light source, an angle between a line of sight of
15 the second camera and an optical axis of the second linear laser light source, and an
angle between the line of sight of the second camera and an optical axis of the third
linear laser light source be mutually independently not less than 30 degrees and not
more than 60 degrees.
20 Advantageous Effects oflnvention
[0023]
As described above, according to the present invention, it becomes possible
to measure the surface height of a rigid body to be measured more accurately even if
any one of three types of disturbance of translation in the height direction, rotation
25 around the longitudinal-direction axis, and rotation around the width-direction axis
has occurred during conveyance.
Brief Description of Drawings
[0024]
30 [FIG. I] FIG. I is an explanatory diagram schematically showing the configuration
of a shape measurement apparatus according to an embodiment of the present
PCT/JP2016/062801
10/76
invention.
[FIG. 2] FIG. 2 is an explanatory diagram schematically showing the configuration
of an imaging apparatus included in the shape measurement apparatus according to
the embodiment.
5 [FIG. 3] FIG. 3 is an explanatory diagram schematically showing the configuration
of the imaging apparatus according to the embodiment.
[FIG. 4] FIG. 4 is an explanatory diagram schematically showing the configuration
of the imaging apparatus according to the embodiment.
[FIG. 5] FIG. 5 is an explanatory diagram schematically showing the configuration
10 of the imaging apparatus according to the embodiment.
[FIG. 6] FIG. 6 is an explanatory diagram schematically showing the configuration
of the imaging apparatus according to the embodiment.
[FIG. 7] FIG. 7 is an explanatory diagram schematically showing the configuration
of the imaging apparatus according to the embodiment.
15 [FIG. 8] FIG. 8 is a schematic diagram for describing disturbances that may occur on
a rigid body to be measured.
[FIG. 9] FIG. 9 is a schematic diagram for describing disturbances that may occur on
a rigid body to be measured.
[FIG. 10] FIG. 10 is a schematic diagram for describing disturbances that may occur
20 on a rigid body to be measured.
[FIG. 11] FIG. 11 is a schematic diagram for describing disturbances that may occur
on a rigid body to be measured.
[FIG. 12] FIG. 12 is a block diagram showing an example of the configuration of an
image processing unit of an arithmetic processing apparatus included in the shape
25 measurement apparatus according to the embodiment.
[FIG. 13] FIG. 13 is an explanatory diagram for describing disturbance estimation
processing performed by a disturbance estimation unit according to the embodiment.
[FIG. 14] FIG. 14 is an explanatory diagram for describing disturbance estimation
processing performed by a disturbance estimation unit according to the embodiment.
30 [FIG. 15] FIG. 15 is an explanatory diagram for describing disturbance estimation
processing performed by a disturbance estimation unit according to the embodiment.
PCT/JP2016/062801
11/76
[FIG. 16] FIG. 16 is a block diagram showing an example of the configuration of a
disturbance estimation unit included in the image processing unit according to the
embodiment.
[FIG. 17] FIG. 17 is a block diagram showing an example of the configuration of a
5 disturbance estimation unit included in the image processing unit according to the
embodiment.
[FIG. 18] FIG. 18 is a block diagram showing an example of the configuration of a
disturbance estimation unit included in the image processing unit according to the
embodiment.
10 [FIG. 19] FIG. 19 is a block diagram showing an example of the configuration of a
disturbance estimation unit included in the image processing unit according to the
embodiment.
[FIG. 20] FIG. 20 is a block diagram showing an example of the configuration of a
disturbance estimation unit included in the image processing unit according to the
15 embodiment.
[FIG. 21] FIG. 21 is a block diagram showing an example of the configuration of a
disturbance estimation unit included in the in1age processing unit according to the
embodiment.
[FIG. 22] FIG. 22 is a block diagram showing an example of the configuration of a
20 disturbance estimation unit included in the image processing unit according to the
embodiment.
[FIG. 23] FIG. 23 is a block diagram showing an example of the configuration of a
disturbance estimation unit included in the image processing unit according to the
embodiment.
25 [FIG. 24] FIG. 24 is an explanatmy diagram for describing shape data calculation
processing performed by a shape data calculation unit according to the embodiment.
[FIG. 25] FIG. 25 is an explanatory diagram for describing correction processing
performed by a correction processing unit according to the embodiment.
[FIG. 26] FIG. 26 is an explanatory diagram for describing correction processing
30 performed by a correction processing unit according to the embodiment.
[FIG. 27] FIG. 27 is an explanatmy diagram schematically showing a modification
PCT/JP2016/062801
12/76
example of the imaging apparatus according to the embodiment.
[FIG. 28] FIG. 28 is an explanatory diagram schematically showing a modification
example of the imaging apparatus according to the embodiment.
[FIG. 29A] FIG. 29A is a flow chart showing an example of the flow of a shape
5 measurement method according to the embodiment.
[FIG. 29B] FIG. 29B is a flow chart showing an example of the flow of a shape
measurement method according to the embodiment.
[FIG. 30] FIG. 30 is a block diagram showing an example of the hardware
configuration of the arithmetic processing apparatus according to the embodiment.
10 [FIG. 31A] FIG. 31A is an explanatory diagram for describing Experimental
Example 1.
[FIG. 31B] FIG. 31B is an explanatory diagram for describing Example 1.
[FIG. 31C] FIG. 31C is a graph showing a result of Experimental Example 1.
[FIG. 31 D] FIG. 31 D is a graph showing a result of Experimental Example I.
15 [FIG. 32A] FIG. 32A is an explanatory diagram for describing Experimental
Example 2.
[FIG. 32B] FIG. 32B is an explanatory diagram for describing Example 2.
[FIG. 32C] FIG. 32C is a graph showing a result of Experimental Example 2.
[FIG. 32D] FIG. 32D is a graph showing a result of Experimental Example 2.
20 [FIG. 33A] FIG. 33A is an explanatory diagram for describing Experimental
Example 3.
25
[FIG. 33B] FIG. 33B is an explanatory diagram for describing Example 3.
[FIG. 33C] FIG. 33C is a graph showing a result of Experimental Example 3.
[FIG. 33D] FIG. 33D is a graph showing a result of Experimental Example 3.
Description of Embodiments
[0025]
Hereinafter, (a) preferred embodiment(s) of the present invention will be
described in detail with reference to the appended drawings. In this specification
30 and the drawings, elements that have substantially the same function and structure
are denoted with the same reference signs, and repeated explanation is omitted.
PCT/JP2016/062801
13/76
[0026]
(With regard to overall configuration of shape measurement apparatus)
In the following, first, the overall configuration of a shape measurement
apparatus 10 according to an embodiment of the present invention is described with
5 reference to FIG. 1. FIG. 1 is an explanatmy diagram schematically showing the
configuration of a shape measurement apparatus according to the present
embodiment.
[0027]
The shape measurement apparatus 1 0 according to the present embodiment
10 is an apparatus that, by what is called the light-section method, measures the shape of
a rigid body to be measured by means of a plurality oflight-section lines based on a
plurality of linear laser light beams applied to a surface of the rigid body to be
measured from a plurality of linear laser light sources that·move relative to the rigid
body to be measured along the longitudinal direction of the rigid body to be
15 measured. In the following, a description is given using as an example the case
where the rigid body to be measured is conveyed on a production line.
[0028]
The following description uses, as shown in FIG. I, a space coordinate
system set in a space where the shape measurement apparatus 10 is provided. For
20 convenience of description, the width direction of a rigid body to be measured S is
defined as a C-axis direction (in the space coordinate system), the longitudinal
direction, that is, the conveyance direction of the rigid body to be measured S is
defined as an L-axis direction, and the height direction of the rigid body to be
measured S is defined as a Z-axis direction.
25 [0029]
Here, it is assumed that the rigid body to be measured S focused on in the
present embodiment is an object that can be regarded as not changing in shape or
volume during shape measurement processing like that described below. Thus, for
example, a slab, a thick sheet, and the like, which are semi-finished products in the
30 steel industry, may be treated as the rigid body to be measured S in the present
embodiment. Further, not only a slab or a thick sheet in the steel indushy but also,
5
PCT/JP2016/062801
14/76
for example, a slab, a thick sheet, and the like of various metals other than iron such
as titanium, copper, and aluminum, ceramics, and composite materials may be
treated as the rigid body to be measured S in the present embodiment.
[0030]
The shape measurement apparatus I 0 according to the present embodiment
includes, as shown in FIG. I, an imaging apparatus 100 that applies a plurality of
linear laser light beams to a surface of the rigid body to be measured S and images
the reflected light of linear laser light at the surface of the rigid body to be measured
S, and an arithmetic processing apparatus 200 that performs prescribed image
10 processing on an image captured by the imaging apparatus I 00 and calculates the
three-dimensional shape of the rigid body to be measured S (that is, the surface
height at each position in the L-axis-C-axis plane).
[0031]
The imaging apparatus I 00 is an apparatus that applies tlu·ee linear laser
15 light beams to a surface of the rigid body to be measured S, sequentially linages the
surface of the rigid body to be measured S along the longitudinal direction at each
time corresponding to a prescribed longitudinal-direction interval, and outputs the
captured image (light-section image) obtained by imaging to the arithmetic
processing apparatus 200 described later. The timing of applying linear laser light
20 to the rigid body to be measured S, the timing of imaging the surface of the rigid
body to be measured S, etc. of the imaging apparatus 100 are controlled by the
aritlunetic processing apparatus 200 described later. The imaging apparatus 100
performs one time of imaging processing each time the rigid body to be measured S
moves a prescribed distance (e.g., 1 mm or the like), on the basis of a PLG signal or
25 the like outputted from a pulse logic generator (a PLG, a pulse-type speed detector)
provided in a driving mechanism or the like that controls the conveyance of the rigid
body to be measured S, in association with, for example, the change in the
longitudinal-direction position of the rigid body to be measured S with respect to the
imaging apparatus 100.
30 [0032]
The arithmetic processing apparatus 200 IS an apparatus that performs
5
PCT/JP2016/062801
15/76
image processing like that described below on the light-section image at each time
generated by the imaging apparatus 100 and thereby calculates the three-dimensional
shape of the rigid body to be measured S.
[0033]
In the following, the imaging apparatus 100 and the arithmetic processing
apparatus 200 are described in detail with reference to the drawings.
[0034]

Next, the imaging apparatus 100 included m the shape measurement
10 apparatus 10 according to the present embodiment is described in detail with
reference to FIG. 2 to FIG. 7. FIG. 2 to FIG. 7 are explanatmy diagrams
schematically showing the configuration of an imaging apparatus according to the
present embodiment.
15
[0035]
The imaging apparatus 100 according to the present embodiment mainly
includes, as schematically shown in FIG. 2, three linear laser light sources lOla,
101 b, and 101 c (hereinafter, occasionally referred to as collectively a "linear laser
light source 101 ") each of which emits linear laser light, and two area cameras Ill
and 113. Here, the linear laser light source lOla is an example of a first linear laser
20 light source, the linear laser light source 101 b is an example of a second linear laser
25
light source, and the linear laser light source 101 c is an example of a third linear laser
light source. Fmther, the area camera Ill is an example of a first camera, and the
area camera 113 is an example of a second camera.
[0036]
In FIG. 2 and the subsequent drawings, a description is given using as an
example of a case where the imaging apparatus 100 includes two area cameras; but
the number of area cameras included in the imaging apparatus 1 00 according to the
present embodiment is not limited to this example. A case where the tmagmg
apparatus 100 includes three area cameras is described later.
30 [0037]
The linear laser light source 101 is a device that applies laser light in a linear
~~
~----l
PCT/JP20 16/062801
16/76
shape (linear laser light) to the surface of the rigid body to be measured (hereinafter,
occasionally referred to as simply a "rigid body") S that is a measurement object.
As the linear laser light source I 0 I according to the present embodiment, any light
source may be used as long as it can apply linear laser light to the surface of the rigid
5 body S; and the linear laser light source 101 may be configured using a laser light
source and various lenses such as a rod lens.
[0038]
As the laser light source, a continuous wave (CW) laser light source that
performs laser oscillation continuously may be used, for example. The wavelength
10 of the laser light oscillated by the laser light source is preferably a wavelength in the
visible light range of approximately 400 nm to 800 run, for example. The laser light
source performs the oscillation of laser light on the basis of an oscillation timing
control signal sent fiom the arithmetic processing apparatus 200 described later.
15
20
[0039]
Also in the case where a pulse laser light source that perfmms pulsed laser
oscillation is used as the laser light source, the light source can be treated in a similar
manner to a CW laser light source by synchronizing the oscillation timing of the
pulse laser and the imaging timing of the area cameras Ill and 113.
[0040]
A rod lens is a lens that spreads the laser light emitted from the laser light
source in a circular sectorial plane toward the surface of the rigid body S. Thereby,
the laser light emitted from the laser light source becomes linear laser light, and this
is applied to the surface ·of the rigid body S. From the viewpoint of image
processing in the arithmetic processing apparatus 200 described later, the laser light
25 source is preferably installed such that the circular sectorial plane obtained by the rod
lens is parallel to the Z-axis. In the linear laser light source 101 according to the
present embodiment, a lens other than a rod lens, such as a cylindrical lens or a
Powell lens, may be used as long as it can spread laser light in a circular sectorial
shape.
30 [0041]
On the surface of the rigid body S irradiated with linear laser light, a linear
~I !_-__ j
PCT/JP2016/06280 1
17/76
bright part (in FIG. 2 etc., shown as the black line) is formed. In the imaging
apparatus I 00 according to the present embodiment, since the three linear laser light
sources lOla, !Olb, and !Ole are used, tluee bright parts are formed. These linear
bright pmis are referred to as light-section lines. The reflected light of the light-
5 section line at the surface of the rigid body S propagates up to the area camera, is
formed as an image in an imaging element provided in the area camera, and is
imaged by the area camera.
