Device for measuring the flatness of a surface or the thickness of a flat glass panel

Abstract

The invention relates to a device (1) for measuring the flatness of a surface (2) or the thickness of a flat glass panel, wherein the device (1) comprises at least three light sources (5) which emit light divergently and a camera (6) which is provided for capturing the light reflected from the surface (2) in a directed manner and originally coming from the light sources (5) emitting divergently. The device (1) also comprises at least three light sources (4) which each emit light in a collimated manner along a definedly positioned straight line (7), wherein the camera (6) is also provided for capturing the light reflected from the surface (2) in a scattered manner and originally coming from the light sources (4) emitting in a collimated manner.

Description

[0001] Device for measuring the flatness of a surface or the thickness of a flat glass pane

[0002] The invention relates to a device for measuring the flatness of a surface or the thickness of a flat glass pane.

[0003] The documents EP 0433555 A2, DE 102018202625 A1 and WO 2022214159 A1 show by way of example a known principle for the optical measurement of the geometry of a surface using diffusely backscattered light. A collimated light beam with the smallest possible cross-sectional area, which typically originates from a laser, is directed onto the surface along a straight line defined in space. The surface is imaged by a camera - that is to say a combination of an objective lens and a spatially resolving optical sensor. The point at which the collimated light beam hits the surface appears as a bright point in the camera image. The position of this point in the image is used to calculate the position of the point in space imaged with it.

[0004] According to DE 2853816 A1, a collimated light beam with a known position and orientation in space is sent onto a reflective surface to be measured, and the light beam reflected in a directed manner (i.e. not diffusely) from this surface strikes a point on a light-sensitive detector surface arranged in a known position. From the position of the point of impact on the detector surface, possible pairs of values ​​for position and angular orientation can be calculated back with regard to the point on the surface to be measured from which the beam was reflected. By applying knowledge of further boundary conditions - which is obtained, for example, through further measurements in defined, changed positions - the position and / or the angular orientation can be clearly calculated back.For a simple good / bad assessment of the flatness or, more generally, the shape accuracy of the surface to be measured, the distance between the actual point of impact and the ideal point of impact of the reflected beam on the detector surface can simply be used.

[0005] According to US 5471307 A, an illuminated flat sample surface provided with a regular light-dark pattern is arranged parallel to a surface to be measured which is directionally reflected, and the image of the sample surface reflected on the surface to be measured is recorded by a camera. From the distortions of the actually reflected image compared to the ideal reflected image - known by calculation - conclusions are drawn about deviations in the flatness in the surface areas of the surface to be measured that are involved in the distorted reflection. By repeating measurements under defined, changed conditions, further conclusions are drawn about the extent to which the deviations are transverse displacement and / or rotation of a surface area compared to the ideal position or alignment. EP 1429 113 A1 describes a very similar method to US 5471307 A.

[0006] According to EP 893685 A2, an ideally flat, directionally reflective surface to be measured is illuminated with divergent (i.e., non-collimated) light by several light sources, which are configured as point-like as possible and spaced at defined positions from the surface to be measured. A detector surface located in the immediate vicinity of the light sources determines the positions of the images of the individual light sources reflected on the measuring surface.If the arrangement is chosen so that the light rays which are reflected back from a light source onto the detector surface hit the surface to be measured almost exactly perpendicularly, the degree of offset of the point of impact of the reflected beam on the detector surface compared to the ideal point of impact is predominantly an indication of a deviation in the inclination of the measured surface at the respective reflection point, and not so much a direct indication of a linear displacement of the reflection point.

[0007] Based on EP 893685 A2, the object underlying the invention is to improve a method and a device for measuring the flatness of surfaces such as the surface of flat glass panes, with strict consideration of economic efficiency, in such a way that both the angle of inclination and the absolute position of points on the surface to be measured are detected as directly as possible, and that the flatness measurement becomes less dependent on the reflection properties of the surface being measured.

[0008] To solve the problem, the features adopted from EP 893685 A2 are that a device is used which is designed to illuminate the surface to be measured with divergently propagating light by means of several light sources which are spaced at defined positions from the surface to be measured, and to capture mirror images of the light sources, which are based on directed reflection of the light emitted by the light sources in a scattered manner on the surface to be measured, by means of a camera, so that the value pair of position and orientation of the associated reflection surface area on the surface to be measured can be calculated back from the position of the images of the mirror images of the light sources.

