WO2011147760A1

WO2011147760A1 - Identification of articles

Info

Publication number: WO2011147760A1
Application number: PCT/EP2011/058318
Authority: WO
Inventors: Markus Gerigk; Andreas BÄCKER; Thomas Birsztejn; Ralf IMHÄUSER; Christian Roth; Walter Speth; Simon Vougioukas
Original assignee: Bayer Technology Services Gmbh
Priority date: 2010-05-25
Filing date: 2011-05-23
Publication date: 2011-12-01
Also published as: EP2577622A1; DE102010021380A1

Abstract

The invention relates to the technical field of the safe identification and authentication of articles using characteristic features from the surface of the article. The subject matter of the present invention is a method for producing an identification feature for an article.

Description

Identification of objects

The invention relates to the technical field of secure identification and authentication of objects based on characteristic features of the surface of the object. The subject of the present invention is a method for producing an identification feature for an article.

It is well known that one can identify and authenticate objects based on the unique intrinsic structure of their surface.

BC d'Agraives et al. disclose a method for identifying an article by its surface structure (Surface Topography, A remearkable method for the identification of seals or structures in general, ^3rd Annual Symposium of Esarda of Karlsruhe, 1981, Poster 8.13, Proceedings pages 403-409).

The GB2097979A disclosure describes a method of identifying objects based on surface features. The surface features are detected in a selected surface area of the object to be identified. For this purpose, "meaningful, random singularities of the profile are counted in the form of bumps and measured their heights and distances".

Disclosure WO2000 / 65541A discloses a method and apparatus for authenticating an item based on intrinsic physical features, particularly topographical information.

The disclosure WO2003 / 087991A2 discloses methods and an apparatus for authenticating objects based on the three-dimensional surface structure.

In WO05 / 088533A1 a method is described in which a surface area of an object is scanned with coherent radiation and by means of photodetectors the differently scattered rays at different locations of the surface at different angles are detected. The detected scattered radiation is characteristic of a variety of different materials and is very difficult to mimic, as it is due to random manufacturing. For example, paper-like objects have a manufacturing fiber structure that is unique to each manufactured object. The scattering data on the individual objects can be described as a characteristic fingerprint of the Be stored in a database to identify the item at a later date and / or to be able to authenticate. For this purpose, the object is measured again and the scatter data is compared with the stored fingerprint.

In WO05 / 088533A1, optical scanning of a surface is performed by moving a focused laser beam across the surface (or moving the surface relative to the focused laser beam) and, during movement at a constant measurement frequency, intensity values of the detected scattered radiation at one or more photodetectors Function of time to be detected. An intensity-time signal obtained in this way is impractical as an identification feature since a different signal would result when scanning at a different speed. A direct comparison of recorded at different speeds signals is therefore not possible. On the other hand, a signal which represents the sampling information as a function of the location of the sampling would have the advantage of being independent of the sampling rate and possibly occurring during the sampling speed fluctuations. Usually, however, the scanning signal is not detected directly as a function of the location. Rather, an additional signal is determined separately from the sampling signal as a function of time (measurement frequency), which links the time (measurement frequency) to the location. This is usually done by means of so-called mechanical, optical or magnetic encoder.

By means of such a coder, the conversion of the intensity-time signal into an intensity-location signal takes place.

In the case of WO05 / 088533A1, for example, markers with a constant spacing of 300 microns are used to transform the intensity-time signal into an intensity-local signal (see WO05 / 088533A1 page 23). These markers are optically detected with a separate photodetector. Since the constant measuring frequency (sampling rate) and the distance of the markings are known, the location at which the focused scanning beam was located can be determined at any time. This makes it possible to transform the time-dependent scanning signal with the aid of the coder into a time-independent intensity-location signal.

Finally, the intensity-location signal (possibly after further conversions, filtering and / or data reductions) can be used as a characteristic fingerprint of the surface for later identification and / or authentication. The conversion of the intensity-time signal into an intensity-local signal using an optical encoder described in WO05 / 088533A1 has the disadvantage that markings have to be applied to the surface. Further, in WO05 / 088533A1, an additional photodetector is used to detect the marks and to make a conversion.

