Method and device for geometric code authentication

FIELD: physics, computer engineering.

SUBSTANCE: invention relates to means of authenticating geometric codes on storage media. The method provides authentication of a code with geometric areas whose shapes and/or colours vary according to a message by creating variable geometric areas according to the message. A digital authentication code is generated to provide numeric values. An image of the geometric code areas is then formed, including a part of the digital authentication code in at least some of its geometric areas and/or in at least one space between geometric areas. Disruption of the digital authentication code is also generated.

EFFECT: improved copy protection of information on storage media.

19 cl, 1 tbl, 14 dwg

The present invention relates to a method and apparatus for authentication geometric codes. The present invention is applicable, in particular, to one-dimensional ("1D"), two-dimensional ("2D") and even three-dimensional ("3D") bar codes and Data Matrix codes (registered trademark).

Code Data Matrix is a two - dimensional symbolic barcode with high density, which would produce a large amount of information, namely up to 2335 alphanumeric characters, or 3116 numeric characters on a small area of approximately 1 cm². Code Data Matrix available for use. Code Data Matrix has the form of a matrix of adjacent dots or squares.

Code Data Matrix is compatible with ISO IEC16022. In accordance with this standard character code Data Matrix may contain different levels of reliability, known as detection and error correction (ECC)that allows you to read this symbol even when it is partially damaged or blocked. The standard allows several options for Data Matrix, from variant ESS, which provides no reliability, if the character is damaged, for example a one-dimensional barcode (EAN 13 etc), to option AS providing the highest level of reliability (the character is read in the presence of 20% symbol).

The main area of application of Data Matrix refers to the labelling of very small mechanical is whether electronic items. This, among other things, is used by NASA (National Aeronautics and Space Administration") for the marking of each component in the space shuttles. Data Matrix is often used to mail in some countries, for example Switzerland, and recently in some mobile applications, which is often called a tag (or label). Flashcode (registered trademark) is a private commercial implementation that uses standard Data Matrix.

The standard Data Matrix AS is among the standards GS1 standard (acronym for "Global Standard"), and according to a recent opinion of the French Agency for sanitary safety of medicines AFSSAPS (Agence Francaise de Securite Sanitaire des Produits de Sante) from January 2011, all the drugs that are subject to licensing on the market must contain, in addition to the existing legal information, two-dimensional Data Matrix code containing the predetermined information blocks.

Data Matrix designed to maximize the amount of data that can be stored in the form image, and ensure the performance and reliability of decoding or reader of these data (based on the registered image). However, Data Matrix is not intended to provide reliable data storage, although this problem is becoming more acute.

Thus, decode the Finance code Data Matrix is performed in accordance with an open standard and does not include any cryptographic key to encrypt and/or application of electronic digital signature to the data. However, the stored data messages may be encrypted or provided with a digital signature before they change with the formation of a Data Matrix code. Thus can be provided with originality and integrity of information with the exception of the possibility of falsification of any genuine messages (i.e. change its content) or the statement of claim to the authorship of the original message.

However, cryptographic methods do not provide protection from an exact duplicate or "clone" of the data code Data Matrix. With this many ways to combat fraud protection such exact duplication is very important because the manufacturers of counterfeit products can easily make an exact copy of the document, package or other object that contains the code Data Matrix, if these objects do not contain elements of copy protection. In some applications, associated with tracking and monitoring the passage, products can be tracked throughout the supply chain with the use of contained in the code Data Matrix ID. Thus there may be a duplicate if the identifier is detected more than once, or can be found irregularities in the distribution, if the identifier indicates a product that should be in a different location in the supply chain.

Obviously, a single counter is eh at all levels of the supply chain helps in combating fraud, even if ultimately it is not possible to determine which of the two at first glance identical products is original. However, in most cases, this control system is too expensive or simply impossible to use, because it must be centralized so that two products with the same code, Data Matrix, found in two different places, could be so identified.

For this reason, the holders using the code Data Matrix, often used other means to ensure the authenticity of a document or product. For example, some decisions are based on reliable labels in combination with the authenticator, such as a hologram or optical variable device situated near the Data Matrix code.

