RU2812804C1 - Method and testing device for optical surface testing - Google Patents

Abstract

FIELD: optics.

SUBSTANCE: methods for optical inspection of the object surface. According to the method, a time-periodic pattern with different lighting patterns is generated on the surface by a lighting device of the control device during an image recording sequence, and in the image registration sequence, multiple images of the circuit are recorded on the surface using the image recording device of the control device. The generation of one of the different lighting patterns is respectively synchronized with the registration of one of the images of the circuit so that each image from the image registration sequence is respectively registered with a known lighting pattern from the different lighting patterns. The phase of the pattern is determined from a continuous series of recorded known lighting patterns in at least one point in the image. Defects on the surface are detected by deviations in the lighting pattern recorded in at least one image from the generated lighting pattern. The lighting device and image recording device are positioned at a reflection angle a, while during surface control the object is moved relative to the control device. The duration of the image registration sequence is chosen in such a way that the reflection zone of the sequence, which is defined as the surface area covered in aggregate by the reflection zones in the corresponding images from the image registration sequence, is considered constant.

EFFECT: improving the reliability of monitoring moving surfaces.

12 cl, 18 dwg

Description

The present invention relates to a method and inspection device for optically inspecting the surface of an object, as well as an advantageous use of the method and inspection device. In this method, a time-periodic pattern with different illumination patterns is generated on the surface during an image recording sequence using the illumination device of the control device. During/during the image recording sequence, several images of the circuit on the surface are recorded by the image recording device of the control device.

During this process, the generation of one of the different lighting patterns is respectively synchronized with image registration with respect to one of the circuit images such that each image in the image registration sequence is respectively registered with a known lighting pattern from the different lighting patterns. In other words, this means that exactly one lighting pattern is visible in every image produced by the camera. By synchronizing image registration and pattern generation, it is achieved, in particular, that the illumination pattern does not change during the exposure time during image registration. From a continuous series of recorded known lighting patterns, the phase at at least one point in the image is determined. Since the pattern/periodically different lighting patterns associated with the pattern are known, an image point can be associated with a point in the known pattern. Defects on the surface are detected by deviations in the lighting pattern recorded in at least one image from the generated known lighting pattern. Defects on the surface cause distortions in the known pattern/one known lighting pattern recorded in a single image. This allows defects to be identified and inferred by image evaluation, generally known to those skilled in the art, using suitable algorithms, for example using a suitably suitable computing device. By scanning several surface areas one after another, i.e. By repeated application to different areas of surfaces, it is possible to control the entire surface or selected parts of the surface.

During surface inspection, one of the most important tasks is to detect and classify defects that, due to their topographic characteristics, cause light reflections. Often these defects are not perceived by the eye as topographic defects at all, but only as changes in brightness or shades on the surface. Control while driving is often necessary or at least advantageous. Particularly preferred applications for such surfaces in the context of the present invention will be described later.

In principle, the method according to the present invention is suitable for optical inspection of reflective surfaces. Reflective surfaces in the context of the present invention include both perfectly reflective (ie specular) surfaces and surfaces that, in addition to reflective properties, also exhibit a certain scattering effect. In this case, the criterion is that the surface illuminated by the circuit (also including the circuit projected onto the surface) is optically registered in the image.

A long-established method for surface inspection is deflectometry. It involves recording the reflection image of a known pattern on a surface using a camera and evaluating it in a computer. Surface defects result in circuit distortions on the surface that are detected. If the registration geometry and pattern geometry are known, this can also be used to determine the 3D topography of the surface. Various methods are known to those skilled in the art on how this is accomplished. They are considered known in the context of the present invention and will not be described in detail.

The basic principle of deflectometry is to determine the reflection of a beam of light incident on a surface, that is, to identify the point in the pattern at which the visual beam coming out of the camera (recording device) and displayed on the surface falls. In other words, the deflection of the visual ray on a surface is determined in the reflection, which depends on the direction of the surface normal (a straight line located vertically on the surface at the point of reflection) at the corresponding location (the point of reflection). From the normal field thus determined, the surface topography can then be determined, for example by integration.

Common methods for determining the location of a point are the so-called phase shift methods, in which the circuit used is a periodic circuit and determines at what phase position of the circuit the point being determined is located.

This is in principle different from methods that require a single image of the circuit or which require multiple images.

The advantage of methods for which a single image is sufficient is that they can also be used on a moving surface and thus seem inherently more suitable for inspection of, for example, a web product or in manufacturing processes. However, their disadvantage is that they are more susceptible to defects or require the physical presence of a second circuit in the beam path. WO 98/17971 A1, for example, describes a method for detecting and determining the smallest beam deviations. Essentially, in this case, the camera tracks a stripe pattern. For the described method, one image is sufficient, since the camera's pixel grid is used as a second circuit. However, the disadvantage in this case is that the camera and circuit require very precise adjustments. In an industrial environment, such as manufacturing processes, this is very difficult to achieve or can only be achieved at unreasonable expense.

Methods that use multiple images are significantly more resistant to defects and do not require time-consuming setup. The circuit is displayed and recorded sequentially at several phase positions shifted relative to each other. A particularly simple estimate is achieved by using a strip design with a sinusoidal brightness curve, which is recorded four times with a shift corresponding to a quarter of the period length. But other schemes and continuous series of schemes are also possible. In this case, the phase position in the circuit is obtained from a continuous series of gray values in each pixel. This method is described in detail in relevant textbooks and articles (for example, Gorthi and Rastogi, “Fringe Projection Techniques: Whither we are?”, Proc. Optics and Lasers in Engineering, vol. 48, no. 2, pp. 133-140, 2010 .). However, the disadvantage is that multiple images of the same surface location are required. However, when inspecting films and other web-like products in a manufacturing process/mostly surfaces moving relative to the inspection device, it is practically impossible to take multiple pictures of exactly the same location on the surface because the surface is constantly moving. For example, webs that move at high speed cannot be stopped during production. This problem can be solved by using a control device that moves synchronously with the belt. It must be recognized that such a solution is technically complex and therefore expensive, and also requires a lot of space, which is often impossible to provide, especially in a production environment.

EP 2390656 B1 discloses a method in which the surface of a moving web is monitored, preferably by a single line scan camera. Lighting consists of fast-switching lighting (preferably LED lighting) mounted across the canvas. This lighting consists of individually controllable LEDs or LED modules, with which different lighting patterns can be dynamically generated very quickly. Switching and image registration are synchronized so that images of a surface with different lighting patterns can be recorded in a fast, continuous series. In particular, scanning and switching can be performed so quickly that the distance between image registrations in the feed direction is much smaller than the extent of the pixel in the feed direction. In this way, images can be recorded in almost the same location. However, this method does not ensure registrations in exactly the same place.

