JP2005031039A - Video type three-dimensional position measuring instrument, video type extensometer, and video type width meter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a three-dimensional measuring instrument capable of measuring a three-dimensional position of an object, using a photographed signal of the object with a video camera, and a video type extensometer and a video type width meter capable of eliminating errors along three axial directions of a test piece and caused by rotation around axes thereof, applying it. <P>SOLUTION: Mark bodies M imparted with at least three marks m1-m4 are bonded to the object, the mark bodies M are photographed with the video camera 21, the photographed signals thereof are fetched to acquire positional information of the marks m1-m4, and momentary three-dimensional position information of the centers among the respective marks m1-m4, and rotation angles around the respective axes are calculated by a successive approximation method using the positional information. When applied for the extensometer or the width meter, the mark bodies M are bonded to the test piece W, and an error caused by motion from an initial test start of a test piece W is corrected based on the three-dimensional position information. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an apparatus for measuring a three-dimensional position of an object by photographing the object using a video camera, and an apparatus to which the apparatus is applied. The present invention relates to a video extensometer that measures elongation or width change and a video width meter.

  Conventionally, a video extensometer is known as an extensometer used in a material testing machine to measure the elongation of a test piece. In the video extensometer, two marked lines are attached to the test piece, the test piece is photographed during the test, and the position information of each marked line is obtained by image processing using the video signal. Calculate the elongation between lines.

  In such a video extensometer, the distance between the marked lines varies depending on the imaging magnification of the test piece. The shooting magnification is fixed when the camera type, the lens to be used, and the distance between the camera and the specimen are determined, but the distance between the camera and the specimen varies depending on various factors. This change in distance easily causes measurement errors in this type of extensometer.

The change in the distance between the camera and the specimen is derived from the displacement of the specimen in the optical axis direction of the camera during the test (hereinafter referred to as the displacement in the front-rear direction) and the rotation of the specimen around each axis. However, as a technique to remove the influence of the displacement of the test piece in the front-rear direction, conventionally, a displacement measuring means for measuring the change in the distance between the camera and the test piece is arranged at the position of the camera, and the displacement is measured. The amount of displacement of the test piece in the front-rear direction is measured every time, and the measurement results are used to eliminate errors included in the measurement results of elongation due to changes in the imaging magnification caused by the movement of the test piece in the front-rear direction. The technique to do is known (for example, refer patent document 1).
Japanese Patent No. 3350272 (page 2-4, FIG. 1)

  By the way, in many video extensometers in practical use, it is assumed that the displacement other than the elongation direction, such as the displacement or rotation of the test piece in the front-rear direction during the test, is small, and errors due to these displacements are assumed. Ignored and measured elongation. In this case, the accuracy of the elongation measurement is deteriorated when the test piece is moved back and forth or rotated.

  Further, in the technique disclosed in Patent Document 1 described above, it is necessary to use a device that is not originally required for elongation measurement, which is a displacement measuring means. In the technology, only errors due to the forward and backward movement of each marked line will be corrected, and therefore errors due to rotation around other axes, excluding rotation in the direction in which the specimen tilts toward the camera, are eliminated. I can't do it.

  The present invention is a video three-dimensional measuring apparatus capable of measuring a three-dimensional position of an object using a video signal of the object by a video camera, and a technique of applying the technique to the video three-dimensional measuring apparatus. In addition, a video extensometer that can accurately measure the elongation by eliminating the error caused by the movement of the specimen in the three-axis directions and the rotation around each axis during the test, and the same as above In addition, the object is to provide a video width meter that can accurately measure the change in the width of each test piece by eliminating errors caused by movement and rotation in each axial direction.

