Security marking, readout device and readout method

Abstract

The invention relates to an inspection device for optically inspecting objects for authenticity, in particular for inspecting whether objects comprise a security marking having known fluorescence properties, said security marking comprising a specified machine-readable code. The inspection device comprises an illumination unit which is designed and arranged to illuminate an object to be inspected with electromagnetic radiation, preferably light visible to the human eye, in a first, fluorescence-exciting wavelength range. The inspection device comprises an area sensor having a multiplicity of light-sensitive sensor elements and comprises an imaging optical unit which is designed and arranged to project an image of an object onto the area sensor, the area sensor being designed to output a picture signal which represents an image projected onto the area sensor. The imaging optical unit comprises an optical filter in the form of a high-pass or low-pass filter, in particular a cut-off filter, which is designed to block, in particular absorb, electromagnetic radiation in the first, fluorescence-exciting wavelength range and to transmit light in a second wavelength range, the second wavelength range comprising wavelengths which correspond to an expected fluorescence, and the second wavelength range not overlapping with the first wavelength range. The inspection device is designed to detect the input of an object, to inspect images of the object which are represented by picture signals of the area sensor, and to reject an object if one or more inspection criteria are not met, or to accept said object if the inspection criterion or the inspection criteria are not met. The inspection device is also designed to identify a security marking on an object and to inspect a found security marking with respect to the fluorescence properties of said security marking.

Description

[0001] Eisenführ Speiser

[0002] Berlin, November 7, 2025

[0003] Our reference: DB 1275-03WO JVO / adp

[0004] Addressee / owner: DPG Deutsche Pfandsystem GmbH

[0005] Official file number: New registration

[0006] DPG Deutsche Pfandsystem GmbH

[0007] Luisenstraße 46, 10117 Berlin

[0008] Security marking, reading device and reading procedure

[0009] The invention relates to a method and a device for testing objects with a security marking applied to them, as well as a system with several testing devices.

[0010] Such methods and devices are used, for example, in a deposit system to ensure that a deposit can be paid out upon return of a deposit-bearing item – e.g., packaging – even if the item is not returned to the place where it was originally purchased and where the deposit was originally paid. Since the deposit value, for example in the case of beverage packaging, can be several times the value of the packaging itself, it is known to provide security markings that are not easily forged and are intended to make the authenticity of, for example, a deposit mark verifiable.

[0011] In general, security markings are used in a variety of ways to make counterfeiting more difficult and to offer the best possible assurance of the authenticity of a document, product, banknote, or the like. Security markings are used on deposit-bearing products because the deposit value is typically higher than the value of the packaging itself.

[0012] *20250556852* In particular, it is known to apply a security marking – e.g., by printing it – to an outer covering of the packaging or a label or band of the packaging, which may be made of, for example, plastic, sheet metal, or cardboard. This security marking may be composed of various parts made of different materials, e.g., printed with different colors, so that the security marking ultimately has several adjacent fields with different reflection and / or emission properties. The reflection and / or emission properties of the fields may be wavelength-dependent, so that the optical appearance of these fields – in particular their relative brightness (intensity of the reflected light) in relation to each other – may depend on the wavelength range in which the security marking is illuminated.

[0013] Known security markings, for example, feature a contrast field with a comparatively high reflectivity across a broad illumination wavelength range, encompassing, for instance, visible and infrared light. Such a contrast field can be formed by a material or paint containing titanium dioxide, which appears white to the human eye in daylight. Security markings are also known that feature a second field, a dark field, which exhibits low reflectivity across the broad wavelength range compared to the contrast field and is formed, for example, by a material or paint containing carbon black, which appears black. A third field of a known security marking is a security field that exhibits different reflection properties in at least one a priori known illumination wavelength range than in another a priori known illumination wavelength range.For example, the safety field can have low reflectivity in a first illumination wavelength range visible to the human eye. In a second, different illumination wavelength range, visible or invisible to the human eye, the safety field, however, has higher reflectivity – or vice versa. The safety field is formed by a safety paint, preferably applied to a broadband highly reflective substrate, which exhibits higher absorption over a larger, visible wavelength range than over a second, longer wavelength range.In the second, longer wavelength range of light, the safety paint may, for example, exhibit high transparency, so that the reflective properties of the safety field are formed by the substrate beneath the safety paint, or the safety paint itself may have a comparatively high reflectivity in the second, longer wavelength range of light. Between the first and second wavelength ranges of light, the reflection, transmission, or absorption properties change by more than 50% in a transition wavelength range, starting from the lower of the two. The transition wavelength range is significantly narrower than either the first or the second wavelength range individually, and may be, for example, only one-fifth, one-tenth, or less than either the first or the second wavelength range individually.

[0014] In many cases, in addition to a security marking, an identification mark is also provided, which identifies the product type, manufacturer, or banknote, etc., e.g., indicating the manufacturer, product, or value of the banknote. The identification mark can be, for example, a GTIN represented in the form of a barcode or a QR code. GTIN stands for Global Trade Item Number, i.e., a globally recognized item number.

[0015] In addition to character components typically printed in black or in a dark color across a broad wavelength range, the background—that is, the area surrounding the dark-colored character components—is light, and especially white, to ensure good contrast. Typically, the light, particularly white, background of the identification marking's character components is formed by the unprinted substrate on which the dark character components are also printed. This is regularly the same substrate that also forms the light contrast area of ​​the security marking. Therefore, the contrast area (or areas) of the security marking and the background around the dark-colored character components of the identification marking regularly share the same light, e.g., white, color.White color, because the background color—that is, the color adjacent to the printed elements—is usually the color of the surface not printed with any characters. This surface can itself be white ink applied—for example, printed—to a substrate such as a transparent film or the metal of a beverage can. However, the background can also be the substrate itself—for example, white paper.

[0016] The dark-colored elements of the identification marking and the dark areas of the security marking are usually printed with the same broadband dark color, e.g., soot black, on a light background.

[0017] The reflectivity of each field for a given wavelength depends on the ink used to print it on the background, or on the background material itself, which, for example, can form part of the security marking. Typically, the ink used to print the contrast field(s) is a broadband reflector, while the ink used to print the dark field(s) is a broadband absorber. The contrast field, in particular, can also be formed by the background itself if it is broadband reflector, such as white paper.

[0018] The ink used to print the safety markings onto the light background—hereinafter referred to as "safety ink"—has a higher absorption rate in a first wavelength range than in a second wavelength range. Accordingly, the safety ink has a higher reflectivity or higher transparency, or both, in the second wavelength range. If the safety ink has a higher reflectivity in the second wavelength range than in the first, the intensity of the reflected light in the second wavelength range is always greater than the intensity of the reflected light in the first wavelength range.If the safety paint has higher transparency in the second wavelength range than in the first, the intensity of the reflected light in the second wavelength range will be greater than the intensity of the reflected light in the first wavelength range, provided the substrate beneath the safety paint has sufficiently high reflectivity in this second wavelength range. If the safety paint itself already has high reflectivity in the second wavelength range, the reflectivity of the substrate beneath the safety paint becomes less important or even irrelevant.

[0019] One way to test a security marking of the type described is to illuminate the security marking first with light in a first illumination wavelength range and second with light in a second illumination wavelength range.

[0020] German patent applications DE 43 19 555 and DE 102 47 252 each describe a safety field printed with an ink that strongly absorbs visible light and is transparent to infrared light. The reflectivity of the safety field in the infrared wavelength range is determined by the substrate beneath the ink. The substrate is white, so that under infrared light the safety field appears just as white as the substrate and its surroundings. This is because the ink used to print the safety field is intended to be transparent to infrared light and therefore invisible, allowing the substrate beneath the ink to be visible under infrared light.Whether the safety marking itself is visible or invisible in this case depends on whether the background beneath the ink used to print the safety marking is the same color as the surrounding area. It is also possible that the background beneath the ink used to print the safety marking is a different color than the surrounding area. If the ink used to print the safety marking appears the same as the surrounding area in a first illumination wavelength range, the safety marking may only become visible when illuminated in a second illumination wavelength range where the ink used to print the safety marking is transparent.

[0021] Another test method focuses on determining the degree to which the safety field reflects more strongly in the second illumination wavelength range than in the first. This method has the advantage that the test of the safety field does not need to reference the environment of the safety field or any reference field that has the same color as the substrate on which the color for the safety field is printed.

[0022] The other testing method has the disadvantage that the intensity of the light reflected by the safety field depends on the intensity of the illumination – hereinafter also referred to as illuminance. The illuminance, in turn, depends not only on the intensity of a light source, but also, for example, on the distance between the safety field being tested and the light source, or on the angle at which the light from the light source strikes the safety field.

[0023] The security mark is typically located on a valuable item, such as a beverage container with a deposit. When the valuable item is handed over, for example, a deposit-bearing beverage container is returned, the value (i.e., the deposit) is paid out, and the disbursing entity calculates the payout amount – to whom, for example, is determined by an identification mark such as a standard GTIN (Global Trade Item Number).

[0024] Since there have been cases where attempts have been made to combine an authentic security mark with a false identification mark in order to calculate the deposit paid out to the wrong entity, WO 2020 / 039091 A1 proposed checking whether the authentic security mark and the identification mark are printed on the same background, or whether the color of the background of the security mark differs from the color of the background of the identification mark.

[0025] The invention is based on the objective of providing an improved testing device and an improved testing method.

[0026] According to the invention, this problem is solved by a test device comprising an area sensor with a plurality of light-sensitive sensor elements and an imaging optic designed and arranged to project an image of an object and – if present on the object – a security marking and / or an identification mark onto the area sensor. During operation, the area sensor provides image signals representing the image projected onto the area sensor and is connected to an electronic image evaluation unit for processing the image signals, which may include a preprocessing unit and an evaluation unit. The evaluation unit may, in turn, include a pre-evaluation unit and an evaluation unit.The test device is designed to detect the input of an object, to check images of the object based on the image signals, and to reject an object if one or more test criteria are not met, or to accept it if the test criterion or criteria are not met.

[0027] According to a first aspect of the invention, a testing device for optically checking objects for authenticity, in particular for checking whether objects have a predetermined, machine-readable optical code-encompassing security marking with known fluorescence properties, is proposed, wherein the testing device comprises the following components: an illumination unit configured and arranged to illuminate an object to be tested with electromagnetic radiation – preferably light visible to the human eye – in a fluorescence-stimulating excitation wavelength range; an area sensor with a plurality of light-sensitive sensor elements; and an imaging optic configured and arranged to project an image of an object onto the area sensor, wherein the area sensor is configured to output an image signal representing an image projected onto the area sensor.and an optical filter in the form of a high-pass or low-pass filter, in particular an edge filter, configured to block electromagnetic radiation in a blocking wavelength range encompassing the excitation wavelength range, in particular to absorb it, and to transmit light in a transmission wavelength range, wherein the transmission wavelength range includes wavelengths corresponding to expected fluorescence and wherein the transmission wavelength range does not overlap with the excitation wavelength range.

[0028] The terms "fluorescence-exciting wavelength range", "excitation wavelength range" and "fluorescence-exciting excitation wavelength range" are used synonymously in the following and refer to a wavelength range of electromagnetic radiation, in particular of light, in which fluorescence of an authentic security color can be excited by means of electromagnetic radiation.

[0029] The test device is designed to detect the input of an object, to check images of the object represented by image signals from the area sensor, and to reject an object if one or more test criteria are not met, or to accept it if the test criterion or criteria are met.