The aim of the present invention is to provide a reliable option for monitoring moving surfaces, which in particular can be implemented in a simple manner even in an industrial environment, for example in manufacturing processes.

According to the present invention, this goal is achieved using a method according to claim 1 and a control device according to claim 12.

In the method described at the beginning, it is provided that the lighting device and the image recording device are located at a reflection angle (respectively with respect to a surface normal vertically aligned on the surface in the reflection zone). The expression "reflection angle" means that the edge rays of an image point (i.e., visual rays emanating from the edge of an image point) are reflected at reflection points on the surface and mark the visible portion of the lighting pattern (pattern region) at the image point. In other words, the reflection of the circuit's illumination pattern on the surface is displayed precisely at the image points of the image recording device. The camera (as a recording device) is thus aimed precisely at the circuit (ie the lighting device, which can be embodied, for example, in the form of a lighting line).

On a moving object, the angle of reflection does not change, provided that the shape of the surface and its location relative to the stationary control device do not change. This is true for a flat surface or a slightly curved surface if the curvature is on average constant and the direction of the normal to the surface (at least relative to the direction of the visual rays) changes only slightly. For example, this may be the case in the case of a wavy surface structure, where the change in direction of the surface normal is small. The expression "small" means that the change is only such that a portion of the circuit remains visible at the image point. Therefore, the circuit portion must be suitably wide in the feed direction. Once this is no longer possible, the method according to the present invention cannot be used with a fixed control device. However, in this case, the control device according to the method proposed in the present invention can be suitably moved in a reflective arrangement along a curved surface.

Since, due to the known periodicity of the change in the direction of the normal to the surface, it is possible to mechanically track the monitoring device such that the reflection conditions are met or the reflection angle is within the detection area by using the planar illumination device and the recording device, and the image points are selected according to the periodically occurring reflection angle , this method can also be used for curved surfaces.

Regardless, the present method can be used to determine the topology of defects, provided that changes in the circuit can be clearly detected by the inspection device.

In order to be able to inspect large areas of a surface or provide continuous inspection, for example, during production according to the present invention, the object, and thus the surface of the object, is moved during surface inspection relative to the inspection device, preferably in a defined/carefully controlled direction of movement.

For the phase shift techniques described at the beginning, it is indeed necessary that the images belonging to a sequence of images always record the same location on the surface. Since in this case a moving surface, such as a web of material moving relative to the recording device, is monitored, this is not possible. However, in order to be able to apply this method and detect the phase of the circuit, the duration of the image recording sequence is chosen to be short enough so that the reflection area of the sequence is considered constant. The reflection zone of a sequence is defined as the total surface area covered by the reflection zones/registered in the corresponding images from the image registration sequence. Simply put, the images of an image acquisition sequence are acquired one after the other at such a rate that the movement from the first to the last image of the image acquisition sequence is so small that the captured surface area (reflection zone) can still be considered essentially the same surface location.

The surface area collectively covered by the reflection zones in the corresponding images from the image acquisition sequence is obtained by combining all the reflection zones of all individual images that were recorded during the image acquisition sequence into a common area, which is then called the sequence reflection zone. In this case, this surface area can be considered at least approximately constant if the reflection zones of all images from the image registration overlap by at least 40% or more, preferably by at least 60%. However, these values should not be understood as fixed values, but as usual standard values which the skilled person can adapt, possibly experimentally, to the relevant conditions. In principle, the methods show good results provided that, due to the optical conditions, clearly less than one length of the circuit period is mapped onto one image point. Concave surface curvature, which, due to the concave mirror effect, maps large areas of the circuit onto an image point, is particularly critical. To detect defects, an overlap area of 40% to 70% should be sufficient, and when assessing the normal to the surface (i.e. assessing surface topology), an overlap area of 60% to 80% is sufficient. Depending on the shape of the surface and the type of defects encountered, other areas may also appear, which an expert, when setting up an appropriate inspection device, can identify and/or predetermine based on the teaching of the present invention, possibly empirically through experimental measurements. In other words, according to the present invention, it is proposed to select the duration of the image registration sequence so as to register the images registered within the image registration sequence in chronological order so quickly one after another that the path of the surface shift due to the movement of the object from the first image to the last image The sequence of image registration is so small that the reflection zones of the first image and the last image can be considered the same area on the surface. Compared to the measurement made when the object is stopped, a measurement error arises, which is smaller the better the above condition is met.

The area of reflection on a surface that is captured at an image point (at the minimum resolution defined by a camera pixel or perhaps a combination of multiple camera pixels) is predetermined by the acquisition geometry (distance, acquisition angle) and the acquisition optics. Due to the placement of the recording device and the lighting device at a reflection angle relative to the surface normal, a change in the angle of one of the two devices must therefore be reproduced for the other device as well. This makes changes to the reflection angle relatively expensive. The same applies to corresponding changes in the recording optics. The size of the reflection zone and/or the portion of the circuit displayed in the reflection zone according to the present invention can be changed or adjusted in a relatively simple manner by the distance to the recording device and/or the lighting device. It should be recognized that this also requires a change in the design of the control device.

According to the present invention, other parameters are easier to influence when performing the method proposed in the present invention. Next, suitable parameters when performing the method will be described. In order to adapt the duration of the image recording sequence depending on the predetermined speed and direction of motion of the object so that the reflection area of the sequence can be considered constant, according to a preferred embodiment of the present invention, one or more of the measures listed below can be taken.

Thus, when performing the method according to an embodiment of the present invention, it is possible to provide for setting the image point size.

In the simplest case, the size of the image point can correspond to the pixel resolution of the camera (used as an image recording device). This represents, for a given camera distance and predetermined camera focal length, the highest possible resolution. The higher the camera resolution, the smaller the reflection area associated with the image point on the surface and the smaller the defects detected on the surface. One option for changing the image point size is to change the camera's pixel resolution. The pixel resolution of the digital image recording camera preferred according to the present invention is predetermined by the photographic chip used as the recording sensor of the camera, on which, during the exposure time, individual pixels (sensor pixels) capture (integrate) light incident on that pixel. By reducing the resolution, the image point size can also be obtained by combining multiple camera sensor pixels to form a single image point. Also, one point in an image can be called a pixel. But an image pixel and a sensor pixel are different if multiple sensor pixels are combined to form a single image pixel.