  In order to achieve the above object, the video three-dimensional measuring apparatus of the present invention is a mark body that is attached to an object, the mutual positional relationship is known, and the positional relationship is the movement of the target object. Sometimes a mark body with at least three invariant marks attached, a video camera that captures the mark body attached to the object, and a video signal from the video camera, and the optical axis ( image processing means for obtaining the position information of each mark in the two axis (x, y axis) directions orthogonal to each other on a plane orthogonal to the (z axis) direction, and the position information of each mark by the image processing means. By using numerical calculation, either or all of the coordinates in the x, y, z axis direction of the center between the marks and the rotation angle of the mark body around the x, y, z axis are calculated. Equipped with 3D position calculation means Characterized by that (claim 1).

  Further, the video extensometer of the present invention is an apparatus for photographing a test piece to which the mark body according to claim 1 is attached and which has two marked lines with a predetermined distance in the load direction. Alternatively, a plurality of video cameras and image processing means for capturing image signals from the video cameras, performing image processing according to claim 1, and obtaining positional information of the respective marked lines in the x and y axis directions 2. The three-dimensional position calculation means according to claim 1 using the position information of each mark by the image processing means, and the elongation of each marked line by using the position information of each mark by the image processing means. The calculation result of the elongation by the elongation calculation means, and the calculation result of the elongation by the elongation calculation means, either the displacement of the mark body in the x, y or z axis direction by the three-dimensional position calculation means and the rotation angle around each of these axes. Or use all Characterized by that it comprises a correction arithmetic means for (claim 2).

  Here, in the video extensometer according to the present invention, the mark body is attached to each of the positions corresponding to two or more marked lines of the test piece, and the image processing means outputs a photographing signal from the video camera. The position information of each mark on the two or more mark bodies is obtained, and the three-dimensional position calculation means performs displacement in the x-, y-, and z-axis directions for the two or more mark bodies, and A rotation angle around each axis is calculated, and instead of the elongation calculation means and the correction calculation means, two or more of each of the three-dimensional position calculation results of the two or more mark bodies by the three-dimensional position calculation means are calculated. It can also be set as the structure (Claim 3) provided with the elongation calculating means which calculates the distance between mark bodies and calculates | requires elongation.

  The video width meter according to the present invention includes a video camera for photographing the test piece to which the mark body according to claim 1 is attached, and a video signal from the video camera. 2. The three-dimensional image according to claim 1, wherein the three-dimensional image processing means performs image processing and obtains position information of each edge along the load direction of the test piece, and position information of each mark position by the image processing means. The position calculation means, the width dimension calculation means for calculating the distance between the edges by using the position information of each edge by the image processing means, and the calculation result of the width by the width dimension calculation means as the above 3 It is characterized by comprising correction calculation means for correcting by using any or all of the displacement of the mark body in the x, y and z axis directions by the dimension position calculation means and the rotation angles around these axes. It is (claim 4).

  According to the video-type three-dimensional measuring apparatus of the present invention, a predetermined mark body is attached to an object and photographed with a video camera, and only the photographing signal is used to move the object in three axial directions. The rotation around each axis can be measured, and three-dimensional measurement of various objects is possible with a simple configuration.

  Further, according to the video extensometer and video width meter of the present invention, the extensometer or the video camera inherently required for the width meter is used as a single measuring instrument, so that the test piece can be moved in the three-axis directions. Errors due to movement and rotation around these axes can be eliminated, and accurate elongation or width measurement can be performed at low cost.

  Based on a simple configuration using only a video camera as a measuring instrument, the effects of the displacement of the specimen in the three axes and the rotation around each axis during the test are eliminated, and the specimen is always stretched accurately. A video-type stretcher / width meter that can measure the width and width.

  FIG. 1 is a configuration diagram of the main part of an embodiment of the present invention, and is a diagram showing a side view in the vicinity of a test piece W mounted on a material testing machine and a block diagram showing a configuration of a video type elongation / width meter. is there. Moreover, FIG. 2 is an AA arrow enlarged view of FIG.