[0030] The testing device is also designed to check whether a detected security marking has a machine-readable optical code that can be interpreted when illuminated with electromagnetic radiation in a fluorescence-exciting excitation wavelength range for authentic security paint, wherein the intensity of the illumination with electromagnetic radiation in the excitation wavelength range is several times greater – at least twice as much – than the radiation incident on an object to be tested in the transmission wavelength range of the optical filter.

[0031] The testing device can be configured to generate a signal rejecting the object if the authenticity check is negative, and / or to generate a signal accepting the object if the authenticity check is positive. Preferably, the optical filter is part of the imaging optics and is a low-pass filter whose transmission wavelength range includes wavelengths larger than its blocking wavelength range.

[0032] The optical filter can be arranged in such a way that it can be selectively removed from or inserted into the beam path of the imaging optics, so that it is selectively ineffective or effective.

[0033] Preferably, the test device has an electronic image evaluation unit that is operatively connected to the area sensor in order to receive at least indirectly raw or pre-processed image signals from the area sensor, wherein the image evaluation unit has a code reading component that is configured to identify and interpret machine-readable codes represented by image signals.

[0034] The testing device is preferably configured such that it generates an authentically characterizing signal of an object to be tested if the code reading component identifies a machine-readable code in the image of a security marking-like image component represented by the image signals when illuminated with electromagnetic radiation encompassing the excitation wavelength range and with an effective optical filter.

[0035] According to a second aspect of the invention, the testing device is also configured to take two photographs of an object to be tested, wherein the object is illuminated for one photograph with electromagnetic radiation in the fluorescence-stimulating excitation wavelength range and for another photograph with electromagnetic radiation in the transition wavelength range in which the absorption properties of an authentic security color change significantly.

[0036] In the latter case, the test device is preferably designed to check a respective image of the security marking recorded with electromagnetic radiation in the transition wavelength range with regard to the contrast between light and dark character components by generating a contrast value representing the contrast between light and dark character components and comparing this with a reference value.

[0037] Preferably, the test device is designed to first detect a safety marking in the images of an object projected onto the area sensor, and to trigger a rejection of the object if no image of a safety marking is found in the images of the object projected onto the area sensor.

[0038] Preferably, the test device is designed to check, after detecting a security mark in the images of the object projected onto the area sensor, whether the signal-to-noise ratio or the dynamic range of the image signals representing the images of the object projected onto the area sensor exceeds a predefined value, and to reject the object to be tested if the signal-to-noise ratio or the dynamic range of the image signals representing the images of the object projected onto the area sensor does not reach or exceeds the corresponding predefined value.

[0039] Preferably, the test device is designed to always reject an object if the signal-to-noise ratio or the dynamic range of the image signals representing the images of a currently tested object projected onto the area sensor reaches or exceeds the specified value.

[0040] Another aspect is a method for optically checking objects for authenticity, whereby the method comprises the following steps:

[0041] Recording the entry of an object to be checked,

[0042] Projecting an image of an object onto an area sensor and generating and outputting at least one image signal representing an image of the object projected onto the area sensor,

[0043] Search for an image of a security mark in the image of the object projected onto the area sensor.

[0044] Determining the signal-to-noise ratio or a signal dynamic range value of the image signal representing the image of the object,

[0045] Verification of the authenticity of the security marking based on the intensity values ​​of the security marking components represented by the image signal. This verification preferably includes at least one fluorescence test, in which a detected security marking is illuminated with electromagnetic radiation suitable for inducing fluorescence in an authentic security color. Furthermore, the verification may also include a reflection test, in which the reflective properties of security marking components are examined at predetermined wavelengths. The fluorescence test and the reflection test can be performed sequentially and in any order.If a security mark being checked fails one of the authentication checks, a corresponding signal indicating failure of the authentication check is generated, which may at least indirectly result in the rejection of an object being checked.

[0046] The evaluation unit is designed to apply various test criteria to an image of an object represented by image signals. These test criteria include:

[0047] The presence of a safety mark on the object projected onto the area sensor, and / or a sufficient signal-to-noise ratio, i.e., fulfilling a specified minimum criterion, or a dynamic value of the intensity value dynamic of an image signal representing an object that exceeds a specified value.

[0048] Verifying the authenticity of a detected security mark involves a fluorescence test with the following steps:

[0049] Illuminating an object to be tested with electromagnetic radiation in an excitation wavelength range suitable for exciting the fluorescence of an authentic security paint,

[0050] Capturing an image of an object to be inspected with an image acquisition unit in which a filter blocking electromagnetic radiation in the excitation wavelength range prevents the capture of electromagnetic radiation in the excitation wavelength range, and checking whether the image of the security marking captured with electromagnetic radiation in the excitation wavelength range contains a decodable optical code.

[0051] Preferably the method additionally includes the step

[0052] Generating a signal indicating a failed test if, when searching for an image of a security mark, no image of a security mark is bound and / or if the signal-to-noise ratio or the signal dynamic range value of the image signal representing the image of the object does not meet a specified minimum criterion and / or if a detected security mark fails the authenticity test.

[0053] The examination of a discovered security mark can therefore include the following:

[0054] Illuminating an object to be tested with electromagnetic radiation in an excitation wavelength range suitable for exciting the fluorescence of an authentic security paint,

[0055] Capturing an image of an object to be inspected with an image acquisition unit in which a filter blocking electromagnetic radiation in the excitation wavelength range prevents the capture of electromagnetic radiation in the excitation wavelength range, and

[0056] Rejection of an object to be inspected if an optical code of the security marking is not recognizable and / or not decodable in an image taken using electromagnetic radiation in the excitation wavelength range.

[0057] In a particularly preferred variant, verifying the authenticity of a discovered security mark includes the following steps:

[0058] Illuminating a detected security mark with electromagnetic radiation in an excitation wavelength range that excites fluorescence in authentic security paint, checking whether the security mark contains an interpretable machine-readable optical code when illuminated with electromagnetic radiation in an excitation wavelength range that excites fluorescence in authentic security paint, and, if no interpretable machine-readable optical code is found, generating a signal indicating that the security mark has failed the authenticity check.

[0059] These steps are part of a fluorescence test, i.e., a test to see if a marking contains components that fluoresce when authentic security ink is used under specified conditions.

[0060] Additionally, verifying the authenticity of a detected security mark preferably includes a reflection test with the following steps:

[0061] Illuminating the discovered security marker with electromagnetic radiation in the transition wavelength range, and

[0062] Determining the ratio of incident illumination intensity to the intensity of light reflected by character components printed with security ink in authentic security markings, and checking whether the ratio of incident illumination intensity to the intensity of light reflected by character components printed with security ink in authentic security markings meets a predefined test criterion.

[0063] Checking whether the ratio of incident illumination intensity and the intensity of light reflected by character components printed with security ink in authentic security markings meets a predetermined test criterion may include a comparison with a predetermined reference value.

[0064] In the event of a failed reflection test, a signal indicating a failed authentication test is preferably also generated, leading to the rejection of the object being tested. If a security mark passes both the fluorescence test and the fluorescence test, a signal indicating a passed authentication test is preferably generated.

[0065] The advantages of an additional reflection test are explained below in connection with Figures 12 to 16.

[0066] Regarding a sufficient signal-to-noise ratio, a rejection occurs if no image of the object under test, represented by image signals, exhibits a sufficient signal-to-noise ratio and / or sufficient signal dynamics. The latter is the case, for example, if the dynamic range of the intensity value does not reach or exceeds a corresponding predefined value.

[0067] The testing device is preferably part of a return system for deposit-bearing packaging. Accordingly, the objects to be tested are, for example, packaging which – if they are genuine deposit-bearing packaging – should bear a security marking that must be located and verified.

[0068] Preferably, an assessment of the authenticity of a detected security mark is only carried out after testing against the test criteria.

[0069] The presence of a security marking and an identification mark on the object projected onto the area sensor, and a sufficient signal-to-noise ratio of an image signal representing an object that meets a specified minimum criterion, has shown that these test criteria are met.

[0070] For checking the test criteria

[0071] Provided that a security mark is present on the object projected onto the area sensor, and that the signal-to-noise ratio of an image signal representing an object is sufficient (i.e., meets a predetermined minimum criterion), the test device preferably includes a pre-evaluation unit connected to a separate evaluation unit. The pre-evaluation unit and the evaluation unit are part of the image evaluation unit. The evaluation unit is configured to evaluate representative intensity values ​​derived from the image signals representing an image of a security mark for the purpose of authenticity verification.

[0072] Thus, the test device is preferably designed to use the pre-evaluation unit to check whether the aforementioned test criteria are met, and only if these are met, to use the evaluation unit to check the authenticity of the detected and recognized security marking.

[0073] A test device for the optical verification of objects for authenticity is thus proposed. The test device comprises an area sensor with a plurality of light-sensitive sensor elements and an imaging optic. It is configured and arranged to project an image of an object onto the area sensor, and the area sensor is configured to output at least one image signal representing an image projected onto the area sensor. The area sensor is connected to an electronic image evaluation unit. The test device is configured to detect the input of an object, to verify images of the object, and to reject an object if one or more test criteria are not met, or to accept it if the test criterion(s) are not met.For this purpose, the image processing unit is designed to apply at least one of the aforementioned test criteria to an image of an object represented by image signals. If one of these test criteria is not met, a rejection is triggered.

[0074] One test criterion is the readability or non-readability of a machine-readable optical code depending on the presence or absence of fluorescence in the aforementioned fluorescence test. Specifically, the test device verifies whether a detected security marking has a machine-readable optical code that can be interpreted when illuminated with electromagnetic radiation in a wavelength range that excites fluorescence in authentic security paint, where the intensity of the electromagnetic illumination in the excitation wavelength range is several times greater – at least twice – than the radiation incident on the object being tested in the transmission wavelength range of the optical filter.An optical code is readable if it can be decoded using known methods by converting the information encoded by the image components of the optical code into a digital signal representing the information, specifically a string of characters. A character similar to an optical code is unreadable and therefore undecipherable if an image of the character cannot be converted into a corresponding digital signal representing a string of characters.

[0075] It is particularly preferred if the testing device is designed to first detect a safety marking in the images of an object projected onto the area sensor, and to trigger a rejection of the object if no image of a safety marking is found in the images of the object projected onto the area sensor.

[0076] The pre-evaluation unit can be configured to identify different fields in an image of a security marking as planar components of the security marking with approximately the same brightness (i.e., with approximately the same intensity values ​​of the pixels), to generate a representative intensity value for each of the identified fields in the image of the security marking, and to transmit the representative intensity values ​​thus generated for the different fields of the security marking to the evaluation unit.

[0077] Preferably, the evaluation unit is designed to verify the authenticity of the security marking based on representative intensity values.

[0078] An optical inspection procedure for optically verifying the authenticity of objects comprises the following steps:

[0079] Recording the entry of an object to be checked,

[0080] Illuminating the object to be tested with illumination that is expected to excite fluorescence, i.e., with fluorescence-exciting electromagnetic radiation in the excitation wavelength range,

[0081] Projecting an image of an object onto an area sensor and generating and outputting at least one image signal that represents an image of the object projected onto the area sensor in the form of an intensity value or color image under a given illumination,

[0082] Searching for an image of a security mark in the image of the object; determining the signal-to-noise ratio and / or the intensity value dynamics of the intensity value or color image representing the image of the object, represented by the image signal.

[0083] Verification of the authenticity of the security marking based on intensity values ​​represented by the image signal for various area components of the security marking.

[0084] The search for an image of a security marking with a machine-readable optical code in the image of the object to be inspected, projected onto the area sensor, is preferably carried out with illumination that is not blocked by the optical filter or without using the optical filter.