In one embodiment, setting the image point size can be accomplished by combining multiple acquisition sensor pixel(s) of the acquisition device to form a single image pixel. In one embodiment, the number of pixels to be merged in the direction of movement of the object and across the direction of movement of the object can be selected in different ways according to the present invention. It may be advisable, by selecting a lower resolution, to increase the size of the reflection zone in the direction of movement of the object in order to obtain greater coverage of the reflection zones of the individual images respectively in one image acquisition sequence. As a result, the reflection zone of the sequence in the direction of movement of the object increases. In the transverse direction to it, higher resolution can be maintained. The resolution across the direction of motion of the object and its surface is determined solely by the registration geometry, i.e. essentially the size of the image dots (limited by the pixel size of the recording sensor of the image recording device applied to the smallest possible extent), the focal length of the lens and the viewing distance. Movement does not affect resolution across the direction of movement.

Motion blur occurs in the longitudinal direction of motion. Because the camera, during image acquisition, integrates all the light at one image point (an image pixel that is not necessarily the same as the recording sensor pixel) that falls on that image point during exposure, the tracked surface imaged per image point , increases in the direction of movement. In relation to the moving surface (also called the reflection zone associated with the image point), the image point appears to be stretched out in length, so to speak. The expressions “longitudinal” and “transverse” in this context refer to the direction of movement and do not necessarily coincide with the direction of the rows and columns of the cameras. For an oblique viewing angle, each pixel appears obliquely stretched accordingly relative to the direction of the camera's rows and columns.

In a continuous series of images (during an image registration sequence) that is registered for the multi-image phase shift method, the same location on the surface must actually be imaged at each image point (image pixel) in all images. However, when several images are recorded one after another, they shift relative to each other with respect to the moving surface. Therefore, to compensate for this, the present invention takes measures that can cause the reflection area of different images to be tracked as approximately the same location on the surface. Changing the image point size can facilitate this in the manner described above.

According to the present invention, an additional measure may be to set the duration of the image recording sequence during execution of the method. The duration of the image acquisition sequence, i.e., in other words, the time required to acquire all images of one image acquisition sequence, determines, for a predetermined object/surface speed, how far the surface area corresponding to the reflection zone of the first image moves up to registering the last image. From this follows the size of the reflection zone of the sequence and the overlap, which must be established according to the present invention, of the reflection zones of the individual images. In general, it can be said that the greater the overlap, the shorter the duration of the image registration sequence.

In addition to restrictions on the maximum scanning frequency of the recording sensor and the shortest possible exposure time of the recording device, it is possible to adapt the scanning frequency (defined as the frequency of sequential image acquisitions) and/or exposure time. The shorter the exposure time, the sharper the recorded image (reduced motion blur) and the faster images can be sequentially recorded (scanning speed). Reducing the exposure time can be achieved by increasing the brightness of the pattern generated on the surface and/or opening the aperture of the recording optics. By increasing the brightness/increasing the aperture opening (usually determined by smaller aperture numbers in the optics), exposure time can be reduced. Therefore, it makes sense to use a lighting device with high but adjustable light intensity.

Suitable lighting devices can be made from individually dimmable LEDs which, when dimmed individually, allow a pattern to be created, and when dimmed together, allow the overall light intensity to be adjusted. In general, it may be preferable to operate the lighting device at maximum light intensity and reduce exposure time until suitably exposed images are recorded.

Therefore, according to the present invention, when adjusting the duration of an image acquisition sequence, at least one of the following variables can be adapted: image exposure time, brightness of the pattern generated on the surface, scanning frequency of the acquisition sensor, and/or the number of images per image acquisition sequence. It is also possible to adapt all or several variables.

Of course, as such, the duration of the image registration sequence can also be changed according to the present invention by changing the number of images per image registration sequence, and reducing the image registration sequence can be achieved by reducing the number of images, and vice versa.

Moreover, according to the present invention, the measurement sensitivity can be influenced by selecting the illumination distance (while also considering the distance between the recording device and the circuit) and viewing angle. Longer distances, as well as flatter viewing angles (i.e. flatter relative to the surface; vertical to the surface will be steepest) result in higher sensitivity. In particular, with partially reflective surfaces in both a specular and reflective manner, it may be particularly preferable to select a flatter viewing and illumination angle (eg <30°) and/or a maximum illumination distance. According to the present invention, the maximum lighting distance may mean that the available space is used to accommodate the lighting device. The illumination distance (distance between the illumination device and the surface) may, for example, be selected to be greater than the distance between the recording device and the surface, with typical values ranging from one to, for example, ten times. The person skilled in the art selects values, possibly experimentally adapted to the respective application, whereby according to the basic idea of the present invention, the sensitivity is in many cases increased by smaller illumination angles and viewing angles and/or a greater illumination distance (between the recording device and the lighting device).

The purpose of recording multiple images is to determine the phase of the pattern in order to identify the position of a known lighting pattern at a registered point in the image. This will allow surface defects to be detected by circuit distortions on the surface. According to one embodiment, for example, three images may be registered. For example, the generated circuit can be periodically shaped asymmetrically, so that the phase of the circuit can be uniquely determined from the three images. Alternatively, the circuit can also be periodically shaped into a symmetrical shape while acquiring images asymmetrically, for example by varying the scanning/image acquisition frequency between different images within the same image acquisition sequence.

However, one application preferred by the present invention involves scanning with at least or exactly four images within the same image acquisition sequence. The actual pattern may be, for example, a sinusoidal luminance distribution that is recorded in an identical scanning sequence at four different phase positions. From this, the phase of the circuit in each of the images can be determined in a simple and accurate manner. For example, the phase shift between phase positions in a recording sequence of successive images may be as little as VA of the length of the circuit period. But other phase shifts between images of the image acquisition sequence are also possible.

According to a further aspect of the present invention, the illumination pattern may be generated by the illumination device such that the visible portion of the illumination pattern recorded at image points related to images recorded during each respective image acquisition sequence can be considered constant.

An illumination pattern region (pattern region) visible at image points during an image acquisition sequence can be considered constant provided that the pattern region remains still visible at the image point and the recorded intensity of the pattern region does not change significantly. This may be assumed, for example, if the recorded intensity during the image recording sequence changes by no more than 10%, preferably by no more than 8%, and especially preferably by no more than 4%, or other specified criterion is maintained. In this case, the criteria already discussed above are also generally applicable.