It would be desirable to be able to convert an intensity-time signal into an intensity-local signal without the need for additional markers and / or optical components solely for conversion.

Starting from the known prior art, therefore, the technical task of providing a method for optical scanning and recording an optical scanning signal, which manages without magnetic, mechanical or conventional, operating by means of markers optical encoder.

Surprisingly, it has been found that the intrinsic structures of many surfaces not only give a characteristic signal for identification and authentication but can themselves act to convert an intensity-time signal into an intensity-site signal.

Surprisingly, it has been found that in the optical scanning of the surface of a multiplicity of different objects, characteristic, recurring structures are present in the scanning signals, which permit a conversion of the time-dependent scanning signal into a time-independent signal without additional coder. Apparently, the surfaces of many objects, in addition to the already known from the prior art unique fine structure that generates characteristic scattered radiation, a further, longer-wave, characteristic structure, which is also referred to below as waviness. If an electromagnetic beam is passed over a surface at a constant speed and the surface is scanned at a constant measuring frequency, this waviness leads to a corresponding modulation of the scanning signal. This modulation allows a link between the measurement frequency and the path traveled during the scan.

Thus, it is not necessary to use markers or separate encoders to transform a time-dependent sample signal into a time-independent fingerprint.

The time-dependent sampling signal already carries all information about the conversion. The subject of the present invention is therefore a method for producing an identification feature of an article, comprising at least the following steps:

(a) directing an electromagnetic beam to a surface of the object,

(b) moving the electromagnetic beam and the object relative to each other such that the electromagnetic beam travels a path on the surface,

(c) picking up a portion of the radiation reflected from the surface of the article during relative movement as a function of time to obtain at least one scanning signal;

(d) extracting a time dependent ripple function from a sample signal, (e) transforming a sample signal into a time independent signal using the

Welligkeitsfunktion,

(f) determining an identifier from the time-independent signal,

(g) optionally associating the identifier with the item.

An object is understood to mean any solid body. The surface of the body separates it from the surrounding medium (mostly air).

By an identification feature is meant a characteristic of the object information that can be used for identification and / or authentication of the subject. The identifier is virtually a fingerprint of the item. In the present case, the identifying feature is information derived by optical methods from the characteristic surface structure of the article. The identification feature is preferably storable and machinable. By storable is meant that the identifier can be taken up again at a later date, for example for comparison purposes. By machine processing is meant that the identifier can be machine read and subjected to various computational and / or memory operations with a machine.

Identification is understood to mean a process that serves to uniquely recognize an object. Authentication is the process of verifying (verifying) an alleged identity. The authentication of objects is the statement that they are authentic - that is, they are unchanged, not copied and / or not faked originals. In steps (a) to (c), a scan is made of a surface area of the object for which an identification feature is to be generated.

The scanning of a surface area is carried out optically, that is using at least one source of electromagnetic radiation and at least one detector for electromagnetic radiation (also referred to as a photodetector). The radiation may be coherent or non-coherent. The radiation is preferably non-coherent if disturbing interference phenomena such as speckle patterns are to be avoided.

Preferably, electromagnetic radiation from the range of visible light or from the infrared range (380 nm to 2.5 μιη) is used.

The electromagnetic radiation may be poly- or monochromatic; preferably it is monochromatic.

By means of an electromagnetic beam, a surface area of an object is scanned. Part of the radiation reflected by the surface is detected by means of at least one photodetector. The detected signals contain information about the surface structure of the object. The surface structure of an article is unique and can be used to identify and / or authenticate the article.

Surface structure is the three-dimensional structure of the surface of an object understood (topography). The terms surface texture and topography are used synonymously here. For the purpose of creating an identifying feature, it is not necessary to take a three-dimensional topography (see below), for example, taking a surface profile is sufficient instead. A surface profile is the profile that results from the (imaginary) intersection of a surface of an object with a given plane (see, for example, DIN EN ISO 4287: 1998, Figure 2).