[0042]
In the following description, the light-section line obtained by the linear
10 laser light source lOla is referred to as a light-section line La, the light-section line
obtained by the linear laser light source 10 I b is referred to as a light -section line Lb,
and the light-section line obtained by the linear laser light source !Ole is referred to
as a light-section line Lc. Further, the light-section lines La, Lb, and Lc may be
collectively refened to as a "light-sectionline L." Here, the light-section line La is
15 an example of a shape-measuring light-section line. Further, the light-section line
Lb and the light-section line Lc are examples of a correcting light-section line; for
example, the light-sectionline Lb conesponds to a first correcting light-section line,
and the light-sectionline Lc corresponds to a second correcting light-sectionline.
20
25
[0043]
Here, the linear laser light source I 0 I according to the present embodiment
is installed on the conveyance line so as to satisfy all the following three conditions,
as illustrated in FIG. 2.
• The light-section line La and the light -section line Lb have an intersection
point A.
· The light-section line La and the light-section line Lc have an intersection
point B.
· Both the light-sectionline Lb and the light-sectionline Lc be parallel to the
L-axis, and the light-section line Lb and the light-section line Lc exist in mutually
different width-direction positions on the surface of the rigid body S.
30 [0044]
In what is called the light-section method, only the light-section line La
PCT/JP2016/062801
18/76
shown in FIG. 2 is used and the surface heights of longitudinal-direction positions of
the rigid body S irradiated with the light-section line La are calculated, and the
obtained surface heights are arranged in a row in the longitudinal direction in
accordance with the relative movement between the rigid body S and the imaging
5 apparatus (for example, the conveyance of the rigid body S); thereby, the surface
height of the whole rigid body S can be found. However, in the case where
disturbance occurs during the conveyance of the rigid body S, the surface height
obtained by the light-section method using one light-section line is an apparent
surface height including the disturbance, and is a measurement value including an
10 error different from the true surface height.
[0045]
Thus, in the shape measurement apparatus 1 0 according to the present
embodiment, as described in detail below, the light -section line Lb extending in the
longitudinal direction of the rigid body S is added, and the relationships between
15 points of longitudinal-direction positions on the light-section line Lb and surface
height changes derived fi·om disturbance are approximated with a straight line.
After that, in the shape measurement apparatus 1 0 according to the present
embodiment, the value of the approximate straight line at the longitudinal-direction
position where the light-section line La exists (that is, the intersection point A of the
20 light-section line La and the light-section line Lb) is uniquely established as the
surface height change derived from disturbance of the light-section line La. Here, in
the shape measurement apparatus 1 0 according to the present embodiment, since the
measurement object is a rigid body, the change in the apparent surface height due to
disturbance from the surface height after disturbance removal (that is, the change in
25 the apparent surface height due to disturbance from the true surface height) changes
in a straight line along the longitudinal direction. Thus, by approximating the
measurement values at points on the light -section line Lb with a straight line, there is
provided an effect of absorbing the variation in the value due to measurement errors.
By the addition of the light-section line Lb like this, it becomes possible to uniquely
30 find the magnitude of two kinds of disturbance of vertical movement in the Zdirection
(the value of the approximate straight line takes a fixed value regardless of
5
PCT/JP201G/OG2801
19176
the longitudinal-direction position; that is, the slope of the approximate straight line
is 0) and rotation around the C-axis (the approximate straight line has a fixed slope
with respect to the longitudinal-direction position).
[0046]
In the shape measurement apparatus 10 according to the present
embodiment, the light-section line Lc is finther added to a width-direction position
different from the light-section line Lb, and processing similar to the processing of
the light-section line Lb is performed. Thereby, in the shape measurement apparatus
I 0 according to the present embodiment, it becomes possible to specify a relationship
10 between the change in the surface height derived from disturbance and the widthdirection
position, and it also becomes possible to obtain the magnitude of rotation
around the L-axis.
[0047]
Thus, in the shape measurement apparatus 10 according to the present
15 embodiment, by using tlu·ee light-section lines like the above, the surface height of
the rigid body to be measured can be measured more accurately even when any one
of tlu·ee types of disturbance of translation in the height direction, rotation around the
longitudinal-direction axis, and rotation around the width-direction axis has occurred
during conveyance.
20 [0048]
Although FIG. 2 and the subsequent drawings show the case where the
light-section line La and the light-section line Lb are orthogonal and the case where
the light-sectionline La and the light-section line Lc are orthogonal, the arrangement
of light-section lines (that is, the arrangement of linear laser light sources I 0 I) is not
25 limited to the cases shown in these drawings. That is, the following description
similarly holds also in the case where the light-section line La, and the light-section
line Lb and the section line Lc are not orthogonal. This is because, in the present
embodiment, the magnitude of disturbance at the intersection point A and the
intersection point B is calculated using the approximate straight line mentioned
30 above as described in detail below, and it is not necessary for two light-section lines
to be orthogonal. Fmther, although FIG. 2 and the subsequent drawings show the
PCT/JP2016/062801
20/76
case where the surface of the rigid body S is flat, the following description is not
limited to the case shown in these drawings, and similarly holds also in the case
where the surface of the rigid body S is not flat. The reason will be described
separately.
5 [0049]
The specific length of the light-seetionline Lis not patiicularly limited, and
the length may be determined such that the luminance distribution of the lightsection
line is uniform on the surface of the rigid body S, as appropriate. Further,
the width-direction position of the light-section lines Lb and Lc is not patiicularly
10 limited, either; the position may be set such that the light-section lines Lb and Lc exist
on the surface of the rigid body S, whatever width the rigid body S conveyed on the
conveyance line has.
[0050]
In the area cameras 11 1 and 113, a lens having a prescribed focal distance
15 and an imaging element such as a charge-coupled device (CCD) or a complementary
metal oxide semiconductor (CMOS) are installed. The area cameras 111 and 113
image the light-section line that is the reflected light of linear laser light applied to
the surface of the rigid body S each time the rigid body S moves a prescribed
distance, and generate a light-section image. After that, the area cameras 111 and
20 113 output the generated light-section image to the arithmetic processing apparatus
200 described later.
[0051]
Here, the area cameras Ill and 113 are controlled by the arithmetic
processing apparatus 200 described later, and a trigger signal for imaging is
25 outputted from the arithmetic processing apparatus 200 each time the rigid body S
moves a prescribed distance. The area cameras 111 and 113 image the surface of
the rigid body S irradiated with linear laser light in accordance with a trigger signal
outputted from the aritlm1etic processing apparatus 200 and generate a light-section
image, and output the generated light -section image to the arithmetic processing
30 apparatus 200. Thereby, N (N being an integer of 2 or more) captured images are
outputted from each of the area cameras 111 and 113 to the arithmetic processing
PCT/JP2016/062801
21/76
apparatus 200.
[0052]
As schematically shown in FIG. 3, the linear laser light source lOla is
installed on the conveyance line in such a marmer that a plane including the linear
5 laser light emitted from this light source is perpendicular to the L-axis-C-axis plane
(in other words, in such a mmmer that the optical axis of the linear laser light source
lOla is substantially parallel to the Z-axis). In the case where this installation
condition is not satisfied, the linear laser light illuminates a pmiion different in the
longitudinal-direction position of the rigid body S due to disturbance described later,
10 and it is difficult to perform accurate measurement of the surface shape. Also for
the linear laser light sources JO!b and !Ole, for a reason similar to the above, each
light source is installed on the conveyance line in such a manner that, as shown in
FIG. 4, a plane including the emitted linear laser light is perpendicular to the L-axisC-
axis plane (in other words, in such a manner that the optical axis of the linear laser
15 light sources lOlb and !Ole is substantially parallel to the Z-axis).
[0053]
Even in the case where light sources are installed in the above mmmer, the
irradiation position of linear laser light is not strictly the same when the rigid body S
is rotated around an axis parallel to each of the light-section lines La, Lb, and Lc by
20 disturbance (for example, when the rigid body S is rotated around the C-axis for the
light-section line La and around the L-axis for the light-section lines Lb and Lc).
However, assuming that the hue surface height of the rigid body S changes smoothly
and that the amount of rotation of the rigid body S is not large, it can be surmised
that linear laser light illuminates the same position of the surface of the rigid body S
25 even when such a rotation has occurred. In pmiicular, when attention is focused on
the surface shape of a rigid body S with a large mass, such as a slab or a thick sheet,
it can be said that the latter assumption is appropriate.
[0054]
In regard to the optical positional relationship between the linear laser light
30 source lOla and the area camera III, as shmm in FIG. 3, the magnitude of the angle
a 1 between the optical axis of the area camera Ill and the optical axis of the linear
~l ;.j
PCT/JP2016/062801
22/76
laser light source lOla (in other words, the Z-axis) in the L-axis-Z-axis plane may be
set to an arbitrary value. However, the magnitude of the angle a 1 is preferably set to
approximately 30 degrees to 60 degrees. If the angle a 1 < 30 degrees, the amount of
movement of the light-section line La in the camera's visual field is small for the
5 same height change, and the resolution along the height direction is reduced. On
the other hand, if the angle a 1 > 60 degrees, the area camera 111 is distant from the
regular reflection direction of the linear laser light source lOla, and the light-section
line La photographed by the area camera 111 is dark; hence, a higher power laser is
needed in order to perform photographing with the same brightness.
10 [0055]
The area camera 111 is preferably installed such that, as shown in FIG. 5,
the optical axis of the area camera 111 projected on the L-axis-C-axis plane and the
light -section line La are orthogonal to each other. Thereby, it becomes possible to
equalize the resolution (the length corresponding to one pixel (unit: mm)) in the C-
15 axis direction of the light-section line La viewed from the area camera 111.
However, as mentioned above, the light-section line La and the light -section lines Lb
and Lc may not be orthogonal. That is, the light -section line La may not be parallel
to the width direction (the C-axis direction). This is because, as described above,
the light -section line La, and the light -section lines Lb and Lc may not be orthogonal
20 to calculate the amount of disturbance at the intersection point A and the intersection
point B.
[0056]
Here, as schematically shown in FIG. 5, the imaging area ARI of the area
camera Ill is set such that the entire light-section line La is included in the imaging
25 visual field.
[0057]
In regard to the optical positional relationship between the linear laser light
sources !Olb and 101c and the area camera 113, as schematically shown in FIG. 4,
the magnitude of the angles a2 and a3 between the line of sight of the area camera
30 113 to the light-section lines Lb and Lc in the C-axis-Z-axis plane and the optical axes
of the respective linear laser light sources !Olb and !Ole (in other words, the Z-axis)
5
PCT/JP2016/062801
23/76
may be set to an arbitrary value similarly to the angle a 1• However, for a similar
reason to the angle a 1, the magnitude of each of the angles a2 and a.3 is preferably set
to approximately 30 degrees to 60 degrees.
[0058]
Similarly to the relationship between the light-section line La and the area
camera 111, the light-section line Lb in the L-axis-C-axis plane and the optical axis of
the area camera 113 projected on the L-axis-C-axis plane are preferably orthogonal
to each other. At this time, the light-section line Lb and the light-section line L, are
parallel to each other; thus, when this condition holds for the light-section line Lb, the
10 condition is automatically satisfied also for the light-section line Lc.
[0059]
Here, as schematically shown in FIG. 6, the imaging area AR2 of the area
camera 113 is set such that the intersection point A and the intersection point B are
included in the imaging visual field. Here, although FIG. 6 shows the case where
15 the entire light-section lines Lb and Lc are included in the imaging visual field,
disturbance estimation processing described later can be performed as long as at least
the intersection point A and the intersection point B are included in the imaging
visual field. In order to increase the accuracy of disturbance estimation processing
described later, it is preferable that the entire light-section lines Lb and Lc be included
20 in the imaging visual field.
[0060]
The imaging timing of the area cameras 111 and 113 is set such that, for
example as schematically shown in FIG. 7, a pmtion of the rigid body S being
irradiated with light-section lines in common (hereinafter, refened to as a "common
25 irradiation pmtion") exists in captured images of the area camera 113 captured at
mutually adjacent times (for example, the imaging time of the i-th image (i being an
integer of 1 or more) and the imaging time of the i+l-th image). This is because, in
the arithmetic processing apparatus 200 according to the present embodiment, the
magnitude of disturbance is calculated with a focus on the light-section lines Lb and
30 Lc in the common irradiation p01tion, as described in detail below. Although FIG. 7
shows the case where the surface of the rigid body S is flat and disturbance has not
lkl
,R--t-l,
5
PCT/JP2016/062801
24/76
occurred between consecutive two images, the common irradiation pmiion exists
also in the case where the surface of the rigid body S is not flat or in the case where
disturbance has occurred between consecutive two images.
[0061]
Hereinabove, the configuration of the imaging apparatus I 00 according to
the present embodiment is described in detail with reference to FIG. 2 to FIG. 7.
[0062]

Next, disturbance occurring on the rigid body to be measured S and the
10 captured image (light-section image) caph1red in association with the disturbance are
specifically described with reference to FIG. 8 to FIG. II. FIG. 8 to FIG. 11 are
schematic diagrams for describing disturbance that may occur on the rigid body to be
measured.
15
[0063]
The shape measurement apparatus 10 according to the present embodiment
measures the surface height of the rigid body S on the occasion when a rigid body
such as a slab or a thick sheet is continuously conveyed or on other occasions. Here,
during the conveyance of the rigid body S, there are various factors of measurement
error, such as vibration derived fi'om a driving mechanism provided on the
20 conveyance line etc.
25
[0064]
In the shape measurement apparatus I 0 according to the present
embodiment, attention is focused on the following three factors as the factor of
measurement error, as shown in FIG. 8.
S)
(I) translation in the Z-axis direction (the height direction of the rigid body
(2) rotation around the L-axis (the longitudinal direction of the rigid body S)
(3) rotation around the C-axis (the width direction of the rigid body S)
Hereinafter, these three factors of measurement error may be collectively
30 referred to as dishn·bance.
[0065]
PCT/JP2016/062801
25/76
With a rigid body S having a flat surface as an object, the change of the
captured image photographed by the area camera 113 occurring depending on the
presence or absence of disturbance will now be described with reference to FIG. 9 to
FIG. II.
