JPS63150610A - Measuring method for sectional shape - Google Patents

Abstract

PURPOSE:To measure the sectional size of an object without contacting by providing two size measuring means which are each constituted by arranging a light source and an image sensor opposite each other in a radial direction about the object and measure the image side of the section of the object. CONSTITUTION:When the sectional size of the object M with the rectangular section specified by plural vertexes P1-P4 is measured, light sources A'-D' and image sensor cameras A-D are arranged opposite each other about the object M in radial directions. Then the cameras B and C of two size measuring means which constitute one vertex measuring instruments for measuring vertexes P1 and P3 are installed almost axially symmetrically about a straight line V and the cameras A and D of the other vertex measuring instrument which measure vertexes P2 and P4 are installed centrally axially symmetrically about the position of the light source D'. Consequently, the coordinates of the vertexes P1 and P3 are found through the cameras B and C and the coordinates of the vertexes P2 and P4 are found through the cameras A and D; and the lengths of diagonals P1P3 and P2P4 are measured from the coordinates of the vertexes P1 and P3, and P2 and P4 to calculate the sectional shape size of the object M.

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Industrial field of application The present invention relates to a cross-section for measuring the dimensions of a cross-sectional shape such as steel A having a cross-section specified by a plurality of vertices such as a rectangle produced in the steel industry etc. The present invention relates to a method for measuring a shape, and particularly to a method for measuring a cross-sectional shape suitable for non-contact measurement in a production process line.

(0) As a method for measuring the dimensions of the cross section of backstripes and conventional steel materials,
As disclosed in Publication No. 759, Japanese Patent Application Laid-open No. 61-258105, etc., a round steel is used as an object, parallel light from a laser light source is aimed at the object, and a one-dimensional image sensor is used to capture the light while the light is blocked. Although measurement methods are known, they cannot be applied directly to cross sections such as rectangles. Furthermore, this method requires two sets of large lenses for receiving parallel light and light with a sufficient width to accommodate the object and its path variations, but large lenses with a diameter of 100 mm or more are expensive, and Since a mechanism for aligning the optical axis of parallel light is required, the manufacturing cost of the device increases, and it cannot be applied to large diameter devices.

Therefore, the conventional dimension measuring method described above is mainly applied to relatively small objects such as wire rods and objects having a round cross section.

(c) Purpose of the Invention The present invention has been made in view of the current situation, and is a method for measuring the cross-sectional shape of an object having a cross-sectional shape specified by a plurality of vertices, such as a rectangle or an H-shape, in which the cross-sectional dimensions of an object can be determined without contact. The method is intended as a method.

(d) Structure 1 of the invention The above objects are achieved by the present invention as follows.

That is, when measuring the cross-sectional dimension of an object having a cross-sectional shape specified by a plurality of vertices, the present invention provides a light source and an image sensor in the radial direction of the object.
The two sets of dimension measuring means are arranged oppositely to each other to measure the projected dimensions of the cross section of the object. A vertex measuring device is arranged so that the tip axis is perpendicular to the axis of the object to form a two-dimensional coordinate system, and the coordinates of the vertex are measured. This method of measuring cross-sectional shape is characterized by measuring everything and measuring cross-sectional dimensions.

The present invention focuses on the apex in measuring the cross-sectional shape,
This apex is described in the above-mentioned Japanese Patent Application Laid-Open No. 50-92759.

Using a technically established dimension measuring means known in Japanese Patent Application Laid-Open No. 61-258105 and the like and which can be used as is with a commercially available image sensor camera, a two-dimensional coordinate system is constructed and its coordinates are measured; Since the required cross-sectional dimensions are determined, the present invention enables stable measurement of a wide range of cross-sectional shapes such as rectangular, 2-trapezoidal, )-1, and T-shaped blades, which were previously difficult to measure, using established technology. At the same time, it is also possible to measure large cross sections such as large diameter steel materials.

