WO2022122392A1

WO2022122392A1 - Method for producing a series of forge-proof packages, series of forge-proof packages, authentication method, and authentication system

Info

Publication number: WO2022122392A1
Application number: PCT/EP2021/082916
Authority: WO
Inventors: Stefan BORGSMÜLLER; Bernd Terhalle; Steffen Scheibenstock; Eduard Ziegler
Original assignee: Scribos Gmbh
Priority date: 2020-12-11
Filing date: 2021-11-25
Publication date: 2022-06-16
Also published as: CN116601685A; DE102020215752A1; EP4260301A1

Abstract

The invention relates to a method for producing a series of forge-proof packages (7) in that the series of packages (7) is divided into batches, each package (7) is assigned a batch-specific batch number (6), and each package (7) is assigned a package-individual series number (9). The batch-specific batch numbers (6) are encoded in order to form a machine-readable, batch-specific first graphical code (3), the batch-specific first graphical code (3) is Fourier-transformed into batch-specific two-dimensional Fourier patterns (2, 2', 21, 21', 22, 22'), the batch-specific Fourier patterns (2, 2', 21, 21', 22, 22') are printed on the respective packages (7) assigned the batch numbers (6) in a first printing method step, the package-individual series numbers (9) are encoded in order to form machine-readable package-individual second graphical codes (4) which are printed on the assigned packages (7) in a second printing method step, and the batch-specific Fourier pattern (2, 2', 21, 21', 22, 22') and the second graphical code (4) together form a forge-proof two-dimensional security feature (1, 1') for each package.

Description

Method for manufacturing a series of anti-counterfeiting packaging and a series of anti-counterfeiting packaging, an authentication method and an authentication system

The invention relates to a method for producing a series of counterfeit-proof packaging. The invention also relates to a series of anti-counterfeiting packaging made according to the manufacturing method. The invention also relates to a method for authenticating a package, a series of packages divided into batches, and an authentication system.

DE 10 2017 206467 A1 describes a method for producing a security feature in which a first graphic code is provided and a second graphic code is provided in the form of a Fourier pattern, and the first and second graphic codes are combined concatenated by inserting the Fourier pattern into the outer border of the first graphic code.

DE 10 2017 206466 A1 also describes a method for producing a security feature in which a machine-readable graphic code is made available, the graphic code is embedded in a two-dimensional discrete complex function, the two-dimensional discrete complex function is Fourier transformed and becomes a two-dimensional image is binarized. The binarized Fourier image is printed with a pixel resolution that is large enough to print enough structures to to enable inverse Fourier transformation, which enables a reconstruction of the machine-readable graphic code.

A disadvantage of the two security features mentioned is that a large number of packages marked individually for the package can only be produced with considerable effort.

It is therefore the object of the present invention to provide a method for producing a series of forgery-proof packaging which avoids the disadvantage mentioned above.

It is also an object of the present invention to provide a series of counterfeit-proof packaging that avoids the disadvantage mentioned above.

It is a further object of the present invention to provide a method for authenticating a package of a batched series of packages.

Finally, the object of the invention is to provide an authentication system with which the authentication method mentioned can be carried out.

The object is achieved in its first aspect by a method having the features of claim 1.

The manufacturing process involves a series of anti-counterfeiting packaging.

Under packaging here is a general term to understand both the

Sales packaging of a product, for example a toothpaste tube also the outer packaging, for example the folding box of the product. The packaging can also be a part of a packaging, for example a label that is stuck onto an outer packaging.

First, the series of packaging is divided into preferably disjunctive batches. Each package in a batch is given a batch-specific batch number. This is assigned to each packaging. Furthermore, each packaging is assigned a packaging-specific serial number. A batch can easily include several hundred, thousand, ten thousand or hundreds of thousands of packages, while the individual serial number of each package is assigned precisely and individually. The packaging-specific serial number of each packaging is unique and is not repeated in the series.

The batch number as well as the serial number can be a sequence of numbers, sequences of letters, character sequences or a combination of the characters mentioned above. The batch and serial numbers are pieces of information that can be used to identify the batch or the individual packaging. It can also be a two-dimensional character or image arrangement.

The lot-specific lot numbers are encoded in machine-readable lot-specific first graphic codes. The packaging-specific serial numbers are encoded in packaging-specific, machine-readable second graphic codes. The first graphic codes and/or the second graphic codes are preferably conventional machine-readable codes such as 2D barcodes, in particular data matrix codes or QR codes. It can also be one-dimensional barcodes, but also trill codes, quickmark codes, shot codes, etc. General the machine-readable graphic codes include optoelectronically readable symbols that consist of bars or dots of different widths and gaps in between with the highest possible contrast.

In the production method according to the invention, the batch-specific first graphic code is Fourier-transformed into a batch-specific two-dimensional Fourier pattern. Identical batch-specific first graphic codes are Fourier-transformed into identical batch-specific two-dimensional Fourier patterns, while different batch-specific first graphic codes are Fourier-transformed into different batch-specific two-dimensional Fourier patterns. A Fourier pattern is a two-dimensional image, e.g. B. is shown in Fig. 4a. Fourier patterns are hardly perceptible to the viewer as such, as they appear as a kind of satin gray background. Identity or difference of Fourier patterns can only be determined for the human observer when they are placed next to each other, i.e. with a direct comparison.

According to the invention, the batch-specific Fourier patterns are printed in a first printing process step on the packaging assigned to the batch number, and the packaging-specific, machine-readable second graphic codes are printed on the packaging assigned to them in a second printing process step.

The batch-specific two-dimensional Fourier pattern and the packaging-specific second graphic codes can be printed one on top of the other, side by side, partially overlapping on the packaging.

The generation of the Fourier pattern is first disclosed below. The machine-readable first graphic code is embedded in a real amplitude function of a two-dimensional, discrete complex function G(fx,fy) with an fx frequency coordinate and an fy frequency coordinate. For this purpose, the code is positioned in a two-dimensional, preferably square, image template, with the x and y values of the image template being interpreted as fx and fy frequencies.

In principle, complex numbers or complex functions can be represented as the sum of the real and imaginary parts or sum of a real function and an imaginary function or in polar coordinate notation as the product of an amplitude function and phase function.

