Method for optically inspecting an object, and corresponding inspection device

Abstract

The invention relates to a cost-effective method for optically inspecting an object and to a corresponding inspection device, wherein the inspection device comprises an illumination unit and a matrix camera, wherein the object and the inspection device carry out a relative movement with respect to one another at a specified movement speed in an advance direction, wherein the at least one illumination unit illuminates the object constantly with an illumination pattern, wherein the illumination pattern comprises N1 (N1 ≥ 2) light / dark stripes which are spaced apart from one another in the advance direction by an interval LAB and have a length LST and a multiplicity of light / dark structures arranged next to one another, wherein each light / dark structure forms a period B1 in the transverse direction, wherein two adjacent light / dark stripes in the advance direction each have a phase shift in the transverse direction by a width B1 / N1, wherein the at least one camera generates a multiplicity of image recordings of the illuminated region of the object in an image recording sequence, in such a way that each image recording of the image recording sequence simultaneously captures all N1 light / dark stripes of the illumination pattern over a given width in the transverse direction, and wherein the camera makes the data from the image recordings available to a computing unit at an appropriate interface, for the purpose of inspecting the object.

Description

[0001] Method for optical inspection of an object and corresponding inspection device

[0002] The invention relates to a method for the optical inspection of an object and a corresponding inspection device.

[0003] The optical inspection of an object that moves relative to an inspection device is a frequently and popular method in production because it allows the quality of the object to be assessed without contact. The object can in particular consist of a transparent material such as glass or structured glass, for example a glass pane or a glass ribbon, but can also have a non-transparent material, for example a painted or otherwise applied at least partially reflective surface. Of particular importance for the inspection can be damage to the surface, for example scratches, and / or, in the case of transparent material, inclusions inside the transparent material, for example air pockets or dust. Deposits on the surface (e.g. dust particles) also play a role. The inspection can also be carried out to determine the dimensions and / or optical parameters of the object.

[0004] The optical inspection of objects made of transparent or reflective materials, such as glass or structured glass, using a camera as the optical recording device, presents the problem that both bright-field and dark-field illumination are required to contrast the various types of defects, as different types of defects can often only be distinguished using one or the other. Absorbing defects, such as inclusions, can be detected using bright-field illumination, where the illumination device shines through the transparent material and the camera is positioned on the other side of the transparent material. Scattering or deflecting defects, such as scratches, are more likely to be detected using dark-field illumination.With dark-field illumination, an area adjacent to the camera's field of view is illuminated with light, ideally allowing the camera to see a dark field without any defects on the surface. However, if there are defects or deposits on the surface, these will deflect light into the camera's image, allowing the defects to be detected. In this case, it is also referred to as dark-field illumination.

[0005] For transparent materials, a transmitted-light arrangement is usually used, although sometimes, for reasons of accessibility, a reflected-light arrangement is also used. For non-transparent materials, a reflected-light arrangement must always be used. The invention relates to both a reflected-light and a transmitted-light arrangement.

[0006] From the document DE 10 2010 021 853 A1, a device and a method for the optical inspection of an object are known, in which a line scan camera and an illumination device are provided, which has a plurality of individually switchable illumination elements, for example light-emitting diodes (LEDs) or groups of jointly switchable light-emitting diodes (LEDs). Furthermore, a computing unit is provided and configured to control the individual illumination elements differently for the plurality of images in the one resolution range, i.e., to switch them on and off. This makes it possible to achieve various illumination configurations, so that, due to the flexibly switchable illumination device in this configuration, several images can be achieved in one resolution range with the respective desired, different illumination and recording conditions.In particular, the lighting elements of the lighting device are arranged in exactly one lighting line aligned transversely to the transport direction, wherein the computing unit is configured to control the lighting elements in the lighting line according to a predetermined, i.e. adjustable, lighting cycle with different lighting configurations. Preferably, the number of controls of the lighting cycle corresponds exactly to the number of images in the intended resolution range, i.e. the lighting cycle is selected such that the lighting cycle has been carried out or processed once when the object has been transported further in the transport direction by the desired resolution.For three, four, five or six images in the resolution range, the illumination line is therefore controlled three, four, five or six times in one illumination cycle and switched according to the desired illumination configuration. With the known method and device, individual illumination configurations are quickly switched back and forth, and the line scan camera is synchronized accordingly. The rapid switching of the illumination involves considerable effort in the control electronics used for the individual illumination elements, which is also cost-intensive. A method for the optical inspection of a surface is also known from the document DE 10 2020 109 945 A1, in which a temporally periodic pattern with different illumination patterns is generated on the surface during an image acquisition sequence and recorded using an inspection device.

[0007] Document US 2016 / 0048969 A1 discloses an image acquisition system for capturing at least one image of a sample surface using a projector and a camera to determine a height profile of the sample surface. The sample, the camera, and the projector move relative to each other. The projector projects a pattern onto the sample, which may have subpatterns in the form of four rectangular fields. The detection areas of the camera may correspond to the fields, and information about the height profile of the sample is obtained from the images captured by the camera. An inspection within the meaning of the present invention is not performed.

[0008] The task is therefore to create an inspection facility whose construction is more cost-effective and less complex. The task also consists in specifying a corresponding inspection procedure.

[0009] The above object is achieved by a method for optical inspection having the features of claim 1 and an inspection device having the features of claim 8.