In this regard, according to a preferred aspect of the present invention, the length of the circuit period in the lighting circuit can be chosen in such a way that, depending on the surface topology in the direction of the stroke of the circuit, the change in intensity can be considered to be fairly constant, in other words, this means that the change intensity does not exceed the criterion appropriate for the relevant circumstances. The choice of criterion can be determined by a specialist, possibly experimentally when installing the system and specific circuits.

The topology of a surface is determined, in particular, by its curvature, which entails a change in the direction of the normal to the surface. The direction of the normal to the surface is related to the angle of reflection. It is therefore possible, by means of the topology of the surfaces to be examined by the method of the present invention, to determine from the resulting reflection angle for a known location of the inspection device which portion of the illumination pattern is imaged at an image point during a certain duration of the image acquisition sequence. By first determining the period length, the lighting design can be refined to meet the above criteria. Therefore, the method can be flexibly applied to specific control tasks.

According to a further aspect of the method of the present invention, provision may be made for generating a periodic pattern along the direction of motion of the object, across the direction of motion of the object, or alternately along and across the direction of motion of the object.

When the pattern is located along the direction of the object's movement, the motion blur and shift of the reflection zone already discussed for the curved surface being monitored will overlap with the shift of the pattern section viewed through the image point due to the change in the reflection angle and with the change in intensity for the same reason , since the intensity of the circuit changes in this direction.

For a pattern across the direction of motion, the reflection zone also changes. However, since the pattern along the direction of the object's shift has the same intensity, changing the reflection angle does not necessarily result in a change in intensity. The intensity measured at an image point remains the same provided that the image point captures the same area of the circuit and the curvature of the surface does not lead to a shift of the section of the circuit captured at the image point across the direction of motion.

According to the present invention, this difference can be taken into account during the above-described adaptation of the period length of the circuits depending on the alignment of the circuit along or across the direction of movement of the object. The length of the circuit period, in particular for circuits located along and transverse to the direction of movement, according to the present invention can particularly preferably be different.

In addition, the known curvature of the surface of an object at a specific surface area to be inspected can also be used according to the present invention to refine suitable criteria to distinguish a surface without defects from a surface with defects and/or to correct the deviation caused by the known (expected) shape surfaces in the evaluation of recorded images as part of defect detection.

By generating patterns alternately along and across the direction of motion, different defects, in particular directional defects in a surface, can be systematically captured in a more reliable manner.

In one embodiment of the method of the present invention, the recording device may be focused such that the lighting pattern recorded in the image is blurred.

This can be achieved, for example, by focusing the recording device not on the circuit, but on a surface or other specific point. By predetermining specific aperture and focus settings, the depth of focus/depth of field can also be selectively selected according to the present invention to display the lighting pattern in the image so that it is blurry but the surface is in focus. As a result, the sharp distribution of brightness appears washed out. Thus, for example, a sharp pattern consisting simply of alternating separated light/dark regions can be displayed as an approximately sinusoidal luminance curve. In this case, a particularly simple lighting device can be used without the need for additional optical elements to generate the desired brightness curve. In addition, the brightness distribution becomes less sharp, which in particular can have a positive effect on curved surfaces, as well as related effects when shifted areas of the circuit are mapped onto images recorded in the image acquisition sequence.

In many cases, the surface to be tested is not perfectly specular, but reflects semi-diffusely. Although the reflection is directional, it scatters at a relatively large spatial angle, which means that the two-dimensional reflectance distribution function (BRDF) has a scattering group of medium width. This also results in a fairly useful blurring of the circuit brightness distribution in images, provided the diffusion group remains narrow enough to result in sufficient modulation of the specular pattern in the image captured by the camera, and possibly work with a reflective, although not perfectly specular, , surfaces. This surface property can also be used to achieve an effect similar to that achieved by the described unfocused camera setup in the diagram. However, this (additional) effect must be taken into account during an unfocused installation, since part of the blur (desirable in this case) is in any case created by the surface itself.

On the other hand, the surface must be specular enough to still allow the circuit to be observed. Therefore, for surfaces with relatively little specular reflection, it is preferable to select a viewing and illumination angle that is as flat as possible and increase the illumination distance.

It may be particularly advantageous if, during the surface inspection performed according to the present invention using deflectometry processes, the three-dimensional topography of the object's surface is determined. If the registration geometry and the circuit geometry are known, as in the method proposed according to the present invention, then the 3D topography of the surface can also be determined. There are several options for performing this task. In deflectometry, the deflection of a beam of light incident on a surface is determined by determining the point in the pattern at which the visual beam emitted by a camera (recording device) falls and is mirrored (reflected) on the surface. Therefore, the deflection of the visual ray is determined, which depends on the normal to the surface at the corresponding location. From the surface normal field thus created, the surface topography can be determined, for example by integration.

A particularly preferred use of the above-described method or parts thereof and/or the control device described below in this document is the result of control of a product in the form of a web, for example, in the production process or after its manufacture, or, in particular, machined, curved or flat surfaces.

An important specific typical example is the inspection of FCCL film during or after production. FCCL (Flexible Copper Foil Laminate) films are a key material for flexible printed circuit boards. FCCL films typically have a thickness of approximately 100-150⋅μm and contain, for example, a polyamide core (usually a plastic film) that is laminated on one or both sides of the surface with a copper film. During lamination, folds may be formed which need to be detected using the method of the present invention. During surface inspection it may also be desirable to detect lamination defects, in particular so-called lamination folds 4 (as schematically shown in FIG. 1) or internal folds 5 (as schematically shown in FIG. 2). In the case of lamination folds, small folds were created in the material, which were then pressed back together during the lamination process. Inner pleats are formed from folds in the inner plastic film that have been laminated.

Both types of defects are very difficult to detect with the naked eye, since the films are very thin and, therefore, the folds do not have much effect on the surface. Defects are detected only by observing the direct reflection of light on the surface of the films. This is compounded by the fact that copper film reflects semi-diffusely. In the case of other laminated films, the appearance is important and is negatively affected by such defects, despite the small topographical characteristic.

