RU2556310C2 - Device for remote measurement of geometric parameters of profiled objects - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: device relates to inspection technology and may be used to measure geometric dimensions of profiled objects. The device comprises a TV camera 2, fixed on a platform 1, rotated in a horizontal (angle α) and vertical (angle β) planes. On the platform surface there is a laser range finder 3, the optical axis of which is parallel to the optical axis of the camera, a sensor of azimuthal angles 4, which generates signals proportionate to angles α and β. The platform is rotated manually by a special micrometric mechanism, which is not shown. Outputs of the TV camera, the range finger and angle sensor are connected to the input of the processing device 5 connected to a video monitoring device 6.

EFFECT: simplified design and procedure of measurements with preservation of accuracy.

2 dwg

Description

The invention relates to instrumentation and can be used to measure the geometric dimensions of pipes, blades of gas turbine engines and other specialized engineering products - external and internal diameter, wall thickness, cross-sectional area.

Known devices for measuring the geometric parameters of objects, including pipes, containing a television camera and processing devices based on the formation of an image of a controlled object on the surface of a CCD camera matrix with further data transfer to a computer and software processing of the results [a.s. SU 1837160, MKI G01B 21/30, bull. No. 32, 1993; A.S. SU 2052768, MKI G01B 17/00, bull. No. 22, 1995; A.S. SU 1657960, MKI G01B 21/10, bull. No. 23, 1991; A.S. SU 1675664, MKI G01B 11/02, bull. No. 33, 1991; A.S. SU 1716327, MKI G01B 21/20, bull. No. 8, 1992; RU patent No. 2163395, G06K 9/52, 2000], utility model patent No. 32261, B.I. No. 25 dated 09/10. 2003.

A disadvantage of the known devices is the low accuracy associated with optical image distortion at the boundaries of the receiving matrix, as well as the dependence of the image size on the distance of the camera lens to the controlled object. This makes them difficult to use in operational control devices in cases where the distance to the controlled object is not known or not fixed.

The prototype of the claimed invention is a television device containing four (six) cameras, grouped in pairs, spaced at fixed distances in orthogonal coordinates and connected to a video monitoring device, the output of which is connected to a scaling device and a processing device (RF patent for utility model No. 73069, b. and. No. 13 dated May 10, 2008).

The disadvantages of the prototype are the design complexity associated with the presence of four (six) cameras at the same time, the complexity of the measurement procedure associated with tuning to the object, since the distance between the cameras is rigidly fixed, the overall object may not fall into the capture zone of all cameras at the same time.

The task: to simplify the design and measurement procedure, while maintaining accuracy.

The solution to this problem is achieved by the fact that in the known device containing a camera connected via an input device to a video monitoring processing device, the camera is placed on a rotary platform, on which an additional laser rangefinder and an azimuth angle sensor are installed, the outputs of which are connected to the inputs of the processing device.

This allows you to create a complete image of the object on the screen of the video monitoring device at different shooting angles, to determine the scale of a flat image at each coordinate, to take measurements regardless of the relative orientation of the camera and the object, taking into account distortions arising in the optical system and the video monitoring device.

The invention is illustrated by drawings, where figure 1 shows the structural diagram of the proposed device, and figure 2 - image of the controlled object obtained on the screen of a video monitoring device at different angles of inclination of the turntable.

The device consists of a camera 2 mounted on a platform 1, capable of rotating in horizontal (angle α) and vertical (angle β) planes. A laser range finder 3 is also installed on the surface of the platform, the optical axis of which is parallel to the optical axis of the camera, as well as an azimuthal angle sensor 4, which generates signals proportional to the angles α and β.

The rotation of the platform is set manually by a special micrometric mechanism, which is not shown. The outputs of the camera, rangefinder and angle sensor are connected to the input of the processing device 5 connected to the video monitoring device 6. Figure 1 also shows a monitored object S, which shows three arbitrary points A, B, C used for calibration. The figure 2 shows the image of the controlled object on the screen of a video monitoring device at three different viewing angles. Moreover, in Figure 2a shows an object image hover laser rangefinder to point _{A [S A = S ((} α A, β A)]; in Figure 2b - hover rangefinder to point _{B [S B = S (α} B, β B )];, in figure 2c, when the range finder is hovering over the point C [S _C = S (α _C , β _C )].

In figure 2, the coordinates µ, η determine the points on the plane of the photomatrix of the camera and, therefore, the video monitoring device. Figure 2 shows that at different angles of shooting the image of the object is formed at different points of the photomatrix and, therefore, is displayed at different points on the screen of the video monitoring device.