For optical scanning, the known method of dynamic laser focusing can be used (see, for example, Wochenblatt fur Papierfabrikation, ISSN0043-7131, Volume 117, April 1989, No. 7, pages 271 to 274). In the case of dynamic laser focusing, a laser is focused onto the surface by means of a lens. The lens can by means of a servomotor be moved perpendicular to the surface (in the z-direction). A sensor detects the respective z-position of the lens in a focused position and thus provides the topography information while the sample is moved through an xy-table under the lens.

However, the high accuracy with which the topography of the surface can be detected by means of dynamic laser focusing is not necessary for the purpose of producing a fingerprint. Surprisingly, it has been found that the prone mechanical readjustment of the lens can be dispensed with.

The scanning is preferably carried out without mechanical readjustment of the lens. FIG. 1 shows schematically how the scanning of a surface area can also be performed with the aid of a scanning beam.

Figure 1 shows the surface 1 of an article as well as an arrangement comprising a source of electromagnetic radiation 2 and a plurality of detectors 5 for electromagnetic radiation. The surface 1 is shown greatly enlarged for reasons of clarity in comparison to the radiation source 2 and the detectors 5. From the radiation source, a scanning beam 3 can be sent to the surface 1 of the object. The object is moved relative to the array of radiation source and detectors (indicated by the thick black arrow). The scanning beam passes over the surface. The scanning beam is reflected by the surface in accordance with the law of reflection. Depending on the curvature of the surface, the reflected radiation 4 passes into one of the detectors. In this way, the surface can be scanned and a scanning signal recorded.

Instead of the large number of individual detectors, it is also conceivable to use a correspondingly large detector (CCD, CMOS camera).

The irradiation (scanning) of the surface can take place at an arbitrary angle of almost 0 ° (if reflection still occurs) up to 90 ° relative to the mean surface level. The detection of the reflected radiation can also be carried out at an arbitrary angle of almost 0 ° to 90 ° relative to the mean surface level. Depending on the surface condition, it may be useful to detect directly reflected radiation (specular reflection) or scattered radiation (diffuse scattering). This can be determined by simple routine experiments. Decisive factors include the achieved signal-to-noise ratio, the reproducibility and the required positioning accuracy. Furthermore, it was surprisingly found that the determination of a characteristic fingerprint does not have to cover the complete surface structure during the scan. The surface texture of many objects is so rich in features that a fraction of it suffices for identification and / or authentication. This means that, in principle, a single one is sufficient in the arrangement in FIG. 1 instead of the multiplicity of detectors. This one detector then no longer detects any curvature of the surface but only the signals that are sent from the surface towards the detector. Surprisingly, however, the scanning signal detected by the detector is sufficient to produce a characteristic fingerprint for the purpose of identification and / or authentication.

In a preferred embodiment, the scan is taken along a single, preferably straight line. This means that the scanning beam is guided once in one direction (along a single line) over the surface of an object to pick up a scanning signal. Scanning along a single line can be much faster than scanning along multiple lines, for example, parallel to each other.

The direction of movement of the scan should be chosen so that it is not perpendicular to the waviness of the surface, since otherwise no correlation between the location and time of the scan can be made. The direction of the ripple can be determined empirically.

As the size of the scan area decreases, it becomes increasingly difficult to retrieve the corresponding area detected during the first scan in a later scan for the purpose of identification and / or authentication. This problem can be solved by using a linear beam profile for scanning. Surprisingly, it was found that even then a scanning signal and a characteristic fingerprint can be determined for the purpose of identifying and / or authenticating an object when the beam profile is widened transversely to the direction of movement. This is shown schematically in FIG. 2: A region 7 of a surface 1 of an object is irradiated by means of a source of electromagnetic radiation 2. Part of the reflected radiation 4 is picked up by a detector to pick up a scanning signal. The object is moved relative to the radiation source and detector assembly (represented by the thick black arrow). In the surface plane is a line-shaped beam profile, the longer extension is transverse to the direction of movement. By widening the beam profile in the direction transverse to the direction of movement, the problem of positioning is solved. Instead of a thin line (having a width corresponding to the extension of the dot-shaped beam profile), a wide area (having a width corresponding to the longer extension of the line-shaped beam profile) is scanned. This wide range can be found correspondingly easier in a later scan.