  The test piece W is mounted on the material testing machine by gripping the upper and lower ends thereof by the gripping tools 1a and 1b of the material testing machine, and one gripping tool 1a is displaced in a direction away from the other gripping tool 1b. Is given, a tensile load is applied to the test piece W. On the surface of the test piece W, two marked lines S1 and S2 are attached in advance at a predetermined distance in the tensile direction, and a mark body M as shown in FIG. Is done.

  In this example, the mark body M is a substantially square paper or film, and is attached to the test piece W in a very small area at the center. On the mark body M, rhombus marks m1 to m4 are printed at the four corners. Each of the marks m1 to m4 has the same shape and size, and is printed at positions that form four vertices of a square. As described above, the mark body M attached to the test piece W substantially at one point is relative to the marks m1 to m4 when the test piece W is displaced in the three-dimensional direction or rotated around each axis. There is no displacement, and therefore, each of the marks m1 to m4 moves substantially as a rigid body by the movement of the test piece W.

  The video type stretcher / width meter 2 is mainly configured by a computer 22 in which a CCD camera 21 and a capture board for capturing a photographing signal from the CCD camera 21 are installed. The CCD camera 21 is set so that the two marked marks S1 and S2 and the mark body M of the test piece W are within the visual field from the start to the end of the test.

  The computer 22 is installed with software for performing image processing using a photographic signal from the CCD camera 21 and software for performing various calculations using position information obtained by the image processing. For convenience of explanation, the functions of the installed main software are shown in block diagrams.

  A photographing signal from the CCD camera 21 during the test is taken into the image processing unit 22a, and the image processing unit 22a follows the positional information of the two marked lines S1 and S2 and the load direction of the test piece W. The position information of the left and right edges and the position information of the marks m1 to m4 on the mark body M are obtained.

  The positional information of the two marked lines S1 and S2 is sent to the elongation calculating unit 22b. In this elongation calculating unit 22b, the load direction (y direction) between the marked lines S1 and S2 is the same as the conventional video extensometer. Is calculated every moment, and the elongation between the marked lines S1 and S2 is calculated.

  The position information of the left and right edges of the test piece W is sent to the width calculation unit 22c, and the width calculation unit 22c calculates the distance between the left and right edges (distance in the x direction) every moment, and the test piece during the test. The change of the width dimension is calculated every moment.

  In the three-dimensional position calculation unit 22d, the position information of the marks m1 to m4 of the mark body M is taken in, and the three-dimensional position of the test piece W is calculated. The calculation method will be described below.

  As shown in FIG. 3, it is assumed that a certain mark m is observed by the CCD camera 21. At this time, the distance in the optical axis direction between the focal point A and the mark m of the CCD camera 21 is z, the distance in the horizontal direction in the plane B perpendicular to the optical axis is x, and the distance in the vertical direction in the plane B is also the same. Let y be y.

FIG. 4 is a diagram showing FIG. 3 in consideration of the image of the CCD camera 21. The mark m is captured as m ′ on the imaging surface C of the CCD camera 21. At this time, the horizontal distance from the center of the mark image m ′ on the imaging surface C of the CCD camera 21 is x _p , and the distance in the vertical direction is y _p .
At this time, x, y, z and x _p, the relationship of y _p is

となる。

It becomes.

  Here, S is a parameter that depends on the lens to be used and the number of pixels of the CCD camera 21, and is uniquely determined if the lens and the camera are determined.

Now, as shown in FIG. 5A, it is assumed that a mark body M as a rigid body exists in a three-dimensional space, and a plurality of marks m1 to m4 are attached on the mark body M. A coordinate system [a ₀ ] having an origin at the same point as point A in FIG. 4 is taken, and the X ₀ -Y ₀ -Z ₀ direction is taken to coincide with the xyz direction in FIG. Here, (the center between m1 to m4) center of the mark M with the A point vector to m ₀ and r ₀ (vector), each vector from the center m ₀ of the mark M with the respective marks m1 to m4 L ₁ (vector), L ₂ (vector), L ₃ (vector), and L ₄ (vector), and vectors from the origin A to the marks m1 to m 4 are r ₁ (vector) and r ₂ (vector), respectively. ), R ₃ (vector), r ₄ (vector) (in FIG. 5B, only r ₂ is shown as a representative).