The essence of the device is as follows. The device is installed in the field of view of the camera. Three arbitrary points A, B, C located on a controlled surface (plane, line) are selected on the object. Using the turntable, a laser range finder is aimed at point A, the angles of the position of the turntable corresponding to this point α _A , β _A are recorded, the distance from the camera to point A is measured, the object is shot at the obtained angles, an image is obtained S _A = S (α _A , β _A ) (Fig. 2a). Then alternately induce rangefinder to points B and C to give the image of the object at angles corresponding to the selected points S _B = S ((α _B, β _B), S _C = S (α _C, β _C). Thus obtained three flat images of a three-dimensional product. The obtained images and laser measurements make it possible to conduct a complete analysis of the geometric dimensions of the product. Each image has its own scale on the screen. Determining this scale is one of the intermediate tasks of the measurement procedure. To determine the scale of each image, it is necessary to correlate the The total dimensions of the segments AB, AC and BC with their lengths on each screen.The actual sizes of these segments in space are determined by the formulas.

$$AB=\sqrt{{\left(x_A-x_B\right)}^2+{\left(y_A-y_B\right)}^2+{\left(z_A-z_B\right)}^2},\left(\mathrm{one}\right)$$

$$AC=\sqrt{{\left(x_A-x_C\right)}^2+{\left(y_A-y_C\right)}^2+{\left(z_A-z_C\right)}^2},\left(2\right)$$

$$BC=\sqrt{{\left(x_B-x_C\right)}^2+{\left(y_B-y_C\right)}^2+{\left(z_B-z_C\right)}^2},\left(3\right)$$

Where $$x_A={\rho }_A\cos {\beta }_A\cos {\alpha }_A,\left(\mathrm{four}\right)$$

$$x_B={\rho }_B\cos {\beta }_B\cos {\alpha }_B,\left(5\right)$$

$$x_C={\rho }_C\cos {\beta }_C\cos {\alpha }_C,\left(6\right)$$

$$y_A={\rho }_A\cos {\beta }_A\cos {\alpha }_A,\left(7\right)$$

$$y_B={\rho }_B\cos {\beta }_B\cos {\alpha }_B,\left(8\right)$$

$$y_C={\rho }_C\cos {\beta }_C\cos {\alpha }_C,\left(9\right)$$

$$z_A={\rho }_A\sin {\beta }_{A}10$$

$$z_B={\rho }_B\sin {\beta }_B\mathrm{eleven}$$

$$z_C={\rho }_C\sin {\beta }_{C}12$$

In formulas (4) - (12), the quantities ρ _A , ρ _B , ρ _{C are} determined by the laser range finder; values α _A , α _B , α _C , β _A , β _B , β _C - angle sensor.

On the screen of the video monitoring device, space segments AB, AC and BC located in space are converted into segments A'B ', A'C' and B'C 'located on the plane of the screen (see figure 2). The length of these segments on the display screen is determined through their coordinates according to the formulas:

$$A\text{'}\text{'}B\text{'}\text{'}=\sqrt{{\left({\mu }_A-{\mu }_B\right)}^2+{\left({\eta }_A-{\eta }_B\right)}^2}\left(13\right)$$

$$A\text{'}\text{'}C\text{'}\text{'}=\sqrt{{\left({\mu }_A-{\mu }_C\right)}^2+{\left({\eta }_A-{\eta }_C\right)}^2}\left(\mathrm{fourteen}\right)$$

$$B\text{'}\text{'}C\text{'}\text{'}=\sqrt{{\left({\mu }_C-{\mu }_B\right)}^2+{\left({\eta }_C-{\eta }_B\right)}^2}\left(\mathrm{fifteen}\right)$$

In order to measure a real object from its flat image, it is necessary to determine the scale, that is, the price of dividing one image pixel by the coordinates µ, η, corresponding to the real dimensions of the object in space. To do this, we introduce the scaling factors in coordinates and compare the real segments (1) - (3) with their flat images on the screen (13) - (15). From the conditions A'B '= A'B' and AC = A'C 'we obtain the following equations:

$$\{\begin{array}{c}A{B}^2={\left(x_A-x_B\right)}^2+{\left(y_A-y_B\right)}^2+{\left(z_A-z_B\right)}^2=R_{\mu }{\left({\mu }_A-{\mu }_B\right)}^2+R_{\eta }{\left({\eta }_A-{\eta }_B\right)}^2\\ A{C}^2={\left(x_A-x_C\right)}^2+{\left(y_A-y_C\right)}^2+{\left(z_A-z_C\right)}^2=R_{\mu }{\left({\mu }_A-{\mu }_C\right)}^2+R_{\eta }{\left({\eta }_A-{\eta }_C\right)}^2\end{array}\left(16\right)$$

where R _µ , R _η are scaling coefficients with respect to the coordinates µ, η.