[0009] As an improvement according to the invention, it is proposed to provide a plurality of light sources on the said device, which emit collimated light along defined straight lines onto the surface to be measured, and to design the camera of the device to also record the light diffusely scattered at the respective point of impact of the collimated light rays, so that the position of the respective point of impact on the surface to be measured can be calculated from the image taken by the camera.The difference between the two measurements - once directed reflected light from a divergent light source, and once scattered reflected light from a collimated light source - results are achieved from which the actual shape of the measured surface can be determined much better and more directly than if only one of the two measuring principles were used with an increased number of measuring points. A further advantage is that the device can be used to measure both ideally directed reflecting surfaces and ideally diffusely reflecting surfaces, as well as all surfaces with any combination of both directed and diffuse reflecting.Thirdly, the device also provides information about the extent to which the area to be measured reflects in a directed and scattered manner, which also allows information about the area to be measured to be recorded by sensors that goes beyond the location and orientation of partial areas.

[0010] These advantages are achieved under very advantageous economic conditions, since many of the essential parts of the device - power supply, camera, data processing device, data transmission device, circuit board which carries the individual elements - can be used jointly for both measuring principles and therefore only have to be present once for each device.

[0011] The invention is explained in more detail with reference to drawings.

[0012] Fig. 1: shows a highly stylized side view of an exemplary device according to the invention during its intended measuring use with collimated light sources. Light rays are symbolized by dashed or dotted lines. Fig. 2: shows the arrangement of Fig. 1 in the same way as Fig.

[0013] 1 When used as intended with divergent light sources. Light rays are symbolized by dotted lines.

[0014] Fig . 3 : shows in the same way as Fig . 1 and the arrangement of

[0015] Fig . 1 and Fig . 2 during the intended measuring application with divergent light sources when used to measure the thickness of a flat glass .

[0016] The drawings show an exemplary device 1 according to the invention and an object which is arranged at a distance from the device 1 and whose surface side facing the device 1 is the surface 2 whose flatness is to be measured with the aid of the device 1.

[0017] The device 1 comprises a circuit board 3, at least three light sources 5 which emit light in a divergent manner, at least three light sources 4 which emit light in a collimated manner, as well as a camera 6, as well as further parts (not shown) which are required for the operation of the light sources 5, 4 and the camera 6, such as a power supply, electrical lines and a control unit at least for switching the light sources and the camera, and for controlling the transmission of the image data captured by the camera 6.

[0018] Ideally, the at least three light sources 5, which emit light with divergent propagation, are arranged at equal angular distances from one another along the circumference of an imaginary circle. Likewise, the at least three light sources 4, which emit light with collimated propagation, are ideally arranged at equal angular distances from one another along the circumference of an imaginary circle. Ideally, the circle on whose circumference the light sources 5 are located and the circle on whose circumference the light sources 4 are located are arranged coaxially to one another, and the common axis of the two circular surfaces ideally runs through the center of the detector surface of the camera 6, and the detector surface of the camera 6 is aligned normal to this axis. Ideally, the surface 2 to be measured is arranged such that, if it were perfectly flat, it would lie exactly parallel to the plane of the said circles and thus normal to the said axis.Particularly clearly understandable relationships and simple formulas for the mathematical conversion of the measurement results into a Cartesian coordinate system are obtained if, in addition, exactly four light sources 5, 4 are used.

[0019] In the state according to Fig. 1, only the light sources 4, which emit collimated light along a precisely defined straight line 7, are in operation. The light from these light sources 4 illuminates the surface 2 only at a single small partial area 8, and is generally partly absorbed there, partly reflected directively, and partly reflected scattered. The positions of the partial areas 8, onto which a collimated light source 4 shines, are ideally selected with respect to the positions of the respective illuminating light sources 4 and the camera 6 such that the light reflected away from the partial areas 8, which originally comes from one of the collimated light sources 4, is directed past the camera 6, and therefore cannot be captured by the camera 6.This means that light emitted by one of the collimated light sources 4 can only reach the camera 2 as light that is scattered and reflected from one of the partial surfaces 8 of the surface 2 to be measured. Since the straight line 7, along which the light from the light source 4 floods the surface 2, is arranged in a precisely defined manner in space, and since the camera can detect exactly the direction from which the said scattered light arrives at its detector surface, it is therefore possible to clearly calculate where in space the respective scattered reflecting partial surface 8 of the surface 2 to be measured is located. If the surface 2 to be measured reflects almost exclusively directionally, light that is emitted in a collimated manner by the light sources 4 can only reach the camera 6 with a comparatively very low intensity.In these cases, a high intensity of the radiation of the light sources 4 must be set and also a very high light sensitivity of the camera 6 must be set, at least to the spectral range in which the light sources 4 radiate.