The vector from the origin A of the [a ₀ ] coordinate to each mark mi (where i = 1 to 4) is

と表される。ここで、［ａ₀ ］座標で表現される成分として、

It is expressed. Here, as a component expressed by [a ₀ ] coordinates,

と表す。

It expresses.

On the other hand, as shown in FIG. 5B, a coordinate system having the center m0 of the mark body M as the origin is taken as the [a ₁ ] coordinate system. The X ₁ -Y ₁ -Z ₁ direction of [a ₁ ] is as illustrated. At this time, the vectors from m0 to the marks m1 to m4 expressed in the [a ₁ ] coordinate system are C ₁ (vector), C ₂ (vector), C ₃ (vector), and C ₄ (vector). Here, if the four marks m1 to m4 are all arranged at the same distance on the X ₁ and Y ₁ coordinates from the center m0, the component expressed on the [a ₁ ] coordinates is

となり、

And

となっている。

It has become.

Here, C _i and L _i originally represent the same vector, but C _i is represented by [a ₁ ] coordinates, and L _i is represented by [a ₀ ] coordinates.

A transformation matrix indicating rotation from the [a ₁ ] coordinate system to the [a ₀ ] coordinate system is represented by A _x A _y A _z .

つまり、［ａ₁ ］座標上のＣ_i （ベクトル）に座標変換を表す回転行列Ａ_x Ａ_y Ａ_z を乗じることにより［ａ₀ ］座標上のＬ_i （ベクトル）として表現される。

That is, it is expressed as L _i (vector) on the [a ₀ ] coordinate by multiplying C _i (vector) on the [a ₁ ] coordinate by the rotation matrix A _x A _y A _z representing coordinate transformation.

Here, when a 3-2-1 Euler angle is used as a rotation matrix indicating coordinate transformation, for example, if A _x A _y A _z is shown,

と表され、θ_x はＸ₀ 軸を中心とした回転角、θ_y はθ_x 回転後のＹ₀ 軸を中心とした回転角、θ_z はθ_x ，θ_y 回転後のＺ₀ 軸＝Ｚ₁ 軸を中心とした回転角を示している。

Is expressed as, theta _x is the rotational angle around the X ₀ axis, theta rotation angle _y is around the Y ₀ axis after rotation θ _x, θ _z is θ _x, Z ₀ axis after theta _y rotation = The rotation angle around the Z ₁ axis is shown.

The above is a basic equation for performing the three-dimensional measurement of the mark body M. Hereinafter, x ₀ , y ₀ , z ₀ , the center m ₀ of the mark body M in the [a ₀ ] coordinate system is determined by a successive approximation method. In addition, the rotation angles with respect to the rotation axes θ _x , θ _y , and θ _z around these axes are calculated by the successive approximation method (Picart's successive approximation method). In the calculation, the calculation result gradually approaches a correct value by repeatedly performing the following first to fourth steps.
As the first step, from the above equations (7) and (8),

となり、同様に、

And similarly,

となる。
よって、これらの（９），（１０）および前記した（３）より、

It becomes.
Therefore, from these (9), (10) and the above (3),

となる。
ここで、［ａ₀ ］座標系における原点Ａから各マークｍ_i へのベクトルｒ_i の成分を、

It becomes.
Here, the component of the vector r _i from the origin A to each mark m _i in the [a ₀ ] coordinate system is

