WO2020126926A1 - Procédé d'étalonnage pour améliorer la vérification de motifs d'authentification par l'intermédiaire d'appareils d'acquisition d'images numériques - Google Patents

Procédé d'étalonnage pour améliorer la vérification de motifs d'authentification par l'intermédiaire d'appareils d'acquisition d'images numériques Download PDF

Info

Publication number
WO2020126926A1
WO2020126926A1 PCT/EP2019/085129 EP2019085129W WO2020126926A1 WO 2020126926 A1 WO2020126926 A1 WO 2020126926A1 EP 2019085129 W EP2019085129 W EP 2019085129W WO 2020126926 A1 WO2020126926 A1 WO 2020126926A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
recording
distance
recording device
authentication
Prior art date
Application number
PCT/EP2019/085129
Other languages
German (de)
English (en)
Inventor
Oliver Weiss
Jan Vorbrüggen
Original Assignee
Schreiner Group Gmbh & Co. Kg
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Schreiner Group Gmbh & Co. Kg filed Critical Schreiner Group Gmbh & Co. Kg
Publication of WO2020126926A1 publication Critical patent/WO2020126926A1/fr

Links

Classifications

    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
    • G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
    • G07D7/16—Testing the dimensions
    • G07D7/162—Length or width

Definitions

  • the application relates to a method for checking authentication patterns by digital image recording devices, in particular a calibration method to improve this check.
  • Authentication patterns for example noise patterns, are often printed on the surface of goods or other objects or their packaging in order to be able to check the authenticity, the origin of the original manufacturer or other circumstances relating to the authenticity of the goods.
  • Authentication patterns are used in many areas of technology for counterfeit-proof labeling of original products, original packaging, ID cards or other documents that can be potentially forged or otherwise imitated, reused or manipulated by unauthorized persons.
  • documents such as seals, ID cards and access authorizations, medicinal products and vehicle parts are examples of genuine technical products or commercial goods, but the problem exists with products of all kinds. Labels for labeling any products or their packaging (such as medication containers) are often imitated.
  • Authentication patterns to protect against imitation are printed as an image of an original, digital template and contain the finest print structures that can only be reproduced with loss due to digital image capture and printing technology copying. Therefore, only the original copies of authentication patterns printed with the authorization of the original manufacturer, i.e. the real print copies, still have a sufficient similarity to the original digital template, with which the authentication pattern, whether genuine or fake, is compared when its authenticity is checked .
  • the degree of similarity determined by comparison with the original digital template which is often referred to as the Q value or quality value of the respective printed authentication pattern, provides information as to whether the authentication pattern is genuine or imitated.
  • the limit value for the minimum percentage of agreement with the original digital template, so that an authentication pattern can be considered to be genuine is determined and set by the original manufacturer through tests.
  • the readers intended for reading out have a camera with a precisely set shooting distance (object distance from the lens system of the camera) when scanning or photographing, as well as a spacer for pressing on an authentication pattern, which ensures precise adherence to this shooting distance.
  • Some manufacturers of authentication patterns also allow the use of selected smartphone models for scanning or photographing authentication patterns;
  • the application programs provided for this ensure that an approximate shooting distance is maintained, or at least a focus setting or zoom function normalizes or compensates for the shooting distance, which is fundamentally variable on a smartphone.
  • the degree of similarity of authentication patterns can also be determined with a smartphone taking into account the image scale or the shooting distance;
  • the recording parameters of the smartphones can be read out when photographing an authentication pattern, for example the focal length, and taken into account when checking the authentication pattern.
  • an improved method for checking authentication patterns by retrieving calibration data is provided by digital image recording devices.
  • the calibration data retrieved are used to check and / or determine the actual size of a printed or physical authentication pattern as part of the authenticity check.
  • calibration data can be called up from a previous or previous calibration, in which a reference image is used to carry out a calibration of an image recording device, for example a smartphone.
  • Calibration data are calculated and saved for authentication tests at a later date.
  • the physical or figurative reference image is photographed under different recording conditions, and the digital images created are used to perform a calibration between different recording parameters, which, although they are actually already matched to one another on commercially available smartphones, nevertheless deviate from the factory setting, in particular can lead to an improved adjustment, with the result that the actual size of an object later photographed with this (or an identical smartphone), in particular an authentication pattern, can be determined using an accuracy that is at least as high, but often still higher Factory-set smartphones. This increases the precision of a later verification of an authentication pattern.
  • the calibration data obtained in this way can contain, for example, an optimal combination of these and / or other recording parameters, for example assignment between optimally combined values of two or more recording parameters; for example as a mathematical function or as a tabular assignment of these values.