5 [0066]
Although FIG. 9 to FIG. 11 make illustration with a focus on the case where
the surface of the rigid body Sis flat, the entire following description is not limited to
the case shown in FIG. 9 to FIG. 11, and similarly holds also in the case where the
surface of the rigid body S is not flat. This is because, in the case where the surface
10 of the rigid body S is not flat, although the light-section line itself is a curved line,
the change of the light-section line depending on the presence or absence of
disturbance changes in a straight line along the longitudinal direction like in the case
of being flat.
15
[0067]
First, in the case where disturbance like the above has not occun·ed between
two images at mutually different two times (for example, the i-th captured image and
the i+ 1-th captured image), the position of each light-section line L does not change
between the captured images. However, in the case where translation in the Z-axis
direction has occmTed as disturbance at the time of capturing the i+ 1-th image, as
20 shown in FIG. 9 the light-section lines La, Lb, and Lc translate in the vertical
direction in the image by a mutually equal amount. Further, in the case where
rotation around the L-axis has occurred as disturbance at the time of capturing the
i+l-th image, as shown in FIG. I 0, the slope and length of the light-section line La
change, and the light-section lines Lb and Lc translate in the image by mutually
25 different amounts. Further, in the case where rotation around the C-axis has
occurred as disturbance at the time of capturing the i+ 1-th image, as shown in FIG.
11 the slope of the light -section lines Lb and Lc changes.
[0068]
Thus, in the arithmetic processing apparatus 200 described in detail below,
30 the change in the surface height (the change in the Z-coordinate) derived from
disturbance that has occurred on the rigid body S is calculated at each imaging time
PCT/JP2016/062801
26/76
by comparing consecutive two images obtained by the area camera 113. After that,
the surface height that is obtained fi·orn the light -section image of the area camera
111 and in which a measurement error due to disturbance is superimposed is
corrected on the basis of the obtained change in the surface height derived from
5 disturbance (in other words, the magnitude of disturbance), and the true surface
height is outputted.
[0069]

Next, the aritlunetic processing apparatus 200 included in the shape
10 measurement apparatus 10 according to the present embodiment is described in detail
with reference to FIG. 1 and FIG. 12 to FIG. 26. FIG. 12 is a block diagram
showing an example of the configuration of an image processing unit of an
aritlunetic processing apparatus included in a shape measurement apparatus
according to the present embodiment. FIG. 14 and FIG. 15, and FIG. 17 to FIG. 23
15 are explanatory diagrams for describing disturbance estimation processing performed
by a disturbance estimation unit according to the present embodiment. FIG. 16 is a
block diagram showing an example of the configuration of a disturbance estimation
unit included in an image processing unit according to the present embodiment.
FIG. 24 is an explanatory diagram for describing shape data calculation processing
20 performed by a shape data calculation unit according to the present embodiment.
25
FIG. 25 and FIG. 26 are explanatory diagrams for describing correction processing
performed by a correction processing unit according to the present embodiment.
[0070]
[With regard to overall configuration of arithmetic processing apparatus J
Returning to FIG. 1 again, the overall configuration of the arithmetic
processing apparatus 200 included in the shape measurement apparatus I 0 according
to the present embodiment is described.
The arithmetic processing apparatus 200 according to the present
embodiment mainly includes, as shown in FIG. 1, an imaging control unit 201, an
30 image processing unit 203, a display control unit 205, and a storage unit 207 ..
[0071]
PCT/JP2016/062801
27/76
The imaging control unit 201 is configured with, for example, a central
processing unit (CPU), a read-only memory (ROM), a random access memmy
(RAM), a communication device, etc. The imaging control unit 20 1
comprehensively controls the processing of imaging the rigid body S performed by
5 the imaging apparahts I 00 according to the present embodiment.
[0072]
More specifically, when starting the imaging of the rigid body S, the
imaging control unit 201 sends to the imaging apparahts 100 a control signal for
starting the oscillation of the linear laser light source 10 I. When the imaging
10 apparatus I 00 stmis the imaging of the rigid body S, the imaging control unit 201
sends a trigger signal for stmiing imaging to the area cameras I I I and I I 3 each time
it acquires a PLG signal sent at regular intervals Jiom a driving mechanism or the
like. that controls the conveyance of the rigid body S (for example, a PLG signal
outputted each time the rigid body S moves I mm or at other times).
15 [0073]
The image processing unit 203 is configured with, for example, a CPU, a
ROM, a RAM, a communication device, etc. The image processing unit 203
acquires imaging data generated by the area cameras Ill and 113 (that is, captured
image data related to the light-section image), performs image processing described
20 below on the imaging data, and calculates the height of the entire surface of the rigid
body S as three-dimensional shape data. On finishing the processing of calculating
the surface height of the rigid body S, the image processing unit 203 transmits
information on the obtained calculation result to the display control unit 205 and the
storage unit 207, or transmits the information to various devices etc. provided outside
25 the shape measurement apparatus 10.
[0074]
The image processing unit 203 is described in detail later.
[0075]
The display control unit 205 is configured with, for example, a CPU, a
30 ROM, a RAM, an output device, a communication device, etc. The display control
unit 205 performs display control at the time of displaying the measurement result of
PCT/JP2016/062801
28/76
the rigid body S transmitted from the image processing unit 203 on an output device
such as a display included in the arithmetic processing apparatus 200, an output
device provided outside the arithmetic processing apparatus 200, or the like.
Thereby, the user of the shape measurement apparatus 1 0 can grasp the measurement
5 result concerning the three-dimensional shape of the rigid body S on the spot.
[0076]
The storage unit 207 is an example of a memory device included in the
aritlunetic processing apparatus 200, and is configured with, for example, a ROM, a
RAM, a storage device, etc. In the storage unit 207, calibration data related to the
10 light -section line L used for image processing performed in the image processing unit
203 are stored. Fmiher, in the storage unit 207, information on design parameters
of the shape measurement apparatus 10, such as information showing the optical
positional relationship between the linear laser light source 101 and the area cameras
111 and 113 included in the imaging apparatus 1 00 and infonnation transmitted from
15 a higher-level computer provided outside the shape measurement apparatus 10 (for
example, a management computer that generally manages the conveyance line, or the
like), is also stored. Fmihe1~ in the storage unit 207, various parameters and reports
on processing still in progress that need to be saved when the arithmetic processing
apparatus 200 according to the present embodiment performs some kind of
20 processing (for example, measurement results transmitted from the image processing
unit 203, and calibration data, various databases, programs, etc. stored beforehand)
are recorded, as appropriate. The imaging control unit 201, the image processing
unit 203, the display control unit 205, a higher-level computer, etc. can freely
perform data read/write processing on the storage unit 207.
25 [0077]
Details of calibration data stored in the storage unit 207 are described later.
[0078]
[With regard to configuration of image processing unit]
Next, the configuration of the image processing unit 203 included in the
30 arithmetic processing apparatus 200 is described with reference to FIG. 12 to FIG. 26.
The image processing unit 203 according to the present embodiment
5
PCT/JP2016/062801
29/76
includes, as shown in FIG. 12, an imaging data acquisition unit 211, a disturbance
estimation unit 213, a shape data calculation unit 215, a correction unit 217, and a
result output unit 219.
[0079]
The imaging data acquisition unit 211 is configured with, for example, a
CPU, a ROM, a RAM, a communication device, etc. The imaging data acquisition
unit 211 acquires the imaging data of the light-section line outputted from the area
cameras Ill and 113 of the imaging apparatus I 00 (that is, image data related to the
light-section image). When the imaging data acquisition unit 211 acquires, from the
10 area camera 113, imaging data related to the light-section lines Lb and Lc used as the
correcting light-section line (in other words, imaging data in which the imaging area
AR2 in FIG. 6 is imaged), the imaging data acquisition unit 211 outputs the imaging
data to the disturbance estimation unit 213 described later. Further, the imaging
data acquisition unit 211 outputs imaging data related to the light-section line La used
15 as the shape-measuring light-section line (in other words, imaging data in which the
imaging area AR1 in FIG. 5 is imaged) from the area camera 111 to the shape data
calculation unit 215 described later.
[0080]
Further, the imaging data acquisition unit 211 may associate, with imaging
20 data related to the light-section line acquired from the imaging apparatus 100, time
information on the date and time at which the imaging data are acquired and on other
matters, and may store these pieces of information as history information in the
storage unit 207 or the like.
25
[0081]
The disturbance estimation unit 213 is configured with, for example, a CPU,
a ROM, a RAM, etc. The disturbance estimation unit 213 is a processing unit that
estimates the magnitude of disturbance that has occurred on the rigid body S using
the imaging data of the correcting light -section line imaged by the area camera 113
(that is, the light-section lines Lb and Lc).
30 [0082]
More specifically, the disturbance estimation unit 213 performs, on the
PCT/JP20 16/062801
30/76
captured image obtained from the area camera 113, height change value acquisition
processing of acquiring, from height measurement values related to the surface
height of the rigid body S acquired at different two times for the same position of the
rigid body S, a height change value derived from disturbance at the position. At
5 tllis time, the disturbance estimation unit 213 performs the height change value
acquisition processing mentioned above on a plurality of points of different
longitudinal-direction positions of the light-section line Lb, and performs the height
change value acquisition processing mentioned above on a plurality of points of
different longitudinal-direction positions of the light-section line Lc. After that, the
10 disturbance estimation unit 213 estimates the amount of height fluctuation
superimposed on shape data calculated by the shape data calculation unit 215
described later, using the height change value at the intersection point A obtained
from the light-section line Lb and the height change value at the intersection point B
obtained fi"om the light-section line Lc.
15 [0083]
Disturbance estimation processing in the disturbance estimation unit 213 is
described in detail later.
[0084]
On finishing disturbance estimation processing described in detail below,
20 the disturbance estimation unit 213 outputs the obtained result of disturbance
estimation to the correction milt 217 described later. Further, the disturbance
estimation unit 213 may associate, with data showing the obtained result concerning
disturbance estimation, time information on the date and time at which the data are
generated and on other matters, and may store these pieces of information as history
25 information in the storage unit 207 or the like.
[0085]
The shape data calculation unit 215 is configured with, for example, a CPU,
a ROM, a RAM, etc. The shape data calculation unit 215 expresses the threedimensional
shape of the surface of the rigid body S on the basis of the captured
30 image concerning the light-section line La at each time generated by the area camera
Ill, and calculates shape data in which a measurement error derived from
PCT/JP2016/062801
31/76
disturbance is superimposed.
[0086]
The processing of calculating shape data in the shape data calculation unit
215 is described later.
5 [0087]
On finishing the processing of calculating shape data described below, the
shape data calculation unit 215 outputs the obtained shape data to the correction unit
217 described later. Further, the shape data calculation unit 215 may associate, with
the obtained shape data, time information on the date and time at which the shape
10 data are generated and on other matters, and may store these pieces of information as
history information in the storage unit 207 or the like.
[0088]
The correction unit 217 is configured with, for example, a CPU, a ROM, a
RAM, etc. 1l1e conection unit 217 subtracts the amount of height fluctuation
15 calculated by the disturbance estimation unit 213 mentioned above from the shape
data calculated by the shape data calculation unit 215, and thereby conects the
measurement error derived from disturbance. Thus, the true shape data related to
the rigid body S in which the measurement error associated with disturbance that
may occur on the rigid body S is removed are generated.
20 [0089]
Correction processing in the correction unit 217 is described later.
[0090]
On finishing correction processing described below, the correction unit 217
outputs the corrected shape data to the result output unit 219 described later.
25 [0091]
The result output unit 219 is configured with, for example, a CPU, a ROM,
a RAM, an output device, a communication device, etc. The result output unit 219
outputs to the display control unit 205 information on the surface shape of the rigid
body S outputted from the correction unit 217. Thereby, information on the surface
30 shape of the rigid body S is outputted to a display unit (not shown). Further, the
result output unit 219 may output the obtained result concerning surface shape
PCT/JP2016/062801
32/76
measurement to an external device such as a process computer system for production
management, or may usc the obtained measurement result to create various record
files. Further, the result output unit 219 may associate information on the surface
shape of the rigid body S with time information on the date and time at which the
5 information is calculated and on other matters, and may store these pieces of
information as histmy information in the storage unit 207 or the like.
10
(0092]
o With regard to disturbance estimation processing in disturbance estimation unit
213
In the following, disturbance estimation processmg performed in the
disturbance estimation unit 213 is described in detail with reference to FIG. 13 to
FIG. 23.
First, before describing disturbance estimation processing, calibration data
used in the disturbance estimation processing are described.
15 (0093]
0 With regard to calibration data
As mentioned above, calibration data related to the light -section line L used
for disturbance estimation processing in the disturbance estimation unit 213 and
shape calculation processing in the shape data calculation unit 215 are stored
20 beforehand in the storage unit 207. The calibration data stored beforehand in the
storage unit 207 include two kinds of calibration data of first calibration data and
second calibration data.
[0094]
The first calibration data are calibration data necessary to convett the
25 amount of change in the position of the light-section line on the captured image
captured by the area cameras Ill and 113 (unit: pixels) to the amount in the real
space (unit: a unit oflength such as nun or m; in the following, a description is given
using the unit of mm).
30
[0095]
The first calibration data are data calculated from the normal imaging
resolution (nun/pixel) of the area camera and the angles u1, u2, and u3 between the
PCT/JP201G/OG2801
33/76
line of sights to the light-section lines L3 , Lb, and L, and the Z-axis direction.
However, in the present invention, the rigid body S that is the measurement object
gets nearer or fariher on the optical axis of the camera; therefore, the imaging
resolution and the angles a 1, a2, and a3 are not strictly a constant but are a value
5 varying with the height of the rigid body S. Thus, in the case where the height
change of the rigid body S that is the measurement object is large, a calibration curve
expressing a relationship between the position of the light-section line in the captured
image and the height in the real space is needed. Hereinafter, the first calibration
data are referred to as a calibration curve. The calibration curve is set for each of
10 the light -section lines La, Lb, and L,.
(0096]
The first calibration data can be calculated by calculation, or can be
obtained by actual measurement.