とおく。ここで、ｘ_i ′，ｙ_i ′ｚ_i ′は（１１）式の右辺に現れるｒ_i （ベクトル）の成分表示となっている。このとき、図４のマークｍをｍ_i と考えると、（１），（２）式のｘ，ｙ，ｚをｘ_i ′，ｙ_i ′，ｚ_i ′の対応をとると、ｘ→ｘ_i ′，ｙ→ｙ_i ′、およびｚ→ｚ_i ′＝ｚ₀ ＋ｚ_i ′となっている。ｘ_piおよびｙ_piを各マークｍｉのＣＣＤカメラ２１の撮像面上の位置とすると、（１），（２）から

far. Here, x _i 'and y _i ' z _i 'are component displays of r _i (vector) appearing on the right side of the equation (11). At this time, when the mark m in FIG. 4 is considered to be m _i , if x, y, and z in the expressions (1) and (2) correspond to x _i ′, y _i ′, and z _i ′, then x → x _i ′, y → y _i ′, and z → z _i ′ = z ₀ + z _i ′. _Assuming that x _pi and y _pi are positions on the imaging surface of the CCD camera 21 of each mark mi, from (1) and (2)

となる。（１２）より、

It becomes. From (12)

となる。
（１１），（１３）より、

It becomes.
From (11) and (13),

となる。この式は、ＣＣＤカメラ２１の撮像面上の４つのマークｍｉの位置をｚ方向の違いを補正して平均したものが各マークｍｉの中央ｍ０の［ａ₀ ］座標系の座標（ｘ₀ ，ｙ₀ ，ｚ₀ ）であることを示した式となっている。

It becomes. This equation, [a _0] coordinate system of the coordinate (x ₀ of the central m0 positions of four marks mi on the imaging surface that average by correcting the difference in z-direction of each mark mi of the CCD camera 21, y ₀ , z ₀ ).

Using this equation (14), x ₀ and y ₀ are calculated. At this time, x _pi and y _pi use the values obtained by obtaining the positions of the marks m1 to m4 by image processing using the photographing signal by the CCD camera 21. Further, S is determined from the lens information of the CCD camera 21 being used (fixed value). In the first calculation of the iterative calculation by the successive approximation method, z ₀ is calculated as an almost correct value as will be described later from the result of calibration for obtaining the photographing magnification of the CCD camera 21. Use the value. From the second time onward, the previous calculation result is used. Further, z _i is calculated as 0 for the first time, and the previous calculation result is used for the second time and thereafter. Each time z ₀ and z _i are repeatedly calculated with the first to fourth steps as one set, they approach the correct values.

Referring to the method for obtaining z ₀ used in the first calculation described above, as shown in FIG. 6, the mark position y _pA [unit: pixel] on the image of the CCD camera 21 is obtained from the equation (2).

となる。ここで、ｙ_A は３次元空間における実際のｙ方向位置［ｍｍ］で、ｚ_A は同じく３次元空間における実際のｚ方向位置［ｍｍ］であって、Ｓは前記したようにレンズの固有倍率である。

It becomes. Here, y _A is the actual y-direction position [mm] in the three-dimensional space, z _A is also the actual z-direction position [mm] in the three-dimensional space, and S is the intrinsic magnification of the lens as described above. It is.

The measured mark position y _pA [pixel] can be converted into the dimension y _m [mm] by the following equation. For example, if the total number of pixels in the y direction of the CCD camera 21 is 640 and the dimension is y _L [mm],

In this equation (16), the term applied to y _pA on the right side is a coefficient obtained by calibration in advance, and is a conversion coefficient from pixel on image to actual distance [mm]. If this is set to K, the equation (16) becomes

（１５），（１７）より、

From (15) and (17),

となる。ここで、ｙ_A ＝ｙ_m （実距離＝計測距離）であるので、（１８）式から、

It becomes. Here, since y _A = y _m (actual distance = measurement distance), from equation (18),

としてｚ_A を求めることができる。このｚ_A をｚ₀ の初期値として使用する。

Z _A can be obtained as This z _A is used as the initial value of z ₀ .