The joint solution of these equations will give the values R _µ , R _η , which determine the metrological correspondence between the actual dimensions of the object in space with its flat image on the screen. From system (16) we find:

$$\begin{array}{cc}{R}_{\mu }=\frac{\Delta \mu }{\Delta },& R_{\eta }=\frac{\Delta \eta }{\Delta }\end{array},\left(17\right)$$

Where $$\Delta =\left|\begin{array}{cc}\begin{array}{c}{\left({\mu }_A-{\mu }_B\right)}^2\\ {\left({\mu }_A-{\mu }_C\right)}^2\end{array}& \begin{array}{c}{\left({\eta }_A-{\eta }_B\right)}^2\\ {\left({\eta }_A-{\eta }_C\right)}^2\end{array}\end{array}\right|={\left({\mu }_A-{\mu }_B\right)}^2{\left({\eta }_A-{\eta }_C\right)}^2-{\left({\eta }_A-{\eta }_C\right)}^2{\left({\mu }_A-{\mu }_C\right)}^2,\left(\mathrm{eighteen}\right)$$

$$\begin{array}{l}\Delta \mu =\left|\begin{array}{c}{\left(x_A-x_B\right)}^2+{\left(y_A-y_B\right)}^2+{\left(z_A-z_B\right)}^2{\left({\eta }_A-{\eta }_B\right)}^2\\ {\left(x_A-x_C\right)}^2+{\left(y_A-y_C\right)}^2+{\left(z_A-z_C\right)}^2{\left({\eta }_A-{\eta }_C\right)}^2\end{array}\right|=\\ =\left[{\left(x_A-x_B\right)}^2+{\left(y_A-y_B\right)}^2+{\left(z_A-z_B\right)}^2\right]\cdot {\left({\eta }_A-{\mu }_C\right)}^2-\\ \left[{\left(x_A-x_C\right)}^2+{\left(y_A-y_C\right)}_2+{\left(z_A-z_C\right)}^2\right]\cdot {\left({\eta }_A-{\eta }_B\right)}^2\left(19\right)\end{array}$$

$$\begin{array}{l}\Delta \eta =\left|\begin{array}{c}{\left({\mu }_A-{\mu }_B\right)}^2+{\left(x_A-x_B\right)}^2+{\left(y_A-y_B\right)}^2{\left(z_A-z_B\right)}^2\\ {\left({\mu }_A-{\mu }_C\right)}^2+{\left(x_A-x_C\right)}^2+{\left(y_A-y_C\right)}^2{\left(z_A-z_C\right)}^2\end{array}\right|=\\ ={\left({\mu }_A-{\mu }_B\right)}^2\cdot \left[{\left(x_A-x_C\right)}^2+{\left(y_A-y_C\right)}^2+{\left(z_A-z_C\right)}^2\right]-\\ {\left({\mu }_A-{\mu }_C\right)}^2\cdot \left[{\left(x_A-x_B\right)}^2+{\left(y_A-y_B\right)}^2+{\left(z_A-z_B\right)}^2\right]\left(\mathrm{twenty}\right)\end{array}$$

Then any arbitrary size MN between arbitrary points on the controlled object is determined by the formula

$$MN=\sqrt{R_{\mu }{\left({\mu }_M-{\mu }_N\right)}^2+R_{\eta }{\left({\eta }_M-{\eta }_N\right)}^2,\left(21\right)}$$

In this case, the size calculation can be performed according to any of the images of figure 2.

The invention allows to almost completely eliminate the influence of distortions of the controlled object relative to the optical axis of the camera. Moreover, if the camera captures the entire image of the object, the processing device can be programmed not only to measure some size, but also to analyze the shape, calculate the center of gravity, the moment of resistance, etc. The device can significantly improve the accuracy of operational measurements of the geometric parameters of objects, making them independent of the relative position of the camera and the controlled object when used in a wide range of measurements.

Claims (1)

A device for remote measurement of the geometric parameters of profile objects containing a camera and a laser range finder, the outputs of which are connected to a processing device, characterized in that the camera and a laser range finder are placed on a two-axis rotary platform equipped with an azimuthal angle sensor, the output of which is connected to the processing device, the optical axis The cameras are parallel to the laser beam of the rangefinder.

Images