When inspecting curved surfaces, such as for example painted containers or car bodies, the inspection device is programmed according to a preferred embodiment, for example by means of a suitable processing unit, and guided along the curved surface such that both the lighting device and the recording device are held at a reflection angle to surfaces. In this case, the control device is moved in this way relative to a substantially stationary object. This creates movement of the object/object surface relative to the control device. As earlier in this description, this type of relative motion is referred to when referring to an object moving relative to the control device. The most important thing is to look for the smallest flat topographic imperfections as often as possible on surfaces that are curved in a way that may adversely affect the appearance or function of the surface. It is often useful to also measure such defects in three dimensions, i.e. determine the 3D topology of the surface and defect.

The present invention also relates to an inspection device for optically inspecting the surface of an object, as well as its use for the above-described applications. The monitoring device is equipped with a lighting device and a recording device, which are aligned with respect to each other in such a way that a visual ray emanating from the recording device falls on the lighting device as a visual ray reflected from the surface when the normal to the surface, located vertically on the surface at where the visual ray hits, bisects the angle between the outgoing visual ray and the reflected visual ray. Therefore, in other words, the recording device and the lighting device of the monitoring device are located at a reflection angle relative to the surface. The lighting device is configured to generate a time-periodic pattern with different lighting patterns during an image recording sequence, and the recording device is configured to register images of the patterns reflected on the surface synchronously with generating lighting patterns during the image recording sequence. The control device further includes a computing unit for controlling the control device and for evaluating the recorded image, wherein the processor of the computing unit is configured to execute the above method or parts thereof.

According to a preferred embodiment of the control device proposed according to the present invention, the lighting device contains individually controllable light elements arranged in rows or in a matrix. Also preferably, the recording device may comprise a recording sensor for recording images displayed on the recording sensor via recording optics, wherein the recording sensor comprises individual sensor pixels (camera pixels) arranged in rows or in a matrix.

The lighting device can be made, for example, in the form of a lighting line, which is preferably located transversely or along the supply direction (the direction of movement of the object/surface relative to the control device). The lighting line, consisting of individually controllable lighting elements arranged in a line, can consist of several LEDs located next to each other, or of LED modules that can be individually switched in sync with image acquisition. The lighting device is used to generate a rapid continuous series of periodic patterns necessary for the phase shift process. The recording device can also be made, for example, in the form of a single-line scan camera, which can also be assembled, if necessary, from several single-line scan camera modules located next to each other. With this arrangement, the resulting image field of a single-line scan camera is a line on the surface (the so-called scanning line). This scanning line may be aligned transverse to the direction of relative motion of the surface and also has a certain very small width in the direction of motion compared to its length (transverse to the direction of motion), which depends on the pixel resolution of the single-line scan camera.

The lighting line can be so long (across the direction of travel) as to cover the entire width of the surface being inspected (or the desired area of inspection on the surface) at the angle of reflection. When the camera and lighting are positioned the same distance from the surface, the lighting line on each side should be longer by approximately half the width of the scan line of an individual single-line camera than the scan line on the surface that is visible to all cameras, and for other distances it should be longer or less depending on the specific case.

The width of the lighting line (in the direction of travel) can determine the maximum angle to the surface that can still be measured in this arrangement. If the angle to the surface becomes greater than the maximum angle to the surface, the camera's visual beam reflected by the surface no longer falls on the lighting and the camera sees nothing.

This method is also suitable for use with a line scan (matrix) camera. In this case, the scanning line becomes the image field, since the width in the direction of movement becomes significantly larger. The width of the lighting line can also increase accordingly in the direction of travel. In one embodiment, a lighting matrix may be used instead of a lighting line. It consists of several individual LEDs or LED modules that are arranged in multiple seamlessly connected lighting lines, all of which are independently switchable in sync with image acquisition. In this way, it is also possible to change the width of the lighting line in a simple manner, so that several lighting lines are switched in the same way.

The lighting matrix can be used to switch not only patterns across the web direction, but also patterns along the web direction. The reason this is an advantage is that deflectometry processes primarily measure angles to the surface/surface normal, namely in the direction of the periodic pattern. Thus, when using a lighting line, only angles transverse to the direction of motion can be measured, while when using a lighting matrix, all directions can be measured, preferably two directions: along and across the direction of motion.

On graphic materials:

in fig. 1 shows a schematic cross-section of an object with an inspected surface containing a first typical defect;

in fig. 2 shows a schematic cross-section of the object shown in FIG. 1 with the surface being inspected containing a second typical defect;

in fig. 3a is a top view of an inspection device according to an embodiment of the present invention for inspecting a flat surface;

in fig. 3b is a side view of the control device of FIG. 3a.

Object 1 shown in FIG. 1 and 2, the surface 10 of which is to be inspected by the inspection device according to the present invention is an FCCL film used as a raw material for printed circuit boards. It is a laminated film 1 composed of three layers, with the middle plastic film 3 serving as the middle layer onto which the outer copper films 2 are laminated. The surface 10 of the object 1 is generally inspected for surface defects.

This surface inspection is also used to detect lamination defects, in particular so-called lamination folds 4 (FIG. 1) and internal folds 5 (FIG. 2). In the case of lamination folds 4, small folds were formed in the material, which were then compressed again during the lamination process. The inner folds 5 are formed due to folds that have formed in the inner plastic film 3, which have been laminated.

In fig. 3b shows a side view of the control device 9 with a lighting device 8 and a recording device 7. The lighting device 8 shows a time-periodic circuit 13 with different lighting patterns 130 that illuminates the surface 10 of the object 1 (see also the top view in FIG. 3a). The lighting circuit 130 contains a brightness distribution 14 . This also results in the generation of the pattern 13 on the surface 10. The registration device 7 registers the pattern 13 on the surface 10 in the image.

The registration device 7 also contains a recording sensor 11, which generates an image with several image points 12. Thanks to the optics of the recording device, which is not shown, the visual rays 15 emanating from (each) image point 12 are reflected on the surface 10 and fall as reflected visual rays 19 onto the lighting device 8 onto the circuit 13 generated therein. The edge rays of these visual rays 15, 19 are graphically plotted in FIG. 3b. The edge rays emanate from the edges of the image point 12 and limit the reflection zone 17 on the surface 10. All visual rays 15 emanating from the image point 12 at the reflection angle a and incident on the surface lie in the reflection zone 17 on the surface 10 and are also reflected at the reflection angle a from the surface in the form of reflected visual rays 19. They fall on the lighting device 8 in section 17 of the circuit, since according to the arrangement proposed in the present invention, the registration device 7 and the lighting device 8 are located at a reflection angle a relative to the surface 10.