Now, after calculating x ₀ and y ₀ by the equation (14), the rotation angle θ _z around the z-axis is calculated as the second step.
(8) Multiply A _x ^-1 and A _y ^-1 on both sides

よって、（２０）式のｉ＝１〜４を用いて以下の式を導出できる。

Therefore, the following equation can be derived using i = 1 to 4 in the equation (20).

ここで、前記と同様に、座標変換を示す回転行列として３−２−１オイラー角を使用した場合、Ａ_y ^-1Ａ_x ^-1およびＡ_z は、

Here, in the same way as described above, when 3-2-1 Euler angles are used as a rotation matrix indicating coordinate transformation, A _y ⁻¹ A _x ⁻¹ and A _z are

となる。

It becomes.

  Now, the left side of the equation (21) is calculated. From equation (3)

となる。ここで、（１３）式から、

It becomes. Here, from equation (13):

であるので、

So

となる。よって（２１）式の左辺は、ｘ_pi，ｙ_pi，ｚ₀ ，ｚ_i ，θ_x ，θ_y ，Ｓを用いて計算できる。

It becomes. Therefore, the left side of the equation (21) can be calculated using x _pi , y _pi , z ₀ , z _i , θ _x , θ _y , and S.

That is, when calculating equation (26) and multiplying the result by A _y ⁻¹ A _x ⁻¹ , the left side of equation (21) is

が求められる。

Is required.

  Next, the right side of equation (21) is calculated. From Equation (6) and Equation (23),

となる。

It becomes.

  From the equations (27) and (28), the sin θ term and the cos θ term are summarized from the corresponding components on both sides of the equation (21).

となり、これによりθ_z を求めることができる。

Thus, θ _z can be obtained.

X _pi used for the calculation of the _{θ z, y pi, S,} z 0, z i is the same as the first step. Also, θ _x and θ _y are set to 0 in the first calculation, and the calculation results from the previous one are used from the second time onward.

Next, as a third step, z ₀ , θ _x , and θ _y are calculated.
From equation (8)

であり、また、

And also

と近似するととともに、

And approximating

とおくと、（６）式に示すように、Ｃ_xi，Ｃ_yiはｉ＝１〜４によりｋあるいは−ｋとなる。よって

Then, as shown in the equation (6), C _xi and C _yi become k or −k when i = 1 to 4. Therefore

となる。
（３）式より、

It becomes.
From equation (3)

となり、（６）式と（１３）式を用いるて上式の第１，第２成分を表すと、

When the first and second components of the above equation are expressed using the equations (6) and (13),

となる。この（３２）式のｘ_i ，ｙ_i ，ｚ_i に（３１）式を代入して整理すると、

It becomes. Substituting equation (31) for x _i , y _i , and z _i in equation (32),

となる。

It becomes.

The equations (33) and (34) have i = 1 to 4, which is a total of eight equations. If the eight equations are regarded as equations for z ₀ , θ _y , θ _x , each coefficient is a _ij and the right side is b _i , the equations (33) and (34) are

と表すことができる。
ここで、（３５）式の両辺にＡ^T （転置行列）をかけて、

It can be expressed as.
Here, A ^T (transpose matrix) is applied to both sides of equation (35), and

更にこの（３６）式に（Ａ^T Ａ）^-1（（）^-1は逆行列）をかけると、

Furthermore, when (A ^T A) ⁻¹ (() ⁻¹ is an inverse matrix) is multiplied by this equation (36),

が得られる。

Is obtained.

Z ₀ , θ _y , and θ _x are calculated from the equations (33), (34), and (37). X _pi used in this case, y _pi, S, z _i is the same value as the first step, Cx _i, Cy _i is x, y distance of the mark body M, (6) k or as formula - k (fixed value), and x ₀ , y ₀ , and θ _z use the results of the first and second steps. For the nonlinear terms of θ _x , θ _y (θ _x ² , θ _x ^3, etc.), the first calculation is 0 as in the second step, and the calculation results from the previous one are used for the second and subsequent times.