[0010] In particular, the object is achieved by the method for optically inspecting an object by means of an inspection device, wherein the inspection device has at least one illumination unit and at least one matrix camera, wherein the object and the inspection device execute a relative movement with a predetermined movement speed in a feed direction to one another.The at least one lighting unit constantly illuminates the object with a respective lighting pattern, wherein the lighting pattern has at least N1 (N1 > 2) light-dark stripes spaced from one another in the feed direction by a distance length LAB, with a length LST in the feed direction and with a plurality of identical light-dark structures arranged directly next to one another in the transverse direction, wherein each light-dark structure forms a period B1 in the transverse direction and consists, for example, of a light field and a dark field arranged next to it in the transverse direction, wherein two light-dark stripes adjacent in the feed direction each have a phase shift in the transverse direction by a width B1 / N1. The object is constantly illuminated by the lighting unit with the lighting pattern.In addition, the at least one matrix camera generates a plurality of image recordings in an image recording sequence of the area of ​​the object respectively illuminated by the illumination unit such that with each image recording of the image recording sequence all N1 light-dark stripes of the illumination pattern are recorded simultaneously over a predetermined width in the transverse direction, wherein the image recording sequence of this matrix camera is synchronized with the movement speed in the feed direction with respect to the illumination pattern such that the relative movement of object and inspection device between two immediately successive image recordings in the feed direction is a maximum of L = LAB + LST, and wherein the matrix camera makes the data of the image recordings available to a computing unit at a corresponding interface so that the computing unit can carry out the inspection of the object on the basis of the data of the image recordings of the illumination pattern.

[0011] The at least one lighting unit constantly illuminates the object in a predetermined area, each with an illumination pattern, i.e. the respective lighting unit shines in the direction of the object with constant light intensity during the inspection and without switching off or switching over. The illumination pattern is generated either by the light-emitting part of the lighting unit itself or by shining a grating through which the illumination pattern is generated. The illumination is in the visible wavelength range. The illumination pattern is radiated onto the surface of the object using non-coherent light. The matrix camera takes a plurality of images of the illuminated area of ​​the object, i.e. it generates an image recording sequence of the images of the illumination pattern that the camera views in reflection (incident light arrangement) or transmission (transmitted light arrangement) in relation to the object.The digitized image data generated by the matrix camera is transmitted from the matrix camera to the processing unit (computer, processor, or similar), where it is made available via a corresponding interface on the matrix camera. The processing unit then performs the desired inspection of the object based on this data, i.e., it identifies defects, deposits in / on the object from the image data, and / or object parameters such as the refractive index, a height profile, or the object's dimensions. Defects can include, for example, inclusions in the object or on the surface and surface damage such as scratches.

[0012] The matrix camera is also called an area scan camera and records a larger surface area of ​​the object. It can be designed as a CCD camera, for example. The matrix camera records the light intensity of a large number of pixels from the illuminated area, which are arranged in rows and columns, i.e. in a matrix. For this purpose, the matrix camera has a light-sensitive element (e.g. a CCD sensor) for each pixel. The size of the area recorded by each light-sensitive element determines the resolution of the matrix camera. If, as described below, an object region is assigned to the area recorded by the matrix camera through spatial synchronization, each pixel of the matrix camera can be assigned a volume and / or surface element of the object. The matrix camera is also arranged such that it looks perpendicular to the object, so that it has a sharp view of the entire specified object region.Several matrix cameras can be arranged next to each other in the transverse direction in order to inspect a particularly wide object.

[0013] As explained above, the object and the inspection device move relative to each other at a predetermined speed in a feed direction. The transverse direction runs transversely, preferably perpendicular to the feed direction. For example, the object is moved past the inspection device on a corresponding belt or as web material at a predetermined speed, and the inspection device does not move. Alternatively, only the inspection device moves, or both the inspection device and the object move simultaneously. It is advantageous for the evaluation of the image data if the relative movement of the object and the inspection device represents a uniform movement, i.e., a movement at a constant speed. This simplifies the evaluation of the image data for the inspection of the object.

[0014] The illumination pattern generated by an illumination unit has N1 (N1 > 2) light-dark stripes spaced apart from one another in the feed direction by a distance LAB, wherein two adjacent light-dark stripes in the feed direction each have a phase shift in the transverse direction by a width B1 / N1. B1 is the period of the light-dark structures (i.e., the width of a single light-dark structure in the transverse direction) that form the light-dark stripe. In one embodiment, the period B1 of the light-dark stripes is greater than or equal to 100 pm (B1 > 100 pm), for example greater than or equal to 200 pm (B1 > 200 pm). In another example, the illumination pattern has N1 = 6 light-dark stripes and a period of B1 = 0.25 mm, so that two light-dark stripes arranged next to each other in the feed direction are shifted from each other in the transverse direction by a width of 0.25 mm / 6 = 41.7 pm (phase shift).Each light-dark stripe is composed of a plurality of identical light-dark structures arranged side by side in the transverse direction. The same light-dark structure (i.e. the same element, composed of a predetermined sequence of at least one light field and at least one dark field) is used for all light-dark stripes. Each light-dark stripe accordingly has a periodic intensity progression in the transverse direction with the period B1, with each period containing at least one light field (high light intensity) and one dark field (low light intensity). For example, the light-dark structure consists of a light field and a dark field arranged directly (immediately) next to it in the transverse direction.In the example, each light-dark stripe consists of a plurality of identical light fields and identical dark fields that alternate (light field - dark field - light field - dark field and so on). The progression of the light intensity in the transverse direction can, for example, have a sinusoidal or rectangular profile. In one embodiment, the width in the transverse direction and the length in the feed direction are the same for the light field and the dark field, but other designs can also be used. For example, the light field can be wider (in the transverse direction) than the dark field or vice versa, or the light field can be longer (in the feed direction) than the dark field or vice versa. All N1 light-dark stripes are of the same design (all light-dark stripes are composed of the same light-dark structures), but are offset in the transverse direction by the width B1 / N1 from the adjacent stripe.Each light-dark stripe has a total length LST in the feed direction. Two adjacent light-dark stripes are spaced from one another by a constantly light or a constantly dark stripe (spacer stripe) extending transversely across the entire width of the illumination pattern, wherein the distance from the end of the first light-dark stripe to the beginning of the adjacent second light-dark stripe (in the feed direction) is LAB. The length of the spacer stripe in the feed direction is LAB. The same applies analogously to the distance of the third light-dark stripe from the adjacent second light-dark stripe (in the feed direction), etc. In one embodiment, the length LST of a light-dark stripe is at least three times the length of the period B1.The spacer strips allow the light intensities of adjacent light and dark stripes to be separated during image capture, thus minimizing interference.