  • FIG. 1 shows an object shown only schematically, which has a printed copy of a noise pattern used for authentication
  • FIG. 2 shows a typical arrangement when recording a digital image of this print copy of the noise pattern using a smartphone or other image recording device
  • FIG. 3 shows a schematic sequence of lossy reproduction steps in the production and, if necessary, later duplication of one for authentication purposes, the noise pattern in printed and / or digital form
  • FIGS. 4A to 4E show various template and reproduction stages shown in FIG. 3 using the example of an example Authentication noise pattern
  • FIG. 5 shows an exemplary reference image for the measurement-related determination of calibration data
  • FIGS. 6 and 7 show an exemplary embodiment with regard to the procedure for creating digital images from the reference image in order to obtain calibration data
  • FIG. 9 shows the best possible assignment of the recording distance and varied focus settings obtained from the interpolation curves in FIG. 8, and an interpolation curve here from.
  • the checking of authentication patterns for example noise patterns, which represents the preferred application of the proposed method, is explained with reference to FIGS. 1 to 4E.
  • the method has other uses; for example quality measurements on lettering images or on other printed (i.e. typographically realized and thus embodied as a printed copy) motifs, e.g. on a Siemens star or on another visual, e.g.
  • Authentication patterns consist of areas that are black or at least dark and white or light.
  • noise patterns consist of a large number of small, mostly square, elementary surfaces, of which approximately half are black and the other half are white and the distribution of light and dark contrast surfaces is apparently chosen at random.
  • QR codes that are made up of relatively large, individually recognizable elementary surfaces, for example to open an internet link and / or call up information by graphing them with a smartphone etc.
  • QR codes that are made up of relatively large, individually recognizable elementary surfaces, for example to open an internet link and / or call up information by graphing them with a smartphone etc.
  • the high density of the contrast surfaces in the real printed copy of the authentication pattern guarantees that the printed pattern on the object, its packaging or its label, even if the most modern equipment is used, for example for scanning or otherwise digitally capturing the printed noise pattern, is never reproducible without loss.
  • the loss of correspondence is such that when digital photography or scanning is performed only once, namely at least a certain degree of similarity (for example 70 or 80%; depending on the authentication pattern) is exceeded.
  • the imitator must obtain at least one genuine print copy of an authentication pattern and record it digitally, ie take photographs or scan it and, after processing the image accordingly, use it to produce one or more print copies of a quasi "second generation" type, which imitation should serve. If such a second-generation print copy is photographed or digitally recorded a second time to check its authenticity, there is an additional loss of information due to the repeated photographing and the subsequent, repeated printing process for the imitated print copy. As a result of the now further reduced degree of similarity, the second-generation printed copy can be recognized as an imitation and, if necessary, a forgery.
  • the user of a reader for authentication patterns, a scanner intended for this purpose or a smartphone with appropriate software equipment can, for example, use the electronic key to access the Internet and / or other data to download or generate the original digital noise pattern, i.e. Obtain a digital copy of the printed copy to regenerate this original digital copy on his smartphone or other image capture device and compare his digital photo with it.
  • the calculable degree of similarity also called Q-value or quality value and often expressed as a percentage (corresponding to theoretically 100% with complete agreement), provides information as to whether the photographed printed copy is genuine (ie a first-generation print copy) or imitated (ie a copy) Print copy of at least the second generation).
  • a limit value (Qmin) is often specified, which indicates the minimum required degree of similarity (for example, 75%) so that the printed copy can be considered genuine; Digital photos of printed copies with a lower degree of similarity to the electronic original indicate that the photographed printed copy is not an original printed copy of the authentication pattern.
  • FIG. 1 shows a schematic top view of the surface 101 of an object 100 which is provided with an authentication pattern A (here a noise pattern).
  • the noise pattern A can be, for example, a security feature 10 which serves to check the authenticity of the object, ie to check whether this specific object is not an unauthorized copy, forgery or other imitation.
  • the authentication pattern A can be used to retrieve information, for example via the Internet, to view and, if necessary, change a database, to exchange other information or to notify the manufacturer, etc. serve. However, this is not necessary.
  • the authentication pattern A can be used to grant or obtain access authorization, for example in a building or to an ATM, or for another authorization of the owner or user.
  • the article 100 provided with the authentication pattern A can be a document or ID or any other article 100 or a packaging thereof, alternatively also a label stuck onto or onto the article 100 or to be stuck on, ie intended for later stucking.
  • the authentication pattern A is a printed pattern, ie a print example P; For example, a contrast pattern of monochrome, dark or black, approximately square pixel areas and corresponding recesses gene between them.
  • the authentication pattern A can also be etched, lasered or engraved, for example in a metallic surface towards or below a surface, but from this it is visible and / or digitally detectable.