The reflection angle a is defined as the angle between the incident visual rays 15, 19 (emanating from the image point 12)/outgoing (reflected from the surface 10) and the associated normal 16 to the surface. The surface normal 16 belonging to the visual ray 15, 19 extends vertically to the surface at the reflection point 170 at which the visual rays 15, 19 are incident on the surface 10.

In fig. 3a specifically shows a line of the recording sensor 11 of the recording device 7 that runs along the width of a surface 10 such as a web product moving in the moving direction as an object 1 such as an FCCL film. The recording device 7 can be made in the form of a single-line scan camera having only one touch line of recording sensors 11, or in the form of a line-scan camera with several such touch lines. The image point 12 may be formed from one or more sensor pixels. By means of optics, which are not shown, the image point 12 of the recording device (camera) captures the reflection zone 17 on the surface 10. The visual rays 15 are deflected on the surface 10 and capture the circuit area 18, which is formed by the area of the circuit 13/the corresponding lighting circuit 130 of the circuit 13 at the moment registration of images. In the example shown in FIG. 3a and 3b, the lighting device is designed as a lighting line aligned transversely to the direction 6 of movement of the surface 10.

In fig. 3b shows a side view of the same arrangement, in which the reflection of the visual rays 15, 19 (graphically plotted as edge rays, as in all figures) with a reflection angle a relative to the normal 16 to the surface is clearly visible. The graphically plotted edge rays of the visual rays 15, 19 visually represent the size/area of the reflection zone 17 on the surface 10 and the pattern area 18 in the pattern 13.

In fig. 3a and 3b show the state during image recording, and it is assumed that the motion of the surface 10 moving in the moving direction can be neglected if the exposure time is short during image recording. If this is not the case, then the recorded images exhibit some motion blur, which can be counteracted by reducing the exposure time (provided the lighting is bright enough).

As already described, multiple images are registered in chronological order using the method according to the present invention during the image registration sequence. Since the surface moves during the image recording sequence in the motion direction 6, the image point 12 no longer sees the same surface area in the corresponding reflection zone 17 of the sequentially recorded images. Instead, the reflection zones 17 on the surface 10 are shifted relative to each other in sequentially recorded images.

This is shown in FIG. 4a and 4b, on which the shift 61 of the surface 10 between the first and last image registration in the image registration sequence is graphically plotted. The reflection zone 17a is graphically plotted as the reflection zone of the first image registration, and the reflection zone 17b is graphically plotted as the reflection zone of the last image registration of the image registration sequence, each of which is depicted as a hatch rotated by 90°. In an overlapping area, two hatches overlap each other. The entire reflection zone 17 in all images of the registration sequence is enlarged accordingly (relative to the total covered surface 10 relative to the reflection zones of the individual registrations). This effect is basically similar to the motion blur already discussed, with the difference being that the entire reflection area is integrated into one image. This gives the image a blurry appearance to the extent that motion blur needs to be recognized at all.

Since the registration geometry does not change for a flat surface, the shift of the surface 10 does not have any effect on the circuit region 18; this remains unchanged throughout the acquisition sequence, with, as previously described, the generation of circuit lights being, of course, phase-shifted. However, this is not shown in FIG. 4a for the sake of clarity.

In fig. 4b shows the same situation as in FIG. 4a in side view. The surface normals 16a during image registration a were at the same time in the same position as the surface normals 16b during image registration b, which is shown here as instantaneous position registration. Due to the flat surface 10, the alignment of the surface normals 16a and 16b is the same, causing the circuit area 18 to also remain unchanged.

In fig. 3c and 4c show the arrangement of the control device 9, where the lighting device 8 contains a lighting line aligned along the direction 6 of movement of the surface 10. This can be achieved using a linear lighting device (with a suitably aligned line) or using a matrix lighting device controlled by an appropriate way. Due to the flat surface, this arrangement also produces a situation comparable to that shown in FIG. 3a, 3b and 4a, 4b. Please refer to the description above for details.

In fig. 3d and 4d show an arrangement of the control device 9 similar to that shown in FIG. 3c and 4c, where not only the lighting line of the lighting device 8, but also the touch line of the recording sensor 11 are aligned along the direction 6 of movement of the surface 10. The recording device can suitably be made in the form of a single-line camera (with only one touch line) or in the form matrix camera (with several sensor lines located next to each other). Due to the flat surface, this arrangement also produces a situation comparable to the arrangement shown in FIG. 3a, 3b, 3 and 4a, 4b, 4c. Please refer to the description above for details.

The situation is different when the surface is not actually flat. This is shown in FIG. 5a, 5b, 5c and 5d/6a, 6b, 6c and 6d. The views and arrangements correspond to the views and arrangements discussed with reference to the views and arrangements related to FIGS. 3a, 3b, 3c and 3d/4a, 4b, 4c and 4d. Therefore, taking into account the general description, reference should be made to the above. Due to the curvature of the surface 10, which affects the alignment of the surface normals 16, 16' and affects the reflections of the visual rays 15, 19, different circuit regions 18a, 18b are obtained as a consequence of different images of the image acquisition sequence.

In fig. 5a, 5b, 5c and 5d show the situation for one image, respectively, eg the first image of a sequence of images. Fig. 5a essentially corresponds to Fig. 3a, with the sides of the surface 10 depicted in a curved manner designating the curvature of the surface 10 as running transverse to the direction 6 of motion. Due to the curvature of the surface 10, the visual rays in the top view are not reflected as a straight line, but are deflected at the reflection point 170, 170'. Accordingly, the reflected visual rays 19 are incident on the circuit 13 at a circuit portion 18 that lies at a different location than the circuit region 18 of FIG. 3a. In fig. 5b accordingly shows that the surface normals 16 and 16' are differently aligned at reflection points 170, 170' (and therefore labeled differently). Consequently, the reflection angles α, α' are also different.

In fig. 6a and 6b show a reflection zone 17a (for visual rays 15, 19 reproduced in FIGS. 5a, 5b during recording) and a reflection zone 17b (for visual rays 15, 19 reproduced in FIGS. 6a, 6b), together with an overlapping portion 171. Circuit portions 18a and 18b and their overlapping portion 181 are shown accordingly.

The image point 12 is illuminated in the recording sensor 11 by a section 18, 18a, 18b of the circuit limited by the edge visual rays 15 (before the specular reflection on the surface 10)/19 (after the specular reflection on the surface 10), and this section 18, 18a, 18b is displayed on circuit 13 in zones 17, 17a, 17b of reflection of surface 10 in registration device 7. However, each of the visual rays 15 is deflected in accordance with the surface normals 16, 16', 16a, 16a'/16b, 16b' present at that location.