The method according to the invention is based on making the machine-readable first graphic code available as an amplitude function of a two-dimensional discrete complex function G(fx,fy). The amplitude function preferably has either the function value 0 or the function value 1 via the two frequency coordinates fx and fy. The black coordinate points of the code positioned in the image template receive the value 1 and the white coordinate points the value 0.

A suitable phase function e ^i(p (fx, fy) is preferably added to the real amplitude function by multiplication.

The task of the phase function is to smooth the frequency spectrum of the amplitude function. The phase function e ^i(p (fx, fy) can be a random phase. The first graphic code for generating the phase function is preferably initially designed as a random gray value image. The outlines of the gray value image correspond to the first graphic code, except that the values do not lie at zero (white) and one (black) as in the construction of the amplitude function, but are random gray values between white and black. The gray values are assigned to numbers between 0 and 2TT. If the gray value is black, the Phase 2TT, and if the gray value is white, the phase is zero. The other gray values are assigned an angle between 0 and 2TT (radians), depending on the gray level. The blacker, i.e. darker, the color, the larger the angle. On in this way the random gray value image can be unambiguously converted into a phase function e ^i(p (fx, fy), and by multiplying the amplitude function by the phase function n the complex-valued function G(fx,fy) is formed.

However, other phase functions can also be added to the real amplitude function.

The two-dimensional discrete complex function G(fx,fy) is then Fourier transformed and the resulting Fourier transformed g(x,y) is binarized into a two-dimensional image. For binarization, the real part of the Fourier transform g(x,y) can be determined and binarized using a threshold value. The real part of the Fourier transform again contains gray levels. Here, binarization of an image means that each pixel of the image whose gray level is above the threshold value is assigned the value 1 and each pixel whose gray level is below the threshold value is assigned the value 0. 10% binarization then means that 10% of the pixels are black and 90% of the pixels are white. 50% binarization then means that 50% of the pixels are black and 50% of the pixels are white, etc. Alternatively, the real part or the phase of the Fourier transform g(x,y) can also be determined and binarized using a threshold value. Other binarization options are known from the prior art (Goodman, JW, Introduction to Fourier Optics, McGraw-Hill (New York) (1996)).

The real part of the Fourier transform is preferably binarized and the batch-specific Fourier pattern is formed by it.

The Fourier pattern is preferably printed with a binarization of less than 50 percent, preferably less than 20 percent, particularly preferably less than 10 percent.

The batch-specific Fourier pattern, which is printed on the packaging in the first printing process step, is understood here as the binarized real part of the Fourier transformation of the function G(fy,fy).

In a particularly preferred embodiment of the manufacturing method according to the invention, the first printing method step has a higher printing resolution than the second printing method step, and the batch-specific Fourier patterns are printed with a higher printing resolution than the packaging-specific second graphic code.

The first printing process step can be selected from gravure printing, offset printing, screen printing or flexographic printing, while the second printing process preferably takes place by means of a digital printing process from the group inkjet printing, thermal transfer printing, laser printing, laser engraving. The first step of the printing process is carried out using a classic printing process, and the Fourier pattern is printed on the packaging with a high effective resolution. The Fourier pattern contains the hidden, machine-readable first graphic code, which contains at least the batch number as information. In classic printing processes, a printing plate or a printing cylinder is first produced, which is then used to print a series of similar packaging. First of all, the prints are the same on all packaging in the series.

In contrast to this known procedure, the invention proposes in one aspect to already diversify the first printing method step and thus the Fourier pattern. For this purpose, the series is divided into batches. These are disjoint sets of the series. If, for example, several printing presses are provided for the production of the series, each printing press could be assigned its own individual printing plate with a batch-specific two-dimensional Fourier pattern that contains the batch number as information. It would also be conceivable for the printing plates, which contain the batch-specific Fourier pattern, to be changed regularly within a printing press and each receive a different batch number as information. If several packages are printed simultaneously with one printing plate, several batch-specific Fourier patterns could be arranged on one printing plate. If, for example, two printing machines are used to produce one hundred thousand packages, and the printing plates are changed once during production on each printing machine, and there are print templates for twenty packages on one printing plate, then the series can have 2 x 2 x 20 = 80 different batch numbers be generated. Typically stand by classic printing processes, high print resolutions are available, e.g. B. 2,000 dpi or 4,000 dpi.

In the second printing process step, a low resolution open machine-readable second graphic code is printed onto the package using a digital printing process. The machine-readable graphic code contains at least one packaging-specific serial number as information. Individual packaging means that the number is only used once in the series of packaging and is therefore unique. The serial number is preferably selected from a large range of numbers or generated cryptographically so that potential forgers are unable to guess a valid serial number.

In addition to the serial number, the second graphic code preferably contains a URL with which a user can connect to an authentication server. Such a URL could be, for example, https://www.authserver.com/, preferably the serial number is part of the URL, for example https://www.authserver.com/serialnumber/12345, where 12345 is the serial number. Thus, by scanning the machine-readable second graphic code with a smartphone, a consumer can connect to the authentication server via an Internet browser.

Favorably, the second printing process step takes place after the first printing process step. A reverse order is also conceivable. The second printing process step can be carried out in the same printing machine as the first printing process step if the printing machine provides printing units using classic printing methods and printing units using digital printing methods. However, the second printing process step can also take place on a different printing press than the first printing process step. With that they can Printing machines can also be made available in different rooms or even different cities. It is also conceivable that the first printing process step takes place at a producer for packaging and the second printing process step is carried out at a later point in time at the manufacturer of the product, for example on the packaging line.

The batch-specific Fourier pattern is preferably printed with an effective resolution of at least 600 dpi. To do this, a printer must be used that has a higher print resolution, preferably a significantly higher print resolution of 1,000 dpi, 2,000 dpi or even 4,000 dpi. All intermediate values are also disclosed here. It has been shown that commercially available printers have a print resolution of less than 600 dpi, so that when the batch-specific Fourier pattern is photographed and the photographed Fourier pattern is printed out again, so much information is lost that a reconstruction of the batch-specific first graphic code is no longer possible. If the Fourier pattern is to be copied and the information contained therein is to be retained, then at least the effective resolution of the Fourier pattern must be retained during the copying process. If, for example, during the copying process either the scanner or the printer used have a lower resolution than the effective resolution of the Fourier pattern, the Fourier pattern will only be transferred incompletely and the graphic batch-specific code contained therein will be damaged and thus no longer readable. It has been shown that with an effective resolution of at least 600 dpi, copy protection against commercially available office copiers is guaranteed.