The scanning with a linear beam profile according to Figure 2 corresponds to an averaging over a plurality of scanning signals, resulting from the sampling with a point-shaped beam profile along a plurality of closely spaced and parallel lines. A linear beam profile is defined here as follows: Usually, the intensity in the cross-sectional center of the radiation is highest and decreases toward the outside. The intensity can decrease evenly in all directions - in this case there is a round cross-sectional profile. In all other cases there is at least one direction in which the intensity gradient is greatest and at least one direction in which the intensity gradient is smallest. In the following, the beam width is understood to mean the distance from the center of the cross-sectional profile in the direction of the smallest intensity gradient, at which the intensity has dropped to half of its value in the center. Furthermore, the beam thickness is understood to be the distance from the center of the cross-sectional profile in the direction of the highest intensity gradient, at which the intensity has dropped to half of its value in the center. A linear beam profile refers to a beam profile in which the beam width is greater than the beam thickness by a factor of more than 10. Preferably, the beam width is greater than the beam thickness by a factor of more than 50, more preferably by a factor of more than 80.

In a preferred embodiment, the beam thickness is in the range of the mean groove width of a profile element of the present surface (for the definition of the average groove width, see DIN EN ISO 4287: 1998).

For the scanning of most objects, such as objects made of paper, the beam thickness is usually in the range of 20 μιη to 100 μιη.

To compensate for torsion tolerances of the scanning beam, the jet thickness is preferably in the range of 30 μιη to 80 μιη, more preferably in the range of 40 μιη to 70 μιη, most preferably in the range of 50 μιη to 60 μιη. The beam width is preferably in the range of 2 mm to 6 mm, particularly preferably in the range of 3 mm to 5 mm. As explained above, the beam width is to find a compromise between signal-to-noise ratio and positioning accuracy.

It is known to the person skilled in the art how a corresponding beam profile can be generated for example by means of optical elements. Optical elements are used for beam shaping and focusing. In particular, lenses, diaphragms, diffractive optical elements and the like are referred to as optical elements.

In a scan, the scanning device and the object whose surface is to be scanned are preferably moved at a constant distance relative to each other. When using a linear beam profile for scanning a surface area, the beam width is transverse to the direction of movement. The angle between the direction of movement and the direction of the beam width is preferably between 10 ° and 90 °, more preferably between 45 ° and 90 °, most preferably between 70 ° and 90 °.

It is conceivable both a movement of the scanning device relative to the object and a movement of the object relative to the scanning device.

The movement may be continuous at a constant rate, accelerating or decelerating, or discontinuous, i. e.g. gradually. Preferably, the movement is carried out at a constant speed.

The radiation intensity incident on at least one detector is detected as a function of time. Usually, measuring signals are recorded and updated at a constant measuring frequency.

As already described, it is possible to detect several scanning signals in parallel by means of a corresponding number of detectors. It is also conceivable to scan a surface with several beams at the same time. When using multiple detectors for parallel recording of multiple scanning signals, it may be useful to detect directly reflected radiation and scattered radiation side by side. Our own experiments have shown that the structures in a scanning signal used to detect a ripple function and the structures used to produce a characteristic fingerprint appear more clearly in different reflection angle ranges (see below). Accordingly, the scanning signals mentioned in steps (d) and (e) of the method according to the invention can be one and the same scanning signal; However, it is also conceivable that the scanning signals mentioned in steps (d) and (e) are different scanning signals. In a preferred embodiment, a scanning signal is generated in step (c) by means of a detector. From this sampling signal, a ripple function is extracted in step (d). Then, the ripple function in step (e) is used to generate a time independent signal from the same sample signal.

In another preferred embodiment, in step (c) two scanning signals are detected by means of two different detectors. From one of the sample signals a ripple function is determined, which is then applied to the other sample signal to generate a time independent signal. Preferably, the different scanning signals are signals of differently reflected radiation, i. in one case of directly reflected radiation and in the other case of diffused radiation. The directly reflected radiation is particularly suitable for generating a ripple function, while the diffusely scattered radiation is particularly suitable for generating an identification feature.