In fig. 5a, 5b, 5c and 5d show the situation in the first image of the sequence. In addition, the camera pixel 12 in the image sensor 11 is illuminated by a pattern portion 18 limited by the edge rays 15 (before surface reflection)/19a (after surface reflection), which portion 18 is mapped onto the pattern 13 in the area 17a of the surface 10 in camera. However, in this case, the visual beam 15 is deflected in accordance with the surface normals 16a/16b present at that location. The situation in the respective last image of each image registration sequence is shown in FIG. 6a, 6b, 6c and 6d. In this case, portion 18b of the circuit 13 is displayed on portion 17b on the shifted surface 10 at image point 12. In this case, the surface normals 16b, 16b' are important for specular reflection of the edge rays 15 emanating from the camera. Since they are different from those present in the first image (FIGS. 5a, 5b, 5c and 5d), the portion of the lighting circuit 130 in the lighting device 8 that is visible/displayed at image point 12 is also shifted. In general, throughout the image sequence from the first to the last acquisition, the image point 12 moves along a portion 17 of the surface 10 in FIG. 6a, 6b, 6c and 6d and thus throughout the entire pattern area 18 of pattern 13. Image point 12 sees a region that is located both in the reflection zones 17a and 17b of the surface 10 and in the pattern regions 18a and 18b of pattern 13. She, i.e. image point 12 does not see areas that throughout the entire sequence of images are present only in 17a or 17b/18a or 18b.

However, it should be noted that the proportions in Fig. 3a, 3b, 3c, 3d, 4a, 4b, 4c, 4d, 5a, 5b, 5c, 5d, 6a, 6b, 6c, 6d are not realistic. Also, the cross-hatched cutoff portions 171, 181, respectively, do not correspond to realistic variables but are for illustrative purposes and to facilitate understanding only. In fact, at least the lighting circuit 13/circuitry must have a much longer wavelength compared to the image size of the image dot 12, such that the image dot 12 covers only a small fraction of the wavelength. If the proportions associated with size were realistic, this principle could not be understood from graphic material.

As described earlier, in the image registration sequence that is registered for the multi-image phase shift process, the same location of the surface 10, i.e. the same reflection zone 17 should in fact be displayed in all images at each image point 12. When several images are recorded one after another, they shift relative to each other with respect to the moving surface 10. What is critical to assessing whether the location recorded by image point 12 during an image sequence can still be considered "about the same location" in the context of the present invention ultimately depends on how the periodic pattern 13 is displayed on the surface 10 in the recording device 7 changes during the image sequence. In turn, this depends, on the one hand, on the actual circuit 13 (lighting circuit 130) and its distance from the surface 10, on the other hand, on the reflection zone 17, which is mapped onto the image point 12 throughout the entire image sequence, and on how this area changes (reflection zone 17). The area of the reflection zone depends on the optical pixel resolution (i.e., the area that is imaged in the viewing plane on a single pixel), exposure time, duration of the exposure sequence, and motion speed (i.e., how far the surface 10 moves throughout the entire image sequence) . Changes in section 18 of the circuit depend on the topography of the surface (in particular, on changes in the normal to the surface).

If it is necessary to carry out a phase shifting process, the circuit 13 and the image point 12 (also in the case of a fixed surface 10) must correspond to each other in such a way that in the part of the lighting circuit 130 covered by the image point 12 in the lighting circuit 130, the brightness can be considered almost constant /average brightness actually represented the brightness measured at point 12 of the image. Also, the brightnesses are allowed to vary to such an extent that the brightness for the required minimum deviation on the surface (caused by the defect being detected) changes sufficiently that the monitoring device 9 is able to perceive it. The first occurs if the surface 10 covered by the image point 12 as the reflection zone 19 can be considered almost flat. If this is not the case, topographic measurement becomes impossible without additional information; all that can still be detected is the presence of a deviation on the surface. In addition, the lateral resolution (i.e., the size of the area on the surface) must be adjusted so that the smallest surface deviations that will be detected during inspection are still resolved.

For the moving surface 10, it must also be taken into account that during the image recording sequence, a larger area (the entire reflection zone 17) on the surface 10 of FIG. 4a, 4b, 4c, 4d, for example 6a, 6b, 6c, 6d, is covered by the image point. This affects the lateral resolution. If the surface is further curved, one image point additionally covers a larger area 18 of the pattern in the pattern 13. This affects the depth resolution. If the surface 10 moves during image registration in the image registration sequence, the decisive factor is how the corresponding visual beam 15, 19 of the image point 12 passes over the lighting circuit 130 (instantaneous registration of the circuit 13).

In the case of a flat surface 10, this effect does not occur at all, as shown in FIG. 3 and 4. Thus, errors caused by the display of different recording sections of the circuit in the recording sequence images do not occur. However, this only applies if there are no measurement errors in a no-interference situation. Once there is any defect on the surface (or if it is curved in any way), this is no longer applicable. Therefore, the case shown in FIG. 5 and 6 also occurs in case of measurement errors.

Thanks to the method according to the present invention and the corresponding control device, the system is designed in such a way that the above conditions are also met for the exposure time/total acquisition time for the complete image acquisition sequence. In this regard, the images of the image registration sequence are recorded in chronological order one after another so quickly that the displacement of the surface 10 during registration is so small that each image point 12 covers an area (reflection zone 17) on the surface 10 that can still be read constant. In addition, the length of the period of the circuit 31 is chosen such that the portion along which the visual beam 15, 19 of the registration device 7 passes, mirrored or reflected on the surface during registration of the image registration sequence, can still be considered constant/error arising from For this, there is less than the required depth resolution.

The more curved the surface 10 is, the faster the images need to be recorded and the more the wavelength of the circuit 13 needs to be increased. However, both conditions need to be met only for those areas on the surface 10 that are actually being monitored. In most cases, they represent structurally defect-free surface areas and areas in which flat topographic defects with a long wavelength exist. Moreover, most surfaces have very small, mostly very vertical, topographic defects. For these defects, in most cases the conditions can no longer be met, and this mainly applies to the static case. All that can be done in this case is to detect these defects (detect the defect), but measuring them (measuring the topography) is no longer possible.

Very high image acquisition rates are required for this method to achieve the required lateral resolution for the entire image acquisition sequence. This in turn requires very short exposure times, which in turn requires very bright lighting.