[0015] The matrix camera is aligned and configured in such a way that in a single image acquisition of the image acquisition sequence it simultaneously captures the light created by the constant illumination with the illumination pattern in relation to all N1 light-dark stripes over a predetermined width (for example the width of the object). During the subsequent image acquisition of the image acquisition sequence, due to the above-described synchronization of relative movement and creation of an image acquisition by the matrix camera, i.e. readout of the matrix camera, a relative movement of the length L between the inspection device and the object in the feed direction occurs, which has a maximum of L = LAB + LST. With a relative movement of the length L = LAB + LST, in a predetermined section of the object, the next image acquisition of all N1 light-dark stripes will contain the adjacent light-dark stripe, shifted transversely by B1 / N1, in this predetermined section of the object.This means that a specific area / section of the object is captured by at least N1 images taken by the matrix camera, and these at least N1 images are all used to inspect the object after the data from the at least N1 images, which were previously digitized, have been forwarded to the computing unit. The matrix camera can further be configured such that at least one light-sensitive element of the matrix camera captures the light of a field of the illumination pattern (i.e., either a bright field or a dark field of the light-dark structure). In one embodiment, two or more light-sensitive elements of the matrix camera can capture the light of a field. With regard to the feed direction, the matrix camera can also be configured such that at least one light-sensitive element, for example two or more light-sensitive elements, capture the light of a field along a field.

[0016] The method according to the invention achieves, on the one hand, a long exposure time for each location on the object, since at least N1 images (images of N1 light-dark stripes from each location) are created for each location on the object. This is explained in more detail below. This results in a better signal-to-noise ratio, or the illumination intensity of the illumination unit can be reduced compared to the conventionally created single image, while maintaining the same signal-to-noise ratio. This results in lower energy consumption, which is accompanied by lower cooling requirements for the illumination unit. On the other hand, the use of constant illumination of the object by the illumination unit, which does not require switching operations during inspection of the object, requires less effort and equipment.Control electronics for quickly switching individual elements (LEDs) of the lighting unit during the inspection are not required. This makes the inspection less complex and more cost-effective.

[0017] In one embodiment, each real location of the object is assigned identical image recording areas of the matrix camera for each of the N1 light-dark stripes. This means that the computing unit knows exactly from which real location of the object (object point) the light captured by each light-sensitive element of the matrix camera originates. This results in spatial synchronization of the illumination unit and the object with respect to image recordings generated by the matrix camera at different times. The assignment of the real location and the pixels of the N1 different image recordings can be determined by the computing unit, for example, through a transformation and the known uniform relative movement of the object and inspection device at a relative speed (e.g., by means of the specified or measured feed speed of the object), and can be assigned to the pixels of the respective image recording of the matrix camera.In other words, in this exemplary embodiment, the recorded intensity values ​​of each pixel or pixel group of the matrix camera from different image recordings are assigned to one another and to a real location of the object (synchronized). The assignment is performed in such a way that each real object location is assigned N1 light intensity values ​​that were recorded at different times with respect to N1 different light-dark stripes. Furthermore, the assignment is characterized by the fact that the specific real object location was located in the relevant area of ​​the respective light-dark stripe at the time the pixel or pixel group assigned to it was recorded. This means that light intensity values ​​of pixels recorded at different times are assigned both to one another and to a real location of the object. Each light-dark stripe represents a specific type of illumination.

[0018] Due to the setup of the lighting unit described below and the design of the matrix camera, in one embodiment an image of the object can be taken simultaneously with brightfield illumination and darkfield illumination, which are recorded for the same object location, but at different times. The pixels of the matrix camera images that capture the light generated by the illumination with the bright field of the light-dark structure can be used for the brightfield image, and the pixels of the matrix camera images that capture the light generated in the dark field of the light-dark structure can be used for the darkfield image. In this case, the images generated by the recording device of the recorded area are separated if necessary and combined to form the desired images, e.g. a brightfield image, a darkfield image, a reflection image, or a transmission image.For example, a bright-field transmission image, a dark-field transmission image, a reflection image, and data for two different deflections (see below) can be generated. Within the scope of the object inspection, this exemplary embodiment can therefore perform defect detection based on the data from the image recordings originating from the dark fields of the light-dark structures (dark-field image) and / or defect detection based on the data from the image recordings originating from the bright fields of the light-dark structures (bright-field image). With respect to the above-mentioned transmission image and the reflection image, at least two illumination units, each with an associated illumination pattern, can be provided in the inspection device, with a first illumination unit being used for the transmission image and a second illumination unit for the reflection image.The first illumination unit is located on the side of the object opposite the matrix camera, i.e., the object is positioned between the first illumination unit and the matrix camera. The light emitted by the first illumination unit passes through the (at least partially transparent) object and is captured after passing through the matrix camera, with the matrix camera viewing the object along the axis of the first illumination unit. In contrast, the second illumination unit is located on the same side of the object as the matrix camera. The light from the second illumination unit reaches the matrix camera after being reflected off the surface of the object.In one embodiment, the first and second illumination units are arranged with respect to the matrix camera such that the image recordings of the transmission image and the reflection image are arranged laterally next to one another on the matrix camera, so that they can be distinguished by their location on the screen of the matrix camera. The matrix camera records reflection image sequences and transmission image sequences simultaneously, as described above, synchronized with the relative movement of the object and the inspection device, and is focused on the object in both cases. Consequently, in this embodiment, the matrix camera captures an illumination pattern reflected on the surface of the object and / or an illumination pattern transmitted through the object with the image recordings in the illuminated area.Alternatively or additionally, an off-axis transmission image sequence can be captured by the matrix camera in an analog manner when the matrix camera looks past the axis of the first illumination unit at a small angle. Accordingly, in one exemplary embodiment, the computing unit is configured to compile at least two images of the object from the multiple recordings taken in the illuminated area of ​​the object and to evaluate them individually or in combination for the presence of defects or with regard to optical parameters of the object. In this case, the at least two images are compiled from the corresponding recordings of a part of the illuminated area or the entire area of ​​the object. For example, a bright field image and a dark field image of the object or a reflection image and a transmission image of the object are compiled.These images can then be evaluated individually or in combination by the computing unit to determine the position and extent of the defect and / or to determine the type of defect. For example, the combined evaluation of the reflection image and the transmission image can be used to differentiate between open and closed bubbles in a glass. In particular, with bright-field illumination, the matrix camera looks at an illuminated field; in the transmitted-light arrangement, it looks through the transparent material and directly into the illumination unit. In the bright field, defects are imaged as contrasting areas. With dark-field illumination, the matrix camera, in a defect-free case, looks into the darkness next to the illumination and therefore sees nothing. If a defect / deposit occurs, defects / deposits are detected by the computing unit based on light reflections in the dark environment.