In step (d) of the method according to the invention, as already mentioned, a filtering of a time-dependent scanning signal for determining a time-dependent ripple function takes place. As described above, a plurality of objects in the surface profiles in addition to the known fine structures that can be used for identification and / or authentication, characteristic, longer-wave structures. These can be used to correlate between time and place of the scan.

In mechanical engineering, the surface structure of a workpiece is treated as a shape deviation from the ideal geometric surface defined in the design. According to DIN EN ISO 4760, shape deviations are divided into six orders. The division into different orders of the shape deviation is based on the knowledge that the shape deviations of different orders have (and can) different origins. Thus, a shape deviation 1st order is referred to as a shape deviation and as possible causes of causes deflection and guide errors are specified in the machine tools. A 2nd-order shape deviation is referred to as waviness, and possible sources of origin are vibrations during production. A Shape deviation 3rd, 4th and 5th order is referred to as roughness and as possible causes of origin of the tool-cutting edge shape, the feed, chip formation, crystallization processes, chemical agents and corrosion are specified.

The assignment of existing in a measured surface profile shape deviations to the orders mentioned in practice by applying one or more profile filters. This approach is based on the model concept that a surface profile can be described from the superimposition of sine waves of different amplitude, wavelength and phase. Accordingly, the different shape deviations according to DIN EN ISO 4287: 1998 assigned wavelength bands. The present invention is based on the discovery that in a multiplicity of different objects there are wavelength bands in the surface profile which have characteristic, recurring structure which can be used to transform a time-dependent into a location-dependent signal. Besides, there are wavelength bands in the surface profile that have a characteristic structure that can be used as a fingerprint of an object.

The structures suitable for transformation lie in a longer wavelength range than the structures that can be used for identification. The structures suitable for the transformation are referred to here as a ripple function, since they often result in the sense of DIN EN ISO 4760 from a shape deviation of the second order, which is referred to as waviness. However, the terms undulation and ripple function in the sense of the present invention are not restricted to the terms of DIN EN ISO 4760. The term ripple should also not be understood to mean only periodic structures below. Instead, the ripple function in the scanning signal is characterized by structures which, on average, occur at specific intervals, which are characteristic of the respective object. Step (d) is for extracting the ripple function from the sampling signal. One possibility of extraction is the application of profile filters described in the standards DIN EN ISO 4287: 1998 and DIN EN ISO 11562: 1997.

A profile filter separates the profile into long-wave and short-wave components. The phase-correct Gauss filter defined in DIN EN ISO 11562: 1997 has established itself in the field of surface metrology and can also be used in step (d) of the method according to the invention. Roughness, waviness and primary profile equipment (see ISO 3274) use three Gaussian filters with the same transmission characteristics but different wavelengths: λβ-ΡΓθίΐΙιϊΙΐβπ defines the transition from roughness to proportions with even shorter wavelengths present on the surface. c -profile filter: defines the transition from roughness to waviness λί-ΡΓθίϊΙίϊΙΐβπ defines the transition from waviness to proportions with even longer ones Wavelengths that are present on the surface.

The ripple profile is the profile created by successively applying the λί and c profile filters to the primary profile (see DIN EN ISO 4287: 1998).

In a variety of articles studied, it has been found that there are structures in a short wavelength band that can serve to produce a characteristic fingerprint. For many objects, these structures would be referred to as roughness in the sense of DIN EN ISO 4760. These are usually structures that are due to the nature of the material that forms the object. This can e.g. be the fibrous structure of objects made of paper.

Furthermore, there are structures that lie in a longer wavelength band and that can be used to correlate the location and time of the sample. In the case of many objects, these structures would fall within the meaning of DIN EN ISO 4760 under the definition of waviness, which is usually impressed on it during the treatment or processing of the object.

Accordingly, the ripple function in a plurality of objects can be extracted from the time-varying sample signal by using two profile filters (Xf and λΰ-ΡΓθίϊΙίϊΙΐβΓ).