  • the authentication pattern A can be provided on flat surfaces as well as curved surfaces, for example of a vehicle part or a pharmaceutical container.
  • the object 100 can have a print 102 on its surface 101, which, for example, has inscriptions in the form of alphanumeric characters and images, graphics or other graphic or visual elements, in particular if the object is or comprises a label.
  • the authentication pattern A can be an additional, separate element of such printing 102; alternatively, the authentication pattern A can also be combined with alphanumeric or other, for example graphic, elements of the printing 102, for example combined with it; so the noise pattern can form part of a more complex labeling or printing pattern or vice versa. For example, actually homogeneous partial areas of alphanumeric or graphic elements can be further subdivided by the noise pattern without the legibility or clarity of these elements being significantly impaired.
  • the surface 101 of the object 100 In contrast to the original, digital image template DO (cf. FIG. 3), which defines the noise pattern, the surface 101 of the object 100 only bears a printed copy P, ie an objectively and permanently embodied that is produced by a printing technique or another method Image of the digital template of the noise pattern. If the print copy P is an authentic, original print copy PI of the first generation, it was printed directly by the manufacturer or authorized dealer based on the original digital template DO. Otherwise, however, it is a printed copy P2 of the second generation, which represents an unauthorized copy or an indirect copy of an original printed copy PI of the first generation created by the manufacturer for test purposes only.
  • Such a copy P2 has greater deviations from the original digital template DO, which cannot be seen with the naked eye, but can be determined by photographing or scanning and then calculating a degree of similarity Q. At first, however, it is uncertain whether there is an authorized printed copy PI or an imitated P2.
  • FIG. 2 shows a typical arrangement for creating a digital image B from an existing print copy P of an authentication pattern A, which is used for authentication and in particular serves as security feature 10.
  • the photographic copy P is photographed or scanned, for example, with the aid of a smartphone 60, with a built-in or separate digital camera 70 or with another image recording device 50, for example with a scanner 80 or a specially designed reader 90.
  • those for reading, ie pictorial capture and evaluation of authentication patterns are constructed and placed on or pressed onto them for photography.
  • FIG. 3 schematically shows the relationship between different reproduction steps, starting with the original digital template DO, a first-generation printed copy PI originally produced from it, which embodies the authentication pattern A (for example a noise pattern serving as a security feature).
  • PI becomes a digital image B
  • B l created, which has deviations from the digital template DO and the digital image B1 itself, inherent deviations resulting from the printing of the printed copy PI; Both are inevitable, but manageable, so that when comparing the digital image Bl with the original DO, a quality or Q value, ie a degree of similarity Ql of the digital image Bl, can be achieved or demonstrated, which lies above a predefinable reference or limit value Qmin .
  • Q1 can have a value of 80%.
  • a digital template Dl created in this case for imitation purposes
  • the test person creates a digital image B2, which in this case is, however, only a second generation digital image and all deviations from the printed copy P2 second generation, which resulted from the two print reproduction, contains and can only have a lower degree of similarity Q2 ⁇ Ql.
  • a comparison with a limit value Qmin will show that the checked print copy P2 is not a printed copy PI authorized by the manufacturer or otherwise authorized.
  • the calculation of the quality grade Q serves as the basis for a result or an evaluation result E (Q), which indicates whether the photographed print copy P is an authentic, original copy or a printing copy of such. This prevents or at least makes it more difficult to output imitations of original prints as original prints.
  • FIGS. 4A to 4E show different submission and reproduction levels using the example of an authentication pattern A, which is used, for example, for checking the authenticity. It is designed as a noise pattern here, but any other two-dimensional motif can also be selected, including one that is colorful instead of black and white; this also applies to all other figures and other exemplary embodiments of this application.
  • Figure 4A shows schematically (in a greatly enlarged detail view) an authentication pattern A as an original, geometrically defined digital template DO, which serves as a template for the printing or other production of a security feature 10 on at least one object to be authenticated and for example in a database of the Manufacturer saved and can be used or restored by a reader or smartphone with the help of an access code or key for comparison purposes.
  • FIG. 4B to 4E show different forms of reproduction and further forms of template that can be produced from this original digital template DO.
  • FIG. 4B shows a print copy PI of the authentication pattern A, which is created directly from this original template DO according to FIG. 4A, as it is embodied permanently and against the surface or another area of a label, a packaging part or any object, for example as printing.
  • FIG. 4B shows a section of this first-generation printed copy PI of a photograph taken with a microscope. Irrespective of the distortions caused by the microscope image, it can be guessed that the very first printing reproduction PI entails certain information losses or deviations from the original digital template DO.
  • FIG. 4C also shows a printed copy PI created directly from the original template DO according to FIG. 4A; but now in the form of a scan Bl obtained with a reader.