[0019] The areas of the object covered by the images can be completely separate from each other or overlapping.

[0020] The inspection of the object performed by the processing unit based on the image data can be implemented as a computer-implemented process, i.e., a process performed using a processing unit (computer). For example, based on the image data, defect detection methods are used to identify defects as local deviations from the image signal that would be generated by a defect-free object.

[0021] In one embodiment, the illumination pattern additionally has at least N2 (N2 > 2) transverse line structures lying next to one another in the feed direction, wherein each transverse line structure comprises a bright line and a dark line which extend over the entire width of the illumination pattern, wherein the length (period) ML of each transverse line structure in the feed direction has a maximum value LMAX which results from the product of the relative speed between the inspection device and the object (e.g. feed speed of the object in the case of a stationary inspection device) and 4 times the time period for the generation of an image recording (= reciprocal value of the recording frequency) by the matrix camera.The resolution of the matrix camera is further configured such that it captures the light of the area of ​​the object illuminated by the N2 transverse line structures, specifically such that a measuring line on the dark line, a measuring line at a first transition between the dark and the light line, a measuring line on the light line, and a measuring line at a second transition between the light and the next dark line are captured. The course of the light intensity in the feed direction of a transverse line structure can, for example, have a sine or rectangular profile. Local synchronization is also carried out for these four measuring lines, i.e. the computing device knows the location on the object at which each pixel of the respective measuring line was recorded. The images of the four measuring lines, taken at four different times, are each assigned to the corresponding location on the object.The assignment may require a transformation different from the transformation of the light-dark fringes described above. The measurement lines provide measurement information for each location that is 90° phase-shifted with respect to the information from the light-dark fringes recording. From this, the phase or deflection angle can be determined using the well-known dynamic moire algorithm. From this, four different independent optical values ​​can be calculated, with the transverse direction representing the x-direction and the feed direction the y-direction: dAx / dx, dAy / dy, dAx / dy, dAy / dx, where dAx, dAy represent the deflection (distortion) in the x- or y-direction, respectively, and dx, dy is the corresponding path difference. From this, the refractive power in the x-direction and / or y-direction can be determined in a known manner.Accordingly, in the exemplary embodiment, the phase and deflection angles in the transverse direction and / or in the feed direction can be determined from a plurality of locations on the object surface based on the N1 consecutively acquired images of the light-dark fringes and the at least (N2-1)x4 acquired images of the transverse line structure, and from this, for example, the refractive power in the transverse direction and / or in the feed direction can be determined. This means that if N2 = 2, at least 4 images are required.

[0022] This results, for example, from the determination of four intensity values ​​in relation to an illumination period a deflection angle <t>« arctan <t>= (I1 - I3) / (I2 - I4), where I1, I2, I3, I4 represent intensity values ​​determined by the matrix camera at the respective location in this order over one illumination period. From this, the refractive power can be determined, which is proportional to the phase shift A <t> = <t> 2 - <t>1 is.

[0023] In one embodiment, the period length ML of the transverse line structure in the feed direction is equal to the period length B1 of the light-dark stripe in the transverse direction. This ensures that the deflection in the feed and transverse directions can be measured with the same accuracy.

[0024] The lighting unit can have a plurality of individually switchable lighting elements, for example light-emitting diodes (LEDs) or groups of jointly switchable light-emitting diodes (LEDs). In this case, the lighting pattern is generated by the entirety of the lighting elements themselves. An individually switchable lighting element can therefore be made up of a single lighting fixture, for example an LED, or of a plurality of lighting fixtures combined in a group and switchable together, for example a group of several LEDs. With this design of the lighting unit, the computing unit is preferably configured to control the individual lighting elements differently, i.e. to switch them on and off, whereby the control does not occur during the inspection of the object, but rather before or after the inspection. The bright fields / lines are realized by switched-on LEDs and the dark fields / lines by switched-off LEDs.This allows for flexible illumination configurations, for example from the same side of the object, from different sides of the object, at different angles to the object and / or with light of different wavelengths.

[0025] Alternatively or additionally, the lighting unit can include a light source (e.g., a fluorescent tube) that completely illuminates the area to be illuminated. A grid is arranged between the light source and the object, creating the desired lighting pattern. The bright fields / lines are created by corresponding slits in the grid, and the dark fields / lines are created by corresponding bars.

[0026] The illumination unit can provide different illumination patterns. This allows for the targeted realization of additional illumination channels on the matrix camera, particularly as bright-field or dark-field illumination, as illumination from different sides of the object, and / or as illumination with light of different wavelengths. However, since dark-field illumination does not occur immediately at the transition from an illuminated to an unilluminated illumination element, it may be advantageous according to the invention to combine a total of several switched-off, i.e., dark, illumination elements next to one another.