In the case where the first calibration data are calculated by calculation, the
15 focal distance f of the lens installed in the area cameras Ill and 113, the distance a
from the lens to the measurement object (that is, the rigid body S), and the distance b
from the imaging element provided in the area cameras Ill and 113 to the lens are
used. More specifically, the first calibration data can be calculated by finding a
magnification m expressed by Formula 1 03 by an image formation formula
20 expressed by Formula 101 below using the above parameters.
25
(0097]
(0098]
Image formation formula: 1/f= 1/a + 1/b ... (Formula 101)
Magnification: m = b/a .... (Formula 1 03)
Here, when the pixel size of the imaging element provided in the area
cameras Ill and 113 is denoted by d (nun), the imaging resolution D (mm/pixel) is a
value expressed by Formula 105 below. The imaging resolution D is an imaging
resolution in a plane perpendicular to the line of sight; thus, when the angle between
the line of sight and the normal direction is a degrees, the amount of ve1iical
30 movement H (mm) of the measurement object corresponding to one pixel is a value
expressed by Formula I 07 below.
5
PCT/JP2016/062801
34/76
[0099]
D = d/m ... (Formula I 05)
H = D/sin a . ... (Formula 1 07)
[0100]
The amount of vertical movement H of the measurement object
corresponding to one pixel obtained in the above manner serves as a conversion
factor for convetiing the amount of change of the light-section line (unit: pixels) on
the captured image captured by the area cameras Ill and 113 to the amount in the
real space (unit: e.g., mm). Thus, values given by Formula 107 above on the basis
10 of the optical positional relationships between the area cameras Ill and 113 and the
15
light -section lines La, Lb, and Lc corresponding to the area cameras Ill and 113 can
be used as calibration curves ca, Cb, and c" (that is, the first calibration data) related
to the light -section lines La, Lb, and Lc, respectively.
[0101]
In the case where the first calibration data are actually measured, a
calibration plate is prepared and installed· in a reference plane with a coordinate in
the height direction of Z = 0, and light-section images are captured by each of the
area cameras 111 and I 13 while the calibration plate is translated in the Z-axis
direction by /I.Z [mm]. After that, the amount of movement /I.Zimg [unit: pixels] of
20 the light-section line Lon a pixel basis in the obtained captured image of each of the
area cameras I 11 and 113 may be actually measured for a plurality of points, and a
calibration curve /I.Z = C(/I.Zimg) may be created (provided that C(/I.Zimg) expresses a
function with /I.Zimg as a variable). Thereby, calibration curves ca, ch, and c"
related to the light-section lines La, Lb, and L0 , respectively, can be obtained.
25 [0102]
Next, the second calibration data are described with reference to FIG. 13.
The second calibration data are data showing the amount of movement
(unit: pixels) in the horizontal direction in the image corresponding to the
conveyance distance (unit: a unit of length such as mm or m) of the rigid body S in
30 the real space between consecutive two imaging times on the real space shm\~1 in
FIG. 13. The second calibration data are set for each of the light-section lines Lb
PCT/JP20 16/062801
35/76
and Lc. As described later, the amount of movement in the ve1iical direction of the
same point on the rigid body S can be compared between consecutive two
photographed images by translating the captured image captured by the area camera
113 in the horizontal direction (a direction corresponding to the L-axis direction in
5 the real space) to the extent of the amount of movement mentioned above. Thus,
the second calibration data are calibration data used to estimate the magnitude of
disturbance.
[0103]
The second calibration data also can be calculated by calculation, or can be
10 obtained by actual measurement.
As described above, the second calibration data are data showing how many
pixels the conveyance distance ~s (~s shown in FIG. 13) in the real space of the rigid
body S in the period when consecutive two photographed images are generated
corresponds to in the generated captured image. Thus, in the case where the second
15 calibration data are calculated by calculation, the imaging resolution D calculated by
Formula 1 OS above may be calculated for both of the light-section lines Lb and Le,
and the set value of the conveyance distance ~s in the real space may be divided
using the obtained imaging resolutions Db and De. That is, when the amount of
movement in the horizontal direction related to the light-sectionline Lb is denoted by
20 ~L b and the amount of movement in the horizontal direction related to the lightsection
line Leis denoted by ~Le, these values can be calculated by Formula 109 and
Formula Ill below.
25
[0104]
[0105]
~Lb = ~s/Db ···(Formula 109)
~L e =~siDe ··· (Formula Ill)
In the case where the second calibration data are actually measured, a
calibration plate may be installed in a reference plane of Z = 0, like in the case where
the first calibration data are actually measured, and captured images may be
30 generated while the calibration plate is translated by ~s [ mm] in the L-axis direction.
After that, the obtained captured images may be analyzed to measure the amounts of
h---i
r: __ i
PCT/JP2016/062801
36/76
movement 11L b and 11L c in the horizontal direction in the captured image.
[01 06]
Hereinabove, two kinds of calibration data used in the image processing unit
203 according to the present embodiment are described.
5 [0107]
0 With respect to coordinate system used in disturbance estimation processing
Next, a coordinate system used in disturbance estimation processmg 1s
specifically described with reference to FIG. 14 and FIG. 15.
In disturbance estimation processing performed m the disturbance
10 estimation unit 213 according to the present embodiment, image processing is
performed using a coordinate system fixed to the captured image captured by the
area camera 113. That is, in the light -section image generated by the area camera
113, a direction corresponding to the longitudinal direction of the rigid body S (that
is, the horizontal direction of the light-section image) is defined as an X -axis
15 direction, and a direction orthogonal to the X-axis direction (that is, the height
direction of the light-section image) is defined as a Y-axis direction.
[01 08]
Fmiher, a position in the height direction that is in a captured image of the
area camera 113 captured while a flat surface of a calibration plate or the like is
20 placed in a position of Z = 0 and where the light -section line Lb is imaged is taken as
a reference position of a Y-coordinate yb for the light-section line Lb (that is, a
position of yb = 0), and a reference position of an X-coordinate Xb is set at the left
end of the captured image. As a result, the X-coordinate Xb for the light-section
line Lb is defined along the extending direction of the light-section line Lb, and the
25 X-axis direction Xb and the Y-axis direction yb for the light-section line Lb are
defined as shown in FIG. 14.
[0109]
Similarly, a position in the height direction that is in a captured image of the
area camera 113 captured while a flat surface of a calibration plate or the like is
30 placed in a position of Z = 0 and where the light-sectionline Lc is imaged is taken as
a reference position of a Y-coordinate yc for the light-section line Lc (that is, a
PCT/JP2016/062801
37/76
position of yc = 0), and a reference position of an X-coordinate Xc is set at the left
end of the captured image. As a result, the X-coordinate Xc for the light-section
line Lc is defined along the extending direction of the light-section line Lc, and the Xaxis
direction xc and theY-axis direction yc for the light-section line Lc are defined
5 as shown in FIG. 15.
[0110]
A coordinate system can be defined similarly also in a captured image of the
area camera 111 captured while a flat surface of a calibration plate or the like is
placed in a position of Z = 0. That is, a position in the height direction where the
10 light -section line La is imaged is taken as a reference position of a Y-coordinate y•
for the light-section line La (that is, a position ofYa = 0), and an X-coordinate Xa for
the light-section line La is defined along the extending direction of the light-section
line La, with the left end of the captured image as a reference. A specific example
of the coordinate system for the light-section line La is mentioned later with
15 reference to FIG. 24.
[0 111]
In the following description, the case where "height" is mentioned expresses
a value in the vettical direction in the captured image, that is, the y•, yb, and yc_
coordinates (unit: pixels), and the value in the case where the "height" in the captured
20 image is converted to the real space (unit: mm) by the calibration curves ca, Cb, and
C" is expressed as "the height in the Z-coordinate" or the like.
[0112]
0 With regard to details Qf disturbance estimation processing
Next, disturbance estimation processing performed m the disturbance
25 estimation unit 213 is described in detail with reference to FIG. 16 to FIG. 23.
In the disturbance estimation unit 213 according to the present embodiment,
the height change value derived from disturbance (that is, the amount of change in
the Z-coordinate in the real space) in a portion of the surface of the rigid body S
existing on the light-section lines Lb and Lc is calculated on the basis of the captured
30 image captured by the area camera 113 in which the light-section lines Lb and Lc are
present.
,__i
PCT/JP2016/062801
38/76
[0113]
o Overview of disturbance estimation processing
In a light-section method like that proposed by Patent Literature 1 above,
surface height measurement is performed at different times for a plurality of points of
5 different longitudinal-direction positions on the light-section line, the difference in
surface height measurement result between points (that is, a change derived from
disturbance) is used as it is for the calculation of the magnitude of disturbance.
However, in the light-section method performed in the shape measurement apparatus
I 0 according to the present embodiment, the relationship between the longitudinal-
10 direction position of each point on the light-section line Lb (that is, the value of the
Xb-coordinate) and the change in the value of the yb_coordinate derived from
disturbance at each of these points is specified by disturbance estimation processing
performed by the disturbance estimation unit 213, using a plurality of captured
images captured at different times. After that, the disturbance estimation nnit 213
15 approximates the distribution along the Xb-direction of the amounts of change in the
Yb-coordinate with a straight line. By using the approximate straight line, the
disturbance estimation unit 213 can accurately calculate the amount of change in the
value of the yb -coordinate at the Xb -coordinate corresponding to the intersection
point A shown in FIG. 2, while suppressing the variation in the value due to the
20 measurement error at each point on the light-section line Lb. After that, the
disturbance estimation unit 213 converts the amount of change in the value of the yb_
coordinate expressed on a pixel basis to the amount of change in the Z-coordinate in
the real space (that is, the amount of height fluctuation derived from disturbance), by
using a calibration curve Cb like that described above.
25 [0114]
30
Also the change in the Z-coordinate derived from disturbance at the
intersection point B shown in FIG. 2 can be found similarly to the above by focusing
on the light-section line Lc in place of the light-section line Lb.
[0115]
Next, assuming that the C-coordinate in the real space (that is, the width
direction of the rigid body S) is a reference, the amounts of change in the ZPCT/
JP2016/062801
39/76
coordinate at the intersection point A and the intersection point B calculated in the
above manner may be plotted on a plane, with the C-coordinate on the horizontal
axis and the amount of change in the Z-coordinate on the vetiical axis. Since the
measurement object focused on in the shape measurement apparatus 10 according to
5 the present embodiment is a rigid body, the amounts of change in the Z-coordinate at
points in the width direction of the rigid body S located between the intersection
point A and the intersection point B are supposed to change in a straight line in the
real space. Therefore, when a straight line passing through the amounts of change
in the Z-coordinate at the intersection point A and the intersection point B is
10 imagined on the C-axis-Z-axis plane like the above, the amounts of change in the Zcoordinate
at points in the width direction of the rigid body S located between the
intersection point A and the intersection point B can be expressed in the real space.
Thus, by finding a straight line like the above on the C-axis-Z-axis plane, the
disturbance estimation unit 213 can find the change in the Z-coordinate derived from
15 disturbance at width-direction positions connecting the two intersection points.
[0116]
The above is an overview of disturbance estimation processing performed in
the disturbance estimation unit 213; the disturbance estimation unit 213 that performs
disturbance estimation processing includes, as shown in FIG. 16, a common
20 irradiation pmiion disturbance estimation unit 221 and an intersection point position
disturbance estimation unit 223.
[0 117]
o With regard to common irradiation pmiion disturbance estimation unit 221
The common irradiation portion disturbance estimation unit 221 is
25 configured with, for example, a CPU, a ROM, a RAM, etc. The common
irradiation pmiion disturbance estimation unit 221 is a processing unit that specifies,
of the processing briefly mentioned in the overview of disturbance estimation
processing mentioned above, the relationship between the longitudinal-direction
position of each point on the light-section lines Lb and Lc (that is, the value of the Xb-
30 coordinate and the Xc -coordinate) and the change in the value of the yb -coordinate
and the yc_coordinate derived from disturbance at each of these points, using a
!----!
PCT/JP2016/062801
40176
plurality of captured images captured at different times.
[0118]
The common irradiation portion disturbance estimation unit 221 performs
the processing of calculating the change value of the value of the yb_coordinate and
5 the yc -coordinate derived from disturbance like the above, on the common
irradiation p01iion shown in FIG. 7. In the following, the processing performed by
the common irradiation portion disturbance estimation unit 221 is described in detail
with reference to FIG. 17 to FIG. 20.
10
[0119]
As described with reference to FIG. 7, when a moving rigid body S is
imaged by the area camera 113, a region imaged in common (that is, the common
irradiation portion shown in FIG. 7) exists in consecutive two captured images (for
example, the i-th captured image and the i+l-th captured image). Therefore, when
the i-th captured image captured by the area camera 113 is translated by L'lL b in the
15 negative direction of the Xb -axis on the basis of the second calibration data, the Xbcoordinate
of the common irradiation portion of the i-th image and the Xb -coordinate
of the common irradiation p01iion of the i+ 1-th image can be caused to coincide.
Similarly, also for the light-section line Lc, when the i-th captured image captured by
the area camera 113 is translated by L'lL c in the negative direction of the Xc -axis on
20 the basis of the second calibration data, the Xc -coordinate of the common irradiation
portion of the i-th image and the Xc -coordinate of the common irradiation portion of
the i+ 1-th image can be caused to coincide. Since the common irradiation portion is
the same position on the rigid body S, the hue surface height of the common
irradiation p01iion in the real space is the same. Therefore, by equalizing the X-
25 coordinate and then comparing the Y-coordinate of the common irradiation p01iion in
the i-th image and the Y-coordinate of the common irradiation potiion in the i+ 1-th
image, it becomes possible to estimate the magnitude of disturbance that has
occurred on the rigid body S at the time of capturing the i+ 1-th image.
30
[0120]
More specifically, the common irradiation portion disturbance estimation
unit 221 calculates the surface height after disturbance removal in the i+ 1-th captured
PCT/JP2016/062801
41/76
image and a height change (hereinafter, referred to as a "disturbance component")
due to a disturbance component in the i+ 1-th image, using an apparent surface height
(hereinafter, referred to as an "apparent height") including a disturbance component
that is obtained from the i+ 1-th captured image and the surface height after
5 disturbance removal in the common irradiation portion in the i-th captured image.