  • the loss of information compared to the original digital template is much greater here.
  • a scan of a reader FIG. 4C
  • a microscopic image as in FIG. 4B; at least for reproduction purposes, the microscope is the more suitable image recording device.
  • Figures 4B and 4C do not show the printed copy of the first generation PI itself, but only digital recordings of it, as is already evident from the different levels of information loss.
  • the first generation printed copy PI itself cannot be shown here (simply because of the much smaller real dimensions), as it e.g. is firmly printed and embodied on the object, for example on an original, original packaging.
  • the microscope image from FIG. 4B is not yet referred to as a first-generation digital template, because the microscope image from FIG. 4B must first be subjected to computational image changes before (as shown in FIG. 4D below) a usable (eg, higher-contrast) digital template second generation Dl can arise.
  • FIG. 4D shows such a second-generation digital template D1, which was calculated from the microscopic image from FIG. 4B and which can now be used for repeated printing reproduction, which then results in the second printing reproduction P2.
  • the printed copy of the second generation P2 itself cannot be represented either, since it is printed and embodied on the other object, for example on a reproduced, no longer original packaging.
  • FIG. 4E shows a digital image B2 (here a microscopic digital image) recorded with an image recording device 50 (see FIG. 2), which was recorded by the second-generation printed copy P2.
  • this second generation digital image clearly has greater information losses than the first generation digital image, which was also recorded with a reader (FIG. 4C); when comparing D1 and / or D2 with the original digital template DO (FIG. 3), the respective loss of information can be calculated as a degree of similarity or a Q value.
  • a manufacturer, dealer, buyer, customer or other (test) person creates a digital image B by photographing or scanning the printed copy P without knowing whether the present printed copy is an original printed copy of the first generation PI or a is a second-generation P2 copy that has only been created by reproducing.
  • This comparison is only possible with the comparison with the original digital template DO, which provides the degree of similarity Q.
  • the degree of similarity Q of the digital image with the original digital template DO is used to evaluate whether the printed copy P is an authentic copy authorized by the manufacturer PI or only an imitation P2 created by repeated reproduction.
  • the procedure of this application makes this assessment more reliable and objective than before.
  • the comparison of a digital image B with the original digital template DO from FIG. 4A is conventionally carried out according to criteria based on the structure of the digital template D created for testing; DO result.
  • Appropriate algorithms are used which, for example, make a comparison in blocks of nxn pixels between the present digital image B and the original digital template DO.
  • Q 100%
  • realistic values are, for example, at a value of 80%.
  • the maximum possible degree of similarity can for example at a value of 70%.
  • a threshold Qmin is usually defined in the middle between the two values (here at 75%), which serves as a differentiation threshold between authorized, genuine PI and imitated print copies P2.
  • a calculated degree of similarity Q of a compared digital image B 1 or B2 is always associated with a certain, ultimately unavoidable uncertainty, since neither the conditions in the printing or other production of the printed copy (first and possibly also second generation) nor the conditions in the creation of the Digital images (Bl; possibly also B2) are known, in particular are not individually traceable or can only be separated from one another.
  • the procedure for checking authentication patterns explained with reference to FIGS. 1 to 4E presupposes that the pattern available as a printed copy has a standardized size, i.e. Area dimension of the pattern and associated with it a standardized density of the pattern structures, so that the planned reproduction losses provide reliable information about the authenticity of the authentication pattern.
  • the manufacturers or users of authentication patterns for the smartphone models approved for authentication pattern testing specify that a certain shooting distance, a certain focus setting (in particular no minimum focus setting, which can usually be read as a dimensionless number) and / or a certain focal length (in the case of a smartphone with a camera capable of manual zooming) is maintained; only if the size specifications for these recording parameters (combination) are observed is an image with a predetermined imaging scale and thus with compliance with a predetermined, not artificially enlarged object or object size (in particular the printed copy P of the authentication pattern A) guaranteed.
  • the autofocus of a smartphone usually optimizes the sharpness of the picture, regardless of how far the photographed object is from the camera.
  • the focus setting resulting from the autofocus procedure can usually be read out.
  • smartphones also provide a value for the focal length of the currently set optical system, i.e. the distance between the main plane of the objective lens and the plane of the sharpest image.
  • the focal length is not always available, and even in these cases it is uncertain whether the value provided for the focal length also corresponds to the actual conditions, as is known from numerous counterexamples.
  • FIG. 5 shows an exemplary reference image R that can be used for the calibration method; it does not need to be an authentication pattern itself, only its actual size is relevant.