It can be determined in each case from the present data by applying different filters with different transmission characteristics and / or different wavelengths empirically, which filters must be applied to the in the scanning with a uniform, non-accelerated movement and with a constant sampling rate characteristic "ripple structures "and" fingerprint structures "to win. Likewise, it can be empirically determined in which reflection angle range the "fingerprint structures" and in which reflection angle range the "waviness structures" emerge more clearly. These investigations then reveal which reflected radiation (directly reflected and / or diffusely scattered radiation) and how many detectors are required to record the reflected radiation. In step (e), the ripple function is used to convert a time-dependent sample signal to a time-independent signal.

This step will be explained in more detail by way of example; It should be noted, however, that there are further possibilities to convert a time-dependent sampling signal by means of the ripple function into a time-independent signal.

It is assumed that there is a periodic ripple on the surface. In a sample with a uniform, non-accelerated motion and at a constant sampling rate, corresponding periodic structures are formed in the time-dependent sampling signal. If the same surface is scanned at a constant scanning rate but with an accelerated motion, the time-dependent scanning signal results in a structure whose wavelength decreases with the speed of the movement. To eliminate the effect of accelerated motion from the time-dependent signal, the scanning signal would have to be stretched along the time axis in the accelerated motion portions in a corresponding manner to regain a periodic structure. Accordingly, one possibility for converting the time-dependent scanning signal into a time-independent signal is to stretch and / or compress the determined ripple function in sections along the time axis such that two adjacent extrema (maxima or minima) are at a constant distance from each other. The result is a corrected, periodic ripple function. Subsequently, the time-dependent scanning signal is to be stretched and / or compressed in sections in the same way along the time axis. Thus, the time-dependent signal is projected onto the periodic structure of the corrected ripple function. The result is a time-independent scanning signal. This is preferably normalized by e.g. the distance of two extrema in the corrected ripple function an arbitrary value of e.g. 100 and divides the segmental stretched and / or compressed, former time axis of the sample signal into units of 100 * number of periods in the corrected ripple function.

If an object is scanned again at a later time and a standardized, time-independent scanning signal is generated from the time-dependent scanning signal in an analogous manner as described above, these are largely identical except for measurement errors and positioning inaccuracies, even if the later scan is performed with a another speed is done than the earlier scan. The identification feature can then be generated from the preferably normalized, time-independent scanning signal.

The described procedure is also successful when an article has a ripple that is not strictly periodic. The fact that the time-dependent scanning signal is transformed in the same way in each scan, always results in a largely identical signal that can be used for comparison purposes.

It is important that the same area is recorded for the most part during each scan. The applications WO09 / 097975A1, WO09 / 097980 and DE 0200923536.1 describe ways in which the area which was scanned during the so-called first detection can be found again in later scans for comparison purposes.

In a preferred embodiment of the method according to the invention, a marking on the object is used as trigger for the beginning of the scanning. For this purpose, the scanning beam is guided over the surface of the object and a part of the radiation reflected by the surface is detected by means of a photodetector. The mark on the surface of the object causes a change in the signal picked up by the photodetector. This signal change initiates the acquisition of the sample signal, i. from the occurrence of the signal change, the time-dependent sampling signal is recorded.

It is also conceivable to use a corresponding marking also for the end of the recording of the scanning signal by recording the scanning signal until a characteristic signal change stops the recording process.

The marking may, for example, be a sharp change in contrast which results, for example, from a transition of a black print to a white print. Due to the high absorption of the black printing, the intensity of the reflected radiation arriving at the photodetector is low. In the transition from black printing to white printing, the intensity of the reflected radiation increases abruptly, which can be used as a trigger to trigger the recording of the scanning signal.

Preferably, markers already present on the article are used. For this purpose, for example, optical codes (barcode, matrix code), logos, fonts but also edges are suitable. As already described above, the preferably standardized, time-independent scanning signal can be used directly as an identification feature. In this case, that will Identification feature of the preferably normalized, time-independent sampling signal in step (f equated.