The effective resolution of the printed batch-specific Fourier pattern is determined by the resolution of the printer and by the positioning of the batch-specific first graphic code within the Image template determined. The image template has half the width fx imit and half the height fy imit. Both values indicate the pixel numbers of half the width and half the height of the image template. The special positioning of the batch-specific first graphic code within the image template in the fx and fy direction defines the greatest horizontal distance fx_max of the batch-specific first graphic code from the center of the image and the greatest vertical distance fy_max of the batch-specific first graphic code from the center of the image.

Quotients fx_max/fx_limit and fy_max/fy_limit are formed. The effective resolution in the x direction is the product of the quotient fx_max/fx_limit multiplied by the resolution of the press ( as a formula: (fx_max/fx_limit) * resolution ), and the effective resolution in the y direction is the product of the quotient fy_max / fyjimit multiplied by the resolution of the press ( as a formula: (fy_max/fy_limit) * resolution ). Effective resolution is the greater of effective x-direction resolution and effective y-direction resolution. The batch-specific first graphic code is preferably positioned so far outside on the original image that the values fx_max and fy_max are so large that the effective print resolution is above 600 dpi.

In its second aspect, the object is achieved by a series of counterfeit-proof packaging with the features of claim 15, which is manufactured using one of the manufacturing methods mentioned above. What was said about the method also applies to the series of forgery-proof packaging and should also be disclosed in connection with the series.

According to the invention, a series of packages is divided into batches and each

Packaging assigned a batch-specific batch number, and each

In addition, the packaging will have a packaging-specific serial number assigned. The lot-specific lot numbers are encoded in machine-readable lot-specific first graphic codes, and the lot-specific first graphic codes are Fourier-transformed into lot-specific two-dimensional Fourier patterns. In a first printing process step, the batch-specific Fourier patterns are each printed on the packaging assigned to the batch number, with the batch-specific Fourier patterns preferably being printed with an effective resolution of at least 600 dpi.

The packaging-specific serial numbers are encoded in packaging-specific, machine-readable second graphic codes that are printed on the associated packaging in a second printing process step. The resolution of the second printer with which the second printing process step is carried out can be lower, possibly significantly lower, than the resolution of the first printer with which the first printing process step is carried out.

The packaging-specific, machine-readable second graphic code is preferably openly readable. In this application, openly readable is understood to mean that it is printed on the packaging in such a way that it is detected with a conventional reading algorithm downloaded onto a commercially available smartphone that has a camera and the information it contains, for example the Serial number, can be read. The reading algorithm can be a commercially available QR code scanner, data matrix code scanner or other corresponding scanner.

The batch-specific Fourier pattern and the second graphic code together represent a forgery-proof two-dimensional security feature according to the invention for each package. The batch-specific Fourier pattern and the second graphic code may be placed one on top of the other, side by side, partially overlapping on the packaging.

The Fourier pattern is conveniently binarized. Binarized Fourier patterns have the advantage that they only have the values white and black and can therefore easily be printed with printing machines. A 20% binarization has proven advantageous. The 20% binarization still contains enough information and can be printed with little ink.

The object is achieved in its third aspect by a method for authenticating a packaging of a series of packaging divided into batches with the features of claim 20.

The authentication method is advantageously carried out using an authentication system which is described further below. The authentication method is suitable for implementation with one of the series of counterfeit-proof packaging mentioned above, which are manufactured using one of the manufacturing methods mentioned above.

According to the invention, the authentication method is carried out using a mobile terminal device which has a camera. The mobile end device can be a commercially available smartphone and the camera can be a commercially available camera integrated into smartphones.

An image of the two-dimensional security feature of the packaging is captured with the camera of the mobile device. The image is then evaluated. The captured image is supplied to a second reading algorithm for the second graphic code, and the second reading algorithm reads a packaging-specific serial number from the second graphic code.

The captured image is subsequently, simultaneously or beforehand, inverse Fourier transformed and the inverse Fourier transformed image is fed to a first reading algorithm for the batch-specific first graphic code. The reading algorithm reads the batch number from the batch-specific first graphic code.

The read batch number and the read serial number are authenticated. It is determined whether the batch number and serial number and their assignment to one another are valid.

First, batch-specific batch numbers and packaging-specific serial numbers are assigned to one another as pairs and preferably stored in a database. Advantageously, the serial numbers are not formed consecutively, but are cryptographically encrypted, so that a counterfeiter cannot simply think up or guess a valid serial number.

The packaging-specific serial number is compared with the serial numbers stored in the database, and the batch-specific batch number is also compared with the batch-specific batch numbers stored in the database. If both numbers match individually and as a pair, the security token is authenticated. The packaging with the security feature is then the original packaging. If the batch number cannot be read, for example because it is a blurred photocopy of the Fourier pattern of the security feature, the security feature will not be authenticated. If the serial number has already been queried one or more times, the serial number is preferably no longer authenticated in the next query. A serial number may preferably only be queried once or a certain number of times and is then blocked or invalidated.

The inverse Fourier transform program required for the authentication process, the first reading algorithm and the second reading algorithm may all or some of them be arranged on the mobile terminal. However, it is also conceivable that one or more of these programs are stored on an authentication server. The database which has stored valid batch number and serial number pairs is preferably also stored on the authentication server.

The serial numbers read out for the individual packaging are advantageously fed to the authentication server and the serial numbers read out are compared with the valid serial numbers stored in the database on the authentication server, and the serial numbers read out are authenticated if they match one of the valid serial numbers stored. The batch numbers read out are preferably also sent to the authentication server and compared with valid batch numbers stored on the authentication server.

It is also checked whether the assignment of the stored serial number and the stored batch number matches the assignment of the batch number and the serial number that have been read. Are the individual numbers correct? and the assignments also match, an authentication signal is sent to the mobile terminal device. The connection between the mobile terminal device and the database server can be established via a conventional, preferably wireless connection such as a W-LAN connection to the Internet, a 3G/4G/5G connection to the Internet or a similar connection.