[0085] If no image of a security mark is found in the intensity value or color image of the object's image projected onto the area sensor, or if the signal-to-noise ratio and / or the intensity value dynamics of the intensity value or color image representing the object's image do not reach or exceed a specified minimum value, or if the verification of the security mark's authenticity does not result in the security mark meeting the verification criteria for authenticity, the object being tested will be rejected, and in the case of a seized item, no pawn payment will be made.

[0086] In particular, an object to be inspected will be rejected if an optical code of the security marking cannot be decoded in the image captured under – essentially exclusively – fluorescently exciting illumination and the application of an optical filter (edge ​​filter) that blocks the fluorescently exciting illumination in front of the area sensor. To verify the authenticity of the security marking, it is always illuminated with electromagnetic radiation in the excitation wavelength range – i.e., with fluorescently exciting illumination – and the optical filter that blocks the fluorescently exciting illumination – i.e., the edge filter – is applied in front of the area sensor to capture the image of the security marking.Fluorescence-stimulating illumination is illumination with electromagnetic radiation in the excitation wavelength range in which an authentic safety paint can be excited to fluorescence; "fluorescence-stimulating" and "excitation wavelength range" thus refer to the fluorescence properties of an authentic safety paint.

[0087] The intensity value or signal dynamics of the intensity value or color image representing the image of the object describes the contrast range of the intensity value or color image, i.e., the ratio of highest to lowest intensity values ​​(signal values) in the respective intensity value or color image.

[0088] Preferably, several intensity value or color images of an object under test are recorded in a sequence, and the object is rejected if none of the intensity value or color images representing the object exhibits a sufficient signal-to-noise ratio and / or sufficient intensity value dynamics. If an object is rejected due to insufficient signal-to-noise ratio and / or insufficient intensity value dynamics, the test is terminated, and the authenticity of the security marking is no longer verified based on intensity values ​​for various areas of the security marking represented by the image signal. This relieves the burden on a separate evaluation unit, preferably one specifically designed for verifying the authenticity of the security marking based on intensity values ​​for various areas of the security marking represented by the image signal.To verify the authenticity of the security marking based on intensity values ​​represented by the image signal for various area components of the security marking, the evaluation unit may execute a more complex test algorithm.

[0089] Preferably, an object to be inspected is rejected if no image of a security marking is found in the intensity value or color image of the object projected onto the area sensor. In this case, the inspection is also terminated, and the authenticity of the security marking is no longer verified based on intensity values ​​for various area components of the security marking represented by the image signal.

[0090] Preferably, a completeness check is also performed for the security marking. This check verifies whether the security marking is at least approximately complete. If, for example, less than 80% of a detected security marking is recognized, it is rejected. In this case, further testing is only carried out if at least 80% of the security marking's area is recognized. The completeness check is therefore preferably performed by determining the area. Since the depicted security markings are not always shown to the same scale, the area can be checked relative to the marking by using, for example, a fully known marking component as a reference. According to a preferred variant of the completeness check, it is thus verified whether a minimum area fraction of a detected security marking has actually been found.The minimum area coverage can be, for example, 80% or 90% of the area of ​​a complete security mark. This prevents fraud, such as when halves of a genuine security mark are affixed to two different packages.

[0091] According to a further preferred embodiment, the pre-evaluation unit is configured as a shape evaluation unit to evaluate a first image signal representing an image of an identification mark in such a way that the pre-evaluation unit outputs a value or text represented by a machine-readable optical code such as a barcode, a data matrix code or QR code, i.e. decodes the barcode, data matrix code or QR code.

[0092] The area sensor is preferably part of a broadband-sensitive intensity image acquisition unit. For this purpose, the area sensor is sensitive at least in the fluorescence wavelength range and the second wavelength range and is configured to deliver intensity images – preferably intensity value images – of the article.

[0093] Preferably, the evaluation unit is configured to identify an image of a security marker in intensity value images acquired by the intensity image acquisition unit. This identification of the security marker image is preferably performed by the pre-evaluation unit, provided the evaluation unit includes a pre-evaluation unit and an evaluation module.

[0094] The testing device preferably comprises an image acquisition unit with an area sensor containing light-sensitive sensor elements, which are preferably arranged in a matrix. The image acquisition unit with area sensor serves to capture an image projected onto the sensor in two dimensions. For this purpose, optics are positioned in front of the area sensor, which project the image of the respective safety marking as sharply as possible onto the area sensor.

[0095] It is also possible to use an area sensor whose sensor elements are not sensitive to illumination in the fluorescence-excitating (excitation) wavelength range. In this case, the optical filter of the imaging optics can be omitted.

[0096] Accordingly, in a preferred embodiment, the test device comprises at least one illumination unit designed and arranged to illuminate a viewing area of ​​the image acquisition unit simultaneously or alternately with light in the fluorescence-excitating wavelength range and with light in the transition wavelength range in which the security ink has an average reflectivity that is higher than in the first wavelength range and lower than in the second wavelength range. More than two wavelength ranges may also be provided, and an illumination scenario may consist of illumination with light from a single wavelength range or with light composed of a combination of different wavelength ranges. The viewing area of ​​the image acquisition unit refers to the space in which the security marking of a package is located when its image is sharply projected onto the area sensor.Two or more lighting modules can also be provided, one for light in the fluorescence-stimulating wavelength range and one for light in the transition wavelength range, which are switched on alternately or operated simultaneously.

[0097] The lighting unit is preferably designed such that the viewing area of ​​the image capture unit, in which an evaluative package is located, is illuminated so uniformly that the intensity difference across the viewing area is at most 25%.

[0098] Furthermore, the lighting unit is preferably arranged such that the illumination angle, relative to a surface normal of the safety marking to be illuminated, is between 20° and 45°.

[0099] The image acquisition unit with associated optics for imaging a safety marking to be evaluated on the area sensor is preferably designed such that 1 mm 2The safety marking is detected by at least four complete sensor elements (pixels). The invention will now be explained in more detail with reference to an exemplary embodiment and the figures:

[0100] Figure 1: shows an example of packaging with a known security marking;

[0101] Figure 2: shows an embodiment of a known security marking;

[0102] Figure 3: shows how a security marking according to the invention can replace a known security marking shown in Figure 2;

[0103] Figure 4: illustrates, by way of example, the reflectivity of the safety paint, possibly on a broadband reflective surface;

[0104] Figure 5: illustrates the principle of fluorescence;

[0105] Figure 6: illustrates the fluorescence of the safety paint in the fluorescence wavelength range when the fluorescence is excited by illumination with electromagnetic radiation in the excitation wavelength range;

[0106] Figure 7: illustrates, by way of example and schematically, the transmission properties of an optical low-pass filter;

[0107] Figure 8: illustrates electromagnetic radiation emitted by a broadband light source in an illumination wavelength range that includes an excitation wavelength range, and the resulting fluorescence;

[0108] Figure 9: shows a security marking with a Data Matrix code with security color as a background, in two variants “Normal” with black color for the Data Matrix code (Figure 9a) and “Inverse” with white color for the Data Matrix code (Figure 9b);

[0109] Figure 10: schematically shows the image of a security marking in the form of a Data Matrix code, in which the typically light-colored components of a Data Matrix code are printed with security ink (top, Fig. 10a) and once without security ink (bottom, Fig. 10b); Figure 11: schematically shows the image of a security marking in the form of a Data Matrix code, in which the typically light-colored components of a Data Matrix code are printed with security ink (top, Fig. 11a) and once without security ink (bottom, Fig. 11b);

[0110] Figure 12: illustrates the illumination of a safety marking with electromagnetic radiation in the transition wavelength range where the absorption of the safety paint is lower than in the first wavelength range, but higher than in the second wavelength range, and the resulting reflection of the safety paint;

[0111] Figure 13: illustrates the spectral conditions when a security marking according to the invention is illuminated not only with electromagnetic radiation in the excitation wavelength range, but also with electromagnetic radiation in the transition wavelength range, in which the absorption of the security color is lower than in the first wavelength range, but higher than in the second wavelength range;

[0112] Figure 14: schematically shows the image of a security marking in the form of a Data Matrix code, in which the typically light components of a Data Matrix code are printed with security ink (top, Fig. 14a) and once printed without security ink (bottom, Fig. 14b), when illuminated with electromagnetic radiation in the excitation wavelength range and the image of the security marking when illuminated with electromagnetic radiation in the transition wavelength range;

[0113] Figure 15: schematically shows the image of a security marking in the form of a Data Matrix code, in which the typically light components of a Data Matrix code are printed with security ink (top, Fig. 15a) and once printed without security ink (bottom, Fig. 15b), when illuminated with electromagnetic radiation in the excitation wavelength range and the image of the security marking when illuminated with electromagnetic radiation in the transition wavelength range;

[0114] Figure 16: schematically shows the image of a security marking in the form of a Data Matrix code, in which the typically light components of a Data Matrix code are printed with security ink (top, Fig. 16a) and once printed without security ink (bottom, Fig. 16b), when illuminated with electromagnetic radiation in the excitation wavelength range and the image of the security marking when illuminated with electromagnetic radiation in the transition wavelength range;

[0115] Figure 17: shows different variants of a security marking in the form of a Data Matrix code, which represents further attributes in addition to an identification mark;

[0116] Figure 18: shows a schematic representation of a reading unit according to the invention with a pre-test unit and a main test module;

[0117] Figure 19: illustrates sensor units for fluorescence testing (Figure 19a) and a combined fluorescence and reflection testing (Figure 19b);

[0118] Figure 20: illustrates details of the sensor units for fluorescence testing (Figure 20a) and a combined fluorescence and reflection testing (Figure 20b);

[0119] Figure 21: is a schematic flowchart of a procedure for verifying the authenticity of a security mark;

[0120] Figure 22: illustrates the arrangement of several lighting units and an image capture unit in a take-back device; and

[0121] Figure 23: illustrates a system with multiple test devices connected to a central server.

[0122] Safety marking

[0123] Figure 1 shows an example of packaging 10 in the form of a can with a conventional security mark 12 and an identification mark 24, which in this example is in the form of a barcode. The security mark 12 serves to identify the packaging 10 as packaging for which a deposit must be paid when purchased by a consumer, and which the consumer receives back upon return of the packaging. The security mark is designed in such a way that it is not readily possible to equip packaging for which no deposit has been paid with the security mark. Since the deposit value is greater than the value of the packaging, the party accepting the returned packaging and paying out the deposit would incur a loss in the case of packaging with a counterfeit security mark.

[0124] Figure 2 shows the essential features of the conventional security marking 12, namely a comparatively highly reflective contrast field 14, which surrounds a security field 16, and a signal field 18. The contrast field 14 is highly reflective over a broad wavelength range, particularly in the visible wavelength range of light and in the transition to the infrared wavelength range. The different areas of the security marking 12 are also referred to as fields in this description.

[0125] The safety field 16 has the property that it is weakly reflective in a first, preferably visible wavelength range of light, i.e. strongly absorbing and therefore appears dark.

[0126] In a second, preferably also visible, wavelength range of light, the security field 16 is highly reflective, for example, as highly reflective as the contrast field 14. The security field 16 acquires this property of different reflectivity at different wavelengths because the ink with which the security field 16 is printed has lower absorption in the second wavelength range than in the first wavelength range.

[0127] This means that when viewing the packaging 10 in the first wavelength range, e.g. in normal daylight, the security field 16 is clearly visible as a dark field against a light background, while when viewing in the second wavelength range, for example with the help of a suitable camera, the security field 16 is less visible, since the security field 16 has a higher reflectivity in the second wavelength range, which is similar to that of the contrast field 14.