  • the reference image R in the sense of this application does not only refer to an actually pictorial object, ie a photo, printout, printout or otherwise a flat representation (for example on paper, cardboard or foil), but the reference image R can also be a reference object be or include one; for example (as seen in Figure 5) a ruler with a centimeter scale.
  • a rectangle of known, verifiable edge lengths is shown in FIG. ches (reference) image used; however, it can also be used as a reference object without an objective centimeter scale or instead only the ruler or another object with at least one known dimension can be used.
  • the method now explained is a calibration method, but in a different way than the calibration method used in the context of the production of smartphones.
  • a smartphone would be programmed in such a way that optimized combinations of shooting parameters when photographing, etc. were maintained.
  • the method proposed here is not one that would change or cancel the coordination of recording parameters with one another, which is determined by the manufacturer's internal factory settings;
  • the smartphone used for the calibration process will generally not experience any change - even in the case of photos taken with it in the future.
  • the calibration proposed here does not even change another smartphone of the same design, which will be used later for checking recording parameters, in the way that would normally be expected with a “calibration”;
  • the smartphone will also work with the device's own combinations of recording parameters set by the manufacturer whenever it creates a digital image of an authentication pattern or another object.
  • the present calibration method in particular the use of the retrieved, previously determined and (for example on the Internet or at another storage location of the use of authentication patterns) stored calibration data leads to the fact that if the same or an at least identical smartphone later an authentication pattern available as a print copy photographed and its actual size checked, this check is not based on the device's own manufacturer-made combinations of recording parameters (which are still set up and valid in this smartphone and are used for every digital photo, and although all photos of the printed authentication patterns were recorded with the device's own parameter combinations), but based on the calibration data obtained using the method described below.
  • the smartphone then abstracts from its own operating data or recording parameters, at least for the purpose of checking the size of the printed copy, but continues to work, in particular also during the photographing of the printed copy, unaffected by the calibration data.
  • a recalibration or recalibration of the smartphone does not take place with the calibration data, but rather a change in the calculation basis for the determined, ie - as far as he knows from the smartphone used - actual size (for example the actual length, width and / or Height dimension) of the authenticity pattern checked or to be checked.
  • the authentication pattern can be photographed with the autofocus function switched on; as before - as described below - the reference image or reference object (possibly also reference pattern) R.
  • a plurality of digital images r is created from the reference image or other reference object R from FIG. 5, as shown schematically in FIGS. 6 and 7.
  • all digital images rl,... are created with a constant recording distance, that is, with unchanged object g between the reference image R from FIG.
  • a different value of the focus setting F ie a varied focus or focus value, is set, which is shown in FIG. 7 as an example in the form of natural numbers 4, 5. 6,7,8,9, 10 (which do not represent any reference signs in the actual sense) is shown.
  • the position of the focus plane in front of the camera also changes, that is, the object width at which the sharpest image is actually or would be expected with the focus value selected in each case.
  • These sharpness levels (or levels or distances of maximum sharpness of the associated digital image), that is to say the respectively optimal object widths g, are therefore sometimes closer to or sometimes further away from the camera 70, and only in the ideal case of an optimal coordination of the recording parameters, in particular here Focus value F and the object distance g on top of each other lies the object distance g, which delivers the sharpest digital image, especially with the value for g already automatically set on the device side (in FIG. 7, therefore, according to a focus value of 7).
  • Figure 7 is not drawn to scale; in particular, the distances between the focus levels for different focus values F are greatly exaggerated compared to the overall distance from camera 70 of smartphone 50; 60 (or also from its image sensor 75). However, in practice it can happen that, for example, at small shooting distances, measurable differences, ie offset shifts, occur between the factory-selected shooting distance g and the ideally changed shooting distance based on the calculation explained below.
  • the degree of similarity or quality value Q itself can be used as a measure of its image sharpness, especially since only different photos of the same reference or test object are to be compared with one another, but (at least not yet here) that Authentication pattern with a real digital template.
  • various methods can be used to measure the sharpness of the photos rl, r2, ... of the reference object R; these methods differ in the steepness of the sharpness measure S around the sharpest possible recording. Under the Assuming that you are already close to the sharpest possible image, the steepest dimension in this area is best suited.
  • the Laplace filter is preferably used to calculate the focus; this is calculated, for example, in a discrete form according to:
  • L (x, y) I (x, y) .h (x, y) is calculated and its contribution, in particular its amount or determinant value, is summed up for all pixels distributed over the M rows and N columns of the digital image B1.
  • This sum formation only takes into account the contribution of those pixels for which