[0027] The geometry of the optical imaging, together with the size of the light-sensitive elements (pixels, image points) of the matrix camera, determines the length of the path imaged onto one camera line. Since most matrix cameras have square pixels and the optics are radially symmetrical to the optical axis, the resolution along the feed direction is usually equal to the resolution across the path. It is also possible to achieve other ratios by selecting appropriately asymmetrical optics or using non-square pixels. However, this is expensive due to the special components required and therefore rarely used. Such an adaptation is only possible within narrow limits.

[0028] The resolution in the direction of the object's feed is determined by the speed at which the individual rows and columns of the matrix camera are recorded. In principle, the higher the readout speed, the finer the resolution.

[0029] The inspection device according to the invention can be used both for transparent materials, in particular a structural glass strip, and for non-transparent materials such as painted body surfaces and can be implemented both in a front-light illumination arrangement and in a transmitted-light illumination arrangement.

[0030] In order to adapt the light to specific inspection purposes, an optical light-shaping unit can be arranged between the illumination unit and the object and / or between the object and the matrix camera. The light-shaping unit is constructed in particular from lenses, diffusers and / or microlenses. Such light-shaping units are particularly suitable for inspecting an object made of structured glass, in particular a structured glass strip, in the transmitted-light arrangement. With bright-field illumination for structured glass, the light can shine as homogeneously as possible onto the structured glass from the front, back, right and left, i.e. from all directions, so that the existing structural pattern can be seen as faintly as possible and the defects being sought can be seen with high contrast in the transmitted-light arrangement.On the other hand, dark field illumination requires more directed illumination perpendicular to the transport direction so that the boundaries between the light and dark LED groups are as sharp as possible.

[0031] A particularly advantageous light-shaping unit for this purpose is a cylindrical Fresnel lens arranged between the illumination unit and the object, preferably in combination with a diffuser arranged between the Fresnel lens and the illumination unit. The cylindrical Fresnel lens distributes the light, particularly in the transport direction, and then focuses the illumination unit onto the object to be inspected. The cylindrical Fresnel lens thus projects the illumination unit (e.g., LEDs) into a reduced size image onto the material of the object, particularly the structured glass. The diffuser is useful for somewhat homogenizing the intensity in the transverse direction. The diffuser should preferably have weakly scattering optical properties, since otherwise the boundaries between the individual light and dark fields or lines become blurred, limiting the quality of dark-field detection.

[0032] In one embodiment, the inspection device comprises a plurality of independent illumination units arranged on different sides of the object and / or at different angles to the object. Using illumination units arranged at different angles to the object, for example, textures on a surface can be suppressed to distinguish them from topological defects. Alternatively or additionally, the illumination units can be arranged such that the respective illumination patterns are generated transversely next to one another on the object in order to inspect very wide objects.

[0033] In a further embodiment, the illumination unit can alternatively or additionally comprise multiple illumination elements that emit light in different wavelength ranges. By using light of different wavelengths, the light captured by the matrix camera, reflected by the object, or transmitted by the object can be precisely differentiated as to which illumination element it originates from. This requires that the matrix camera be able to distinguish the incoming light from the different wavelength ranges.

[0034] The above object is achieved in particular by an inspection device for the optical inspection of an object with a lighting unit and at least one matrix camera, wherein the inspection device is configured such that the object and the inspection device perform a relative movement with a predetermined movement speed in a feed direction to one another, such that the lighting unit constantly illuminates the object with an illumination pattern, wherein the illumination pattern has at least N1 (N1 > 2) light-dark stripes spaced from one another in the feed direction by a distance length LAB, with a length LST in the feed direction and with a plurality of light-dark structures arranged next to one another transversely, wherein each light-dark structure forms a period B1 in the transverse direction and consists, for example, of a light field and a dark field arranged next to it in the transverse direction,wherein two adjacent light-dark stripes in the feed direction each have a phase shift in the transverse direction by a width B1 / N1, that the at least one matrix camera generates a plurality of image recordings in an image recording sequence of the area of ​​the object respectively illuminated by the illumination unit such that with each image recording of the image recording sequence all N1 light-dark stripes of the illumination pattern are simultaneously recorded over a predetermined width in the transverse direction, that the image recording sequence of this matrix camera is synchronized with the movement speed in the feed direction in relation to the illumination pattern such that the relative movement of the object and the inspection device between two immediately successive image recordings in the feed direction is a maximum of L = LAB + LST,and that the matrix camera provides the image data at an interface of the matrix camera to a computing unit connected to the matrix camera in such a way that the object can be inspected by the computing unit based on the image data of the illumination pattern. The advantages of this inspection device, which together with the computing unit forms an inspection system, have already been explained above in connection with the method according to the invention. Reference is made to this, also with regard to the exemplary embodiments of the inspection device below.

[0035] In one embodiment, the inspection device is additionally configured such that, for each of the N1 light-dark stripes, identical image recording areas of the matrix camera are assigned to each real location of the object. In particular, each pixel of each image recording is assigned to a volume element and / or surface element of the object. In one embodiment, the inspection device is additionally configured such that, as part of the object inspection, defect detection is performed based on the data from the image recordings that originate from the dark fields of the light-dark structures and / or that originate from the light fields of the light-dark structures.

[0036] In one embodiment, the inspection device is additionally configured such that the illumination pattern additionally comprises a section with at least N2 (N2 > 2) transverse line structures located directly adjacent to one another in the feed direction, each transverse line structure comprising a light line and a dark line extending across the entire width of the illumination pattern, the length of each transverse line structure being a maximum of L / N2 in the feed direction. The section with the transverse line structures is arranged at a distance from the N1 light-dark stripes in the feed direction.

[0037] In one embodiment, the inspection device is additionally configured such that the computing unit determines the phase and deflection angle in the transverse direction and / or in the feed direction from a plurality of locations on the object surface on the basis of the N1 consecutively created image recordings of the light-dark stripes and the at least (N2-1)x4 image recordings of the transverse line structure, and from this, for example, the refractive power in the transverse direction and / or in the feed direction is determined.