In the fourth aspect, the object is achieved by an authentication system having the features of claim 25.

The method mentioned above is carried out with one of the authentication systems described here; the authentication systems described below are preferably also suitable for carrying out one of the authentication methods mentioned above.

The authentication system includes a series of forgery-proof packaging, as described above, a mobile terminal device with a camera and with a transceiver unit, with which the security features of the packaging detected by the camera can be transmitted to an authentication server.

The authentication system also includes a Fourier inverse transformation program with which the batch-specific Fourier pattern can be Fourier inverse-transformed, a first reading algorithm for the batch-specific first graphic code, which can read the batch number from the Fourier inverse-transformed image, a second reading algorithm for the packaging-specific second graphic code, which reads the individual serial number can. The authentication system also includes the authentication server, which is in data communication with the mobile end device and on which the serial number and batch number assigned to each package are stored in association with one another and on which the batch numbers and serial numbers transmitted by the mobile terminal can be authenticated and with which an authentication signal is sent to the mobile terminal is deliverable.

Here, too, what has been disclosed for the series of forgery-proof packaging and what has been disclosed for the method for producing a series of forgery-proof packaging applies mutatis mutandis as well.

In particular, the Fourier inverse transformation program can be downloaded to the mobile terminal device, but it can also be the case that the Fourier pattern captured by the camera is transmitted to the authentication server by means of the transceiver unit. Likewise, the first as well as the second reading algorithm can be stored either on the mobile terminal device or on the database authentication server. However, both reading programs and the Fourier inverse transformation program are advantageously stored on the mobile terminal, so that only the batch-specific batch numbers already read and the individual packaging serial numbers that have been read need to be transmitted to the authentication server via the transceiver unit of the mobile terminal. The numbers can be transmitted with a significantly lower data volume than the Fourier pattern scanned by the camera.

The invention is described using several exemplary embodiments in 18 figures. show: Fig. 1 coding of a batch number in a data matrix code,

Fig. 2 positioning of the data matrix code in an image template,

3a amplitude function in a first embodiment with low effective resolution,

3b phases in a first embodiment with low effective resolution,

4a Fourier pattern as real part of the Fourier transformation of the data matrix code in FIGS. 3a, 3b,

Fig. 4b Fourier pattern of Fig. 4a in a 50% binarization,

4c Fourier pattern of FIG. 4a in a 20% binarization,

5a real part of the Fourier inverse transformation of the Fourier pattern in FIG. 4a,

5b real part of the inverse Fourier transformation of the binarized Fourier pattern in FIG. 4b,

5c real part of the inverse Fourier transformation of the binarized Fourier pattern in FIG. 4c,

6 first embodiment of the security feature according to the invention with the 20% binarized Fourier pattern and a packaging-specific QR code, Fig. 7 inverse Fourier transformation of the security feature in Fig. 6,

Fig. 8 photocopy of the security feature in Fig. 6 with half resolution,

Fig. 9 Fourier inverse transformation of the copied security feature in Fig. 8,

10a amplitude function in a second embodiment with high effective resolution,

10b phases in a second embodiment with high effective resolution,

11a Fourier pattern as real part of the Fourier transformation of the data matrix code in FIGS. 10a, 10b,

Fig. 11 b Fourier pattern in Fig. 11 a in a 50% binarization,

11 c Fourier pattern of FIG. 11 a in a 20% binarization,

12a real part of the Fourier inverse transformation of the Fourier pattern in FIG. 11a,

12b real part of the Fourier inverse transformation of the Fourier pattern in FIG. 11b,

12c real part of the Fourier inverse transformation of the Fourier pattern in FIG. 11c, 13 second embodiment of the security feature according to the invention with the 20% binarized Fourier pattern and packaging-specific QR code

Fig. 14 inverse Fourier transformation of the security feature in Fig. 13,

Fig. 15 photocopy of the security feature in Fig. 13 with half resolution,

Fig. 16 inverse Fourier transformation of the security feature in Fig. 15,

Fig. 17 packaging series divided into four batches,

Fig. 18 Packaging series in Fig. 18 with a batch-specific Fourier pattern and with a packaging-specific QR code printed on three packaging.

1 and 2 show a basic representation of the construction of a security feature 1 according to the invention.

3 to 9 show the construction of a first embodiment of the security feature 1 with a low resolution. 10 to 16 show the construction of a second embodiment of a security feature 1' with a high effective resolution.

A basic structure of the security feature 1, 1' according to the invention is shown in FIGS. 6, 8, 13, 15 and the right three packages of FIG. The security feature 1, 1' according to the invention basically has two components, namely a binarized batch-specific one two-dimensional Fourier pattern 22, 22', which can be seen in Fig. 6 and Fig. 13 as a kind of background noise, but is actually the real part of the Fourier transform of a batch-specific first graphic code 3 brought into mathematical form, which is embodied here as a data matrix code , as well as a packaging-specific, machine-readable second graphic code 4 printed on the batch-specific two-dimensional Fourier pattern 22, 22′, which is embodied here in the form of a QR code.

In principle, both the first graphic code 3 and the second graphic code 4 can be designed in particular as 1D or 2D barcodes, in particular as a data matrix code or QR code. It should be machine-readable, i.e. it should be able to be read using a commercially available reading algorithm, which can be downloaded to mobile devices in the form of an app, for example. According to FIG. 6, the binarized Fourier pattern 22 and the QR code 4 are preferably printed one on top of the other. Due to their clearly different graphic design, however, they do not "interfere" with each other.

1 shows an assignment of a batch number 6, which is called ABCDEF here, to the first graphic code 3, here a data matrix code. The batch number 6 can have almost any shape, it can be a sequence of letters, characters, numbers, bits or a combination thereof. The batch number 6 is coded in a data matrix code 3 according to FIG. In this exemplary embodiment, the data matrix code 3 is the machine-readable first graphic code 3. Both are therefore given the same reference number. The data matrix code 3 is the same for each pack 7 of a batch.

A series of packages 7 is divided into a number of batches. In the example according to FIG. 17, the packages 7 are divided into four batches. per batch is one of the packages 7 shown in fig. Of course, more or fewer batches can also be formed.