[0128] The reflectivity of each field for a given wavelength—and thus the intensity with which light in a given wavelength range is reflected—depends on the ink used to print the field onto the substrate. Typically, the ink used to print contrast field 14, or the contrast fields themselves, is a broadband reflecting ink, while the ink used to print any dark fields 20 and 22 is a broadband absorbing ink. Contrast field 14 can also be formed by the substrate itself if it is broadband reflecting, meaning that contrast field 14 does not necessarily have to be printed.

[0129] The ink used to print security field 16 is also known as security ink and has, for example, higher absorption in the first wavelength range than in the second. Accordingly, the security ink has higher reflectivity or higher transparency, or both, in the second wavelength range than in the first. If the security ink has higher reflectivity in the second wavelength range than in the first, the intensity of the reflected light in the second wavelength range is always greater than the intensity of the reflected light in the first wavelength range.If the security ink has higher transparency in the second wavelength range than in the first, the intensity of the reflected light in the second wavelength range will be greater than the intensity of the reflected light in the first wavelength range, provided the substrate beneath the ink used to print security field 16 has sufficiently high reflectivity in this second wavelength range. If the security ink itself already has high reflectivity in the second wavelength range, the reflectivity of the substrate beneath the ink used to print security field 16 becomes less important or even irrelevant.

[0130] For security reasons, however, it is advantageous if the substrate beneath the security ink has a reflectivity that differs from that of the contrast field 14. This means that the security field 16 can be printed with two colors: first, with a first color having reflective properties that differ from those of the contrast field 14, and then with a second color, such that the second color covers the first. The second color with which the security field 16 is printed is then the security ink, which, as described above, has a higher reflectivity and / or transparency in the second wavelength range than in the first wavelength range. According to the invention, it is provided that a security marking is included that simultaneously functions as an identification mark and comprises a machine-readable optical code.The identification mark can be represented by the machine-readable code, eliminating the need for a separate identification mark. This is illustrated in Figure 3; see Figure 3a), the transition from I. to II. The security mark therefore contains a machine-readable optical code, such as a barcode, a QR code, or a Data Matrix code. The machine-readable optical code contains code components printed with security ink, which can represent, among other things, an identification mark. In addition to the security mark, symbols can be provided, for example, to make it easy for users to recognize which deposit system an object belongs to; see Figure 3b), II. and III.

[0131] The optical code is preferably a 2D code, such as a QR code or a Data Matrix code, with the latter being preferred. Regarding the readability of the optical code, both the optical resolution of the image capture unit with its imaging optics and area sensor must be considered, as well as any deformations of an object, such as a beverage container, on which the optical code is located. The information content of the optical code, i.e., how many bits or bytes the code can represent, is also a factor. Considering that common commercial identifiers can be represented by 40 to 60 bytes, an optimal size for the 2D code is between 15 mm x 15 mm and 20 mm x 20 mm. The lengths of the edges of an at least approximately square 2D code are therefore preferably between 15 mm and 20 mm.

[0132] The security ink exhibits higher absorption over a larger, visible wavelength range of light than over a second, longer wavelength range. In this second, longer wavelength range, the security ink may, for example, have high transparency, meaning that the reflective properties of the security marking components printed with the security ink are at least partially determined by the substrate beneath the ink. Alternatively, the security ink itself may exhibit comparatively high reflectivity in this second, longer wavelength range. Between the first and second wavelength ranges, the reflection, transmission, or absorption properties change by more than 50% in a transition wavelength range, starting from the lower value in each case.The transition wavelength range is significantly narrower than either the first or the second wavelength range individually, and may be, for example, only one-fifth or one-tenth, or even less, than either the first or the second wavelength range individually. This is illustrated in Figure 4, which shows the reflectivity of the safety paint, possibly on a broadband reflective substrate. The dashed line in Figure 4 illustrates the reflection properties of a safety paint on a broadband reflective substrate. It can be seen that the reflectivity of the safety paint is low in the first, shorter wavelength range, increases in the transition wavelength range, and is comparatively high in the second, longer wavelength range.The low reflectivity in the first, shorter wavelength range is due to a correspondingly high absorption of light by the safety paint in this wavelength range. In Figure 4, "R" denotes the reflectivity and "X" the wavelengths; that is, in Figure 4, the reflectivity of the safety paint is plotted against the wavelength.

[0133] Furthermore, the safety paint also exhibits fluorescent properties. This means that the safety paint emits light in a typically very narrow fluorescence wavelength range when illuminated with electromagnetic radiation in a fluorescence-exciting wavelength range. The principle of fluorescence is illustrated in Figure 5. Fluorescence is the ability of certain atoms and molecules to absorb light at a specific excitation wavelength and then re-emit it, with a certain energy loss, as light at a longer wavelength. Figure 5 is a simplified representation of excited emission or fluorescence using an atomic model (p-proton, e-electron), in which a photon with a short wavelength and thus high energy (h*v) excites an electron from the ground state |0> to the higher energy state |2>.The electron can only remain in this state very briefly and, with an intermediate stop, falls into the energy level |1>, where some of the energy is released as heat, before continuing to the ground state |0>. Here, the remaining energy is emitted as a photon, i.e., light with a longer wavelength and therefore lower energy than the excitation energy. The difference between the excitation frequency v or excitation wavelength X and the emission frequency v' or emission wavelength is called the Stokes shift.

[0134] The following wavelength ranges are therefore characteristic of the properties of the safety paint: the first wavelength range in which the safety paint has high absorption, the second wavelength range in which the safety paint has low absorption, the transition wavelength range between the first and the second wavelength range, the fluorescence-exciting wavelength range - also called excitation wavelength range - and the fluorescence wavelength range in which the safety paint fluoresces in the event of excitation in the fluorescence-exciting wavelength range.

[0135] Figure 6 illustrates the fluorescence of the safety paint in the fluorescence wavelength range when the fluorescence is excited by illumination with electromagnetic radiation in the excitation wavelength range. In Figure 6, "I" denotes the intensity of the electromagnetic radiation and "X" the wavelengths; that is, in Figure 6, the intensity of the illumination (solid line) and the intensity of the fluorescence of the safety paint (dashed line) are plotted against wavelength X.

[0136] In order to detect fluorescence, an image acquisition unit is provided which has an imaging optic and an area sensor with matrix-like arranged, light-sensitive sensor elements.

[0137] In addition, a lighting unit is provided for illuminating an object to be tested with electromagnetic radiation in the excitation wavelength range.

[0138] The image acquisition unit is designed such that its sensor elements do not detect electromagnetic radiation in the excitation wavelength range. This can be achieved by the sensor elements themselves being insensitive to electromagnetic radiation in the excitation wavelength range.

[0139] However, an optical filter in the form of a high-pass or low-pass filter, in particular an edge filter, can also be provided, which is configured to block electromagnetic radiation in a blocking wavelength range encompassing the excitation wavelength range, in particular to absorb it, and to transmit light in a transmission wavelength range, wherein the transmission wavelength range includes wavelengths corresponding to expected fluorescence and wherein the transmission wavelength range does not overlap with the excitation wavelength range. Edge filters are special filter glasses or dielectric layers that filter out certain wavelength ranges from the electromagnetic spectrum, e.g., absorb them, while simultaneously transmuting (i.e., allowing) other wavelength ranges to pass through.Between the blocking wavelength range and the transmitting wavelength range, the optical filter exhibits a narrow cutoff wavelength range over which the transmission properties of the optical filter change significantly. The transmission properties of an optical low-pass filter are illustrated schematically and by way of example in Figure 7 using a filter characteristic curve. In Figure 7, "T" denotes the transmissivity of the optical filter and "X" the wavelengths; that is, in Figure 7, the transmissivity of the optical filter (dashed line) is plotted against wavelength X. The optical filter 52 (see Figure 18) preferably blocks light below a wavelength of 650 nm and preferably transmits light above a wavelength of 700 nm.

[0140] The optical filter 52 is preferably part of the imaging optics 42, see Figure 18.

[0141] Figure 8 illustrates how electromagnetic radiation is emitted from a broadband light source in an illumination wavelength range encompassing an excitation wavelength range. To ensure that only the resulting emission—i.e., the fluorescence of the safety ink—is detected, an edge filter is provided. This filter has the properties shown in Figure 7 and absorbs wavelengths below a cutoff wavelength range while allowing wavelengths above the cutoff wavelength range to pass through. The cutoff wavelength range can be a narrowband range, for example, between 600 nm and 700 nm.

[0142] According to the invention, the security marking comprises a machine-readable optical code, e.g., a Data Matrix code as shown in Figure 9. Figures 9a and 9b illustrate how components of the Data Matrix code can be printed with security ink. Machine-readable optical codes such as barcodes, QR codes, or Data Matrix codes typically have light and dark components.

[0143] Figure 9 shows a security mark with a Data Matrix code (GTIN with attributes) and a security ink background, presented in two versions: "Normal" with black ink for the Data Matrix code and "Inverse" with white ink for the Data Matrix code. In the version shown in Figure 9a, the typically light components of the Data Matrix code are printed with security ink, so that when viewed under visible light, they appear almost as dark as the typically dark components. The Data Matrix code in Figure 9a is therefore not machine-readable under visible light, but it is readable under excited fluorescence, as explained below. In the version shown in Figure 9b, the typically dark components of the Data Matrix code are printed with security ink, making them easily visible under visible light.The Data Matrix code shown in Figure 9b is therefore machine-readable under visible light, but not under excited fluorescence, as explained below.

[0144] According to a first aspect of the invention, it is provided that the object to be tested, and in particular a possible security marking on the object, is illuminated with electromagnetic radiation in the excitation wavelength range and that the image of the object and the marking captured by the image acquisition unit in the transmission wavelength range of the optical filter is analyzed.

[0145] Figure 10 schematically shows the image of a security marking in the form of a Data Matrix code, where the typically light-colored components of a Data Matrix code are printed with security ink (top, Fig. 10a) and once without security ink (bottom, Fig. 10b). Due to the fluorescence effect, components of the Data Matrix code formed by the security ink appear light under broadband fluorescent illumination because of the security ink's fluorescence. The edge filter "separates" this fluorescence from the excitation wavelength, so that only components of the security marking printed with fluorescent security ink appear light. If the component is not the security ink, which does not possess this fluorescence effect, the image remains dark overall because the security ink does not fluoresce.An optical code, whose typically light components are printed with authentic security ink that appears dark in the visible wavelength range of light, is therefore only recognizable in the image captured under fluorescent illumination and can be decoded conventionally. The decodability of the security marking under—essentially exclusively—fluorescence-stimulating illumination and the use of an optical filter (edge ​​filter) in front of the area sensor that blocks the fluorescent illumination is thus a property to be tested that indicates the use of authentic security ink (Figure 10a).If, however, the optical code of the security marking is not decodable in the image taken under - essentially exclusively - fluorescently stimulating illumination and the application of an optical filter (edge ​​filter) that blocks the fluorescently stimulating illumination in front of the area sensor, an object to be checked will be rejected (Figure 10b).