  • T only pixels in the region of large local changes in brightness are typically taken into account.
  • another filter such as an edge filter, can also be used, in which different and / or different numbers of neighboring pixels can also be taken into account and multiplied by the above or other numerical values.
  • FIG. 8 shows the values for the image sharpness s of the digital images r of the reference object R calculated in this way, namely for varying focus values F for each fixed recording distance g (see FIG. 7). This series of measurements was repeated in addition, also for differently set recording distances g (here of 60, 67, 72 and 77 millimeters) that were constant during the variation of the focus values F.
  • the image sharpness values obtained are compared with one another and the focus setting belonging to the greatest image sharpness is selected for each series of measurements.
  • This focus setting forms the x coordinate of the vertex of the quadratic or other (polynomial) function and would actually be the optimal focus setting for the selected image distance g.
  • FIG. 9 shows the plot of the recording distances g as a function of the optimal focus distances from FIG. 8 determined in this way.
  • a mathematical function is in turn fitted to this data, e.g. a 3rd degree polynomial.
  • This function now shows the ideal course of the shooting distance g and focus setting F from each other, i.e. the ideally carried out (but not necessarily also actually carried out and implemented by the smartphone) assignment of these two parameters (and in the general case additionally possibly also the focal length f) to one another.
  • This ideally set parameter combination according to FIG. 9 can be used for a
  • the recording distance g and the focus value F can be either a polynomial or other mathematical function or in tabular form as a large number of value pairs or value n-tuples, and / or data derived therefrom serve as calibration data.
  • these data are subsequently identified with the addition “ref” (for “reference”) to separate them from those data and recording parameters to be distinguished from those digital images which are to be evaluated later (as described below) when an authentication pattern or another object (as described with reference to FIGS. 1 to 4E) is photographed with a smartphone that works with the smartphone used to generate reference data least identical or provided with an identical camera.
  • the calibration data should be used in particular for checking the actual size of the authentication pattern or printed copy in order to increase the accuracy of this size determination.
  • the identical smartphone 50; 60 can, for example, call up the calibration data and use it to photograph the authentication pattern A at the recognized or set recording distance with an improved focus setting (as preset on the device side) in order to obtain and / or select an even sharper digital image of the pattern and with the Compare original original artwork.
  • the calibration data can be used (either by using the identical smartphone itself or by a computing unit on the Internet), using the focus setting to calculate the shooting distance to the level of the sharpest image and from this the absolute object size G.
  • this provides an at least as precise, but often more precise value of the actual size of the authentication pattern A than without calling up the calibration data.
  • the method (according to claim 8 ff.) Can be carried out, for example, as follows:
  • Focus setting F is determined or read out or at least recorded.
  • the auto focus function of the smartphone 60 or its built-in camera 70 is preferably used, which usually selects already well set but not necessarily optimal focus value F for the recognized shooting distance from the object or printed copy. This focus setting F selected by the auto focus function is thus determined for the purpose of further calculation or is simply acquired directly.
  • the smartphones, cell phones or other image capture devices 50; 60 have a fixed focal length f, which is mechanically unchangeable at least during the image recording, is the actual object size G to be determined from the recording distance g when photographing the printed copy P or A and the number of pixels n of the object in the image B using the calibration data a ref and g ref can be calculated, namely according to
  • G n / a.
  • This object size of the printed copy calculated in this way is at least as precise, but often even more precise than the value calculated only with the device-specific parameter combinations.
  • the calculated object size of the printed copy is checked, for example by comparing it with a predetermined target size corridor for permissible sizes of printed copies of authentication samples, the decisions and / or measures based thereon (e.g. with regard to the recognition of the sample as genuine or alternatively the rejection of the sample as false) ) hit or displayed with greater reliability than before, which is a safety advantage over
  • Camera model (which according to FIGS. 5 to 9 is carried out at an earlier point in time, for example only once) preferably not only the relationship between the Focus setting and the image distance of the greatest sharpness determined (approximately by
  • Assignment of the ideal image distances per focusing value), but alternatively or additionally also the respectively assigned for a selected (reference) recording distance, i.e. achieved values of the (reference) image resolution are determined and stored as calibration data or as part thereof.
  • the actual image distance can be calculated from the focus setting using the calibration data and then the image resolution a of the digital image of the object to be checked can be calculated using the reference image resolution with an identical (reference) image distance.
  • the actual size G of the object (to be specified in metric or in other length units such as mm, cm, inch, etc.) for the object to be examined in the digital image to be checked can be calculated on the basis of its image size within the digital image (in pixels).