[0038] In one embodiment, the inspection device is additionally configured such that the matrix camera uses the image recordings in the illuminated area to capture the illumination pattern reflected on the surface of the object and / or the illumination pattern transmitted through the object. In one embodiment, the inspection device is additionally configured such that at least two of the multiple image recordings generated in the illuminated area are combined and evaluated individually or in combination with regard to the presence of defects.

[0039] In one embodiment, a lighting unit with a plurality of individually switchable lighting elements arranged in a matrix and / or with a lamp and a grid arranged between the lamp and the object, which generates the light-dark stripes and optionally the transverse line structures, is provided in the inspection device for generating the lighting pattern.

[0040] The invention is used in particular for the inspection of an object in the form of a flat glass strip or structural glass.

[0041] Further advantages, features, and possible applications of the invention are described below with reference to exemplary embodiments and the figures. All described and / or illustrated features form the subject matter of the present invention, regardless of their summary in the claims and their references.

[0042] They show schematically:

[0043] Fig. 1 shows an embodiment of an inspection device according to the invention with the object in a side view,

[0044] Fig. 2 shows a first embodiment of a grid of the illumination unit of the inspection device according to Fig. 1 in a view from above, Fig. 3 shows a section of an illumination pattern generated by the grid according to Fig. 2 in an enlarged view in a view from above,

[0045] Fig. 4 shows a second embodiment of a grid of the lighting unit of the inspection device according to Fig. 1 in a view from above,

[0046] Fig. 5 shows a section of an illumination pattern generated by the grating according to Fig. 4 in an enlarged view from above and

[0047] Fig. 6 the local synchronization of the image recordings.

[0048] Fig. 1 shows an embodiment of an inspection device which inspects an object for possible defects and deposits on the surface, for example a flat glass strip 1. The flat glass strip 1 is illustrated by an arrow, as it moves past the stationary inspection device at a predetermined, constant speed in the feed direction (see arrow 11). The inspection device has a first lighting unit 31, a second lighting unit 32 and a matrix camera 5, wherein the first lighting unit 31 is arranged on the same side of the flat glass strip 1 as the matrix camera 5 and the second lighting unit 32 is arranged on the side opposite the matrix camera 5.

[0049] Each lighting unit 31, 32 has, for example, an LED matrix or a phosphor body as a light source and the grid 33 shown in Figs. 2 and 3, by means of which a lighting pattern is generated on the surface of the flat glass strip (for the first lighting unit 31) and in / on the opposite surface of the flat glass strip (for the second lighting unit 32). The matrix camera 5 observes the object and thereby the respective lighting pattern of the two lighting units 31, 32. The flat glass strip 1 is illuminated during the inspection with a constant intensity and without interruptions, i.e. without switching on or off. The line of sight from the matrix camera 5 to the first lighting unit 31 is illustrated by the arrow 51, while the line of sight from the matrix camera 5 to the second lighting unit 32 is illustrated by the arrow 52.The lines of sight (arrows 51, 52) make it clear that the matrix camera 5 observes the first illumination unit 31 in reflection and the second illumination unit 32 in transmission with respect to the flat glass strip 1. The matrix camera 5, for example a CCD camera, has a screen 53 which contains the light-sensitive elements (e.g. CCD sensors) which record the intensity of the light arriving there in a spatially resolved manner. The light-sensitive elements are arranged in a matrix. The matrix camera 5 looks perpendicularly onto the surface of the flat glass strip 1, so that in the image recording a predetermined width of the flat glass strip (i.e. perpendicular to the feed direction) of the surface of the flat glass strip 1 is sharply recorded. Fig. 1 also shows that the light from the first illumination unit 31 strikes the screen 53 in a lateral direction next to the light from the second illumination unit 32.This allows the reflection signal and the transmission signal to be recorded separately, with the CCD camera being able to read both channels simultaneously.

[0050] The first embodiment of a grating 33 shown in Fig. 2 generates a corresponding illumination pattern 43 in the region of the flat glass strip 1. The grating 33 has four slit strips 33a, 33b, 33c, 33d, which are arranged next to one another in the feed direction (see arrow 11) and generate the corresponding number (N1 = 4) of light-dark strips 43a, 43b, 43c, 43d with light and dark fields. Adjacent slit strips 33a, 33b, 33c, 33d are separated from one another by webs 33p, 33q, 33r extending across the entire width of the grating 33. In this case, a dark field is arranged next to a light field, and the light and dark fields are of equal size. A light and a dark field together form a light-dark structure, with a plurality of adjacent light-dark structures in the transverse direction forming a light-dark stripe 43a, 43b, 43c, 43d. Adjacent light-dark stripes, for example the light-dark stripes 43a and 43b or43b and 43c are each shifted by B1 / 4 in the transverse direction (see arrow 12, B1 is the period of each of the light-dark stripes). The illumination pattern observed in reflection or transmission further has a spacing length LAB between adjacent light-dark stripes 43a, 43b, 43c, 43d, which is, for example, 1.5 mm, while the length of the light-dark stripes 43a, 43b, 43c, 43d LST is 3.5 mm. At this spacing length LAB, the illumination pattern each has a constant dark stripe 43p, 43q, 43r (spacing stripe), i.e., each spacing stripe 43p, 43q, 43r has a length in the feed direction (arrow 11) of LAB (e.g., LAB = 1.5 mm). The width of each spacer strip 43p, 43q, 43r in the transverse direction (arrow 12) corresponds to the width of the light-dark stripes 43a, 43b, 43c, 43d or the width of the pattern 43. Alternatively, the spacer strip could also be designed as a constantly bright stripe.