[0146] Figure 11 schematically shows a representation of a security marking in the form of a Data Matrix code, where the typically light-colored components of a Data Matrix code are printed once with security ink (top, Fig. 11a) and once without security ink (bottom, Fig. 11b). Due to the fluorescence effect, components of the Data Matrix code formed by the security ink appear light under broadband fluorescent illumination because of the security ink's fluorescence. The edge filter "separates" this light from the excitation wavelength.If the security ink does not possess this fluorescent effect, the grayscale image remains dark because the ink does not fluoresce, and the fluorescent-stimulating illumination is not detected due to the optical filter. This occurs when the security marking is illuminated only with fluorescent-stimulating light. The optical filter (edge ​​filter) in front of the area sensor, which blocks the fluorescent-stimulating illumination, prevents electromagnetic rays reflected from the bright components of the security marking from reaching the sensor elements in the excitation wavelength range. An optical code whose dark components in the visible wavelength range are printed with authentic security ink thus appears inverted in its image captured under fluorescent-stimulating illumination and can be decoded conventionally.The decoding capability of the security marking under – essentially exclusively – fluorescently stimulating illumination and the application of an optical filter (edge ​​filter) that blocks fluorescently stimulating illumination in front of the area sensor is therefore a property to be tested that indicates an authentic security color (Figure 11a). If, however, the optical code of the security marking is not decoding in the image recorded under – essentially exclusively – fluorescently stimulating illumination and the application of an optical filter (edge ​​filter) that blocks fluorescently stimulating illumination in front of the area sensor, the object being tested will be rejected (Figure 11b).

[0147] Due to the white color of the Data Matrix code, it can be read without problems by the cash register system or other devices in visible light, since the contrast ratio between light and dark is sufficiently high in the wavelength range of visible light, and code readers typically do not have an optical filter to block visible light. It should be noted here that the testing device is preferably part of a return system for objects, where the objects to be tested are exposed to comparatively little ambient light, so that the intensity of the illumination from the lighting unit is several times greater than the intensity of the ambient light. Therefore, the optical code shown in Figure 11 cannot be decoded if it is printed with a supposedly incorrect security ink that does not fluoresce under the illumination that is actually intended to excite fluorescence, as shown in Figure 11b.

[0148] According to a second aspect of the invention, during the authenticity check, an object to be tested is not only illuminated with electromagnetic radiation in the excitation wavelength range, but also additionally or alternately with electromagnetic radiation in the transition wavelength range, in which the absorption of the security paint is lower than in the first wavelength range, but higher than in the second wavelength range.

[0149] This is illustrated in Figures 12 and 13. For example, two light sources are used to illuminate an object to be tested: a broadband light source to excite the fluorescence and a narrowband light source to illuminate the object in the transition wavelength range. To ensure that only the resulting emission—that is, the fluorescence of the security paint and its reflection in the transition wavelength range—is detected, an edge filter is used. This filter absorbs wavelengths below the cutoff wavelength range and allows wavelengths above it to pass through (a combination of absorption and transmission). The cutoff wavelength range of the edge filter extends over shorter wavelengths (higher frequencies) than the transition wavelength range of the security paint with respect to its absorption properties.

[0150] The illumination of the safety marking with electromagnetic radiation in the transition wavelength range for reflection measurement in the transition wavelength range is preferably carried out separately from the illumination of the safety marking with electromagnetic radiation in the excitation wavelength range – for example, alternately. The advantage of this is that color identification can also be performed by detecting the reflection in the transition wavelength range, for example, if another color also exhibits fluorescence in the same wavelength range. In Figure 12, the reflection properties of a safety color on a broadband reflective substrate are shown by way of example with the dashed line. The reflection properties of the safety color represented by example in Figure 12 correspond to the example in Figure 4.The solid double line in Figure 12 describes an intensity I of the illumination of the security marking, and the dashed double line in Figure 12 describes an intensity I of the light reflected by authentic security paint. The illumination of the security marking with electromagnetic radiation in the transition wavelength range preferably occurs in a narrow wavelength range between 700 nm and 800 nm, particularly between 700 nm and 750 nm. The narrowband light source for illuminating the object in the transition wavelength range thus preferably has an emission spectrum that is preferably located between 700 nm and 750 nm. The broadband light source for exciting the fluorescence, on the other hand, preferably emits in a wavelength range between 400 nm and 650 nm, whereby even shorter wavelengths may also be suitable for exciting fluorescence.

[0151] According to the second aspect, a safety marking to be tested is examined with regard to its fluorescence properties when illuminated in the excitation wavelength range and with regard to its reflection properties when illuminated in the transition wavelength range.

[0152] Figure 14 schematically shows a security marking in the form of a Data Matrix code, where the typically light-colored components of a Data Matrix code are printed with security ink (top, Fig. 14a) and once without security ink (bottom, Fig. 14b). The fluorescence properties of the security marking are tested as described for Figure 10. Due to the fluorescence effect, the security ink appears bright under broadband illumination covering the excitation wavelength range. The edge filter separates the fluorescence radiation from the excitation wavelength.If the safety paint is not a supposedly incorrect, but in fact inaccurate, safety paint that lacks this fluorescence effect, the image of the safety marking captured under fluorescent-exciting illumination will appear dark overall, because the supposedly incorrect safety paint will not fluoresce. Additionally, after the fluorescence image, an image of the safety marking is captured under electromagnetic radiation in the transition wavelength range. Authentic safety paint then exhibits a relative reflection with an intensity approximately 50% of the illumination intensity. This represents the characteristic inflection point of the reflection curve in the transition range of an authentic safety paint.With a different color, the gray values ​​are significantly different; normally, when illuminated in the transition wavelength range, the grayscale image results in either a very bright or a very dark intensity or grayscale value. In the example shown, the grayscale value is very bright.

[0153] Figure 15 schematically shows the image of a security marking in the form of a Data Matrix code, where the typically light-colored components of a Data Matrix code are printed with security ink (top, Fig. 15a) and once without security ink (bottom, Fig. 15b). The fluorescence properties of the security marking are tested as described for Figure 11. Due to the fluorescence effect, the security ink appears bright under broadband illumination encompassing the excitation wavelength range. This results from the fluorescence of the security ink. The edge filter "separates" the fluorescence radiation from the excitation wavelength. If the security ink is not a supposedly genuine, but actually counterfeit, security ink that does not possess the fluorescence properties of authentic security ink, the image of the security marking remains dark overall because the ink does not fluoresce.In addition, after the fluorescence image is captured, an image of the security marking is acquired using electromagnetic radiation in the transition wavelength range. Authentic security paint then exhibits a relative reflectance with an intensity approximately 50% of the illumination intensity. This represents the characteristic inflection point of the reflectance curve in the transition range of an authentic security paint. With other paints, the gray values ​​are significantly different; normally, the gray value image obtained when illuminated in the transition wavelength range results in a very bright or a very dark intensity or gray value. In the example shown, the gray value is very bright.

[0154] The white color of the Data Matrix code allows it to be read without difficulty by the point-of-sale system or other devices in visible light, since the contrast ratio between light and dark is sufficiently high in the wavelength range of visible light, and code readers typically do not have an optical filter to block visible light. The security marking variant shown in Figures 15 and 16 thus has the advantage that the machine-readable optical code represented by the security marking can be read in ordinary daylight, but can only be read under illumination almost exclusively from fluorescent-stimulating lighting blocked by the edge filter if the security color is authentic.

[0155] It should be noted that the testing device is preferably part of a return device for objects in which the objects to be tested are exposed to comparatively little ambient light, so that the intensity of the illumination by the lighting unit is several times greater than the intensity of the ambient light.

[0156] Figure 16 schematically shows the image of a security marking in the form of a Data Matrix code, where the typically light-colored components of a Data Matrix code are printed with security ink (top, Fig. 16a) and once without security ink (bottom, Fig. 16b). The fluorescence properties of the security marking are tested as described for Figure 11. Figure 16b shows images that result when a purported security ink has similar fluorescence properties to the authentic security ink but different reflective properties. Here, a clear difference is only visible in the image that results when the security marking is illuminated with electromagnetic radiation in the transition wavelength range. The detectability of such false, purported security inks is a further advantage of testing according to the second aspect of the invention and the use of "white" ink for the Data Matrix code.

[0157] In addition to using the 2D code as a security mark with minimal requirements as a data carrier for identification, e.g., the GTIN, further information, i.e., additional attributes, can also be stored within the optical 2D code itself. The identification mark, e.g., the GTIN, is uniquely and inseparably linked to the security mark containing the Data Matrix code; this cannot be covered up.

[0158] Other attributes can include, for example, material type “PET bottle” 01 or can 02, which deposit system(s) are authorized here, e.g. 10 Germany, 11 other country, etc., and 10 / 11, i.e., Germany and another country, or custom coding values ​​for multiple deposit systems.

[0159] However, more data also increases the granularity of the Data Matrix code. This, in turn, affects the resolution when reading and processing the Data Matrix code. Therefore, a good balance must be found regarding how much data is stored in the Data Matrix code.

[0160] The requirements are illustrated in Figure 17 using the Data Matrix code as an example. Figure 17 shows Data Matrix codes with increasing data content. From left to right at the top, the data content increases from 1234567890128 (GTIN) to 1234567890128 / 01 (GTIN with key label or can) to 1234567890128 / 01 / 10 (GTIN with key label or can and key deposit system, here 10 or DE for the German system) and finally to 1234567890128 / 01 / DE. Below is a link to the corresponding website of the deposit system with the same attributes; this may be linked to master data information. However, the increased data also affects the granularity of the Data Matrix code. This, in turn, impacts the resolution during reading and processing of the Data Matrix code. Therefore, a good balance must be found to determine how much data is stored in the Data Matrix code.

[0161] An inspection of an object 10 with a security marking 12 in the sense described above can be carried out with a device such as that shown by way of example in Figure 18.

[0162] In general, a device for testing an article 10 with a security marking 12 preferably comprises the following components: at least one illumination unit 38.1, 38.2 for wavelength-selective illumination of the article 10 in at least one fluorescence-exciting illumination wavelength range; at least one broadband-sensitive intensity image acquisition unit 36 ​​with an area sensor 40, which is sensitive in at least one fluorescence wavelength range up to the long-wavelength red and preferably also in the second wavelength range and is configured to provide intensity images – preferably intensity value images – of the article 10; a control unit 50, which is connected to the illumination unit 38.1, 38.2 and the intensity image acquisition unit 36 ​​and which is configured to control the illumination unit 38.1, 38.2.2. to control such that the article 10 is successively illuminated with light in different illumination wavelength ranges, preferably both exclusively in one illumination wavelength range and with light in a combination of two or more illumination wavelength ranges; furthermore, the control unit 50 is configured to control the intensity image acquisition unit 36 ​​in such a way that it captures an image of the security marking 12 under the respective illumination; and a preprocessing, evaluation and assessment unit 44, 46 and 48 respectively, which are connected to the control unit 50 and the intensity image acquisition unit 36 ​​and are configured to evaluate and check intensity images supplied by the intensity image acquisition unit 36 ​​during operation with regard to the intensity captured in certain image areas.

[0163] The evaluation unit 46, 48 preferably comprises at least two components: a pre-evaluation unit 46 and an evaluation module 48. That is, the evaluation unit 48 is preferably designed as a dedicated evaluation module to which representative intensity values ​​for individual image areas of an image of a security marking are supplied. These values ​​result from the pre-processing and pre-evaluation of the intensity value images. The evaluation module 48 can be designed as a hardware module with dedicated interfaces to the pre-evaluation unit and the control unit. Alternatively, the evaluation module 48 can be designed as a particularly secure software module, e.g., as a software container, in which the software container can only be modified, e.g., replaced or updated, with special access rights and is also encrypted.The software running in the software container embodies, among other things, the test algorithms for evaluating whether the recorded and pre-processed intensity values ​​reflect the properties of a genuine security marking.

[0164] While the preprocessing of the output values ​​supplied by the individual sensor elements of the area sensor 40 into preprocessed intensity values ​​of the individual pixels of the recorded image and their pre-evaluation do not necessarily have to take place in separate units, namely by a preprocessing unit 44 and a pre-evaluation unit 46, the subsequent evaluation of the intensity values ​​of the individual image areas of the image of the security marking 12 by the evaluation module 48 is preferably clearly separated from the pre-evaluation by the pre-evaluation unit 46.