For the phase shift method to be used very advantageously in this context, it is most advantageous for the circuit 13 (ie, each of the lighting circuits 130) to have a sinusoidal luminance curve. This is usually achieved, for example, by using screens or circuits projected onto a surface. Thanks to this, the sine curve can be represented in a very good or ideal way. Unfortunately, the brightness achievable at an economical cost with this lighting is often insufficient, and the possible image rates are limited so that they can only be used in slow processes.

By using an LED line or LED matrix, where individual LEDs or even individual LED modules that consist of several single LEDs can be controlled individually, both the required brightness and the required switching frequency can be realized, synchronized with the cameras' image acquisition. Or several lines can be combined to form a matrix.

In their simplest form, individual LEDs/LED modules can only be switched on or off. This means that only a rectangular luminance curve can be realized, which is only a rough approximation to the actual luminance curve required. This is already enough to perform the phase shift method, but the accuracy is limited. By taking various measures, a better approximation to the required curve can be achieved. The greater the similarity to a sine curve, the higher the accuracy. The light line/light matrix can be modified so that intermediate brightness values for individual LEDs can also be adjusted. Depending on the size of the LEDs or LED modules, a good approximation to the sinusoidal curve can be achieved in this way. This is possible, for example, because the individual LEDs/LED modules are only connected sporadically during the actual exposure time. However, this method is expensive because it requires extremely fast control electronics. A preferred solution according to the present invention causes the circuit to be displayed to be blurred on the camera. This has already been described and will not be repeated in this document.

It should be noted that in the above description, the terms “camera” and “image recording device” are used interchangeably. All features and functions disclosed in relation to the camera are accordingly also applicable to the image recording device and vice versa.

List of reference designations:

1 object

2 copper film

3 plastic film

4 first defect

5 second defect

6 direction of movement

61 shifts

7 registration device

8 lighting device

9 control device

10 surface

11 recording sensor

12 point image

13 scheme

130 lighting scheme

14 brightness distribution

15 visual beam

16 normal to surface

17 reflection zone

170 reflection point

171 sections for cutting off reflection zones of individual images

Section 18 of the scheme

181 cut-off sections of circuit sections in separate images

19 visual rays

α reflection angle

Claims (22)

1. A method for optical control of the surface (10) of an object (1) using a control device (9), in which a time-periodic circuit (13) with different lighting patterns (130) is generated on the surface (10) by means of a control lighting device (8) devices (9) during the image registration sequence and in the image registration sequence register several images of the circuit (13) on the surface (10) using the image registration device (7) of the control device (9), while the generation of one of the different lighting patterns (130) is synchronized respectively with the registration of one of the images of the circuit (13) such that each image from the image registration sequence is respectively registered with a known lighting pattern (130) from the different lighting patterns (130); the phase of the pattern (13) is determined from a continuous series of recorded known lighting patterns (130) at at least one point in the image; defects (4, 5) on the surface (10) are detected by deviations of the lighting pattern (130) recorded in at least one image from the generated known lighting pattern (130); The lighting device (8) and the image recording device (7) are positioned at a reflection angle ; And during surface control (10), the object (1) is moved relative to the control device (9); characterized in that the duration of the image registration sequence is chosen in such a way that the reflection zone (17) of the sequence, which is defined as the surface area covered collectively by the reflection zones (17a, 17b) in the corresponding images from the image registration sequence, can be considered constant, and for Thereby, the lighting circuit (130) is generated by the lighting device (7) so that the portion of the lighting circuit (130) visible at the image points (12) recorded during each image recording sequence can be considered constant by selecting the period length of the circuit (13 ) in the lighting circuit (130) in such a way that, depending on the topology of the surface (10) in the direction of the circuit, the change in intensity does not exceed the selected criterion. 2. Method according to claim 1, characterized in that the size of the image dot (12) is set during execution of the method. 3. The method according to claim 2, characterized in that the size of the image dot (12) is set by combining several pixels of the recording sensor (11) of the recording device (7). 4. The method according to one of the previous paragraphs, characterized in that the duration of the image registration sequence is set during execution of the method. 5. The method according to claim 4, characterized in that when setting the duration of the image registration sequence, at least one of the variables listed below is adapted: - image exposure time; - brightness of the circuit (13) generated on the surface (10); - scanning frequency of the recording sensor (11); - number of images per image registration sequence. 6. The method according to one of the previous paragraphs, characterized in that the periodic pattern (13) is generated along the direction of movement of the object (10) across the direction of movement of the object (10) or alternately along and across the direction of movement of the object (10). 7. Method according to one of the previous claims, characterized in that the recording device (7) is focused in such a way that the lighting pattern (130) recorded in the image is blurred. 8. Method according to one of the previous paragraphs, characterized in that during inspection of the surface (10) the three-dimensional topography of the surface (10) of the object (1) is determined through deflectometry processes. 9. Application of the method according to one of the previous paragraphs. 1-8 to control the product in the form of a web or treated surfaces (10). 10. Control device for optical inspection of the surface (10) of an object (1) with a lighting device (8) and a registration device (7), which are aligned relative to each other in such a way that the visual beam (15) emanating from the registration device (7) as a visual ray (19), reflected on the surface, falls on the lighting device (8) when the normal (16) to the surface, located vertically on the surface (10) at the point of incidence of the visual ray (15, 19), bisects the angle between the output visual beam (15) and the reflected visual beam (19), wherein the lighting device (8) is configured to generate a time-periodic pattern (13) by different lighting patterns (130) during the image recording sequence and the recording device (7) configured to register images of the circuit (13) reflected on the surface (10) during the image registration sequence synchronously with the generation of the lighting circuit (130), wherein the control device (9) contains a computing unit for controlling the control device (9) and for evaluation of registered images, characterized in that the processor of the computing unit is configured to perform the method according to one of claims. 1-8. 11. The control device according to claim 10, characterized in that the lighting device (8) contains individually controllable light elements arranged in a row or in the form of a matrix, and the recording device (7) contains a sensor (11) for registering images displayed on the sensor (11), using recording optics, while the sensor (11) contains individual sensor pixels arranged in a row or in the form of a matrix. 12. The control device according to claim 10 or 11, characterized in that the registration device (7) and the lighting device (8) are located in such a way that a flat viewing angle is provided between the corresponding visual beam (15, 19) and the surface (10) and lighting, less than 30°, and/or between the surface (10) and the lighting device (8) a large lighting distance is provided, ranging from one to ten times the distance between the registration device (7) and the surface (10).