[0121]
When disturbance has occurred on the rigid body S, the yb -coordinate of the
light -section line Lb and the yC -coordinate of the light -section line Lc that are present
in the caphu'Cd image obtained by the area camera 113 change in the manner
10 illustrated in FIG. 9 to FIG. II. FIG. 17 is an explanatmy diagram for describing a
method for calculating a change value in the yb -coordinate derived from disturbance
in the common irradiation pmiion disturbance estimation unit 221. Although FIG.
17 shows the case where translation in the Z-axis direction has occurred as
disturbance between consecutive two images, the entire following description is not
15 limited to the case where translation in the Z-axis direction has occurred as
disturbance, and similarly holds also in the case where rotation around the L-axis has
occurred and the case where rotation around the C-axis has occurred. The reason is
that, in all of the three disturbances, the change in the yb_coordinate and the yc_
coordinate derived from disturbance can be approximated with a straight line because
20 the focused-on measurement object is a rigid body.
[0122]
The common irradiation portion dish1rbance estimation unit 221 performs
processing similar to the processing performed on the light -section line Lb also on the
light-section line Lc. Hence, in the following drawings and description, a
25 description is given using the processing performed on the light-section line Lb
representatively.
[0123]
The common irradiation portion disturbance estimation unit 221 first
executes the following processing on two cap hired images of the i-th image and the
30 i+ 1-th image photographed by the area camera 113 for the Xb -coordinate belonging
to the respective conm1on irradiation portions.
PCT/JP2016/062801
42/76
[0124]
In the following description, the light-section line Lb in the i-th captured
image in the (Xb, Yb) coordinate system is seen as a function ofXb, and is expressed
as yb = Fobsb(i, Xb). Further, in the following, Fobsb(i, Xb) is referred to as an
5 "apparent height" of the light-section line Lb.
[0 125]
As shown in FIG. 9 to FIG. 11, the position of the light-section line in the
captured image changes due to disturbance; the vertical movement of the lightsection
line derived from disturbance of the i-th captured image with the captured
10 image of i = 1st as a reference is expressed as a disturbance component of db(i, Xb).
Here, assuming that a common light -section method is used, this can be seen as a
method in which the magnitude of disturbance is estimated by specifying the vertical
movement of the position of the light-section line in the i+ 1-th captured image with
the position of the light-section line in the i-th captured image as a reference (that is,
15 disturbance is estimated between captured image frames). However, note that the
light-section method according to the present embodiment is, as mentioned above
and described in detail below, a method in which the magnitude of disturbance is
estimated with the position of the light -section line in the 1st captured image as a
reference.
20 [0126]
In view of FIG. 9 to FIG. 11 and the like for reference, the apparent height
of the light-section line Lb in the i-th captured image can be sees as "a value in which
a change in the· position of the light-section line derived from a disturbance
component is added to the surface height that is supposed to be observed in the case
25 where disturbance does not exist." That is, as schematically shown in FIG. 17, the
apparent height of the light-section line Lb of the i-th captured image can be seen as
the sum total of the disturbance component and the surface height after the
disturbance is removed (that is, the surface height that is supposed to be observed in
the case where disturbance does not exist; hereinafter, occasionally referred to as
30 simply "a surface height after disturbance removal"). Since the measurement object
is a rigid body as mentioned from above, the disturbance component db(i, Xb) can be
PCT/JP2016/062801
43176
seen as a linear function, that is, a straight line for Xb.
[0127]
Here, in the disturbance estimation processing according to the present
embodiment, it is assumed that "the disturbance component in the 1st captured image
5 is zero." That is, it is assumed that Formula 121 below holds for all the Xbcoordinates
belonging to the common irradiation portion in the 1st captured image
and the 2nd and subsequent captured images in which the common irradiation
portion in the I st captured image exists.
10
[0128]
db(1, Xb) = 0 ··· (Formula 121)
[0129]
There may be a case where disturbance has been added in the 1st image; in
this case, the surface height outputted by the image processing according to the
present embodiment in the end is a value in which a plane determined by the
15 magnitude of a disturbance component that had already be.en added at the time of
capturing the 1st image is uniformly added to the original surface height. However,
in the case where a reference plane is determined in a rigid body S like a slab of a
steel semi-finished product, the surface height viewed from the reference plane can
be obtained by performing a correction in which a plane is subtracted so that the
20 surface height of the total length and the total width outputted in the end coincides
with the reference plane. I·lence, in the following, a description is given on the
assumption that Formula 121 above holds.
[0130]
At this moment, as shown in FIG. 17, the surface height after disturbance
25 removal of the portion irradiated with the light-section line Lb at the photographing
time of the i-th image can be obtained by subtracting the disturbance component
from the apparent surface height. That is, the surface height after disturbance
removal Hb(i, X b) of the rigid body S irradiated with the light-section line Lb in the ith
captured image can be found in accordance with Formula 123 below.
30 [0131]
PCT/JP201G/OG2801
44/76
[0132]
Further, the disturbance component in the i+ 1-th captured image can be
found by subtracting the surface height after disturbance removal fi·om the apparent
height in the i+l-th captured image. That is, Formula 125 below holds.
5 [0133]
[0134]
Here, the surface height after disturbance removal Hb(i+ 1, Xb) in the i+ 1-th
captured image cannot be measured from the i+ 1-th image alone. However, since
10 the common irradiation portion is the same position on the rigid body S, the surface
height after disturbance removal in the i+ 1-th captured image is equal to the surface
height after disturbance removal in the i-th captured image. Thus, the common
irradiation pmiion disturbance estimation unit 221 according to the present
embodiment uses, as the surface height after disturbance removal Hb(i+ 1, Xb) in the
15 i+ 1-th captured image, a value obtained by translating the surface height after
disturbance removal H\i, Xb) in the i-th image that has already been found by
Formula 123 by Ll.Lb in the conveyance direction (that is, the negative direction of the
Xb-axis) and equalizing the common irradiation portion. That is, the fact that the
relationship expressed by Formula 127 below holds is utilezed.
20 [0135]
[0136]
Therefore, by substituting Formula 127 in Formula 125, the disturbance
component db(i+ 1, Xb) of the i+ 1-th image can be found by Formula 129 below using
25 the apparent height obtained from the i+ 1-th image and the surface height after
disturbance removal of the i-th image.
30
[0137]
[0138]
Further, in Formula 123 above, the parameter i is incremented by one into i
= i + 1, and the disturbance component of the i-th image obtained by Formula 129
fl o;
PCT/JP2016/062801
45/76
above is substituted in the portion of db(i+ 1, Xb); thereby, the surface height after
disturbance removal Hb(i+ 1, Xb) in the i+ 1-th captured image can be found.
[0139]
Tims, the common irradiation portion disturbance estimation unit 221 uses
5 Formula 129 and Formula 123 alternately, with Formula 121 at i = 1 taken as the
initial value, and sequentially increments the value of the parameter i by one; and
thereby can sequentially calculate the surface height after disturbance removal in the
i-th image and the disturbance component in the i+ 1-th image.
10
15
[0140]
In the following, how the processmg of specifYing the disturbance
component in the common irradiation portion by the common irradiation portion
disturbance estimation unit 221 mentioned above is performed in the case where it is
used for a situation like that shown in FIG. 18 is specifically described.
[0141]
In the following, a description is given only for the light-section line Lb; but
this similarly applies also to the light-section line Lc.
[0142]
In FIG. 18, a rigid body Sin which a part of the pmiion to be irradiated with
the light -section line Lb has unevenness is used as a measurement object. At this
20 moment, it is assumed that, as shown in the left half of FIG. 18, translation in the Zdirection
has occurred as disturbance while the 1st and 2nd captured images are
captured.
[0 143]
FIG. 19 is an explanatory diagram for describing processmg based on
25 Formula 129 in the common irradiation portion in the 1st captured image and the 2nd
captured image. As shown in Formula 125 above, the disturbance component db(2,
Xb) in the 2nd captured image is the difference between the apparent height Fobsb(2,
Xb) and the smface height after disturbance removal H\2, Xb) in the 2nd captured
image. On the other hand, as described from above, the surface height after
30 disturbance removal Hb(2, Xb) in the 2nd captured image is a value obtained by
translating the surface height after disturbance removal Hb(l, Xb) in the 1st captured
PCT/JP2016/062801
46/76
image by L'.Lb as shown by the broken line in FIG. 19 (that is, Hb(l, Xb+L'.Lb)).
Here, from Formula 121 above, Hb(l, Xb) is equal to Fob/(1, Xb). Therefore, Hb(l,
Xb+L'.Lb) is equal to Fobsb(l, Xb+L'.Lb). Thus, referring to FIG. 19, the disturbance
component d\2, Xb) in the 2nd captured image is equal to a value obtained by
5 subtracting, from the apparent height Fobsb(2, Xb), a value obtained by translating the
apparent height of the 1st image by L'.Lb. That is, the situation shown in FIG. 19
corresponds to the formula shown as Formula 129 above expressed as a drawing.
[0144]
In the case of FIG. 18, since the disturbance that has occurred on the rigid
10 body S is translation in the Z-direction, the disturbance component (the magnitude
shown by the alternate long and shmt dash line in FIG. 19) db(2, Xb) found by
Formula 129 is fixed regardless of the Xb-coordinate.
[0 145]
Next, it is assumed that, as shown in the right half of FIG. 18, rotation
15 around the C-axis has occurred while the 2nd and 3rd captured images are captured.
In this case, as is clear when the 1st captured image and the 3rd captured image are
compared in FIG. 18, translation in the Z-direction and rotation around the C-axis
have occurred as disturbance on the rigid body S, assuming that the 1st captured
image is a reference.
20 [0146]
25
FIG. 20 is an explanatmy diagram for describing processmg based on
Formula 123 and Formula 129 in the common irradiation portion of the 2nd captured
image and the 3rd captured image.
[0147]
As shown in the pmtion of the right half of FIG. 20, the surface height after
disturbance removal Hb(2, Xb) can be calculated by subtracting, fi·om the apparent
height Fobsb(2, Xb) obtained from the 2nd image, the disturbance component d\2, Xb)
that has already been calculated on the basis of FIG. 19. This relationship is the
relationship expressed by Formula 123 above expressed as a drawing.
30 [0148]
Next, as shown in the pmtion of the left half of FIG. 20, the surface height
PCT/JP2016/062801
47/76
after disturbance removal H\2, Xb) in the conm1on irradiation portion of the 2nd
captured image is translated by L'.Lb so that the common irradiation portion coincides
with the 3rd captured image, and is subtracted from the apparent height obtained
from the 3rd captured image; thus, the disturbance component d\3, Xb) in the 3rd
5 captured image can be calculated.
[0149]
Here, rotation around the C-axis has been added between the 2nd captured
image and the 3rd captured image; when the length of the alternate long and short
dash line shown in the left half of FIG. 20 (that is, db(3, Xb)) is plotted against the X-
10 coordinate Xb, db(3, Xb) forms a straight line having a certain slope.
[0150]
Here, as is clear also from FIG. 20, the disturbance component db(3, Xb) of
the 3rd captured image is a value obtained by subtracting the surface height after
disturbance removal Hb(2, Xb) in the 2nd captured image from the apparent height
15 Fobsb(3, Xb) of the 3rd captured image, and the surface height after disturbance
removal H\2, Xb) in the 2nd captured in1age is a value obtained by subtracting the
disturbance component d\2, X b) of the 2nd captured image from the apparent height
Fobsb(2, Xb) of the 2nd captured image. Therefore, the disturbance component db(3,
Xb) of the 3rd captured inlage can also be seen as an amount based on the
20 disturbance component db(2, Xb) of the 2nd captured image. Similarly, the
disturbance component db(2, Xb) of the 2nd captured image can be grasped as an
amount based on the disturbance component db(!, Xb) of the 1st captured image.
As is clear from such relationships, the disturbance estimation processing according
to the present embodiment specifies the disturbance component db(i, Xb) in the i-th
25 captured image as a result of the accumulation of all the disturbances from the
disturbance in the I st captured image to the disturbance in the (i-1 )st captured image.
[0 !51]
In the case where rotation around the L-axis has occurred as disturbance, the
magnitude of the disturbance component on the light-section line Lb is fixed
30 regardless of the Xb -coordinate like in the case where translation in the Z-direction
has occurred. The disturbance component d0 (i, Xc) on the light-section line Lc
PCT/JP2016/062801
48/76
existing in a different width-direction position in the real space is also fixed
regardless of the coordinate Xc. However, the values of the disturbance component
db and the disturbance component de are different; thus, the fact that rotation around
the L-axis exists can be grasped.
5 [0152]
By the common irradiation portion disturbance estimation unit 221
performing processing like the above, the magnitude of the disturbance component
db(i, Xb) on the light-section line Lb can be calculated using two consecutive captured
Images. By using processing like the above also for the light-section line Lc
10 similarly, the common irradiation portion disturbance estimation unit 221 can
calculate the magnitude of the disturbance component dc(i, Xc) on the light-section
line Lc.
[0153]
The common irradiation portion disturbance estimation unit 221 outputs
15 information on the magnitude of the disturbance component on each of the lightsection
lines Lb and Lc thus calculated to the intersection point position disturbance
estimation unit 223 described later.
20
[0154]
o With regard to intersection point position disturbance estimation unit 223
The intersection point position disturbance estimation unit 223 is configured
with, for example, a CPU, a ROM, a RAM, etc. The intersection point position
disturbance estimation unit 223 is a processing unit that performs, of the processing
briefly mentioned in the overview of disturbance estimation processing mentioned
above, the processing of approximating the distribution along the Xb -direction of the
25 amount of change in the yb_coordinate with a straight line for the light-section line
Lb and approximating the distribution along the Xc-direction of the amount of change
in the yc -coordinate with a straight line for the light-section line Lc, and thus
estimates the magnitude of disturbance at the positions of the intersection points A
and B.