The batch number 6 can also be encoded into any other machine-readable code. For example, the batch number in FIG. 1 can also be encoded in a QR code or a bar code.

Part of the idea according to the invention consists in converting the batch-specific first graphic code 3, here the data matrix code 3, into an associated Fourier pattern 2. For this purpose, the data matrix code 3 is positioned in an empty image template 8 according to FIG. The level is understood as a frequency level. The function is defined in the frequency plane, which is spanned by an fx frequency and an fy frequency. Depending on where the data matrix code is positioned in the fx, fy level, it is formed by higher or lower frequencies. The image template 8 has a size of m×n pixels.

A function G(fx,fy) is now formed from the arrangement of the data matrix code in the empty image template 8 . The image template of m x n pixels forms the domain of definition of the function G(fx,fy). The function G(fx,fy) consists of the product of an amplitude function and a phase function. The amplitude function is shown graphically for the data matrix code 3 in FIG. 1 in FIG. 3a. The amplitude function is zero at the white points and one or some other constant value at the black points. This means that the amplitude function is in the form of a real-valued function with the function values zero and one.

The amplitude function is multiplied by an appropriate phase function e ^i(p (fx,fy). The phase function e ^i(p (fx,fy) can be a random phase act, but other phase distributions are also known in the prior art (Akahori, H., Comparison of deterministic phase coding with random face coding in terms of dynamic range, Appl. Opt. 12, pp. 2336-43 (1973)).

In FIG. 3b, the phase φ(fx,fy) selected here is designed and shown as a random gray value of the data matrix code 3a. The outlines of the phase correspond to data matrix code 3, except that the values are not at zero (white) and one (black), but are random gray values between white and black. Within the data matrix code 3a, each pixel is assigned a random gray value between white and black. The gray values are now assigned to numbers between 0 and 2TT. If this gray value is black then the phase is 2π and if the gray value is white then the phase is zero. The other gray values are assigned an angle between 0 and 2TT depending on the gray level. The blacker the color, the higher the angle. In this way, the random gray-scale image can be unambiguously converted into a phase function, and by multiplying the amplitude function graphed in Fig. 3a by the phase function in Fig. 3b e ^i(p (fx,fy), the complex-valued function G(fx,fy ) educated.

The complex function G(fx,fy) is formed on a domain of definition according to FIG. 2 of m×n pixels, where m denotes the number of pixels in the fx direction and n denotes the number of pixels in the fy direction. In the example, m = n = 512. The complex function G(fx,fy) is Fourier transformed in the usual way, resulting in a new two-dimensional complex function on mxn pixels, the two-dimensional Fourier transformed g(x,y). As an alternative to the Fourier transformation, an inverse Fourier transformation or an inverse Fourier transformation can also be used in this method, since due to the Symmetry conditions between Fourier transformation and inverse

Fourier transformation revealed no differences relevant to the invention.

The real part of the Fourier transformation g(x,y) is referred to here as a two-dimensional Fourier pattern 2 and is shown in FIG. 4a. The two-dimensional Fourier pattern 2 is also batch-specific and has gray values between white and black.

4b and 4c show so-called binarized Fourier patterns 21, 22 of the Fourier pattern 2 of FIG. 4a. Binarization means that either a pixel value 1 or a pixel value 0 is assigned to each pixel of the Fourier pattern 2 in FIG. 4a. Black is used as pixel value 1 and white as pixel value 0. However, two different gray values or two different color values are also conceivable. Various methods from the literature for computer-generated holograms are known for binarization, e.g. the Detour phase method (Goodman, J.W., Introduction to Fourier Optics, McGraw-Hill (New York) (1996)).

A preferred method, which is also used here, is the discrete binarization of the real part of the Fourier transform g(x,y). A threshold value is selected here, and all values of the real part of the Fourier transform g(x,y) that are below the threshold value are assigned to the pixel value 0 and all other values to the pixel value 1. The threshold value can be selected, as in FIG. 4b, such that 50% receive the pixel value 1, ie are black, and 50% receive the pixel value 0, ie are white. Fig. 4b shows a 50% binarized Fourier pattern 21. However, the threshold value can also be selected in such a way that any other desired percentage has the pixel value 1 and the remaining pixels have the pixel value 0. A 20% binarized Fourier pattern 22 is shown in Fig. 4c shown. The percentage of binarization is preferably between 5% and 25% for the current invention.

5a, 5b, 5c show the real part of an inverse Fourier transformation 3, 31, 32 of the Fourier pattern 1, 21, 22 in FIGS. 4a, 4b, 4c. It can be seen that using the real part as a Fourier pattern results in a symmetrical, so-called negative order. It can also be seen that the binarization increases the noise (grey shadow in the background), with low binarization leading to more noise. Nevertheless, the inverse transformed Fourier pattern 3, 31, 32 in FIGS. 5a, 5b, 5c remains in machine-readable form.

6 shows the first embodiment of the security feature 1 according to the invention, in which a QR code 4 is printed on the 20% binarized Fourier pattern 22 according to FIG. 4c.

The QR code 4 is an embodiment of the packaging-specific, machine-readable second graphic code 4, a packaging-specific serial number 9 being encoded in the QR code 4 in accordance with the explanation relating to FIG. Instead of the QR code, any other machine-readable code, in particular a 2D barcode, such as a data matrix code according to FIG. 1 , or any other code can also be selected as the packaging-specific, machine-readable second graphic code 4 . QR code 4 and packaging-specific, machine-readable second graphic code 4 are also given the same reference number, since they coincide in this exemplary embodiment.

Fig. 7 again shows the inverse Fourier transformation of the security feature 1 in

Fig. 6. Due to the Fourier inverse transformation in Fig. 7 between the two reconstructed data matrix code 3 produces a black cross with shadows, which represents the inverse Fourier transformation of the QR code 4 printed in FIG. At this point, an advantage of the invention becomes clear: on the one hand, the QR code 4 can still be read with standard reading devices, although it is in the same area as the binarized Fourier pattern 22 . On the other hand, although the QR code 4 is printed on the binarized Fourier pattern 22 and thus partially covers it, the graphic code 3 in the Fourier inverse transformed can also continue to be read with standard reading devices.

The security feature 1 is printed onto the packaging 7 in two separate printing process steps.