Above) In the present invention, each vertex m11 determining device and each of the dimension measuring means constituting it need only be on a plane perpendicular to the axis of the object, and from the viewpoint of eliminating interference of the light source, it is sufficient to place it on a different plane. It is preferable that they are located on the same plane, and from the viewpoint of data processing and measurement accuracy, it is preferable that they are all located on the same plane. Appropriate selection is made depending on the situation. Further, the origin of the two-dimensional coordinate system does not necessarily need to be the same, but it is preferable to make it the same point on the same plane from the viewpoint of data processing and accuracy.

Preferably, the central axis of the object is located near this origin.

Furthermore, in the present invention, since the object to be measured is illuminated by a plurality of light sources, a scattered light source is preferable as the light source. By using a scattered light source, it is possible to use a light receiving lens with a small diameter, which is a great advantage in terms of system configuration.

The details of the present invention will be explained below using an example of measuring a rectangular cross section. FIG. 1 is an explanatory diagram of the configuration of the above embodiment.

FIG. 2 is an explanatory diagram of the principle of apex measurement.

A, B, C, and D in the figure are one-dimensional image sensors, respectively.
In this example, a commercially available camera, specifically Chiruskiton (trade name) manufactured by Yojin Engineering ■, was used. A'~D
' is a scattered light source that constitutes a means for determining the dimensions aI of the cameras A, B, C, and D, and in this example, a high-frequency lighting fluorescent lamp is used. M is the object to be measured, which in this example is a steel material with a rectangular cross section PI P2 P3 PJ specified by vertices P1 to P4, and is transported on a pass line (not shown) consisting of a figure-7 grooved eye. has been done. Here, let the straight line connecting the vertices P1 and P4 of each diagonal be the straight line connecting the H1 vertices P2 and P3, and let the intersection of H and ■, that is, the center of the cross section, be ○.

The object is guided by a V-shaped grooved roller guide installed on the production line as described above, but the object is guided vertically by a V-shaped grooved roller guide.

with side-to-side fluctuations and twisting. Therefore, in the state when the reference piece for calibration is temporarily installed, the two straight lines V,
Define H and its intersection 0.

2 constituting a vertex measuring device that measures vertex P+, P3
Cameras B and C of the two dimension measuring means are both at the vertex P+,
The position is approximately symmetrical to the straight line V, so that the projection with P3 as both ends can be measured, and its optical axis lies on the same plane perpendicular to the axis of the object, and intersects at the cross-sectional center O at an intersection angle α. Set it up like this. Therefore, from the cameras B and C, due to the effect of back illumination from the respective scattered light sources B' and C', a shaded image blocked by the diagonal lines of the vertices P1 and P3 can be seen, and the vertices P2 and PJ are hidden. Vertex P for cameras B and C
+, a projection with P3 at both ends is detected. The angle α is
Camera B even when object M is twisted.

For both vertices P2. P4 is always hidden, vertex P1
It is necessary that a shadow image of only the diagonal line connecting P3 and P3 can be seen, and in practical terms, it is preferable that α=30°.

Cameras A and D of another vertex measuring device that measure the remaining vertices P2 and PJ measure the aforementioned vertices P+ and PJ with respect to the straight line H.
It can be installed in the same way as cameras B and C for P3 measurement, but
Due to the constraints of the production process, the position of D was point-symmetrical to the position of D', and the other points were installed in the same manner as the apex measuring apparatus for the apexes P+ and P3 described above. In this case, the intersection angle of the optical axes is (1
80-β).

Then, the output signals of the cameras A, B, C, and D are processed by a data processing device including a microcomputer (not shown), and the apex P
+, P2, P3, and PJ coordinates are measured.

Hereinafter, according to Fig. 2, the apex aP where the optical axes intersect at the intersection angle α
An algorithm for determining the coordinates of a vertex P at an arbitrary position using the vertex measuring device of +, P2 will be explained.

Origin 0. In the coordinate system determined by the straight lines H and ■ mentioned above, the coordinates of any vertex P are expressed in polar coordinates (r.

θ).

When two cameras B and C are focused on the origin O,
The objective distance and imaging distance of the light receiving lenses Km and Kn of cameras C and 8 are (am, bm), respectively.

(an, bn).