[0165] As already indicated, the preprocessing includes processing the output signals of the sensor elements of the area sensor 40 into intensity values ​​of the individual pixels of the image captured by the intensity image acquisition unit 36.

[0166] Further evaluation of these intensity values ​​is then carried out by the pre-processing unit 46, which is designed, for example, to determine a signal-to-noise ratio or intensity value dynamics based on the pre-processed output signals, i.e., based on the intensity values ​​of the pixels of the captured image. The pre-processing unit 46 is also designed to segment the image of the security marking 12, i.e., to identify the individual image areas of the security marking image. Subsequently, the pre-processing unit 46 can determine a representative intensity value for each identified image area. As described below, the representative intensity value can, for example, be the median of filtered intensity values ​​of the pixels belonging to a respective image area.The segmentation of the image, the definition of measurement windows for the different image areas, which are images of the different fields of the security marking, and the formation of representative intensity values ​​for each measurement window, and thus for each image area representing a respective field of the security marking 12, is explained in more detail below in the section “Preprocessing and Pre-checking”.

[0167] By dividing the evaluation of the captured sensor output signals between a preprocessing unit 44 and a pre-evaluation unit 46 on the one hand, and the evaluation module 48 on the other, an interface is created between the pre-evaluation unit 46 and the evaluation module 48. This interface allows the representative intensity values ​​for a given measurement window, i.e., for a given image area of ​​the security marking, to be transferred to the evaluation module 48. In other words, the evaluation of the captured signals for the purpose of authenticity verification only takes place in the evaluation module 48.

[0168] The preprocessing and pre-evaluation by preprocessing unit 44 and pre-evaluation unit 46 serve only to process the acquired signals, without evaluating them. The results of this preprocessing and pre-evaluation therefore include, among other things:

[0169] Isolating one of the images of a detected security marking (if one is detected at all) by segmenting the images of an object to be tested, and a signal-to-noise ratio value or a dynamic value for the signal dynamics of the intensity values ​​(intensity value dynamics) for the output signals of the individual sensor elements of the area sensor 40.

[0170] The result of preprocessing or pre-evaluation may be that no image of a security field is detected, or that a signal-to-noise ratio or dynamic range is determined that is lower than the minimum signal-to-noise ratio or dynamic range required for a reliable evaluation. Both of these can lead to the termination of further testing or to the intensity image acquisition unit 36 ​​re-acquiring an image of the article. In any case, if the signal-to-noise ratio or dynamic range is too low, or if no image of a security mark is detected at all, no representative intensity values ​​are transmitted to the evaluation module 48.Rather, a signal is usually sent to the control unit 50, which causes the control unit 50 to transmit a rejection signal to the rest of the control system of a return device, causing the return device to reject an entered item and, for example, not to pay out a deposit.

[0171] If, on the other hand, the signal-to-noise ratio or the dynamic range is high enough at least in the image areas of an image of the security mark that can be assigned to a security mark - and thus a security mark has also been detected in the recorded image - the pre-evaluation unit 46 finally determines a representative intensity value for each of the image areas of the image of the security mark that each represent a field of the security mark 12, which is then transmitted to the evaluation module 48.

[0172] Evaluation module 28 implements a test algorithm that evaluates the representative intensity values. If the representative intensity values ​​match, the algorithm delivers a test result indicating the authenticity of a detected security mark. If the evaluation module 48 and its implemented test algorithm determine that the representative intensity values ​​do not meet the test criteria, no output signal indicating the authenticity of the detected security marks is generated (also referred to here as an acceptance signal). In this case, no signal indicating a positive test result is sent to the control unit 50 via a feedback channel from evaluation module 48 to the control unit 50 – thus, no acceptance signal is generated, but rather, preferably, a rejection signal.A positive test result signal (i.e., the security mark has been recognized as genuine) sent to control unit 50 triggers the return of the tested object or item and, if applicable, a refund of any deposit. Otherwise, a rejection signal is generated, and the entered object or item is rejected, and no deposit is refunded, if applicable. Instead of explicitly generating a rejection signal, the failure to generate an acceptance signal can also result in a rejection.

[0173] The advantage of performing the actual evaluation in a separate evaluation module 48 is that the evaluation module 48 requires only a few interfaces to the rest of the test device and can contain fully encrypted test software that implements the test algorithm. The test software can be part of a software container. Preferably, the evaluation module 48 has a data transmission interface to a central server (see Figure 20) to which, firstly, the test results obtained by the evaluation module 48 during a test of the representative intensity values ​​can be transmitted. Conversely, this interface also allows for an update of the evaluation module 48's software. This can be done, in particular, to implement a different, for example, improved test algorithm in the evaluation module 48.

[0174] The connection between evaluation module 48 and the central server is preferably end-to-end encrypted.

[0175] Ideally, the readout device 30 also has an interface through which the images of a tested security mark, i.e., the pixel images composed of preprocessed intensity values, can be transmitted to the central server. This allows for subsequent evaluation on the central server 62 and optimization of the testing algorithm.

[0176] The evaluation of the intensity value images supplied by the intensity image acquisition unit 36 ​​during operation preferably comprises intensity normalization of an intensity value image obtained from the intensity image acquisition unit and the determination of a representative intensity value for at least one image area within the intensity value image. Preferably, the evaluation includes calculating the differences between two representative intensity values ​​of the same image area that result from illuminating the object with two different illumination wavelength ranges.

[0177] The intensity image acquisition unit 36 ​​preferably comprises an optical blocking filter that blocks light in one of the three different illumination wavelength ranges, so that the intensity image acquisition unit 36 ​​does not detect any light in the blocked illumination wavelength range. Figure 18 shows an exemplary schematic, sketch-like representation of such a readout device 30. The readout device 30 is used to check security markings 12 on packaging such as the packaging 10.

[0178] Essential components of the reading device 30 are one or more illumination units 38.1 and 38.2 for illuminating a package 10 with light in one or more different illumination wavelength ranges and an image acquisition unit 36 ​​for capturing images (preferably intensity value images) of a security marking 12 on the package 10 under the respective illumination. The image acquisition unit 36 ​​serves as an intensity image acquisition unit and comprises an area sensor 40 and optics 42, which sharply projects an image of the surface of the package 10 onto a surface of the area sensor 40. A blocking filter 52 is preferably provided in front of or as part of the optics 42. This filter acts as a low-pass filter, blocking fluorescence-exciting illumination and transmitting light with wavelengths corresponding to the excited fluorescence. The blocking filter is preferably an edge filter, as described with reference to Figure 11.

[0179] To test a security marking which has components printed with security ink exhibiting both characteristic fluorescence properties and characteristic wavelength-dependent absorption properties, as explained in connection with Figures 12 to 16, different or switchable illumination units 38.1 and 38.2 are provided to illuminate a security marking on a package 10 on the one hand with electromagnetic radiation in the fluorescence-exciting wavelength range and on the other hand - preferably alternately - with electromagnetic radiation in the transition wavelength range in which the security ink-printed components of the security marking show neither their maximum nor their minimum absorption.

[0180] Figures 19a and 20a schematically illustrate a sensor unit 56F designed to detect the fluorescence expected of genuine security paint. This sensor unit comprises an illumination unit 38F for illuminating a marking with fluorescent-stimulating electromagnetic radiation, and an image acquisition unit with imaging optics and an edge filter 52 for blocking the fluorescent-stimulating electromagnetic radiation. The illumination unit 38F includes a white light-emitting diode (or several thereof) and a BG39 filter. Figures 19b and 20b schematically illustrate a sensor unit 56FR designed to detect the fluorescence expected of genuine security paint and to detect light reflected from a marking in the transition wavelength range.The sensor unit 56FR includes an illumination unit 38F for illuminating a marker with fluorescent-stimulating electromagnetic radiation and an illumination unit 38R for illuminating a marker with electromagnetic radiation in the transition wavelength range. The sensor unit 56FR also includes an image acquisition unit with imaging optics and an edge filter 52 for blocking the fluorescent-stimulating electromagnetic radiation. The illumination unit 38F includes a white light-emitting LED (or several thereof) and a BG39 filter. The illumination unit 38R includes a red light-emitting LED (or several thereof).

[0181] Not shown is a sensor unit 56R, which only has a lighting unit 38R for illuminating a marking with electromagnetic radiation in the transition wavelength range.

[0182] The reading device 30, which can, for example, be part of a reverse vending machine for beverage packaging, can also include a transport device 32 with which a package 10' can be positioned in front of a reading unit 34 so that it is within the field of view of the image acquisition unit 36 ​​of the reading unit 34. The field of view is indicated in Figure 18 by dashed diagonal lines.

[0183] Instead of an image acquisition unit, an alternative reading device (not shown) can also have several illumination units and several image acquisition units arranged around an input opening of a return device in such a way that a package inserted through the input opening is illuminated and imaged from all sides simultaneously. This is illustrated, for example, in Figure 21 a).

[0184] To test a security marking that includes components printed with security ink exhibiting both characteristic fluorescence properties and characteristic wavelength-dependent absorption properties, as explained in connection with Figures 12 to 16. In the example shown in Figure 19a), sensor units 56F for detecting the fluorescence properties alternate with sensor units 56R for detecting the reflection properties. Figures 21b and 21c show alternative arrangements of sensor units configured similarly to the example shown in Figure 18 for illuminating an object from one side.

[0185] In the embodiment shown in Figure 18, two lighting units 38.1 and 38.2 are provided for illuminating the packaging 10 and, in particular, the security marking 12. These can, for example, correspond to the lighting units 38F and 38R from Figure 20b. The direction of illumination, and thus the angle at which the illumination falls on the packaging 10 to be read, is indicated by dotted arrows. The angle of illumination should be between 20° and 45° with respect to the surface normal of the packaging 10. The lighting units 38.1 and 38.2 of the lighting module are arranged and aligned accordingly.

[0186] The lighting units 38.1 and 38.2 have a plurality of light-emitting diodes (LEDs) as light sources. The lighting units 38.1 and 38.2 are designed to illuminate the packaging 10 with light in one or more of a specified number of illumination wavelength ranges, preferably a comparatively shorter wavelength, preferably fluorescently exciting illumination wavelength range, which is preferably suitable for exciting the fluorescence of a genuine security ink, and a further, somewhat longer wavelength illumination range in which a genuine security ink no longer has its maximum absorption, but the security field has also not yet reached its maximum reflectivity. The comparatively shorter wavelength, fluorescently exciting illumination wavelength range is preferably broadband and can extend over a wavelength range of more than 10 µm, which lies, for example, between 400 nm and 600 nm.The further, somewhat longer wavelength illumination range, in which a true safety color no longer possesses its maximum absorption, but the safety field has also not yet reached its maximum reflectivity, is preferably narrowband and extends only over approximately 10 nm to 40 nm (Full With Half Maximum) and preferably lies between 700 nm and 750 nm. The illumination units 38.1 and 38.2. For this purpose, they possess two or more types of light-emitting diodes (LEDs): a first type of LED emitting light capable of inducing fluorescence; a second type of LED emitting red, visible light in a wavelength range where the ink used to print the security field 16 (if genuine) no longer has its maximum absorption, but the security field has not yet reached its maximum reflectivity; and a third type of LED may also be provided, emitting red, visible light in a wavelength range where the ink used to print the security field 16 is transparent and / or capable of fluorescence. The visible light emitted by the illumination units 38.1 and 38.1The light emitted by module 2 thus falls within one or more of a total of two or more wavelength ranges, each with, for example, a central wavelength in the blue region of the visible spectrum and a central wavelength in the red region of the visible spectrum. The red-emitting LEDs are such that the respective wavelength range of the light they emit is narrowband and has a pronounced central wavelength. The half-value bandwidth of the two wavelength ranges emitted by the lighting modules 38.1 and 38.2 in the transition wavelength range of the safety paint is each less than 50 nm, in particular less than 30 nm.