The 20% binarized Fourier pattern 22 in FIG. 6 is printed onto the packaging 7 in a first printing process step. The first step in the printing process has a print resolution of at least 600 dpi.

To print the 20% binarized Fourier pattern 22, a gravure printing method, offset printing method, screen printing method or flexographic printing method is selected in the first printing method step. These are so-called classic printing processes in which a printing die is formed and the individual packaging rolls over the printing die. The production of a printing matrix is complex, so that for cost reasons a separate printing matrix cannot be provided for each packaging 7, but a printing matrix is made separately for each batch, so that usually several hundred or thousands of packaging 7 with a same batch-specific two-dimensional Fourier pattern according to FIG. 6 are printed. The packaging-specific, machine-readable second graphic code 4, here the QR code 4, is printed in a second printing process step, which can have a significantly lower print resolution than the first printing process step.

The second printing method step is preferably a digital printing method from the group of ink jet printing, thermal transfer printing, laser printing, laser engraving. Digital printing methods make it possible to assign one of the packaging-specific serial numbers 9 to each packaging, which is encoded in one of the packaging-specific second codes 4 . The packaging-specific second code 4 can be printed inexpensively and individually in the second method step due to the selection of the digital printing method.

FIG. 8 shows a photocopy 11 of the security feature 1 in FIG. 6, where the photocopy 11 was only printed out with a printer that has half the resolution of the printer in FIG. At first glance, there are hardly any differences between the photocopy 11 and the original 1, but Fig. 9 shows that the inverse Fourier transformation of the photocopy 11 in Fig. 8 is weaker than the inverse Fourier transformation of the original security feature 1 in Fig. 6. In addition, ghost images, i.e. higher-order Fourier inverse transformations, can be seen. However, the information in the data matrix code 3 is retained and the data matrix code 3 can still be read by machines.

10 to 16 show a second embodiment of a security feature 1' with the associated construction of the security feature T. The security feature T of the second embodiment has a higher effective resolution than the security feature 1 of the first embodiment. In FIGS. 10 to 16, an analogous construction, which was carried out in FIGS. 3 to 9, is shown for the same data matrix code 3. However, the data matrix code 3' in FIG. 10a and the random phase in FIG. 10b are positioned at a different point in the image plane 8 than in FIGS. 3a and 3b. The data matrix code 3' is positioned somewhat further away from the center of the image plane 8. The fx values of the data matrix codes 3' in FIG. 3a and FIG. 10a remain the same, while the fy values in FIG. 10a are larger because the data matrix code 3' is shifted in the fy direction. The same applies to the random phase in Figures 3b and 10b.

The complex function G'(fx,fy) is constructed in a manner analogous to the complex function from the amplitude function and the phase function of FIGS. 3a, 3b. The complex function G'(fx,fy) and the real part of the Fourier transformation g'(x,y) as well as the other reference symbols are marked with a prime. An analog Fourier pattern 2' is produced by Fourier transformation of the amplitude function G'(fx,fy), the real part of which is shown in FIG. 11a. A comparison of the real parts of the Fourier transformation g(x,y) in FIG. 4a and the real part of the Fourier transformation g'(x,y) in FIG. 11a clearly shows that the structures in FIG. 11a have a higher frequency than the structures in Figure 4a. This is due to the fact that the data matrix code 3' is positioned further away from the zero point in the fx, fy plane in Fig. 10a compared to Fig. 3a and consists of higher individual frequencies.

The 50% and the 20% binarized Fourier patterns 21' and 22', which are shown in FIGS. 11b and 11c, are carried out technically in exactly the same way as the binarization in FIGS. 4b, 4c. However, it can be seen in particular in FIG. 11c that the structures of the 20% binarized Fourier pattern 22' are significantly finer than the structures of the 20% binarized Fourier pattern 22 in FIG. 4c. In the inverse Fourier transformation shown in FIGS. 12a, 12b, 12c of the real parts of the Fourier patterns 2', 21' and 22' shown in FIGS. 11a, 11b, 11c, the data matrix code 3', 31', 32' is easy to read , so that the information of the data matrix code 3 is retained.

As in FIG. 6, FIG. 13 shows a security feature 1'. It is formed from the 20% binarized Fourier pattern 22', which is printed onto the packaging 7 in every first printing process step. In the second printing method step, the QR code 4 is again printed on the already printed 20% binarized Fourier pattern 22'. A Fourier inverse transformation of the originally printed security feature 1 in FIG. 13 leads to an easily readable data matrix code 3' in FIG.

FIG. 15 again shows a photocopy 11' of the security feature 1' in FIG. 13, the photocopy 11' being printed with a printer at half the resolution of the print in FIG.

FIG. 16 shows the inverse Fourier transformation of the photocopied security feature 11' in FIG. 15. It can be seen that the reconstructed data matrix code 3' can no longer be read. The information is destroyed. This means that the effective resolution of the security feature 1' in Fig. 13 was sufficient to destroy the information hidden in the 20% binarized Fourier pattern 22' by a photocopy using a conventional printer, while the effective resolution of the Security feature in Fig. 6 was not large enough so that the same photocopy resolution did not lead to the destruction of the information.

The positioning of the batch-specific first graphic code 3, 3' in the original image 8 can be determined using the effective resolution of the Fourier pattern 2, 2'. determine. If the effective resolution is greater than the resolution of the counterfeit printer used, information is lost during photocopying and the batch-specific first graphic code 3, 3' cannot be reconstructed by an inverse Fourier transformation, ie it cannot be read out.

According to FIG. 2, for the special positioning of the batch-specific first graphic code 3 within the original image 8 in the fx and fy direction, the width fx imit and the height fy imit of the original image 8 are determined, as well as the horizontal distance fx_max of the batch-specific first graphic code 3 from the center of the image and the greatest vertical distance fy_max of the batch-specific first graphic code 3 from the center of the image. The corresponding distances are indicated in FIG. 2 by double arrows. In FIG. 2 the overall width of the original image is 512 pixels and the height is also 512 pixels. The fx_max distance here is 75 pixels, fy_max is 168 pixels, fx_limit is 256 pixels and fy_limit is also 256 pixels. The ratio fx_max / fx_limit is thus 75 / 256 = 0.29, and fy_max / fy limit is 168 / 256 = 0.66.