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Editing Of Facsimile Originals (AREA)
  • Image Analysis (AREA)

Abstract

L'invention concerne un procédé pour vérifier des motifs d'authentification (A), ce procédé comprenant au moins les étapes consistant : à créer au moins une image numérique (B) d'au moins un motif d'authentification (A) se présentant sous la forme d'un exemplaire imprimé (P) sous une forme imprimée ou sous une autre forme physique, au moyen d'un d'appareil d'acquisition d'images numériques (50), en particulier au moyen d'un smartphone (60) ; à vérifier l'authenticité du motif d'authentification (A), y compris, à déterminer la taille réelle (G) du motif d'authentification (A) imprimé ou se présentant sous une autre forme physique, la concordance entre la taille (G) déterminée du motif d'authentification (A) et une taille prévue de motifs d'identification authentiques et/ou une bande passante prévue pour la taille de motifs d'authentification authentiques étant vérifiée. Cette invention est caractérisée en ce que la détermination de la taille (G) réelle du motif d'authentification (A) comporte l'extraction de données d'étalonnage qui sont acquises et enregistrées au moyen de cet appareil d'acquisition d'images (50), au moyen d'un appareil d'acquisition d'images (50) de structure identique, ou au moyen d'un appareil d'acquisition d'images (50) au moins équipé d'un appareil de prise de vue (70) de structure identique.
PCT/EP2019/085129 2018-12-20 2019-12-13 Procédé d'étalonnage pour améliorer la vérification de motifs d'authentification par l'intermédiaire d'appareils d'acquisition d'images numériques WO2020126926A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102018133230.0 2018-12-20
DE102018133230.0A DE102018133230A1 (de) 2018-12-20 2018-12-20 Kalibrierungsverfahren zur Verbesserung der Prüfung von Authentifizierungsmustern durch digitale Bildaufnahmegeräte

Publications (1)

Publication Number Publication Date
WO2020126926A1 true WO2020126926A1 (fr) 2020-06-25

Family

ID=69061321

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2019/085129 WO2020126926A1 (fr) 2018-12-20 2019-12-13 Procédé d'étalonnage pour améliorer la vérification de motifs d'authentification par l'intermédiaire d'appareils d'acquisition d'images numériques

Country Status (2)

Country Link
DE (1) DE102018133230A1 (fr)
WO (1) WO2020126926A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116433670A (zh) * 2023-06-14 2023-07-14 浙江舶云科技有限公司 一种图像质量检测方法及检测系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102006051710A1 (de) * 2005-10-28 2007-05-03 Arc Seibersdorf Research Gmbh Längennormierung
US20150116530A1 (en) * 2012-05-30 2015-04-30 Easy Printing Network Limited Article authentication apparatus having a built-in light emitting device and camera
WO2017159608A1 (fr) * 2016-03-16 2017-09-21 凸版印刷株式会社 Dispositif d'identification, procédé d'identification, programme d'identification et support lisible par ordinateur contenant un programme d'identification
US20180276523A1 (en) * 2015-11-30 2018-09-27 Toppan Printing Co., Ltd. Identification method and identification medium

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7054461B2 (en) * 2002-02-15 2006-05-30 Pitney Bowes Inc. Authenticating printed objects using digital watermarks associated with multidimensional quality metrics
ES2424480T3 (es) * 2002-05-14 2013-10-02 Schreiner Group Gmbh & Co. Kg Patrones de autenticación visibles para documento impreso
DE102015106447B4 (de) * 2015-04-27 2024-07-11 Schreiner Group Gmbh & Co. Kg Manipulationssicheres Etikett