[0020] In the state according to Fig. 2, only the light sources 5 which emit light divergently are in operation. The light from these light sources 5 illuminates the surface 2 largely evenly over a large area and is at least partially reflected by it in a directionally direction, i.e. mirrored. The camera 6 is therefore able to take an image of the arrangement of the light sources 5 which emit light divergently, in which image the mirror images of these light sources 5 appear as bright points. Assuming that the normal distance between the plane in which the light sources 5 lie and the plane in which the surface 2 essentially lies is known, value pairs can be calculated from the positions of the said bright points in the mirrored image, which pairs define a possible position in space and a possible angular orientation matching this position with regard to the partial surface 9 which reflects an image of a light source 5.

[0021] In practice, it is to be expected that light is reflected not only directedly but also scatteredly at surface 2, and that thus even scattered light which does not come from the partial area 9 of the surface 2 to be measured is detected by the camera 6. However, the scattered reflected light arrives at the detector surface of the camera 6 with a significantly lower intensity than the directed reflected light. The pixels in the image captured by the camera 6 caused by scattered reflected light can be easily detected and sorted out by data processing due to their low brightness.

[0022] Ideally, the positions of the partial surfaces 9 and the partial surfaces 8 are chosen so that they coincide, in the event that the surface 2 to be measured is ideally flat and ideally precisely positioned, so that a partial surface 9 from which directed reflected light from one of the divergently radiating light sources 5 is reflected onto the camera 6 is also exactly a partial surface 8 from which scattered reflected light from one of the collimated radiating light sources 4 is reflected onto the camera 6. The position of the partial surface 8, which can be calculated from the measuring principle according to Fig. 1, can then be sought in the series of pairs of values ​​calculated from the measuring principle of Fig. 2, and the inclination given in the relevant pair of values ​​can be clearly determined as the inclination of the relevant partial surface 9, 8.

[0023] Even if the two different measurements according to Fig. 1 and Fig. 2 cover the same partial areas 8, 9 on the surface 2 to be measured, they do not need to take place one after the other, but can also take place simultaneously. For this purpose, the light from the divergently radiating light sources 5 only needs to be distinguishable from that from the collimated radiating light sources during or after detection by the camera 6. This distinguishability can be achieved, for example, by different light colors or by different frequencies of intensity fluctuations in the radiation from the light sources.

[0024] If the divergently radiating light sources 5 and the collimated light sources 4 occupy different relevant partial areas 9, 8 on the surface 2 to be measured, from which reflected light is radiated to the camera 6, the spatial resolution of the camera 6 alone makes it possible to distinguish whether the light arriving at the camera 6 from one of the partial areas 9, 8 originally comes from one of the divergently radiating light sources 5 or from one of the collimated light sources 4.

[0025] When using the invention to monitor the flatness of flat glass in a corresponding processing plant, it is best to arrange a whole series of devices according to the invention next to one another so that together they cover at least one width of the flat glass that is usually to be monitored.

[0026] According to an advantageous further development of the invention illustrated by Fig. 3, the device 1 according to the invention, using the divergently radiating light sources 5, is also used to measure the thickness of flat glass. If the surface 2 to be measured according to Fig. 1 and Fig. 2 is the surface of a flat glass pane facing the device 1, then a reflection that can be detected by the camera 6 occurs not only on the surface 2 at the partial surface 9, but also on the underlying second glass surface at the partial surface 10. The camera image can then also be interpreted in such a way that the partial surfaces 8 and 10, or 9 and 10, do not appear as a point or small circle in the image recorded by the camera 6, but either as a short line, or as a narrow oval area, or as two points. The length of the short line or the ratio of length to width of the oval area orthe center-to-center distance between the two points is proportional to the glass thickness. The method works particularly well when divergent light sources 5 are used to measure the glass thickness and when the emitting surfaces of these light sources have a very small diameter. An SMD LED with an emitting surface diameter of approximately 0.3 mm is readily available and very suitable, for example. This allows the thickness of flat glass to be detected with an accuracy of tenths to hundredths of a millimeter using the method outlined in Fig. 3. When evaluating the camera image to determine the glass thickness, the well-known method of so-called sub-pixel interpretation (English: "sub-pixel resolution") should be applied.