With the values fx_max/fx_limit and fy_max/fy_limit and the resolution of the printing press, which can be selected as the resolution of the printing plate exposer when using a printing plate, which is usually 1,000 dpi to 4,000 dpi, the effective resolution of the forier pattern 2 can be determined. With a print resolution of 2000 dpi, the effective resolution in the horizontal direction is 2000 dpi * 0.29 = 580 dpi and in the vertical direction is 2000 dpi * 0.66 = 1320 dpi. This means that when the Fourier pattern is copied with a conventional printer, which usually has a print resolution of 600 dpi, the structures are destroyed to such an extent that the information contained in the Fourier pattern becomes illegible after Fourier inverse transformation, as shown in FIG. Two things must therefore be observed when constructing the security feature 1, 1' according to the invention; on the one hand, the graphic code 3, 3' must be far enough away from the zero point of the fx, fy plane, ie consist of such high frequencies, that the Fourier pattern 2, 2' Fig. 11a is sufficiently fine, and on the other hand, a first printing process step can be selected, which makes it possible to print this sufficiently fine pattern on the packaging 7 even after binarization without any major loss of information. If the first graphic code 3, as in FIG. 3a, is positioned too close to the zero point, the Fourier pattern in FIG. 4a is relatively coarse, and the information encoded in the coarse Fourier pattern can still be completely printed or photocopied using a low-resolution printer will.

Figure 17 shows four packages 7 of a series of packages 7. The series of packages 7 is divided into four batches. Each lot is assigned a lot specific lot number 6 here ABCDEF, JK7MQ8, 90LTXS and PK6HG4. The batch number 6 is encoded in a 20% binarized Fourier pattern 22' in the manner described above, and the 20% binarized Fourier pattern 22' is printed onto the packaging 7 in a first printing process step.

18 shows the second printing process step according to the invention. In the second printing process step, the packaging-specific, machine-readable second graphic code 4 is printed on the batch-specific 20% binarized Fourier pattern 22'. Each packaging 7 thus receives a packaging-specific graphic code 4, which encodes a packaging-specific serial number 9, here GB4Q3, KLP789, 14FVL, and a batch-specific batch number 6, which is identical for all packaging 7 of the same batch. However, the batch-specific two-dimensional Fourier pattern 22' has such a high effective resolution that if it were photocopied with conventional photocopiers, it can no longer be transformed back into the first graphic code 3', but the information would be destroyed as shown in FIG.

The security feature 1' printed on the packaging 7 is scanned with a conventional camera of a mobile terminal device, in particular a smartphone. The QR code 4 is read out by means of a second reading algorithm and the serial number 9 specific to the packaging is determined. The packaging-specific serial number 9 is transmitted to an authentication server by means of a transceiver unit of the mobile terminal. The URL of a database server is also encoded in the second graphic code, so that the serial number 9 specific to the packaging can be transmitted to the database server. A Fourier transformation program is also stored on the mobile terminal. This Fourier transformation program Fourier-transforms the security feature 1' back, and the batch-specific first graphic code 3' that is formed is read out with a first reading algorithm, with the first and second reading algorithms being able to be identical if the code type is the same, and the determined batch number 6 is also transmitted to the authentication server. An authentication database is stored on the authentication server, in which all valid combinations of batch numbers 6 and serial numbers 9 are stored. If the transmitted combination of serial number 9 and batch number 6 is found, a positive authentication signal is sent back to the mobile device. The packaging-specific serial number 9 can then be blocked. If a serial number 9 and no batch number 6 is transmitted to the database server after the security feature 1' has been scanned, it will presumably be a bad photocopy of an original security feature 1' on the packaging 7. Then a negative authentication signal can be sent to the mobile end device are returned. A negative authentication signal is also sent back to the mobile terminal device by the authentication server if the serial number 9 has already been queried one or more times and is blocked. A negative authentication signal is also sent back to the mobile terminal device by the authentication server if the assignment of the individual packaging serial number to the batch number is not correct.

Reference List

1 security feature

1' security feature

2 Fourier patterns

2' Fourier pattern

3 first batch-specific graphic code, data matrix code

3a phase

3' first batch-specific graphic code, data matrix code

3'a phase

4 second graphic code, QR code

6 batch number

7 packaging

8 picture template

9 serial number

11 Photocopy of the security feature

11 ‘ Photocopy of security feature

21 50% binarized Fourier pattern

21 ' 50% binarized Fourier pattern

22 20% binarized Fourier pattern

22' 20% binarized Fourier pattern

31 Real part of the inverse Fourier transform

31 ' Real part of the inverse Fourier transform

32 Real part of the inverse Fourier transform

32' real part of the inverse Fourier transform

Claims

SCRIBOS GmbH, Sickingenstraße 65, 69126 Heidelberg, Germany Patent claims

1. A method for producing a series of forgery-proof packaging (7) by dividing the series of packaging (7) into batches and assigning a batch-specific batch number (6) to each packaging (7) and assigning a packaging-specific serial number (9 ) is assigned, the batch-specific batch numbers (6) are encoded in machine-readable, batch-specific first graphic codes (3), the batch-specific first graphic codes (3) in batch-specific two-dimensional Fourier patterns (2, 2', 21, 21', 22, 22' ) are Fourier transformed, the batch-specific Fourier patterns (2, 2', 21, 21', 22, 22') are printed in a first printing process step on the packaging (7) assigned to the batch number (6), the packaging-specific serial numbers (9) in machine-readable, packaging-specific second graphic codes (4) are encoded on the associated packaging (7) in a second

printing process step are printed, the batch-specific Fourier pattern (2, 2', 21, 21', 22, 22') and the second graphic code (4) together form a forgery-proof, two-dimensional security feature (1, T) for each packaging.

2. The method according to claim 1, characterized in that the first printing method step has a higher printing resolution than the second printing method step and the batch-specific Fourier patterns (2, 2', 21, 21', 22, 22') are printed with a higher printing resolution than the packaging-specific second graphic code (4).

3. The method according to claim 1 or 2, characterized in that the first printing process step is selected from the set of gravure printing processes, offset printing processes, screen printing processes or flexographic printing processes.