[0051] The flat glass strip 1 is continuously illuminated by the two illumination units 31, 32 with the illumination pattern shown in part in Fig. 3. The matrix camera 5 generates images of both illumination patterns at predetermined time intervals t1, t2, t3, t4, specifically across the entire width in the transverse direction (arrow 12) of the respective illumination pattern (see also Fig. 6, which is intended to show the images of a defect 60 in the flat glass strip 1, wherein, for better illustration, each image is shifted to the right compared to the previous image). The matrix camera also captures all four light-dark stripes 43a, 43b, 43c, 43d of an illumination pattern simultaneously with one image. The arrangement on the screen 53 of the matrix camera 5 allows the images to be differentiated. The speed of movement of the flat glass strip 1 in the feed direction (arrow 11 ) is known or is continuously measured.The feed rate is used to assign four image recording areas of the matrix camera 5 to a location x1, x2, x3 or x4 of the flat glass strip 1 in such a way that a time interval (e.g. t2-t1) is maintained between each two consecutively created images, which corresponds to the path L = LAB + LST of the flat glass in the feed direction during this time. With such local synchronization, four images are assigned to each location on the flat glass strip 1, namely one image recorded at the times t1, t2, t3, t4 of each of the four light-dark strips 43a, 43b, 43c, 43d. The assignment is illustrated in Fig. 6 by the dashed arrow.These images, which contain an intensity value for each light-sensitive element (pixel) of the screen 53 of the matrix camera, are transmitted in digitized form via an interface 54 of the matrix camera to a computing unit 7 connected to the matrix camera via a data line, which inspects the flat glass strip 1 based on the images of the illumination patterns and the methods described above. The data from the images are correlated with location information so that each intensity value (pixel) of an image can be assigned a volume element / surface element of the flat glass strip 1. Each volume element forms a surface element on the surface of the flat glass strip, which reflects the light emitted by the first illumination unit and is observed by the matrix camera 5 via line of sight 51.

[0052] Fig. 4 shows a second embodiment of a grid for inspection by means of an inspection device shown in Fig. 1, which has the grid 33 shown in Fig. 2 and, on one side of the grid in the feed direction, three recesses running transversely and parallel to one another, which produce corresponding bright lines a, c, e (see Fig. 5) of the illumination pattern. Between the bright lines a, c, e of the illumination pattern section 44, dark lines b, d running transversely and parallel to one another are provided. Overall, the bright and dark lines a, b, c, d, e form two transverse line structures, each consisting of a bright line and a dark line, as well as an additional bright line. Further dark lines delimit the illumination pattern section 44 in the feed direction. The matrix camera 5 is designed such that it can capture four of the four lines shown in Fig.5 simultaneously records the measuring lines ML illustrated by dashed lines, with one measuring line extending on the bright transverse line of the illumination pattern, two measuring lines between a bright transverse line and a black transverse line (with different sequences of the bright and dark lines), and one measuring line on the black transverse line. As shown in Fig. 4 using the grid 33, 34 and Fig. 3, the length that the flat glass strip 1 moves between two image recordings corresponds approximately to the distance between two adjacent of the four measuring lines ML in the feed direction, so that during the next image recording, a different location on the flat glass strip is illuminated by a measuring line of the section 44 of the entire illumination pattern 43, 44 and recorded by the matrix camera 5.Also with regard to the illumination pattern section 44, the computing unit 7 can assign a volume and area element of the flat glass strip 1 to each intensity value of each image recording, analogously to the above procedure.

[0053] This procedure has the advantage that defect detection during inspection is based on a large number of image data generated from the locations on the flat glass strip 1 that are illuminated by an illumination pattern 43, 44 that alternates between light and dark areas in the transverse direction and by an illumination pattern 43, 44 that alternates between light and dark areas in the feed direction. This image data can also be forwarded by the matrix camera to the computing unit 7, where it can be used to inspect the flat glass strip 1. In particular, the data generated as described above from the recordings of the recording sequence with a modulation of the light source in the transverse direction and in the feed direction can determine the longitudinal and transverse deflection per line element in the longitudinal and transverse directions and thus detect linear refractive power defects in the longitudinal and transverse directions.

[0054] The grating with structures 33, 34 can, for example, have a dimension Q1 of 250 mm in the transverse direction and a length LG of 40 mm in the feed direction. These parameters are shown in Fig. 4. For the light-dark stripes, the vertical slits are approximately 1 mm wide, and the webs running between them are also each 1 mm wide (each in the transverse direction).

[0055] The above method and inspection system can achieve a good signal-to-noise ratio at comparatively low illumination intensity, as each location on the flat glass strip receives four or five times the exposure time (depending on the grid used for illumination) compared to a single image acquisition as with conventional methods. This can reduce energy consumption. Furthermore, the lighting unit does not need to be switched on during inspection. This makes the inspection system more cost-effective and less complex.

[0056] The method can also be used very variably, as it can generate dark-field images of the flat glass strip 1 from the image data of the dark fields or lines of the illumination pattern and generate bright-field images from the image data of the bright fields or lines of the illumination pattern and use them for corresponding defect detection. Furthermore, the method can operate in transmission or reflection. It is also possible to provide an additional illumination unit that generates an off-axis dark field. This illumination unit can also be illuminated analogously to the illumination units 31, 32 described above, with an analogous grating 33, 34 and a constantly glowing light source, so that the corresponding illumination pattern 43, 44 can be realized in the area of ​​the flat glass strip 1 and observed by the matrix camera 5.< / t> < / t> < / t> < / t> < / t>