As a rule, the identification feature in step (f) is generated from the time-independent sampling signal by various mathematical methods such as filtering and / or background subtraction. These mathematical methods eliminate as far as possible random or systematic fluctuations that can result from individual measurements. It is conceivable to remove the ripple function in the time-independent sampling signal by means of corresponding profile filters so that as far as possible only the characteristic structures for identification remain. In step (g) of the invention, the identification feature can be linked to the article. Such a link is typically made on the first scan of an item. The first scan to generate a first identifier is also referred to herein as registration. By means of the method according to the invention, a characteristic fingerprint is generated, which can be used in the form of preferably storable and machine processable data as a unique identifier for the object.

The link in step (g) can be physical or virtual. In the case of a physical linkage, the identification feature can be printed on the article or introduced into the article, for example in the form of an optical code (barcode, matrix code, OCR text or the like). It is also conceivable to associate the article with a sticker which contains the identification feature stored. The attachment of an electronic data carrier to the object, such as an RFID chip on which the identification feature is stored, is conceivable.

In the case of a virtual link, for example, a unique number assigned to the respective object (ID number, batch number or the like) is linked to the identification feature in a database. For example, the identifier may include this number in a header (metadata at the beginning of a file). The link ensures that there is a clear and unambiguous association between the identification feature and the object. The identification feature clearly indicates the associated item.

At a later time, an identifier of the item may be re-generated. This second identifier can be used to identify and authenticate the item. Details can be the following Applications are: WO09 / 097975A1, WO09 / 097974A1, WO09 / 097979A1 and WO09 / 097980A1.

The invention is explained in more detail below by means of examples, without, however, limiting them to them.

Show it:

Figure 1 (a), (b): Schematic representation for the optical scanning of a surface

Figure 2: Schematic representation for the optical scanning of a surface with a linear beam profile Figure 3: Schematic representation of a sensor according to the invention for the scanning of

surfaces

FIG. 1 shows schematically how the scanning of a surface area can be performed with the aid of a scanning beam.

The figure shows the surface 1 of an article as well as an arrangement comprising a source of electromagnetic radiation 2 and a plurality of electromagnetic radiation detectors 5. The surface 1 is shown greatly enlarged for reasons of clarity in comparison to the radiation source 2 and the detectors 5.

From the radiation source, a scanning beam 3 can be sent to the surface 1 of the object. The object is moved relative to the array of radiation source and detectors (indicated by the thick black arrow). The scanning beam passes over the surface. The scanning beam is reflected by the surface in accordance with the law of reflection. Depending on the curvature of the surface, the reflected radiation 4 passes into one of the detectors. In this way, the surface can be scanned and a scanning signal recorded. The surface structure can be determined from the scanning signal. Figure 2 shows a preferred method for scanning a surface. An area 7 of a surface 1 of an object is irradiated by means of a source of electromagnetic radiation 2. Part of the reflected radiation 4 is picked up by a detector to pick up a scanning signal. The object is related to the arrangement of radiation source and Detector moves (represented by the thick black arrow). In the surface plane is a line-shaped beam profile, the longer extension is transverse to the direction of movement.

FIG. 3 shows by way of example a part of a device (sensor) for scanning a surface. This sensor comprises a block 10 with a designated outer surface 15. This designated outer surface - hereinafter referred to as outer surface - is directed at the scanning on the surface of the corresponding object.

The block 10 serves to receive all optical components of the sensor according to the invention. It has at least two passages 11, 12 which converge towards the designated outer surface. The first feedthrough 11 extends at an angle γ with respect to the normal 16 of the outer surface (short outer surface normal) and serves to receive the source of electromagnetic radiation.

A second feedthrough 12 extends at an angle δ with respect to the outer surface normal 16 and serves to receive a photodetector. The amounts of the angles γ and δ are preferably the same.

The amounts of the angles γ and δ are in the range of 5 ° to 90 °, preferably in the range 20 ° to 80 °, more preferably in the range 30 ° to 70 °, most preferably in the range 40 ° to 60 °.

In a preferred embodiment of the sensor according to the invention, one or two further passages 13, 14 are provided, which serve to receive one or two further photodetectors. These are arranged at an angle 8i and / or ε ₂ to the second passage 12. The size of the angle ει and / or e ₂ is 1 ° to 20 °, preferably 5 ° to 15 °.