Next, as a fourth step, x _i , y _i and z _i are calculated from the equation (8). At this time, θ _x , θ _y , and θ _z use the results of the second and third steps, and each component of C _i (vector) is expressed by the expression (6) as the x, y distance of the mark body M. .

When the above first to fourth steps are repeatedly calculated, the calculated value gradually approaches the correct value, and the displacement of the mark body M in the x, y, and z directions and the rotational angles θ _x , θ _y , θ _z can be accurately obtained.

  The calculation result of the three-dimensional position information of the mark body M by the above three-dimensional position calculation unit 22d is sent to the elongation correction calculation unit 22e together with the output of the elongation calculation unit 22b, and at the same time, the width correction calculation together with the output of the width calculation unit 22c. Sent to the unit 22f.

The elongation correction calculation unit 22b uses the three-dimensional position information of the mark body M, and equivalently, the three-dimensional position information of the test piece W to which the mark body M is attached, to test the marked lines S1 and S2. The distance between the marked lines S1 and S2 is accurately corrected in consideration of the displacement in the three-dimensional direction and the rotation in each axial direction from the beginning. This correction calculation means that the distance between the marked lines S1 and S2 is measured in the [a ₁ ] coordinate system, and accordingly, the displacement of the test piece W in the three-dimensional direction from the beginning of the test and the rotation around each axis. Accurate elongation that eliminates the effects of rotation can be obtained.

  The same applies to the width correction calculation unit 22f, and the three-dimensional position information of the test piece W calculated by the width calculation unit 22c using the three-dimensional position information of the mark body M is the three-dimensional direction of the test piece W. It is corrected to an accurate dimension that eliminates errors due to displacement and rotation around each axis.

  Of particular note in the above embodiments is that the position information of the marked lines S1 and S2, the position information of both edges of the test piece W, and the three-dimensional position information of the mark body M are all in one CCD camera 21. Therefore, it is possible to obtain the test piece 3 at a low cost with a relatively simple hardware configuration without using any other measuring device such as a test piece displacement detecting means. It is possible to accurately measure the elongation and width change in which all the errors due to rotation in the dimension direction and around each axis are eliminated.

  In the above embodiment, an example is shown in which four rhombus marks m1 to m4 are attached to the mark body M. However, any number can be used as long as the number of marks is three or more. Also, any shape can be used as long as its center is known.

  Moreover, in the above Example, while the one mark body M was stuck to the test piece W, and the markings S1 and S2 were attached | subjected separately from this, the example which attached | subjected mark lines S1 and S2 was shown, but it is clear from the above Example. In addition, since the three-dimensional position information of the center of the mark body M is obtained, the mark body is attached to each of the two marked lines, and the three-dimensional position information of the two mark bodies is equivalent to the above. By calculating using the technique, the centers of the two mark bodies are substantially regarded as two marked lines, and the distance between the centers of the two mark bodies is calculated from the three-dimensional position information itself of each mark body to expand. You may ask for.

  Furthermore, in the above embodiment, a method of performing processing with one camera has been shown, but it is also possible to use a plurality of cameras. For example, it is possible to use one camera for three-dimensional measurement and use one or two cameras for two marked lines. In the case where a mark body is also used as a marked line, each mark body 1 It is possible to use one camera for each camera and measure using two cameras.

  Furthermore, although the case where both the elongation and the width are measured has been described in the above embodiments, the present invention can of course be applied to the measurement of only the elongation or only the width.

  Furthermore, it is needless to say that the video type three-dimensional measuring apparatus of the present invention can be applied not only to a test piece in a material test but also to a three-dimensional measurement of an arbitrary object. By sticking a required mark body to an arbitrary object, it is possible to know the three-dimensional position information of the object with a simple hardware configuration.