30 [0155]
More specifically, the intersection point position disturbance estimation unit
r-J
Fi
PCT/JP2016/062801
49176
223 approximates the distribution of the magnitude of disturbance along the Xcoordinate
with a straight line using the magnitude of disturbance in the common
irradiation portion calculated by the common itTadiation pmtion disturbance
estimation unit 221, and extrapolates (depending on circumstances, interpolates) the
5 obtained approximate straight line up to the position of the intersection point; thereby,
calculates the magnitude of disturbance at the intersection point A and the
intersection point B. By the approximation with a straight line, the variation
occurring between points on the light-section lines Lb and Lc can be absorbed, and
the value of disturbance at the intersection point A and the intersection point B can be
10 found with better accuracy, as compared to conventional light-section methods
including the invention described in Patent Literature 1 above. After that, the
intersection point position disturbance estimation unit 223 converts the surface
height expressed on a pixel basis to a value in the Z-coordinate (unit: 111111) using the
calibration curves cb and C that are the first calibration data, and calculates the
15 magnitude of disturbance in the Z-coordinate of the intersection points A and B.
[0156)
As mentioned above, the intersection point position disturbance estimation
unit 222 is a processing unit that calculates each of
· the change in the Z-coordinate L1Zb(i) (unit: 111111) derived from the
20 disturbance component at the intersection point A, and
· the change in the Z-coordinate L1Zc(i) (unit: mm) derived from the
disturbance component at the intersection point B,
in the i -th image.
[0157]
25 The reason for finding the disturbance components at the two intersection
points A and B is the following two reasons. The first reason is that the
measurement object is a rigid body, and therefore the disturbance components da(i,
Xa) along the light -section line La in the captured image captured by the area camera
111 and the disturbance components in the Z-coordinate obtained by converting the
30 disturbance components da(i, x•) with the calibration curve c• are a straight line like
in the case of the light-section lines Lb and Lc. The second reason is that the values
n
fl
PCT/JP2016/062801
50176
of the disturbance components at two points on a straight line related to the lightsection
line La can be specified, and thereby the value of the disturbance component
in a place other than the intersection points can be estimated for the light-section line
La.
5 [0158]
In the following, processing performed by the intersection point position
disturbance estimation unit 223 is described in detail with reference to FIG. 21 to
FIG. 23. Although FIG. 21 shows the case where translation in the Z-axis direction
has occurred as disturbance between consecutive two captured images, the following
10 description is not limited to the case shown in FIG. 21, and can similarly apply also
to the case where rotation around the L-axis has occutTed and the case where rotation
around the C-axis has occurred.
[0159]
At this moment, at the imaging time of the i-th captured image, an apparent
15 Z-coordinate including a disturbance component is expressed as zb(i) for the
intersection point A of the light-section line La and the light-section line Lb, and an
apparent Z-coordinate including a disturbance component is expressed as zc(i) for
the intersection point B of the light-section line La and the light-section line Lc.
20
[0160]
Further, as shown in FIG. 21, with the imaging time of the 1st captured
image as a reference, the surface height in the Z-coordinate (that is, the surface
height in the Z-coordinate after disturbance removal) on the assumption that
disturbance does not occur until the i-th image is expressed as zb1(i) for the
intersection point A, and is expressed as zc1(i) for the intersection point B.
25 [0161]
As shown in FIG. 21 and Formula 131 below, the difference between the
apparent surface height Z\i) at the intersection point A in the Z-coordinate and the
surface height after disturbance removal Z\(i) in the Z-coordinate is defined as a
change in the Z-coordinate t.zb(i) due to a disturbance component. Similarly, as
30 shown in Formula 133 below, the difference between the apparent surface height
zc(i) at the intersection point B in the Z-coordinate and the surface height after
5
PCT/JP2016/062801
51/76
disturbance removal zc,(i) in the Z-coordinate is defined as a change in the Zcoordinate
L'.Zc(i) due to a disturbance component.
[0162]
[0163]
L'.Z\i) = zb(i) - zb1(i) .. · (Formula 131)
L'.Zc(i) = zc(i) - zc,(i) ... (Formula 133)
To calculate the change in the Z-coordinate L'.Zb(i) due to a disturbance
component, as shown in FIG. 22, the intersection point position disturbance
estimation unit 223 takes into account how the magnitude of the disturbance
10 component db(i, Xb) outputted from the common irradiation portion disturbance
estimation unit 221 is distributed along the Xb direction. After that, the intersection
point position disturbance estimation unit 223 approximates the distribution of the
disturbance component db(i, Xb) along the Xb direction with a straight line by known
statistical treatment such as the method of least squares. After that, the intersection
15 point position disturbance estimation unit 223 calculates a disturbance component
db(i, A) (unit: pixels) that is the magnitude of the disturbance component at the
intersection point A, using the Xb-coordinate of the intersection point A and the
calculated approximate straight line.
20
[0164]
After calculating the disturbance component db(i, A) (unit: pixels) at the
intersection point A, the intersection point position disturbance estimation unit 223
converts the magnitude of the disturbance component on a pixel basis to the
disturbance component in the Z-coordinate L'.Zb(i) (unit: mm), using the calibration
curve Cb that is the first calibration data.
25 [0165]
Here, when calculating the disturbance component in the Z-coordinate
L'.Zb(i) in the real space, it is important to take into account the fact that the
calibration curve Cb is a curved line and the disturbance component db(i, A) is, as
mentioned from above, a disturbance component with the 1st captured image as a
30 reference. Specifically, to find L'.Zb(i) using a calibration curve Cb like that shown
in FIG. 23, it is necessary to perform conversion from the pixel unit to the mm unit at
PCT/JP2016/062801
52/76
two points on the calibration curve and take a difference in the Z-coordinate.
(0166]
Here, as mentioned from above, the value obtained by adding the
disturbance component d\i, A) to the surface height after disturbance removal Hb(i,
5 A) is the apparent height Fobs\i, A) of the intersection point A in the i-th captured
image. Thus, as shown in FIG. 23, the intersection point position disturbance
estimation unit 223 calculates the apparent surface height zb(i) of the intersection
point A in the Z-coordinate of the i-th image using the apparent height Fobs\i, A) of
the intersection point A and the calibration curve Cb. Further, the intersection point
10 position disturbance estimation unit 223 calculates the surface height after
disturbance removal zb1(i) in the Z-coordinate of the i-th image using the surface
height after disturbance removal Hb(i, A) and the calibration curve cb. After that,
the intersection point position disturbance estimation unit 223 calculates the
difference between the obtained two surface heights, and thereby calculates the
15 disturbance component in the Z-coordinate LI.Zb(i) at the intersection point A.
20
Fm1her, the intersection point position disturbance estimation unit 223 completely
similarly calculates also the disturbance component in the Z-coordinate LI.Zc(i) at the
intersection point B.
(0167]
The intersection point position disturbance estimation unit 223 outputs
information on the magnitude of the disturbance component at the intersection point
A and the intersection point B thus calculated to the correction unit 217.
(0168]
Hereinabove, disturbance estimation processing performed m the
25 disturbance estimation unit 213 is described in detail with reference to FIG. 16 to
FIG. 23.
[0169]
o With regard to shape data calculation processing in shape data calculation unit 215
Next, shape data calculation processing performed in the shape data
30 calculation unit 215 is described in detail with reference to FIG. 24. Although FIG.
24 shows the case where rotation around the L-axis has occurred as disturbance,
~I l--l
PCT/JP2016/062801
53/76
similarly to the above description, the following description is not limited to the case
shown in FIG. 24.
[0170]
In the shape data calculation unit 215, first as shown in FIG. 24, an apparent
5 height Fobs"(i, Xa) (unit: pixels) for the light-section line La in the i-th captured image
is specified with reference to captured image data captured by the area camera Ill
that is outputted from the imaging data acquisition unit 211. Here, as described
above, a coordinate system in the captured image shown in FIG. 24 can be defined
using a captured image of the area camera Ill captured while a flat surface of a
10 calibration plate or the like is placed in a position of Z = 0. That is, a position in the
height direction where the light-section line La is imaged may be defined as a
reference position of a Y-coordinate ya for the light-section line La (that is, a position
of ya = 0), and an X-coordinate Xa for the light-sectionline La may be defined along
the extending direction of the light -section line La, with the left end of the captured
15 image as a reference.
[0171]
Next, the shape data calculation unit 215 converts the apparent height Fobs"
(i, Xa) (unit: pixels) obtained from the i-th captured image to an apparent height in
the Z-coordinate Z(i, Xa) (unit: a unit of length such as 111111), using the calibration
20 curve c• that is the first calibration data stored in the storage unit 207.
[0172]
The apparent height in the Z-coordinate Z(i, x•) thus calculated is a value in
which a change in the Z-coordinate derived from dishn·bance (that is, a measurement
error) is superimposed. The shape data calculation unit 215 outputs information on
25 the apparent height in the Z-coordinate Z(i, Xa) thus calculated to the correction unit
217 described later.
[0173]
o With regard to correction processing in correction unit 217
Next, correction processing performed in the correction unit 217 IS
30 described in detail with reference to FIG. 25 and FIG. 26.
[0174]
PCT/JP20 16/06280 I
54/76
The correction unit 217 according to the present embodiment performs
correction processing using shape data including a measurement error calculated by
the shape data calculation unit 215 (the apparent height in the Z-coordinate Z(i, x•))
and a disturbance component calculated by the disturbance estimation unit 213 (the
5 disturbance component in the Z-coordinate ""zb(i)), and calculates the true surface
height of the rigid body S that is the measurement object. The correction
processing is repeated for all the images captured by the area camera Ill, and
thereby the true surface height is placed one upon another in the longitudinal
direction; as a result, it becomes possible to calculate the true surface height in the
10 whole rigid body S.
[0175]
More specifically, the correction unit 217 first calculates a straight line like
that shown in FIG. 25, using the disturbance components in the Z-coordinate ""Zb(i)
and AZc(i) at the intersection point A and the intersection point B calculated by the
15 disturbance estimation unit 213. As mentioned above, the disturbance component
in the Z-coordinate AZ(i, Xa) along the light-section line La is a linear function (that
is, a straight line) with respect to the coordinate X" because the measurement object
is a rigid body. Therefore, the disturbance component in the Z-coordinate ""Z(i, X")
along the light-section line La can be specified by calculating a straight line
20 connecting the disturbance components in the Z-coordinate ""zb(i) and AZc(i) at the
intersection point A and the intersection point B.
[0176]
Subsequently, as sho\\~1 in FIG. 26 and Formula 141 below, the correction
unit 217 subtracts the change in the Z-coordinate due to disturbance (that is, the
25 disturbance component ""Z(i, X")) from Z(i, X") obtained by the shape data
calculation unit 215, and thereby calculates the true surface height in the Zcoordinate
Zout(i, X").
[0177]
Z0111(i, Xa) = Z(i, X") - AZ(i, X") ··· (Formula 141)
30 [0178]
The correction unit 217 repeats the above processing for all the images
u
II ' ~
PCT/JP2016/062801
55176
captured by the area camera 111 (that is, assummg that the number of images
photographed by each of the area cameras 111 and 113 is N, repeats the processing of
finding Z0111(i, Xa) for i = 1, 2, ···, N), and sequentially arranges the true surface
heights in the longitudinal direction; and can thereby calculate the true surface height
5 of the whole rigid body S.
[0179]
Hereinabove, correction processing performed in the correction unit 217
according to the present embodiment is described with reference to FIG. 25 and FIG.
26.
10 [0180]
An example of the function of the arithmetic processing apparatus 200
according to the present embodiment has been illustrated. Each of the above
stmctural elements may be configured with a general-purpose member or circuit, and
may be configured with hardware specialized for the function of each structural
15 element. A CPU or the like may perform all of the functions of respective structural
elements. Thus, a utilized configuration can be changed as appropriate, according
to the teclmology level at the time of performing the present embodiment.
[0 181]
Note that the computer program for providing each function of the
20 aritlunetic processing apparatus according to the above present embodiment can be
created and implemented in a personal computer or the like. Moreover, a computerreadable
recording medium that contains this computer program can be provided as
well. For example, the recording medium is a magnetic disk, an optical disc, a
magneto-optical disk, a flash memory, or the like. The above computer program
25 may be delivered via a network for example, without using the recording medium.
[0182]
(Modification examples of imaging apparatus)
Next, modification examples of the imaging apparatus I 00 according to the
present embodiment are briefly described with reference to FIG. 27 and FIG. 28.
30 FIG. 27 and FIG. 28 are explanatory diagrams schematically showing a modification
example of the imaging apparatus according to the present embodiment.
PCT/JP2016/062801
56176
[0183]
Although the above description shows the case where the two area cameras
111 and 113 are provided in the imaging apparatus 100, the configuration of the
imaging apparatus 100 according to the present embodiment is not limited to this
5 example.
[0184]
For example, as shown in FIG. 27, three area cameras may be used such that
the light -section line Lb is imaged by an area camera 115 and the light -section line Lc
is imaged by an area camera 117, in combination with the area camera Ill.
10 [0185]
Like in the case where the two area cameras Ill and 113 are used as the
imaging apparatus 100, each of the area cameras 115 and 117 is installed such that,
as shown in FIG. 28, the light-section line Lb and the optical axis of the area camera
115 projected on the L-axis-C-axis plane are orthogonal, and the light-section line Lc
15 and the optical axis of the area camera 117 projected on the L-axis-C-axis plane are
orthogonal. Like in the case where the two area cameras 111 and 113 are used as
the imaging apparatus 100, the imaging area AR3 of the area camera 115 and the
imaging area AR4 of the area camera 117 may be set so as to include the intersection
point A and the intersection point B, respectively, in their imaging visual fields, as
20 appropriate; but it is preferable that the entire light -section lines Lb and Lc be
included in the imaging visual field.
[0186]
Each of the angles u4 and as between the optical axes of the area cameras
and the Z-axis is preferably set to, for example, approximately 30 degrees to 60
25 degrees for a similar reason to the case where the number of area cameras is two.