Unfortunately, the means used are usually expensive and inefficient. The high cost is related to the fact that many authentication methods require advanced technology to create optical effects. The inefficiency stems from the fact that optical effects increasingly can be fabricated with sufficient accuracy at a low cost. In addition, these effects do not provide for its own defence identity. For example, if the set of documents containing authentication tools stolen, it can be applied to arbitrary Data Matrix codes.

To the Data Matrix can be protected from copying by marking it for example, special ink. However, manufacturers of counterfeit products can get special ink more simple way, and this solution does not provide the desired protection while remaining expensive. For this reason, in many applications, Data Matrix codes are applied laser burning.

In the document U.S. 2008/0252066 proposed print color two-dimensional bar code reading and/or authentication which require printed code was illuminated by different light sources and/or spectral filters. Unfortunately, the use of different ink is expensive and difficult to manufacture and require specialized tools to capture images for registration, which limits the possibilities for authentication. In addition, this approach does not provide a high level of protection aimed at counterfeit opponent who can easily find the ink type used to recognize printed codes through the corresponding spectral lighting.

In the document U.S. 2008/110990 asked printhead rotation, the impact of which can then be registered and measured on the basis of the registered image printed barcode. However, from the specified document essentially follows that the proposed method allows the exclusively register copies, which are made by using printing means preventing rotation of the printhead. Thus, this invention provides no real protection against counterfeits, made by the same printing means, and is limited in the sense that it requires the use of a specific print medium, which significantly narrows the scope.

In document WO 2008/003964 ways of introduction of a second level of information in one - and two-dimensional barcodes by changing the information-bearing elements in such a way that they represent a second level of information, for example by increasing or reducing the size of code Data Matrix or by cutting or saving all black elements. This approach eliminates some of the disadvantages of the previous methods, because the second level of information that can be used to authenticate entered when printing. This is a convenient and inexpensive operation. This approach protects against the manufacturers of fake, who doesn't know this method and make a duplicate of the barcode by the playback information is only the first level. However, the second level can be easily and accurately copied by the manufacturer of the counterfeit, which are aware of this level. In addition, in the above-mentioned document States that W is Roy the information level can be copied in high quality printing means, even if provided with the best characteristics of copy protection (p.12, lines 9-12 of this document).

The present invention is to remedy these disadvantages. In particular, the present invention relates to a method to include a second level of information using the same printing method that is used when printing the barcode. In contrast to the invention according to the aforementioned document, the second level of information cannot be copied in the physical and mathematical sense.

In this regard, in accordance with the first aspect of the present invention, a method authentication code with geometric areas, shapes and/or colors which change in accordance with the message, characterized in that it includes:

- step, which is produced in accordance with the specified message code with variable geometric areas with the formation of geometric sections;

- the step that generates a digital authentication code with the formation of numeric values; and

- step on which form the image areas specified geometric code, containing part of the digital authentication code at least in some of its geometrical areas and/or at least one gap between geometric sections.

T is thus, the present invention allows to directly authenticate the two-dimensional barcode based on his images exclusively digital means, while this two-dimensional bar code remains readable.

As described above, the digital authentication codes (DAC) are digital images that after marking media, for example by printing or local modification of the media have characteristics that generally are measured automatically on the basis of the registered image and change when copied. Digital authentication codes are usually based on the deterioration of at least one sensitive to the copying of the signal based on the image elements to be measured and sensitive copying characteristics.

Thus, the geometric area containing part of the digital authentication code has a variable characteristic marking, deteriorating copying geometric area.

Certain types of digital authentication codes can also contain information to identify or track the containing document. Digital authentication codes are perfect for Desk copies. Indeed, they are very inexpensive to produce, simple to implement and can be read by devices which, having means for capturing images, providing the high level of copy protection. Thanks to the implementation of the present invention the code with geometric areas, such as the barcode, is inextricably linked with digital authentication codes.

You can see that the present invention has advantages compared to the simple placement of the next code with geometric areas and digital authentication code. First, this latter approach enables printing of two codes on the document at different times, which requires space and complicates the process of production of reliable documents. Secondly, authentication requires the capture of two images: the image of the digital authentication code and image code with geometric areas, making way read less convenient. Finally, the manufacturer of the counterfeit, which successfully receives the documents, bearing the digital authentication code, to print Data Matrix code or who fails to produce a printed form or a file containing the original digital authentication code, can produce "authentic" documents cloning the original Data Matrix codes associated with the digital authentication code.