Claims

Patent claims 1. A method for the optical inspection of an object (1) by means of an inspection device, wherein the inspection device has at least one illumination unit (31, 32) and at least one matrix camera (5), wherein the object and the inspection device perform a relative movement with a predetermined movement speed in a feed direction, wherein the at least one illumination unit constantly illuminates the object with a respective illumination pattern (43, 44), wherein the at least one illumination unit is configured such that the illumination pattern N1 (N1 > 2) has light-dark stripes (43a, 43b, 43c, 43d) spaced from one another in the feed direction by a distance length LAB, with a length LST in the feed direction and with a plurality of identical light-dark structures arranged next to one another in the transverse direction,wherein each light-dark structure forms a period B1 in the transverse direction and consists, for example, of a bright field and a dark field arranged next to it in the transverse direction, wherein two adjacent light-dark stripes in the feed direction each have a phase shift in the transverse direction by a width B1 / N1, wherein the at least one matrix camera generates a plurality of image recordings in an image recording sequence of the area of ​​the object respectively illuminated by the illumination unit such that with each image recording of the image recording sequence, all N1 light-dark stripes of the illumination pattern are simultaneously captured over a predetermined width in the transverse direction, wherein the image recording sequence of this matrix camera is synchronized with the movement speed in the feed direction with respect to the illumination pattern,that the relative movement of object and inspection device between two immediately consecutive image recordings in the feed direction is a maximum of L = LAB + LST, and wherein the matrix camera stores the data of the image recording, men of a computing unit (7) at a corresponding interface (54) so ​​that the computing unit can carry out the inspection of the object on the basis of the data of the image recordings of the illumination pattern.

2. Method according to claim 1, characterized in that identical image recording areas of the matrix camera are assigned to each real location of the object for each of the N1 light-dark stripes.

3. Method according to one of the preceding claims, characterized in that, within the scope of the object inspection, error detection is carried out on the basis of the data of the image recordings which originate from the dark fields of the light-dark structures and / or error detection is carried out on the basis of the data of the image recordings which originate from the light fields of the light-dark structures.

4. Method according to one of the preceding claims, characterized in that the illumination pattern additionally has a section (44) with at least N2 (N2 > 2) transverse line structures (a, b, c, d, e) lying next to one another in the feed direction, wherein each transverse line structure comprises a light line and a dark line which extend over the entire width of the illumination pattern, wherein the length ML of each transverse line structure in the feed direction is a maximum of LMAX.

5. Method according to claim 4, characterized in that from a plurality of locations on the object surface on the basis of the N1 consecutively created image recordings of the light-dark stripes and the at least (N2-1)x4 created image recordings of the transverse line structure, the phase and deflection angle in the transverse direction and / or in the feed direction is determined and from this, for example, the refractive power in the transverse direction and / or in the feed direction is determined.

6. Method according to one of the preceding claims, characterized in that the matrix camera captures with the image recordings in the illuminated area an illumination pattern reflected on the surface of the object and / or an illumination pattern transmitted through the object.

7. Method according to one of the preceding claims, characterized in that at least two of the multiple image recordings generated in the illuminated area are combined and evaluated individually or in combination with regard to the presence of defects.

8. Inspection device for the optical inspection of an object (1) with an illumination unit (31, 32) and at least one matrix camera (5), wherein the inspection device is configured such that the object and the inspection device perform a relative movement with a predetermined movement speed in a feed direction, such that the illumination unit constantly illuminates the object with an illumination pattern (43, 44), wherein the illumination pattern has at least N1 (N1 > 2) light-dark stripes (43a, 43b, 43c, 43d) spaced from one another in the feed direction by a distance length LAB, with a length LST in the feed direction and with a plurality of identical light-dark structures arranged next to one another transversely, wherein each light-dark structure forms a period B1 in the transverse direction and consists, for example, of a light field and a dark field arranged next to it in the transverse direction,wherein two adjacent light-dark stripes in the feed direction each have a phase shift in the transverse direction by a width B1 / N1, that the at least one matrix camera takes a plurality of images in an image recording sequence of the area illuminated by the illumination unit in each case, th area of ​​the object in such a way that with each image recording of the image recording sequence all N1 light-dark stripes of the illumination pattern are recorded simultaneously over a predetermined width in the transverse direction, that the image recording sequence of this matrix camera is synchronized with the movement speed in the feed direction in relation to the illumination pattern in such a way that the relative movement of object and inspection device between two immediately successive image recordings in the feed direction is at most L = LAB + LST, and that the matrix camera makes the data of the image recordings available at an interface (54) of the matrix camera to a computing unit (7) connected to the matrix camera in such a way that the inspection of the object can be carried out by means of the computing unit on the basis of the data of the image recordings of the illumination pattern.

9. Inspection device according to claim 8, which is additionally configured such that identical image recording areas of the matrix camera are assigned to each real location of the object for each of the N1 light-dark stripes.

10. Inspection device according to one of claims 8 to 9, which is additionally set up such that, within the scope of the object inspection, the computing unit carries out error detection on the basis of the data of the image recordings which originate from the dark fields of the light-dark structures and / or which originate from the light fields of the light-dark structures. 11 . Inspection device according to one of claims 8 to 10, which is additionally arranged such that the illumination pattern additionally has a section (44) with at least N2 (N2 > 2) transverse line structures (a, b, c, d, e) lying next to one another in the feed direction, each transverse line structure comprising a light line and a dark line which extends over the entire width of the illumination pattern, whereby the length ML of each transverse line structure in the feed direction is a maximum of LMAX.

12. Inspection device according to claim 11, which is additionally set up in such a way that the computing unit determines the phase and deflection angle in the transverse direction and / or in the feed direction from a plurality of locations on the object surface on the basis of the N1 successively created image recordings of the light-dark stripes and the at least (N2-1)x4 created image recordings of the transverse line structure and from this, for example, the refractive power in the transverse direction and / or in the feed direction is determined.

13. Inspection device according to one of claims 8 to 12, which is additionally configured such that the matrix camera captures the illumination pattern reflected on the surface of the object and / or the illumination pattern transmitted through the object with the image recordings in the illuminated area.

14. Inspection device according to one of claims 8 to 13, which is additionally designed such that at least two of the multiple image recordings generated in the illuminated area are combined and evaluated individually or in combination with regard to the presence of defects.

15. Inspection device according to one of claims 8 to 14, wherein a lighting unit with a plurality of individually switchable lighting elements arranged in a matrix and / or with a lamp and a grid arranged between the lamp and the object, which generates the light-dark stripes and optionally the transverse line structures, is provided for generating the lighting pattern.

Images