30
The angles a4 and as may be the same value, or may be mutually different values.
In either case, the shape to be found can be measured by the same calculation
processing as the case where one area camera is used.
[0187]
Although FIG. 27 and FIG. 28 show the case where the two area cameras
115 and 117 are provided on one side in the width direction of the rigid body S, it is
~~ \;
PCT/JP2016/062801
57/76
also possible to place the area camera 115 on the lateral side of the light-section line
Lb side of the rigid body S and place the area camera 117 on the lateral side of the
light-section line Lc side of the rigid body S as long as attention is given to the
direction of translation in the disturbance estimation unit 213.
5 [0188]
Further, four or more area cameras may be used by dividing the
photographing visual field for the light-sectionlines La, Lb, and Lc.
[0189]
Hereinabove, modification examples of the imaging apparatus 100
10 according to the present embodiment are described with reference to FIG. 27 and
FIG. 28.
[0190]
(With regard to flow of shape measurement method)
Next, a flow of a shape measurement method performed in the shape
15 measurement apparatus 10 according to the present embodiment is briefly described
with reference to FIG. 29A and FIG. 29B. FIG. 29A and FIG. 29B are flow chm1s
showing an example of the flow of a shape measurement method according to the
present embodiment.
20
[0191]
Before the following description, it is assumed that the first calibration data
and the second calibration data are appropriately generated and are stored in the
storage unit 207 using various methods like the above.
[0192]
First, under the control of the imaging control unit 201 in the arithmetic
25 processing apparatus 200, the imaging apparatus 100 of the shape measurement
apparatus I 0 according to the present embodiment images a rigid body to be
measured S being conveyed with each of the area cameras Ill and 113, and each
camera generates N captured images (step S 1 01 ). The area cameras 111 and 113 of
the imaging apparatus I 00, each time they generate one captured image, output
30 imaging data of the generated captured image to the arithmetic processing apparatus
200.
Ll
PCT/JP2016/062801
58/76
[0193]
On acquiring imaging data from the imaging apparatus 100, the imaging
data acquisition unit 211 of the arithmetic processing apparatus 200 outputs the
imaging data generated by the area camera Ill to the shape data calculation unit 215,
5 and outputs the imaging data generated by the area camera 113 to the disturbance
estimation unit 213.
[0194]
The disturbance estimation unit 213, the shape data calculation unit 215, and
the correction unit 217 initialize the parameter i used for processing performed in
10 these processing units to i = I (step SI03). Subsequently, the disturbance
estimation unit 213, the shape data calculation unit 215, and the correction unit 217
assess whether the value of the parameter i is not more than N, which is the number
of captured images (step S105). In the case where the value of the parameter i is
not more than N, the disturbance estimation unit 213 struts disturbance estimation
15 processing like the above, and the shape data calculation unit 215 starts shape data
calculation processing like the above. Further, the correction unit 217 starts the
standby of the output of data sent from the disturbance estimation unit 213 and the
shape data calculation unit 215. On the other hand, in the case where the value of
the parameter i is more than N, the shape measurement apparatus 10 finishes the
20 shape measurement processing.
[0195]
The disturbance estimation processing in the disturbance estimation unit 213
and the shape data calculation processing in the shape data calculation unit 215 may
be performed in parallel, or the processing in either one processing unit may be
25 performed prior to the processing in the other processing unit, as a matter of course.
[0196]
By a method like that described above, the shape data calculation unit 215
calculates shape data in the real space (the surface height in the Z-coordinate), using
the shape-measuring light-section line (that is, the light-section line La) and the
30 calibration curve c• with reference to the i-th captured image (step Sl07). After
calculating shape data in the real space in relation to the i-th captured image, the
PCT/JP2016/062801
59176
shape data calculation unit 215 outputs information on the obtained shape data to the
correction unit 217.
[0197]
On the other hand, by a method like that described above, the disturbance
5 estimation unit 213 calculates disturbance components of the common irradiation
potiion on the basis of the correcting light-section lines (that is, the light-section lines
Lb and Lc) with reference to the i-th captured image (step S109). After that, the
disturbance estimation unit 213 calculates an approximate straight line using the
calculated disturbance components, and then calculates the disturbance components
10 at the intersection point A and the intersection point B (step Slll). Subsequently,
the disturbance estimation unit 213 converts the disturbance components at the
intersection point A and the intersection point B to the amounts in the real space
using the calibration curves Cb and cc (step S 113). After that, the disturbance
estimation unit 213 outputs infonnation on the magnitude of the obtained disturbance
15 component in the real space to the correction unit 217.
[0198]
The correction unit 217 calculates the disturbance components at positions
of the shape-measuring light-section line by a method like that described above on
the basis of the disturbance components in the real space at the intersection point A
20 and the intersection point B outputted from the disturbance estimation unit 213 (step
Sl15). After that, the correction unit 217 subtracts the disturbance component in
the real space from the shape data in the real space outputted from the shape data
calculation unit 215, and calculates the true surface height (stepS 117).
25
[0 199]
After that, the disturbance estimation unit 213, the shape data calculation
unit 215, and the correction unit 217 update the value of the parameter i to i = i + I
(step S 119), and perform the processing of step S I 05 again.
[0200]
Hereinabove, a flow of a shape measurement method according to the
30 present embodiment is briefly described with reference to FIG. 29A and FIG. 29B.
[0201]
~I !.-j
PCT/JP2016/062801
60/76
(Hardware Configuration)
Next, the hardware configuration of the arithmetic processing apparatus 200
according to an embodiment of the present invention will be described in detail with
reference to FIG. 30. FIG. 30 is a block diagram for explaining the hardware
5 configuration of the arithmetic processing apparatus 200 according to an
embodiment of the present invention.
[0202]
The arithmetic processing apparatus 200 mainly includes a CPU 901, a
ROM 903, and a RAM 905. Furthermore, the arithmetic processing apparatus 200
10 also includes a bus 907, an input device 909, an output device 911, a storage device
913, a drive 915, a connection port 917, and a communication device 919.
[0203]
The CPU 901 serves as a central processing apparatus and a conh·ol device,
and controls the overall operation or a part of the operation of the arithmetic
15 processing apparatus 200 according to various programs recorded in the ROM 903,
the RAM 905, the storage device 913, or a removable recording medium 921. The
ROM 903 stores programs, operation parameters, and the like used by the CPU 901.
The RAM 905 primarily stores programs that the CPU 901 uses and parameters and
the like varying as appropriate during the execution of the programs. These are
20 connected with each other via the bus 907 configured from an internal bus such as a
CPU bus or the like.
[0204]
The bus 907 ·is connected to the external bus such as a PCI (Peripheral
Component Interconnect/Interface) bus via the bridge.
25 [0205]
The input device 909 is an operation means operated by a user, such as a
mouse, a keyboard, a touch panel, buttons, a switch and a lever. The input device
909 may be a remote control means (a so-called remote control) using, for example,
infi-ared light or other radio waves, or may be an externally cmmected apparatus 923
30 such as a PDA conforming to the operation of the arithmetic processing apparatus
200. Furthermore, the input device 909 generates an input signal based on, for
Ll
PCT/JP2016/062801
61176
example, information which is input by a user with the above operation means, and is
configured from an input control circuit for outputting the input signal to the CPU
90 I. The user can input various data to the shape measurement apparatus 10 and
can instruct the shape inspection apparatus 10 to perform processing by operating
5 this input device 909.
[0206]
The output device 911 is configured from a device capable of visually or
audibly notifYing acquired information to a user. Examples of such device include
display devices such as a CRT display device, a liquid crystal display device, a
10 plasma display device, an EL display device and lamps, audio output devices such as
a speaker and a headphone, a printer, a mobile phone, a facsimile machine, and the
like. For example, the output device 911 outputs a result obtained by various
processes performed by the arithmetic processing apparatus 200. More specifically,
the display device displays, in the form of texts or images, a result obtained by
15 various processes performed by the arithmetic processing apparatus 200. On the
other hand, the audio output device converts an audio signal such as reproduced
audio data and sound data into an analog signal, and outputs the analog signal.
[0207]
The storage device 913 is a device for storing data configured as an example
20 of a storage unit of the aritlnnetic processing apparatus 200 and is used to store data.
The storage device 913 is configured from, for example, a magnetic storage device
such as a HDD (Hard Disk Drive), a semiconductor storage device, an optical storage
device, or a magneto-optical storage device. This storage device 913 stores
programs to be executed by the CPU 901, various data, and various data obtained
25 from the outside.
[0208]
The drive 915 is a reader/writer for recording medium, and is embedded in
the arithmetic processing apparatus 200 or attached externally thereto. The drive
915 reads information recorded in the attached removable recording medium 921
30 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor
memory, and outputs the read information to the RAM 905. Fmihermore, the drive
PCT/JP2016/062801
62/76
915 can write in the attached removable recording medium 921 such as a magnetic
disk, an optical disk, a magneto-optical disk, or a semiconductor memory. The
removable recording medium 921 is, for example, a CD medium, a DVD medium, or
a Blu-ray (registered trademark) medium. The removable recording medium 921
5 may be a CompactFlash (CF; registered trademark), a flash memmy, an SD memory
card (Secure Digital Memory Card), or the like. Alternatively, the removable
recording medium 921 may be, for example, an IC card (Integrated Circuit Card)
equipped with a non-contact IC chip or an electronic device.
10
[0209]
The connection pmi 917 is a pmi for allowing devices to directly cmmect to
the arithmetic processing apparatus 200. Examples of the connection port 917
include a USB (Universal Serial Bus) port, an IEEE1394 pmi, a SCSI (Small
Computer System Interface) pmi, an RS-232C pmi, and the like. By the externally
connected apparatus 923 com1ecting to this connection port 917, the arithmetic
15 processing apparatus 200 directly obtains various data from the externally com1ected
apparatus 923 and provides various data to the externally connected apparatus 923.
[0210]
The communication device 919 is a communication interface configured
fi·om, for example, a communication device for cormecting to a conununication
20 network 925. The communication device 919 is, for example, a wired or wireless
LAN (Local Area Network), Bluetooth (registered trademark), a conununication card
for WUSB (Wireless USB), or the like. Alternatively, the communication device
919 may be a router for optical communication, a router for ADSL (Asymmetric
Digital Subscriber Line), a modem for various communications, or the like. This
25 communication device 919 can transmit and receive signals and the like in
accordance with a predetermined protocol such as TCPIIP on the Internet and with
other conmmnication devices, for example. The communication network 925
connected to the communication device 919 is configured from a network and the
like, which is connected via wire or wirelessly, and may be, for example, the Internet,
30 a home LAN, an in-house LAN, infrared communication, radio wave communication,
satellite conmmnication, or the like.
fl il
PCT/JP2016/062801
63/76
[0211]
Heretofore, an example of the hardware configuration capable of realizing
the functions of the arithmetic processing apparatus 200 according to an embodiment
of the present invention has been shown. Each of the structural elements described
5 above may be configured using a general-purpose material, or may be configured
from hardware dedicated to the function of each structural element. Accordingly,
the hardware configuration to be used can be changed as appropriate according to the
technical level at the time of carrying out the present embodiment.
[Examples]
10 [0212]
In the following, the shape measurement apparatus and the shape
measurement method according to the present invention are specifically described
with reference to Examples. Examples shown below are only examples of the
shape measurement apparatus and the shape measurement method according to the
15 present invention, and the shape measurement apparatus and the shape measurement
method according to the present invention are not limited to Examples shown below.
[0213]
In Example 1 to Example 3 shown below, an aluminum sheet in which it is
known that a surface thereof is flat was used as the rigid body to be measured S.
20 The shape measurement apparatus used for shape measurement is the shape
measurement apparatus 1 0 according to the present embodiment like that shown in
FIG. 1 and FIG. 2.
[0214]
In Example 1 to Example 3, while an aluminum sheet like that mentioned
25 above was conveyed 60 mm at a constant speed of 5 mmlsecond, one image was
captured per 0.2 seconds with two area cameras, and 60 captured images were
obtained by each area camera. Before the above, the calibration curves c•, cb, and
cc, and /'I.Lb and /'I.L" were created, and the obtained data were stored in the storage
unit.
30 [0215]
In Examples shown below, during the conveyance of the aluminum sheet,
PCT/JP2016/062801
64/76
three kinds of disturbance (movement in the Z-axis direction, rotation around the Laxis,
and rotation around the C-axis) were added respectively, and the value Z(i, x•)
in which a change in the Z-coordinate derived from disturbance was included and the
true surface height Zout(i, Xa) (i = I, 2, ... , 60) outputted fi"om the arithmetic
5 processing apparatus 200 were compared. The following shows results obtained by
converting the x•-coordinate (unit: pixels) to the C-coordinate (unit: mm) that is in
the width direction of the rigid body S.
10
[0216]
(Example 1)
In Example I, translation in the Z-direction like that shown in FIG. 31A was
added as disturbance during the conveyance of the aluminum sheet. The positions
of the light-section lines are as shown in FIG. 31B. As a result, as shown in FIG.
31 C, it can be seen that changes in the Z-axis direction due to the disturbance have
been superimposed in Z(i, x•), and the surface height of the corresponding portion is
15 not flat. This result indicates that Z(i, X') has failed to express an accurate surface
20
height. On the other hand, as shown in FIG. 31D, it has been found that Zout(i, x•)
(i = 1, 2, ···, 60) is flat and an accurate surface height has been measured.
[0217]
(Example 2)
In Example 2, rotation around the L-axis like that shown in FIG. 32A (the
rotation axis was set to the center position in the width direction of the aluminum
sheet, and the positive direction of the rotation angle was set to clockwise along the
positive direction of the L-axis) was added as disturbance during the conveyance of
the aluminum sheet. The positional relationship between the positions of the light-
25 section lines and the rotation axis is as shown in FIG. 32B. As a result, as shown in
FIG. 32C, it can be seen that changes in the Z-axis direction due to the rotation
around the L-axis have been superimposed in Z(i, x•), and the surface height of the
corresponding pmtion is not flat. This result indicates that Z(i, x•) has failed to
express an accurate surface height. On the other hand, as shown in FIG. 32D, it has
30 been found that Z0u1(i, Xa) (i = 1, 2, ···, 60) is flat and an accurate surface height has
been measured.