You can see that the step of forming the image may contain, for example, printing, demolition material, solid transfer or local the physical or chemical change, for example by heating.

In accordance with individual characteristics as physical uncertainties inherent to image formation, at the step of forming the image specify the magnitude of the error display of the digital authentication code such that greater than the first preset value and less than the second preset value.

As described above, the digital authentication codes contain different elements, taking discrete values. In the case of binary variables, the elements can be represented in black (printed) or white (not printed) element. At the time of registration is determined by the magnitude of the error corresponding to the number of elements that contain an incorrect value.

It should be noted that the magnitude of the error is directly associated with the ratio signal/noise.

For example, the first pre-pre-set value is 10%, and the second preset value is 35%.

In accordance with individual characteristics as physical uncertainties inherent to image formation, at the step of image formation between the two groups of the same code with variable geometric areas, set the deviation display digital authentication code such that is greater than the third pre-fitted the fair value and less than a fourth preset value.

For example, the third preset value is 2%, and the fourth preset value is 45%.

In accordance with individual characteristics as physical uncertainties inherent to image formation, at the step of forming the image set the noise display digital authentication code, such for which the signal-to-noise representation of the digital authentication code is less than a fifth preset value.

In accordance with individual characteristics as physical uncertainties inherent to image formation, at the step of forming the image set the noise to display the digital authentication code, for which the signal-to-noise representation of the digital authentication code is greater than the sixth preset value.

For example, the fifth preset value of the ratio signal/noise is 0.05, and the sixth preset value is 2.63, resulting in the efficiency of registration copy is at least 25% of the optimum efficiency of Desk copies (obtained when the value is 0.56).

For example, the fifth preset value of the signal-to-noise ratio is 0.11, and sixth have been fitted in advance what I value is 1.8, resulting efficiency definition copy is at least 75% of the optimum efficiency of Desk copies (obtained when the value is 0.56).

For example, the fifth preset value of the signal-to-noise ratio is 0.32, and the sixth preset value is 0.93, resulting in efficiency definition copy is at least 90% from the optimal efficiency of Desk copies (obtained when the value is 0.56).

In accordance with the private characteristics of the proposed method further includes:

- the step that determines the conditions for the formation of the specified image; and

a step in which, depending on the forming conditions of the image determine the physical characteristics of the elements at least part of the digital authentication code.

In accordance with individual characteristics at the specified step of generating code with variable geometric areas as changing geometric plots used in generally parallel rectangular blocks, width and/or the distance between them is changed in accordance with the message. Thus, the present invention is applicable to one-dimensional barcodes.

In accordance with individual characteristics at the specified step of the code generation with variable geometrical the ski areas as changeable geometric plots of use are entered into the matrix square sections, color and/or at least one dimension which is changed in accordance with the message. Thus, the present invention is applicable to two-dimensional bar codes.

In accordance with individual characteristics at the specified step of forming the image code with geometric areas containing at least some of the geometric areas of the digital authentication code, as a digital authentication code using the edit form at least one dimension changeable geometric sections.

In accordance with individual characteristics at the specified step of forming the image code with geometric areas containing at least some of the geometric areas of the digital authentication code, as each part of the digital authentication code entered in the geometric part of the code with variable geometric areas, use the rectangular distribution of elements whose size is at least one order of magnitude smaller than the dimensions of the specified geometric area and part of which has a color different from the color of the specified geometric area.

In accordance with private signs on each geometric area containing part of the digital authentication code, the area of these elements is less than a quarter of p is Asadi specified geometric area.

In accordance with the private characteristics of the proposed method additionally includes a step in which the digital authentication code includes encoded information.

In accordance with individual characteristics specified information based on the message and/or message based on the specified information.

Thanks to this improved authentication, because the possibility of change the message without changing the information contained in the digital authentication code, and/or Vice versa.

In accordance with individual characteristics specified information is a measure of the destruction of the digital authentication code due to physical uncertainties that affect the image at the specified step of forming the image.