In the present example, the sensor is suitable for detecting both directly reflected and diffusely scattered radiation during the optical scanning of a surface. The directly reflected radiation is detected by means of a detector in the bushing 12, while the diffusely scattered radiation is detected by means of two detectors in the bushings 13 and 14.

Preferably, all bushings are in one plane to allow a compact design of the sensor. The use of a block with two to four feedthroughs for receiving a radiation source and one or more photodetectors offers the advantage that the optical components can be arranged in a simple manner, but nevertheless in a defined manner relative to one another. Preferably, a stop is located in the passage for the laser. Against this stop, the radiation source is pushed into the bushing, so that it assumes a predetermined fixed position with respect to the block and the photodetectors. The further feedthroughs for receiving photodetectors can also be provided with a stop.

The block can easily be e.g. be made by injection molding of plastic one or two pieces.

The sensor may have a housing into which the block is inserted. In the housing of the sensor, further components are preferably introduced, e.g. the control electronics for the radiation source, signal preprocessing electronics, complete evaluation electronics and the like. The housing preferably also serves to anchor a connection cable with which the sensor according to the invention can be connected to a control unit and / or a data acquisition unit for controlling the sensor and / or for detecting and further processing the characteristic reflection patterns.

Optionally, the sensor may have a window located in front of, behind or in the outer surface protecting the optical components from damage and contamination. Preferably, the window forms the outer surface of the sensor. The window is at least partially transparent at least for the wavelength of the radiation used.

The sensor in Figure 3 is further characterized in that the center axes of the feedthroughs intersect at a point 18 which is outside the block at a distance of 2 to 10 mm from the outer surface. For scanning the surface of an object, the sensor according to the invention is correspondingly guided at a distance above this object, so that the focal point and intersection of the central axes lie on the surface of the object.

In the mentioned distance range of 2 to 10 mm, the positioning of the surface to be scanned of an object with respect to the radiation source and the photodetectors is simple and sufficiently accurate. As the distance between the sensor and the object increases, the angle of the sensor with respect to the surface of the object must become increasingly accurate be adhered to in order to detect a predetermined area of the surface, so that the requirements for positioning increase.

Furthermore, the radiation intensity decreases with increasing distance from the radiation source, so that with an increasing distance between sensor and object the correspondingly reduced radiation intensity arriving at the object would have to be compensated by a higher power of the radiation source.

Reference numerals:

1 surface

2 source of electromagnetic radiation

3 scanning beam

4 reflected beam

5 photodetector

6 linear beam profile

7 scanned area

10 block

11 first implementation for receiving a radiation source

12 second passage for receiving a photodetector

13 further bushing for receiving a photodetector

14 more lead-through for receiving a photodetector

15 outer surface

16 exterior surface normals

18 focus point

Claims

A method of generating an identification feature of an article, comprising at least the following steps:

(a) directing an electromagnetic beam to a surface of the object,

(d) extracting a time dependent ripple function from a sample signal,

(e) transforming a sample signal into a time independent signal using the ripple function,

(f) determining an identifier from the time-independent signal,

(g) optionally associating the identifier with the item.

A method according to claim 1, characterized in that the electromagnetic beam is monochromatic and is in the visible or infrared range.

Method according to one of claims 1 to 2, characterized in that the beam profile is linear on the object.

A method according to claim 3, characterized in that the beam thickness in the range of 20 μιη to 100 μιη, preferably in the range of 30 μιη to 80 μιη, more preferably in the range of 40 μιη to 70 μιη, most preferably in the range of 50 μιη to 60μιη lies.

Method according to one of claims 1 to 4, characterized in that the ripple function is generated by applying two profile filters from the scanning signal, wherein the first profile filter longer-wave components and the second profile filter eliminates shorter-wave components.

Method according to one of Claims 1 to 5, characterized in that the scanning signals mentioned in steps (d) and (e) are one and the same scanning signal.

7. The method according to any one of claims 1 to 5, characterized in that in step (c) two scanning signals are detected, wherein one of the scanning signals a ripple function is extracted, which is then applied to the other sample signal to produce a time independent signal.

A method according to claim 7, characterized in that a scanning signal from directly reflected radiation results, while the other scanning signal results from diffused radiation.