4. The method according to any one of the preceding claims, characterized in that the second printing process step takes place by means of a digital printing process and is selected from the group of ink jet printing, thermal transfer printing, laser printing, laser engraving.

5. Method according to one of the preceding claims, characterized in that the batch-specific Fourier patterns (2, 2', 21, 21', 22, 22') are each printed with an effective resolution of at least 600 dpi.

6. The method according to any one of the preceding claims, characterized in that the Fourier patterns (2, 2', 21, 21', 22, 22') are binarized and the binarized Fourier patterns (21, 21', 22, 22') are printed.

7. Method according to one of the preceding claims, characterized in that the Fourier patterns (21, 21', 22, 22') are binarized to 50%, preferably less than 20%, preferably less than 10%.

8. Method according to one of the preceding claims, characterized in that a URL of an authentication server is also stored in the second graphic code (4).

9. The method as claimed in one of the preceding claims, characterized in that the first and/or second graphic codes (3, 4) are 2D barcodes, in particular QR codes or data matrix codes.

10. The method according to any one of the preceding claims, characterized in that the Fourier pattern (2, 2', 21, 21', 22, 22') and the second graphic code (4) are arranged to overlap one another.

11 . Method according to one of the preceding claims, characterized in that the Fourier pattern (2, 2', 21, 21', 22, 22') and the second graphic code (4) are arranged adjacent.

12. The method according to any one of the preceding claims, characterized in that the batch numbers (6) and the serial numbers (9) are stored in pairs in an authentication database of an authentication server. Method according to Claim 12, characterized in that a mobile terminal device is made available with which the two-dimensional batch-specific Fourier pattern (2, 2', 21, 21', 22, 22') and the packaging-specific second code (4) are detected and a data-conducting connection between the mobile terminal device and the authentication server is made available and authentication software enables a comparison of the data recorded by the mobile terminal device and transmitted to the authentication server and the batch numbers (6) and serial numbers (9) stored in the authentication database. Method according to one of the preceding claims, characterized in that the first printing process step is carried out at a packaging means manufacturer and the second printing process step is carried out on a packaging line. Series of forgery-proof packaging (7), produced according to one of the aforementioned manufacturing processes, the series of packaging (7) being divided into batches and each packaging (7) being assigned a batch-specific batch number (6) and each packaging (7) having a packaging-specific serial number ( 9) is assigned, the batch-specific batch numbers (6) are encoded in machine-readable, batch-specific first graphic codes (3, 3'), the batch-specific first graphic codes (3. 3') are Fourier-transformed into batch-specific two-dimensional Fourier patterns (2, 2', 21, 21', 22, 22'), the batch-specific Fourier patterns (2, 2', 21, 21', 22, 22') are printed in a first printing process step on the packaging (7) assigned to the batch number (6), the packaging-specific serial numbers (9) are encoded in packaging-specific, machine-readable second graphic codes (4), which are printed on the assigned packaging (7) in are printed in a second printing process step, the batch-specific Fourier pattern (2, 2', 21, 21', 22, 22') and the second graphic code (4) a forgery-proof, two-dimensional

Represent security feature (1, 1 ') for each package (7). Series of counterfeit-proof packaging according to Claim 15, characterized in that the resolutions of the printed Fourier patterns (2, 2', 21, 21', 22, 22') are higher than the resolutions of the printed second graphic code (4). Series of counterfeit-proof packaging according to one of Claims 15 or 16, characterized in that the Fourier pattern (2, 2', 21, 21', 22, 22') is binarized. Series of counterfeit-proof packaging according to one of claims 15, 16 or 17, characterized in that the Fourier pattern (2, 2', 21, 21', 22, 22') is printed with an effective resolution of at least 600 dpi. Series of counterfeit-proof packaging according to any one of claims 15 to 18, characterized in that the Fourier pattern (2, 2', 21, 21', 22, 22') and the second graphic code (4) are printed one on top of the other on each packaging (7). Method for authenticating a packaging (7) of a series of packaging (7) divided into batches with an authentication system by capturing an image of a two-dimensional security feature (1, 1') of the packaging (7) with a camera of a mobile terminal device that captured Image is inverse Fourier transformed and the inverse Fourier transformed image is fed to a first reading algorithm for the batch-specific first graphic code (3, 3', 31, 31, 32, 32') and a batch number (6) is read out and the captured image is sent to a second reading algorithm for the second graphic code (4) is supplied and a packaging-specific serial number (9) is read and the batch number (6) and the serial number (9) are checked. Method according to Claim 20, characterized in that the read serial number (9) is fed to an authentication server and is compared with valid serial numbers (9) stored on the authentication server and the read serial number (9) if it matches one of the valid serial numbers (9) stored is authenticated. The method according to any one of claims 20 or 21, characterized in that the read batch number (6) is supplied to the authentication server and on the

Authentication server stored valid batch numbers (6) is compared and the read batch number (6) is authenticated if it matches one of the stored valid batch numbers (6). Method according to one of Claims 20, 21 or 22, characterized in that the security feature (1, 1') is authenticated if both the serial number (9) and the batch number (6) have been authenticated. Method according to one of Claims 20 to 23, characterized in that the security feature (1, 1') is authenticated if the combination of serial number (9) and batch number (6) has been authenticated. authentication system with

- a series of forgery-proof packaging (7) with forgery-proof security features (1, 1 ') according to one of claims 15 to 19,

- A mobile terminal with a camera for capturing the images of the forgery-proof security features (1, 1') and with a transceiver unit with which the data of the security features (1, 1') of the packaging (7) captured by the camera can be transmitted to an authentication server ,

- a Fourier inverse transformation program with which the batch-specific Fourier pattern (2, 2', 21, 21', 22, 22') can be Fourier inverse transformed, - A first reading algorithm for the first batch-specific code (3, 3', 31, 31, 32, 32') which can determine the batch number (6) from the Fourier inverse-transformed image,

- a second reading algorithm for the packaging-specific second graphic code (4), which can determine the packaging-specific serial number (9),

- with the authentication server, which is in data communication with the mobile end device and on which the serial number (9) and batch number (6) assigned to each package (7) are stored and assigned to one another and on which the batch numbers (6) and Serial numbers (9) can be authenticated and with which an authentication signal can be sent to the mobile terminal.