[0218]
(Example 3)
65176
PCT/JP2016/062801
In Example 3, rotation around the C-axis like that shown in FIG. 33A (the
rotation axis was set to the center position in the longitudinal direction of the
5 aluminum sheet, and the positive direction of the rotation angle was set to clockwise
along the positive direction of the C-axis) was added as disturbance during the
conveyance of the aluminum sheet. The positional relationship between the
positions of the light-section lines and the rotation axis is as shown in FIG. 338. As
a result, as shown in FIG. 33C, it can be seen that changes in the Z-axis direction due
10 to the rotation around the C-axis have been superimposed in Z(i, X"), and the surface
height of the corresponding portion is not flat. This result indicates that Z(i, X") has
failed to express an accurate surface height. On the other hand, as shown in FIG.
33D, it has been found that Zout(i, x•) (i = 1, 2, ···, 60) is flat, and an accurate surface
height has been measured.
15 [0219]
The preferred embodiment(s) of the present invention has/have been
described above with reference to the accompanying drawings, whilst the present
invention is not limited to the above examples. A person skilled in the art may find
various alterations and modifications within the scope of the appended claims, and it
20 should be understood that they will naturally come under the technical scope of the
present invention.
25
Reference Signs List
[0220]
10 shape measurement apparatus
100 imaging apparatus
lOla, !Olb, and !Ole
111,113,115,117
linear laser light source
area camera
200 aritlunetic processing apparatus
30 201 imaging control unit
203 image processing unit
PCT/JP2016/062801
66176
205 display control unit
207 storage unit
211 imaging data acquisition unit
213 disturbance estimation unit
5 215 shape data calculation unit
217 correction unit
219 result output unit
221 common irradiation portion disturbance estimation unit
223 intersection point position disturbance estimation unit
10
PCT/JP2016/062801

CLAIMS
Claim 1
A shape measurement apparatus that measures a shape of a rigid body to be
measured by means of a plurality of light -section lines based on a plurality of linear
5 laser light beams applied to a surface of the rigid body to be measured from a
plurality oflinear laser light sources moving relative to the rigid body to be measured
along a longitudinal direction of the rigid body to be measured,
the shape measurement apparatus comprising:
an imaging apparatus that applies three beams of the linear laser light to the
10 surface of the rigid body to be measured moving relatively along the longitudinal
direction and images reflected light of the three beams of the linear laser light from
the surface of the rigid body to be measured at a prescribed longitudinal-direction
inte1val; and
an arithmetic processmg apparatus that performs image processmg on
15 captured images related to the light-section lines imaged by the imaging apparatus
and calculates a surface shape of the rigid body to be measured,
wherein the imaging apparatus includes
a first linear laser light source that emits a shape-measuring light-section
line that is the light-section line extending in a width direction of the rigid body to be
20 measured and is used to calculate the surface shape of the rigid body to be measured,
25
a second linear laser light source that emits a first correcting light-section
line that is parallel to the longitudinal direction of the rigid body to be measured and
crosses the shape-measuring light-section line, and is used to correct an effect of
disturbance acting on the rigid body to be measured,
a third linear laser light source that emits a second correcting light-section
line that is parallel to the longitudinal direction of the rigid body to be measured,
crosses the shape-measuring light-section line, and exists in a width-direction
position of the rigid body to be measured different fi·mn the first correcting lightsection
line, and is used to correct an effect of disturbance acting on the rigid body to
30 be measured,
a first camera that images the shape-measuring light-section line at each
L.J
PCT/JP2016/062801
68/76
time corresponding to a prescribed longitudinal-direction interval and generates a
captured image of the shape-measuring light-section line at each time, and
a second camera that images the correcting light-section lines at each time
corresponding to a prescribed longitudinal-direction interval and generates a captured
5 image of the correcting light-section lines at each time, and
the arithmetic processing apparatus includes
a shape data calculation unit that, on the basis of the captured image of the
shape-measuring light-section line at each time generated by the first camera,
calculates shape data that show a three-dimensional shape of the surface of the rigid
10 body to be measured and in which a measurement error derived from the disturbance
is superimposed,
a disturbance estimation unit that
performs, on a plurality of points of different longitudinal-direction
positions of the first correcting light-section line, height change value acquisition
15 processing of acquiring, fi·om height measurement values related to a surface height
of the rigid body to be measured acquired at different two times for the same position
of the rigid body to be measured, a height change value derived fi·om the disturbance
at the position, using captured images of the first correcting light-section line,
performs the height change value acquisition processing on a
20 plurality of points of different longitudinal-direction positions of the second
correcting light-section line using captured images of the second correcting lightsection
line, and
estimates the amount of height fluctuation derived from the
disturbance superimposed in the shape data, using a plurality of height change values
25 derived from the disturbance obtained from captured images of the first correcting
light-section line and a plurality of height change values derived from the
disturbance obtained from captured images of the second correcting light -section line,
and
a correction unit that subtracts the amount of height fluctuation from the
30 shape data and thereby corrects the measurement error derived from the disturbance.
M !'
Claim2
69/76
The shape measurement apparatus according to claim 1,
wherein the disturbance estimation unit
PCT/JP2016/062801
approximates, with a straight line, height change values derived from the
5 disturbance at a plurality of points on the first correcting light-section line and
estimates a height change value derived from the disturbance at an intersection point
of the straight line and the shape-measuring light-sectionline,
approximates, with a straight line, height change values derived from the
disturbance at a plurality of points on the second correcting light-section line and
10 estimates a height change value derived from the disturbance at an intersection point
of the straight line and the shape-measuring light-section line, and
15
estimates the amount of. height fluctuation by means of a straight line
connecting the height change values derived from the disturbance at the two
intersection points.
Claim 3
The shape measurement apparatus according to claim 1 or 2,
wherein each of the first camera and the second camera performs imaging at
each time corresponding to a prescribed longitudinal-direction interval and generates
20 N (N being an integer of 2 or more) captured images, and
25
the disturbance estimation unit calculates the amount of height fluctuation
on the assumption that the disturbance has not occurred in a 1st captured image.
Claim 4
The shape measurement apparatus according to any one of claims 1 to 3,
wherein an imaging timing of the first camera and the second camera is
controlled so that a common irradiation region that is a portion of the rigid body to
be measured irradiated with the correcting light-section line in common exists in
captured images of the second camera captured at mutually adjacent imaging times,
30 and
the disturbance estimation unit calculates a height change value derived
~I
PCT/JP2016/062801
70/76
from the disturbance for the plurality of points falling under the common irradiation
region of each of the first correcting light-section line and the second correcting
light-section line.
5 Claim 5
The shape measurement apparatus according to claim 4,
wherein, using an apparent surface height including the height change value
obtained from an i+l-th captured image (i = 1, 2, ···, N-1) of the second camera and a
surface height that is obtained fi·om an i-th captured image of the second camera and
10 that is after the height change value in the common irradiation region of the i-th
captured image is removed, the disturbance estimation unit calculates the height
change value in the i+ 1-th captured image and a surface height after the height
change value is removed.
15 Claim6
20
1l1e shape measurement apparatus according to claim 4 or 5,
wherein the disturbance estimation unit calculates the height change value
in an i-th captured image (i = 2;··, N) of the second camera with a 1st captured image
of the second camera as a reference.
Claim 7
The shape measurement apparatus according to any one of claims 1 to 6,
wherein the first linear laser light source, the second linear laser light source,
and the third linear laser light source are provided such that an optical axis of each
25 light source is perpendicular to a plane defined by a longitudinal direction and a
width direction of the rigid body to be measured.
30
Claim 8
The shape measurement apparatus according to any one of claims 1 to 7,
wherein an angle between an optical axis of the first camera and an optical
axis of the first linear laser light source, an angle between a line of sight of the
5
PCT/JP2016/062801
71/76
second camera and an optical axis of the second linear laser light source, and an
angle between the line of sight of the second camera and an optical axis of the third
linear laser light source are mutually independently not less than 30 degrees and not
more than 60 degrees.
Claim 9
A shape measurement method that measures a shape of a rigid body to be
measured by means of a plurality of light-section lines based on a plurality of linear
laser light beams applied to a surface of the rigid body to be measured fi·om a
10 plurality of linear laser light sources moving relative to the rigid body to be measured
along a longitudinal direction of the rigid body to be measured,
the shape measurement method comprising:
an imaging step of imaging reflected light of tluee light-section lines from
the surface of the rigid body to be measured at a prescribed longitudinal-direction
15 interval by applying the tlu·ee light-section lines to the surface of the rigid body to be
measured moving relatively along the longitudinal direction from an imaging
apparatus including
a first linear laser light source that emits a shape-measuring lightsection
line that is the light-section line extending in a width direction of the rigid
20 body to be measured and is used to calculate a surface shape of the rigid body to be
measured,
a second linear laser light source that emits a first correcting lightsection
line that is parallel to the longitudinal direction of the rigid body to be
measured and crosses the shape-measuring light -section line, and is used to correct
25 an effect of disturbance acting on the rigid body to be measured,
a third linear laser light source that emits a second correcting lightsection
line that is parallel to the longitudinal direction of the rigid body to be
measured, crosses the shape-measuring light-section line, and exists in a widthdirection
position of the rigid body to be measured different from the first correcting
30 light-section line, and is used to correct an effect of disturbance acting on the rigid
body to be measured,
PCT/JP2016/062801
72/76
a first camera that images the shape-measuring light-section line at
each time corresponding to a prescribed longitudinal-direction interval and generates
a captured image of the shape-measuring light -section line at each time, and
a second camera that images the correcting light-section lines at
5 each time corresponding to a prescribed longitudinal-direction interval and generates
a captured image of the correcting light-section lines at each time;
a shape data calculation step of, on the basis of the captured image of the
shape-measuring light-section line at each time generated by the first camera,
calculating shape data that show a three-dimensional shape of the surface of the rigid
10 body to be measured and in which a measurement error derived limn the disturbance
is superimposed;
a disturbance estimation step of
performing, on a plurality of points of different longitudinaldirection
positions of the first correcting light-section line, height change value
15 acquisition processing of acquiring, from height measurement values related to a
surface height of the rigid body to be measured acquired at different two times for
the same position of the rigid body to be measured, a height change value derived
from the disturbance at the position, using captured images of the first correcting
light-section line,
20 performing the height change value acquisition processing on a
plurality of points of different longitudinal-direction positions of the second
correcting light-section line using captured images of the second correcting lightsection
line, and
estimating the amount of height fluctuation derived from the
25 disturbance superimposed in the shape data, using a plurality of height change values
derived from the disturbance obtained from captured images of the first correcting
light-section line and a plurality of height change values derived from the
disturbance obtained from captured images of the second correcting light-section
line; and
30 a correction step of subtracting the amount of height fluctuation from the
shape data and thereby correcting the measurement error derived from the
5
10
PCT/JP2016/062801
73/76
disturbance.
Claim 10
The shape measurement method according to claim 9,
wherein, in the disturbance estimation step,
height change values derived from the disturbance at a plurality of points on
the first correcting light-section line are approximated with a straight line and
thereby a height change value derived from the disturbance at an intersection point of
the straight line and the shape-measuring light-section line is estimated,
height change values derived Jiom the disturbance at a plurality of points on
the second correcting light-section line arc approximated with a straight line and
thereby a height change value derived from the disturbance at an intersection point of
the straight line and the shape-measuring light-section line is estimated, and
the amount of height fluctuation is estimated by means of a straight line
15 connecting the height change values derived from the disturbance at the two
20
intersection points.
Claim II
The shape measurement method according to claim 9 or 10,
wherein each of the first camera and the second camera performs imaging at
each time corresponding to a prescribed longitudinal-direction interval and generates
N (N being an integer of 2 or more) cap hired images, and
in the disturbance estimation step, the amount of height fluctuation is
calculated on the assumption that the disturbance has not occurred in a I st captured
25 Image.
Claim 12
The shape measurement method according to any one of claims 9 to II,
wherein an imaging timing of the first camera and the second camera is
30 controlled so that a common irradiation region that is a portion of the rigid body to
be measured irradiated with the correcting light-section line in common exists in
G.:-l
,---,.,
PCT/JP20 16/062801
74/76
captured images of the second camera imaged at mutually adjacent imaging times,
and
in the disturbance estimation step, a height change value derived from the
disturbance is calculated for the plurality of points falling under the cmmnon
5 irradiation region of each of the first correcting light-section line and the second
correcting light -section line.
10
Claim 13
The shape measurement method according to claim I 2,
wherein, in the disturbance estimation step, using an apparent surface height
including the height change value obtained from an i+ I -th captured image (i = I, 2,
···, N-1) of the second camera and a surface height that is obtained from an i-th
captured image of the second camera and that is after the height change value in the
common irradiation region of the i-th captured image is removed, the height change
15 value in the i+ I -th captured image and a smface height after the height change value
20
25
30
is removed are calculated.
Claim 14
The shape measurement method according to claim I2 or 13,
wherein, in the disturbance estimation step, the height change value in an ith
captured image (i = 2, ·· ·, N) of the second camera is calculated with a I st captured
image of the second camera as a reference.
Claim 15
The shape measurement method according to any one of claims 9 to I 4,
wherein the first linear laser light source, the second linear laser light source,
and the third linear laser light source are provided such that an optical axis of each
light source is perpendicular to a plane defined by a longitudinal direction and a
width direction of the rigid body to be measured.
Claim 16
f!
PCT/JP2016/062801
75176
The shape measurement method according to any one of claims 9 to 15,
wherein an angle between an optical axis of the first camera and an optical
axis of the first linear laser light source, an angle between a line of sight of the
second camera and an optical axis of the second linear laser light source, and an
5 angle between the line of sight of the second camera and an optical axis of the third
linear laser light source are mutually independently not less than 30 degrees and not
more than 60 degrees.