[0187] The lighting module and its lighting units 38.1 and 38.2 allow for the setting of specific lighting scenarios. In normal operation, the lighting units 38.1 and 38.2 are operated in such a way that the packaging 10 is illuminated successively with light in only one of several, e.g., two or more wavelength ranges, as described above.

[0188] In general, the light sources, for example the light-emitting diodes, of the lighting units 38.1 and 38.2 are selected to allow the detection of characteristic spectral properties of a safety paint. Since the safety paint has fluorescent properties according to the invention, the lighting units 38.1 and 38.2 contain fluorescent-exciting light sources. If the safety paint is particularly absorbent or particularly transparent for certain wavelengths, it is advantageous if the lighting units 38.1 and 38.2 have light sources that can emit light in these specific wavelength ranges and—for comparison—also in other wavelength ranges.

[0189] By illuminating the packaging 10 - or another object to be tested - with different lighting scenarios, it is possible to take different pictures of the packaging 10 in which the image of the security marking - if one is present - appears differently depending on the wavelength.

[0190] For this purpose, the light backscattered, reflected, and / or emitted from the surface of the packaging 10 under a given illumination is captured by the image acquisition unit 36. The image acquisition unit 36 ​​comprises an area sensor 40 and an imaging optic 42, which projects a sharp image of the surface of the packaging 10 onto a surface of the area sensor 40. A blocking filter is preferably provided in front of or as part of the optic 42, which blocks electromagnetic radiation from the fluorescently exciting illumination.

[0191] The surface of the area sensor 40 is formed by a plurality of light-sensitive sensor elements. These are preferably arranged in a matrix. The sensor elements of the area sensor 40 and the optics 42 are designed such that one square millimeter of the surface of the packaging 10 is imaged onto a partial surface of the area sensor 40 in such a way that the partial surface contains at least four complete sensor elements. Thus, the image scale with which the optics 42 projects an image of the surface of the packaging 10 onto the surface of the area sensor 40 depends on the size of the sensor elements on the surface of the area sensor 40 and the distance between the sensor elements. It goes without saying that the optics 42 are designed to project a sharp image of the surface of the packaging 10 onto the area sensor 40 within the depth of field required by varying packaging diameters.

[0192] The sensor elements of the area sensor 40 are typically broadband light-sensitive, meaning they are sensitive at least in the illumination wavelength ranges of visible light emitted simultaneously or alternately by the illumination units 38.1 and 38.2, and which include wavelengths capable of exciting fluorescence of the safety paint. However, the sensor elements of the area sensor 40 need not be sensitive to the illumination wavelength range used to excite fluorescence. The previously mentioned edge filter for blocking fluorescence-stimulating electromagnetic radiation can be part of the sensor elements. The output value supplied by each individual sensor element of the area sensor 40—also referred to here as the intensity value—corresponds to the total intensity of all light in the various wavelength ranges detected by the respective sensor element.

[0193] The output value supplied by each sensor element of the area sensor 40, also referred to here as the intensity value, is greater the greater the total intensity of the light striking that particular sensor element. The total intensity of the light striking that sensor element is composed of the partial intensities of the light in the various wavelength ranges (illumination and, if applicable, fluorescence) that make up the light striking that particular sensor element. In one possible embodiment, an illumination wavelength range is selected to include short-wavelength light that specifically excites fluorescence within a specific excitation wavelength range, and the optical blocking filter (the edge filter) is designed to block the light in this fluorescence-exciting illumination wavelength range.In this case, the light reflected from the security marking 12 in the fluorescence-stimulating illumination wavelength range is not detected by the image acquisition module, as it is prevented from entering the image acquisition module by the corresponding optical blocking filter. Only any fluorescence that may occur is detected, since it occurs in a different wavelength range—namely, the fluorescence wavelength range. Because the light that would stimulate fluorescence of the security ink is preferably prevented by a corresponding optical blocking filter from reaching the area sensor 40, this light does not contribute to the intensity detected by a single sensor element.If the supposed safety color does not fluoresce, the intensity value image of the safety marking would appear almost completely dark because the optical blocking filter in front of the image detection unit almost completely blocks the light in the fluorescence-stimulating illumination wavelength range.

[0194] In the case where the ink used to print the security field 16 is fluorescent because it has the property of being able to fluoresce and is accordingly excited to fluoresce, the light intensity detected by a respective sensor element – ​​and thus the output intensity value – depends on the intensity of the fluorescence, and possibly also on an additional reflection of the security field 16.

[0195] The optical blocking filter used to block the fluorescence-stimulating illumination does not interfere with the testing of the reflection properties, as this is carried out at longer wavelengths that are not blocked by the optical blocking filter.

[0196] The device shown in Figure 18 performs a test procedure to check an article 10 for the presence and authenticity of a security marking 12 with the following procedure steps.

[0197] The testing procedure includes at least the following steps:

[0198] Recording the input of an object to be tested, repeatedly illuminating the object to be tested with different lighting conditions,

[0199] Projecting an image of the object to be tested onto an area sensor under different lighting conditions and generating and outputting at least one image signal that represents an image of the object to be tested projected onto the area sensor.

[0200] Searching for an image of a security mark with a machine-readable optical code in the image of the object to be inspected projected onto the area sensor,

[0201] Determining the signal-to-noise ratio and / or the signal dynamics of the image signal representing the image of the object,

[0202] Verification of the authenticity of the security marking based on images of the security marking represented by the image signal.

[0203] The search for an image of a security marking with a machine-readable optical code in the image of the object to be inspected, projected onto the area sensor, is preferably carried out with illumination that is not blocked by the optical filter or without using the optical filter.

[0204] If no image of a security mark is found in the intensity value or color image of the object projected onto the area sensor, or if the signal-to-noise ratio and / or the intensity value dynamics of the intensity value or color image representing the object do not reach or exceed a specified minimum value, or if the verification of the security mark's authenticity does not result in the security mark meeting the verification criteria for authenticity, the object being tested will be rejected, and in the case of a seized item, no pawn payment will be made. - M -

[0205] In particular, an object being inspected will be rejected if an optical code of the security marking cannot be decoded in the image captured under – essentially exclusively – fluorescently exciting illumination and the application of an optical filter (edge ​​filter) that blocks the fluorescently exciting illumination in front of the area sensor. To verify the authenticity of the security marking, it is always illuminated with electromagnetic radiation in the excitation wavelength range – i.e., with fluorescently exciting illumination – and an optical filter that blocks the fluorescently exciting illumination – i.e., the edge filter – is applied in front of the area sensor to capture the image of the security marking.

[0206] First, an intensity image of an object or article is acquired using the broadband-sensitive intensity image acquisition unit 36 ​​with the area sensor 40 and its numerous sensor elements. The system then searches for a security marking within this intensity image. Once located, its position is determined. The pixels of the intensity images reflect intensity values ​​that represent the output values ​​of the individual sensor elements of the area sensor.

[0207] The search for a security marking can be performed in an intensity value image taken under any lighting conditions.

[0208] Pre-processing and preliminary testing

[0209] The preprocessing of intensity value images acquired by the intensity image acquisition unit includes at least searching for an image of the security mark in the respective intensity value image and determining the signal-to-noise ratio for at least that area of ​​an intensity value image that represents an image of the security mark.

[0210] If no image of the security marking is found, or if the signal-to-noise ratio is too poor, the intensity value image is discarded. If no better intensity value image exists or can be created for an object to be inspected, the object is rejected, and, for example, no deposit is paid out.

[0211] Before a security marking is inspected, an image of the security marking is sought within a respective intensity value image captured by the intensity image acquisition unit. Known pattern recognition methods and / or trained neural networks can be used to recognize the image of the security marking. Within the image acquisition unit 36, the intensity values ​​of a detected image of the security marking 12, supplied by the area sensor 40, are preprocessed in a preprocessing unit 44. The processed intensity values ​​(output values ​​of the sensor elements of the area sensor 40) are fed to a pre-evaluation unit 46, in which the different areas of the security marking are detected using known pattern recognition and / or segmentation methods.

[0212] One step in the preliminary evaluation process is to check whether the signal-to-noise ratio or the signal dynamics of the image of the security marking is good enough to allow for meaningful further testing. Determining the signal-to-noise ratio or signal dynamics can be done during preprocessing and / or preliminary evaluation. If none of the images of an object to be tested yields an image signal with sufficient signal dynamics or a sufficient signal-to-noise ratio, the object is rejected and further testing is terminated.

[0213] Capturing the different image areas of the image of the security marking 16 - i.e., e.g., this segmentation of a respective intensity value image - also serves to determine the location of the signal field 18 in order to be able to evaluate its intensity values, since the testing of the security marking in the evaluation module may depend on these.

[0214] The evaluation unit 46 is connected to a control unit 50, which, for example, controls the lighting units 38.1 and 38.2 and, if present, also controls the transport device 32 to rotate the packaging 10 using the transport device 32 so that the security marking on the surface of the packaging 10 is within the field of view of the readout unit 34. Image recognition by the evaluation unit 46 is also used for this purpose.

[0215] The control unit 50 also manages deposit refunds and the return of packaging. As already mentioned, the images are captured using the imaging optics 42 and the area sensor 40 and initially preprocessed by the preprocessing unit 44. Preprocessing can, for example, include determining the signal-to-noise ratio. If this ratio is lower than a predefined minimum, the preprocessing unit 44 can, for example, trigger the capture of another image of the object 10 or initiate a rejection, resulting in the object 10 being rejected and no deposit being paid out.

[0216] The preprocessed images, represented by corresponding image signals, are then fed to the pre-evaluation unit 46, which first locates an image of a security mark 12 within the image of the object 10. Once the image of a security mark 12 has been found and its position within the image of the object 10 has been determined, the pre-evaluation unit triggers at least one image acquisition of the security mark under illumination within the excitation wavelength range.

[0217] The image evaluation unit includes, in particular, a code reading component configured to identify and interpret a machine-readable optical code in an image. If the code reading component does not identify an interpretable machine-readable optical code in an image captured under illumination with electromagnetic radiation in the excitation wavelength range, the inspection device triggers a rejection of the inspected object.

[0218] If the verification of all criteria (including presence checks and authentication of the security mark 12) yields the required positive result, a container – i.e., object 10 with the associated security mark – is accepted and the deposit is paid out. Otherwise – if one of the verification criteria is not met – the container is rejected. The acceptance or rejection of containers is carried out by the control unit 50, as can also be seen in Figure 18.

[0219] Procedure for verifying the authenticity of the security mark

[0220] A testing procedure for the optical verification of objects for authenticity includes the following steps:

[0221] Detecting the input of an object to be inspected, projecting an image of the object onto an area sensor, and generating and outputting at least one image signal representing an image of the object projected onto the area sensor.

[0222] Search for an image of a security mark in the image of the object projected onto the area sensor.

[0223] Determining the signal-to-noise ratio or signal dynamic range value of the image signal representing the image of the object, and

[0224] Verification of the authenticity of the security marking based on intensity values ​​of the components of the security marking represented by the image signal.

[0225] The verification of the authenticity of the security marking comprises the process steps shown in Figure 22, namely illuminating (S1) a found security marking with electromagnetic radiation in an excitation wavelength range that can be interpreted as fluorescence excitation for authentic security paint, wherein the intensity of the illumination with electromagnetic radiation in the excitation wavelength range is several times greater – at least twice as much – than the radiation incident on an object to be tested in the transmission wavelength range of the optical filter.