It is a principal part block diagram of the Example of this invention, and is a figure which shows together the side view of the test piece W with which the material testing machine was mounted | worn, and the block diagram showing the structure of a video type extension and width meter. It is an AA arrow enlarged view of FIG. It is explanatory drawing of calculation of the three-dimensional position information of the mark body M by the three-dimensional position calculating part 22d of this invention Example. Similarly it is explanatory drawing of calculation of the three-dimensional position information of the mark body M by the three-dimensional position calculating part 22d of an Example of this invention. In illustration of the calculation of the three-dimensional position information of the mark body M by the three-dimensional position calculating section 22d of the present invention embodiment, (A) is an origin focus of the CCD camera 21 in the illustration of [a _0] coordinate system Yes, (B) is an explanatory diagram of the [a ₁ ] coordinate system with the center of the mark body M as the origin. It is an explanatory view of obtaining the z ₀ used for the first calculation by the three-dimensional position calculating section 22d of the present invention embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1a, 1b Grasp 2 Video type stretcher / width meter 21 CCD camera 22 Computer 22a Image processing part 22b Elongation calculating part 22c Width calculating part 22d Three-dimensional position calculating part 22e Elongation correction calculating part 22f Width correction calculating part M Mark body m1- m4 mark W test piece

Claims (4)

  A mark body that is affixed to an object, the position relationship of which is known, and the position relationship of which is at least three marks that are unchanged when the object moves, and the object A video camera that captures the attached mark body, and a video signal from the video camera, and two axes (x, y) that are orthogonal to each other on a plane that is orthogonal to the optical axis (z-axis) direction of the video camera Image processing means for obtaining the position information of each mark in the (axis) direction, and the position information of each mark by the image processing means, by numerical calculation, x, y, A three-dimensional measuring apparatus comprising a three-dimensional position calculating means for calculating any or all of coordinates in the z-axis direction and momentary rotation angles around the x, y, and z axes of the mark body. .   One or a plurality of video cameras for photographing a test piece to which the mark body according to claim 1 is attached and a predetermined distance is provided in the load direction and two marked lines are attached, and the video camera An image processing unit that captures a photographing signal from the image processing unit to perform image processing according to claim 1 and obtains positional information of each of the marked lines in the x and y axis directions, and a position of each mark by the image processing unit 2. The three-dimensional position calculating means according to claim 1, which uses information, an elongation calculating means for calculating the elongation of each marked line by using the position information of each marked line by the image processing means, and its elongation calculation Correction calculation means for correcting the calculation result of the momentary elongation by the means by using any or all of the displacement of the mark body in the x, y, z axis direction and the rotation angle around each axis by the three-dimensional position calculation means. It is equipped with Video extensometer according to claim.   The mark bodies are attached to positions corresponding to two or more marked lines of the test piece, respectively, and the image processing means captures a photographing signal from the video camera, and each of the two or more mark bodies on the two or more mark bodies. Obtaining the position information of the mark, the three-dimensional position calculation means calculates the displacement in the x, y and z axis directions and the rotation angle around each of the two or more mark bodies, and the elongation calculation. Instead of the means and the correction calculation means, the elongation for calculating the elongation by calculating the distance between the two or more mark bodies from the calculation result of the three-dimensional positions of the two or more mark bodies by the three-dimensional position calculation means. The video extensometer according to claim 2, further comprising a calculation means.   A video camera for photographing a test piece to which the mark body according to claim 1 is attached, and a video signal from the video camera are captured to perform image processing according to claim 1 and a load direction of the test piece. The image processing means for obtaining the position information of the edges on both sides along the line, the position information of each mark position by the image processing means, and the three-dimensional position calculation means according to claim 1 and each of the image processing means Width dimension calculating means for calculating the distance between the edges using edge position information, and the calculation result of the width by the width dimension calculating means for the x, y of the mark body by the three-dimensional position calculating means. A video width meter comprising correction calculation means for correcting using any or all of the displacement in the z-axis direction and the rotation angle around each axis.

Images