The image of vertex P is formed by one-dimensional image sensors of cameras C and B.
It can be determined by the displacement amount from the origin 0 on 1 m, In, and Δpm and Δpn are given by the following formula from the figure, respectively.
α+θ)]xr sin (1/2α10)...■
ΔPn = bn/ Can-r cos (1/2α
-θ)]xr sin (1/2α-θ)...■
The coordinates (r, θ) of the vertex P can be found from the simultaneous equations (1) and (2).

Therefore, in FIG. 1, point PI is obtained from cameras B and C by applying the above algorithm.

P3, camera A, D to point P2. The coordinates of P4 are found,
Using these coordinates, the shape and dimensions of the rectangular cross section are calculated and measured using a conventional method. In this example, the cross-sectional shape is diagonal PI P3,
P2 shall be managed by the length of PJ, and the vertices P+, P
3 and from the coordinates of P2 and P3, the diagonal line PI P3 is obtained based on Vitagoras' theorem.

P2 The length of PJ was measured. The flow of these programs is the same as usual, and the explanation will be omitted.

Incidentally, since the origin O is common to the polar coordinate systems of cameras B and C and cameras A and D, the mutual relationship can be corrected based on the intersection angle of straight line 2 and H.

Normally, in the case of a long object with a rectangular cross section, it is often sufficient for quality control purposes if the dimensions of the two diagonal lines mentioned above are accurately determined, and the intersection angle of the two straight lines V and H mentioned above is 90 ''' is approximated.

In the method of the present invention described above, it is necessary to accurately determine the intersecting angle α of the optical axes of the two cameras, but this calibration work can be easily performed using a calibration apparatus as shown in FIG. Bold 11 is a rotary disk 1''4 rotatably provided on a base plate 13.
The ball )-12 is mounted on the rotating disk H4. Temporarily install this calibration device as a reference piece in the measurement area, and after focusing each camera on the optical axis with respect to pole H1, rotate the rotating disk to rotate the 11 points where the shadows of pole H1° and ball + 12 overlap. By reading the angle for each camera, the intersection angle α is 1 no J
determined.

As explained above, the present invention measures the dimensions necessary for cross-section management by determining the coordinates of the vertices that specify the cross-section, so it can be applied even if the cross-section is trapezoidal, single-mouth, or H-shaped. It has a wide range of applications and can be easily applied even if the cross section is large. Moreover, since it is an optical non-contact measurement, it can be applied to on-line ITI determination in production processes such as steel materials that are transferred at high speed. In this way, the present invention makes a significant contribution to industry.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram that does not show the configuration of the embodiment, FIG. 2 is an explanatory diagram showing a vertex measuring method thereof, and FIG. 3 is a perspective view of the calibration H position of the embodiment. A, B, C, D: Image sensor cameras A', B'
, C'D': Light source M: Target P+, P2
, P3, P4: Vertex Fig. 1

Claims (1)

[Claims] 1. When measuring the cross-sectional dimension of an object having a cross-sectional shape specified by a plurality of vertices, a light source and an image sensor are placed facing each other in the radial direction of the object. It consists of two sets of dimension measuring means configured to measure the projected dimensions of a cross section of an object, and the two sets of dimension measuring means are arranged such that the same apex of the cross section of the object becomes the projection end point and the optical axis of the two sets of dimension measuring means measures the projected dimension of the cross section of the object. Measure all the vertices necessary for specifying the cross-sectional dimension using a vertex measuring device arranged to form a two-dimensional coordinate system with planes perpendicular to the axis of the vertices, and measuring the coordinates of the vertices; A method for measuring cross-sectional shape, characterized by measuring cross-sectional dimensions. 2. The cross-sectional shape is rectangular, and the apex measuring device has dimension measuring means that can simultaneously measure diagonal vertices, and the first aspect of the present invention is Method for measuring cross-sectional shape as described in section. 3. The cross-sectional shape according to claim 1 or 2, wherein all the light sources of the dimension measuring means are on one side of the object, and all the image sensors are on the other side. How to measure. 4. The cross-sectional shape of any one of claims 1 to 3, wherein all of the vertex measuring devices have two-dimensional coordinate systems on the same plane perpendicular to the object and have the same origin. How to measure. 5. Claims 1 to 5, wherein the light source is a scattered light source.
The method for measuring a cross-sectional shape according to any one of item 4.