[0226] An image of the safety marking, illuminated by electromagnetic radiation in the excitation wavelength range, is captured by the image acquisition unit, whereby the image acquisition unit does not detect any electromagnetic radiation in the excitation wavelength range.

[0227] The resulting image of the security marking is checked for the presence of an interpretable, machine-readable optical code (S2). If no interpretable, machine-readable optical code is found, the tested object is rejected. The testing device generates a signal indicating that the security marking has failed the authenticity check (S6).

[0228] Steps S1 and S2 are part of a fluorescence test. If an interpretable, machine-readable optical code is found during the fluorescence test, a reflection test is preferably also carried out, in which a detected security marking is illuminated with electromagnetic radiation in the transition wavelength range (S3), in which an authentic security color reflects only part of the illuminating light and absorbs the rest.

[0229] The ratio of incident illumination intensity and the intensity of the light reflected by the character components printed with security ink in authentic security markings is determined and compared with a predetermined reference value (S4).

[0230] If a tested security mark passes not only the fluorescence test (procedure steps S1 and S2) but also the reflection test (procedure steps S3 and S4), the testing device generates a signal indicating the authenticity of the security mark (S5). Otherwise – after failing the reflection test – the testing device generates a signal indicating that the security mark has failed the authenticity test (S6).

[0231] The order in which a fluorescence test and a reflection test are performed as part of an authenticity check is basically arbitrary.

[0232] System with multiple test devices and a central server

[0233] It is particularly advantageous if various test devices 60 are connected to one or more servers 62, as schematically illustrated in Figure 23. The data acquired by the various test devices 60, in particular the recorded intensity value images and / or the representative intensity values, can then be supplied to a central server 62 in conjunction with information on rejections or acceptances. This server allows all acquired values ​​to be centrally evaluated or assessed. This can, for example, serve to optimize the test algorithm implemented in the respective evaluation module 48, and then to update the test algorithm in the evaluation modules 48 of all test devices 60 connected to the server 62. Typically, the individual test devices 60 are connected to one or more central servers 62 via the so-called cloud, i.e., the internet or other networks.In a system 10 such as the one shown in Figure 23, the server 62 is connected, at least temporarily, to the evaluation module 48 of a respective test device 60 via a data transmission network 64, forming a client-server system with the respective evaluation module 48. The connection between the server 62 and a respective evaluation module 48 is end-to-end encrypted. The server 62 and the respective evaluation module 48 are configured to transmit evaluation results to and receive them. The server 60 is designed to statistically evaluate evaluations received from multiple evaluation modules 48. Specifically, it is advantageous if a respective evaluation module transmits the underlying representative intensity values ​​to the server in addition to the respective evaluation result.The evaluation result obtained by the testing algorithm implemented in the evaluation module, through the analysis of representative intensity values, simply indicates whether a security mark has been assessed as authentic or inauthentic. The representative intensity values ​​belonging to the tested (security) mark are transmitted to Server 62 to enable a more in-depth statistical analysis and, if necessary, the development of an improved testing algorithm. Test parameters implemented by the testing algorithm can be optimized, for example, using a neural network trained on the representative intensity values ​​transmitted to Server 62 – or manually.Accordingly, it is also advantageous if the server 62 and the evaluation modules 48 of the test devices 60 are configured to transfer updates of a software implementation of an evaluation algorithm to the evaluation modules 48 and to install them on the evaluation modules 48.

[0234] According to one embodiment, the server 62 can also be connected to the pre-evaluation units 46 of the test devices 60 and configured to receive intensity value images recorded by the test devices 60 and evaluate them centrally. However, since this causes considerable data traffic, it is preferable if only the respective representative intensity values—i.e., one intensity value per field of the tested marking—are transmitted to the server. Reference character list

[0235] 10, 10' packaging

[0236] 12 safety markings

[0237] 14 Contrast field of the safety marking

[0238] 16 Safety field of the safety marking

[0239] 18 Signal field of the safety marking

[0240] 20 Dark field, namely corner field of the safety marking

[0241] 22 Dark field, namely orientation field of the safety marking

[0242] 24 Identification mark, e.g. barcode, GTIN

[0243] 26 Background of the identification mark, e.g. white background of the barcode

[0244] 28 dark-colored character components of the identification mark, e.g., bars of the barcode.

[0245] 30 Reading device

[0246] 32 Transport equipment

[0247] 34 Reading unit of the reading device

[0248] 36 (Intensity) image acquisition units

[0249] 38.1, 38.2 Lighting unit

[0250] 40 surface sensor

[0251] 42 imaging optics

[0252] 44 Pre-processing unit

[0253] 46 Evaluation unit, pre-evaluation unit

[0254] 48 Assessment unit, assessment module

[0255] 50 control unit

[0256] 52 barrier filters (edge ​​filters) for blocking fluorescent-stimulating lighting

[0257] 54 optical low-pass filter, soft focus

[0258] 56F, 56R Sensor unit with illumination unit and image acquisition unit

[0259] 60 Test device 62 Server

[0260] 64 Data transmission network

[0261] X wavelength

[0262] I Intensity R Reflectivity

[0263] T Transmissivity

Claims

1. Claims 1. Testing device (60) for optically testing objects for authenticity, in particular for testing whether objects have a predefined, machine-readable optical code security marking (12) with known fluorescence properties, wherein the testing device (60) comprises the following components: an illumination unit (38.1, 38.2) configured and arranged to illuminate an object to be tested with electromagnetic radiation – preferably light visible to the human eye – in a fluorescence-stimulating excitation wavelength range; an area sensor (40) with a plurality of light-sensitive sensor elements; an imaging optic (42) configured and arranged to project an image of an object onto the area sensor (40), wherein the area sensor (40) is configured to output an image signal representing an image projected onto the area sensor (40).an optical filter (54) in the form of a high-pass or low-pass filter, in particular an edge filter, configured to block electromagnetic radiation in a blocking wavelength range encompassing the excitation wavelength range, in particular to absorb it, and to transmit light in a transmission wavelength range, wherein the transmission wavelength range includes wavelengths corresponding to expected fluorescence and wherein the transmission wavelength range does not overlap with the excitation wavelength range, wherein the test device (60) is configured to detect the entry of an object, to check images of the object represented by image signals from the area sensor (40), and to reject an object if one or more test criteria are not met, or to accept it if the test criterion or criteria are met.where one test criterion is the readability or non-readability of a machine-readable optical code depending on the presence or absence of fluorescence.

2. Test device (60) according to claim 1, wherein the optical filter (54) is part of the imaging optics and a low-pass filter and the transmission wavelength range includes longer wavelengths than the blocking wavelength range.

3. Test device (60) according to claim 1 or 2, comprising an electronic image evaluation unit which is operatively connected to the area sensor (40) in order to receive at least indirectly raw or pre-processed image signals from the area sensor (40), wherein the image evaluation unit has a code reading component which is configured to identify and interpret machine-readable codes represented by image signals.

4. Test device (60) according to claim 3, which is configured such that it generates an object to be tested as an authentic characterizing signal if the code reading component, when illuminated with electromagnetic radiation comprising the excitation wavelength range and with an effective optical filter (54), identifies a machine-readable code in the image of a security-mark-like image component represented by the image signals.

5. Test device (60) according to at least one of the preceding claims 1 to 4, which is configured to take two photographs of an object to be tested, wherein the object is illuminated for one photograph with electromagnetic radiation in the fluorescence-stimulating excitation wavelength range and for another photograph with electromagnetic radiation in the transition wavelength range in which the absorption properties of an authentic security color change significantly.

6. Testing device (60) according to claim 5, which is designed to test an image of the security marking (12) recorded with electromagnetic radiation in the transition wavelength range with regard to the contrast between light and dark character components.

7. Test device (60) according to at least one of claims 1 to 6, wherein the test device (60) is configured to first detect a safety marking (12) in the images of an object projected onto the area sensor (40), and to trigger a rejection of the object if no image of a safety marking (12) was found in the images of the object projected onto the area sensor (40).

8. Test device (60) according to claim 7, wherein the test device (60) is configured to check, after the detection of a security mark (12) in the images of the object projected onto the area sensor (40), whether the signal-to-noise ratio or the dynamic range of the image signals representing the images of the object projected onto the area sensor (40) exceeds a predefined value and to reject the object to be tested if the signal-to-noise ratio or the dynamic range of the image signals representing the images of the object projected onto the area sensor (40) does not reach or exceeds the corresponding predefined value.

9. Test device (60) according to claim 8, wherein the test device (60) is configured to always reject an object if the signal-to-noise ratio or the dynamic value of the image signals representing the images of a currently tested object projected onto the area sensor (40) reaches or exceeds the setpoint value.

10. Method for optically verifying the authenticity of objects, the method comprising the following steps: Recording the entry of an object to be checked, Projecting an image of an object onto an area sensor (40) and generating and outputting at least one image signal representing an image of the object projected onto the area sensor (40), Search for an image of a security mark (12) in the image of the object projected onto the area sensor (40). Determining the signal-to-noise ratio or signal dynamic range value of the image signal representing the image of the object, and Verification of the authenticity of the security mark (12) based on intensity values ​​of the components of the security mark (12) represented by the image signal 11. Method according to claim 10, further comprising the step Generating a signal indicating a failed test if, when searching for an image of a security mark (12), no image of a security mark (12) is bound and / or if the signal-to-noise ratio or the signal dynamic range of the image signal representing the image of the object does not meet a specified minimum criterion and / or if a security mark (12) found fails the authenticity test.

12. Method according to claim 10 or 11, wherein the verification of authenticity comprises the following steps: Illumination (S1) of an object to be tested with electromagnetic radiation in an excitation wavelength range suitable for exciting the fluorescence of an authentic security paint, Capturing an image of an object to be inspected with an image acquisition unit in which a filter blocking electromagnetic radiation in the excitation wavelength range prevents the capture of electromagnetic radiation in the excitation wavelength range, and Rejection of an object to be inspected if an optical code of the security marking is not recognizable and / or not decodable in an image taken using electromagnetic radiation in the excitation wavelength range.

13. Method according to any one of claims 10 to 12, wherein the verification of authenticity comprises the following steps 51 Illuminating a discovered security marking with electromagnetic radiation in an excitation wavelength range that excites fluorescence in authentic security paint, 52 Check that the security marking contains a machine-readable optical code that can be interpreted when illuminated with electromagnetic radiation in a fluorescent excitation wavelength range for authentic security paint, and S6 If no interpretable machine-readable optical code is found, generate a signal indicating that the security mark has failed the authentication test (S6).

14. The method according to claim 13, wherein the verification of authenticity comprises the following steps 53 Illuminating the discovered safety marking with electromagnetic radiation in the transition wavelength range, 54 Determining the ratio of incident illuminance intensity to the intensity of light reflected from character components printed with security ink in authentic security markings and checking whether the ratio of incident illuminance intensity to the intensity of light reflected from character components printed with security ink in authentic security markings meets a specified test criterion.

15. Method according to claims 13 and 14, comprising the step 55 Generating a signal indicating that the security marking has passed the authenticity test (S5) if the test according to step S2 has shown that the security marking contains a machine-readable optical code interpretable when illuminated with electromagnetic radiation in an excitation wavelength range that excites fluorescence in authentic security ink, and furthermore, the test according to step S4 has shown that the ratio of incident illumination intensity to the intensity of light reflected from character components printed with security ink in authentic security markings meets the specified test criterion.

Images