[0027] The measurement of the glass thickness is also possible with the collimated light sources 4, whereby one challenge lies in capturing the small proportion of light which is scatteredly reflected on the surface of the glass pane facing away from the camera 6.

[0028] The achievable measurement accuracy improves the more acute (i.e. smaller) the angle between the light rays involved in the measurement and the surface 2 is.

[0029] A particularly valuable application of the device according to the invention using the described glass thickness measurement occurs when flat glass panes are to be stored mechanically in a vertical position in compartments that are separated from one another only by fork-like prongs, as closely as possible but still free from damage and without touching one another. In this case, it is necessary to know exactly the space required by the individual panes normal to their plane, whereby this space requirement can be calculated as the sum of the glass thickness and the deflection of the flat glass. Both the glass thickness and the deflection can be measured as described by the device according to the invention. For this application, a device can also be used which differs from the device shown in that it does not have collimated light sources 4, but only divergent light sources 5.

[0030] It is also useful to use the subpixel interpretation method to measure the exact position of the partial areas 8 and 9, or to determine the exact center of these areas - defined, for example, as their center of gravity or as their incircle center.

Claims

Patent claims 1. Device (1) for measuring the flatness of a surface (2) or the thickness of a flat glass pane, the device (1) having at least three light sources (5) which emit light divergently, and a camera (6) which is designed to capture the light which is reflected in a directed manner at the surface (2) and originally comes from the divergently emitting light sources (5), characterized in that the device (1) also has at least three light sources (4) which emit light in a collimated manner along a respective defined positioned straight line (7), the camera (6) also being designed to capture the light which is scatteredly reflected at the surface (2) and originally comes from the collimated light sources (4).

2. Device (1) according to claim 1, characterized in that the light sources (4) which emit light collimated and the light sources (5) which emit light divergently, as well as the camera (6), as well as a single power supply unit which supplies these parts together and a control unit which controls all the light sources (4, 5) and the camera (6) are arranged on a common circuit board (3).

3. Device according to claim 1 or 2, characterized in that the at least three light sources (5) which radiate light in a divergent manner are arranged along the circumference of an imaginary circle at a uniform angular distance from one another, and that the at least three light sources (4) which radiate light in a collimated manner are arranged along the circumference of an imaginary circle at a uniform angular distance from one another, and that the two circles- are arranged coaxially to each other, and that the common axis of the two circular surfaces passes through the center of the detector surface of the camera (6).

4. Device according to claim 3, characterized in that it has exactly four light sources (5) which emit light in a divergent manner and exactly four light sources (4) which emit light in a collimated manner.

5. Method for measuring the flatness of a surface (2) using a device according to one of claims 1 to 4, characterized in that a partial area (9) on the surface (2) from which directed reflected light from one of the divergently radiating light sources (5) is reflected onto the camera (6) is also a partial area (8) from which scattered reflected light from one of the collimated radiating light sources (4) is reflected onto the camera (6).

6. Method for measuring the flatness of a surface (2) using a device according to one of claims 1 to 4, characterized in that the light reflected away from a partial surface (8), which originally comes from one of the collimated light sources (4), is directed past the camera (6).

1. Method according to the combination of claims 5 and 6.

8. Method according to one of claims 5 to 7, characterized in that it is used to measure the flatness of a flat glass surface, wherein in addition to the detection of the reflections on the surface (2) facing the device (1) by the camera (6), reflections are also detected which are from the surface facing away from the device (1) of the flat glass, and the thickness of the flat glass is calculated from the optical information thus recorded.

9. Method according to one of claims 5 to 8, characterized in that it is used for measuring properties of flat glass, and that for this purpose several devices (1) are arranged next to one another in a row.

10. A method for measuring the thickness of a flat glass pane, characterized in that a device for measuring the flatness of a surface (2) is used for this purpose, wherein the device (1) has at least three light sources (5) which emit light divergently, and a camera (6) which is designed to capture the light originally coming from the divergently emitting light sources (5) which is reflected at the surface (2), wherein the surface of the flat glass pane facing the device is used as the surface (2), and wherein the surface of the flat glass pane facing away from the device is used as a second surface which reflects the light originating from the light sources (5) to the camera (6), wherein the camera (6) captures both an image of that partial surface (9) from which light is reflected onto the camera (6) from the side of the flat glass pane facing the camera (6),as well as an image of that partial area (10) from which light is reflected onto the camera (6) from the side of the flat glass pane facing away from the camera (6) is recorded, and the thickness of the flat glass pane is calculated from the distance between these two images.

Images