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102006051710A1 (de) * 2005-10-28 2007-05-03 Arc Seibersdorf Research Gmbh Längennormierung
US20150116530A1 (en) * 2012-05-30 2015-04-30 Easy Printing Network Limited Article authentication apparatus having a built-in light emitting device and camera
US20180276523A1 (en) * 2015-11-30 2018-09-27 Toppan Printing Co., Ltd. Identification method and identification medium
WO2017159608A1 (fr) * 2016-03-16 2017-09-21 凸版印刷株式会社 Dispositif d'identification, procédé d'identification, programme d'identification et support lisible par ordinateur contenant un programme d'identification
EP3432277A1 (fr) * 2016-03-16 2019-01-23 Toppan Printing Co., Ltd. Dispositif d'identification, procédé d'identification, programme d'identification et support lisible par ordinateur contenant un programme d'identification

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116433670A (zh) * 2023-06-14 2023-07-14 浙江舶云科技有限公司 一种图像质量检测方法及检测系统
CN116433670B (zh) * 2023-06-14 2023-08-29 浙江舶云科技有限公司 一种图像质量检测方法及检测系统

Also Published As

Publication number Publication date
DE102018133230A1 (de) 2020-06-25

Similar Documents

Publication Publication Date Title
EP1413972B1 (fr) Examination d'images de personnes
DE102006005617B4 (de) Verfahren zur Bewertung der Qualität eines Bildes, Verfahren zur Herstellung eines Dokuments, Computerprogrammprodukt und elektronisches Gerät
DE69909997T2 (de) Individualisierter fingerabdruckabtaster
EP1693781B1 (fr) Procédé et disposition pour l'enregistrement optique de données digitales biométriques
DE69634583T2 (de) System und Verfahren zum Prüfen von Identifizierungskarten
EP2144036A2 (fr) Procédé et dispositif destinés à la numérisation en 3D d'un objet
DE112020005864T5 (de) Verfahren und Einrichtung zur Verifizierung der Authentizität eines Produkts
WO2006021165A1 (fr) Procede et dispositif d'enregistrement optique de donnees biometriques
WO2020126720A1 (fr) Correction optique par apprentissage automatique
EP1022541A2 (fr) Procédé et dispositif pour déterminer la forme géométrique de matériau en feuilles ou de feuiles empilées
DE112019007066T5 (de) Fälschungssicheres Papierprodukt basierend auf Mikrostrukturerkennung, Herstellungsverfahren sowie Authentifizierungsverfahren dafür
WO2020126926A1 (fr) Procédé d'étalonnage pour améliorer la vérification de motifs d'authentification par l'intermédiaire d'appareils d'acquisition d'images numériques
DE69810528T2 (de) Verfahren und Vorrichtung zur Überprüfung von Wert- oder Sicherheitsgegenständen, insbesondere Banknoten
DE102021133868A1 (de) Mikroskopiesystem und Verfahren zum Modifizieren von Mikroskopbildern im Merkmalsraum eines generativen Netzes
DE102018125780A1 (de) Ähnlichkeitsbestimmung eines von einem Druckexemplar erstellten Digitalbildes mit einer Digitalvorlage
DE102008052248B4 (de) Verfahren und Vorrichtung für die Überprüfung von Linsenstrukturen
DE102011077895A1 (de) Verfahren und Vorrichtung zum Erstellen eines Dokumentenreferenzdatensatzes anhand eines Dokumentes
EP2394250A1 (fr) Procédé et dispositif pour vérifier des documents par utilisation d'une transformation en ondelettes
DE112019007001T5 (de) Fälschungssicheres Produkt mit zufälligen Mikropunkt-Merkmalen sowie Herstellungsverfahren und Verifizierungsverfahren dafür
EP2917897B1 (fr) Vérification de la qualité d'une image pivotante d'une structure lenticulaire
DE112017001026T5 (de) Verfahren und System zur ldentifizierung und Messung eines Defekts, der die Transparenz in einem Substrat für ein Sicherheitsdokument verringert
EP4205091A1 (fr) Procédé pour produire un identifiant numérique d'un exemplaire d'un produit d'impression, ledit exemplaire présentant au moins une image d'impression, téléphone intelligent ou tablette doté(e) d'un tel dispositif et procédé d'utilisation de ce dispositif
DE102022214390A1 (de) Verfahren zur Bestimmung von mindestens einer ersten Abmessung und einer zweiten Abmessung eines Objekts, Erfassungsvorrichtung, Verfahren zum Erfassen von multiplen Objektdatensätzen zumindest eines Objekts
EP4064221A1 (fr) Procédé de personnalisation d'un document d'identification, ainsi que procédé d'identification d'une personne en fonction des caractéristiques faciales biométriques et document d'identification
EP4127670A1 (fr) Procédé de reconnaissance de surfaces

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19829063

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19829063

Country of ref document: EP

Kind code of ref document: A1