For example, this information represents the magnitude of the error signal/noise occurring at the specified step of image formation, or the value of the correlation with the original digital authentication code. This information can provide the expected level of destruction or the level of ultimate destruction, for which the digital authentication code is considered as a copy. The authentication image can be thus performed independently using a reader that is intended to register the image of the digital code authentication is being used is then, put in the code with variable geometric areas, because the information contained in the digital authentication code, indicates the normal level of destruction, and, consequently, the level of destruction, in which damage to the recorded image is a copy of the digital authentication code.

In accordance with the private characteristics of the proposed method additionally includes the step on which measure the destruction of the digital authentication code generated at the specified step of forming the image.

In accordance with individual characteristics at the specified step of measuring the fracture using registration codes errors introduced in the digital authentication code.

As specified measurement or evaluation use, for example, the percentage of correct bits, the correlation coefficient between the original digital authentication code and digital authentication code registered in the image recorded by the image sensor.

In accordance with the private characteristics of the proposed method includes the step in which you determine the output image data is formed, dependent on the destruction of the digital authentication code at the specified step of forming the image.

The object or document bearing the code with variable geometric in what astami can be thus identified, i.e. recognized, even if the code with variable geometric areas and digital authentication code is identical for multiple objects or documents.

In accordance with the second aspect of the present invention, an apparatus for authentication code with geometric areas, shapes and/or colors which change in accordance with the message, characterized in that it contains:

means for generating a specified code with variable geometric areas in accordance with the message, designed to provide geometric sections;

means for generating a digital authentication code, for the formation of numeric values; and

- the means through which is formed the image areas specified code with geometric areas containing at least some of its geometrical areas and/or at least one gap between the geometric areas of the digital authentication code.

In accordance with a third aspect of the present invention, a method of authenticating the registered image code with geometric areas, shape and/or color of which is changed, characterized in that it includes:

- the step that reads the message in the middle of the Fort is Oh and the colors of geometric sections;

- step on which measure the level of destruction of the digital authentication code, presents at least in some geographical areas specified code with geometric areas; and

- the step that determines the authenticity of the specified code with geometric areas on the basis of at least the level of destruction.

In accordance with the fourth aspect of the present invention, an apparatus for authenticating the registered image code with geometric areas having shapes and/or colors, characterized in that it contains:

- means for reading the message contained in the forms and colors of geometric sections;

means for measuring the level of destruction of the digital authentication code, presents at least some of the geometric sections of the code with geometric areas; and

means for determining authenticity code with geometric areas to determine the authenticity of the specified code with variable geometric areas on the basis of at least the level of destruction.

In accordance with the fifth aspect of the present invention is proposed code with geometric areas, shapes and/or colors which change in accordance with the message, characterized in that it is before the hat:

message by means of geometric sections; and

at least in some of its geometric sections of the digital authentication code using the characteristic markings, which varies depending on the digital authentication code.

Benefits, challenges and special characteristics of the proposed device, method and code similar to that briefly described above advantages, objectives and specific characteristics of the proposed authentication method according to the first aspect of the present invention and are not given next.

Other advantages, aims and special features of the present invention will become clear from the lax description below with reference to the accompanying drawings, on which:

figure 1A shows a known Data Matrix code,

figure 1B shows a magnified code Data Matrix, shown in figure 1A,

figures 2A and 3A represent the private implementation of the proposed codes, and increasing portions of these codes are shown by figures 2B and 3B, respectively,

figures 4-7 present in the form of a logical diagram, steps implemented in the private versions of the proposed methods,

figure 8 represents schematically a private implementation of the proposed device,

figure 9A shows a private implementation of the proposed code, and the increase of this code QAR is shown in figure 9B,

figure 10 shows the characteristic of Desk copies, normalized relative to the optimal value, depending on the signal-to-noise.

Throughout the description the terms "imaging" and "print" refer to the manufacturer marks that can be registered, for example, by deposition of ink, material failure, migration of solid particles or a local physical or chemical effects, such as heating.

Although the following description refers to the case of two-dimensional barcodes, the present invention is not limited to this type of marking and printing on the object and is applicable to all types of marking and printing code with geometric areas, shapes and/or colors which change depending on the message, in particular to one-, two - or three-dimensional barcode, which is formed on the surface of objects, and the markings below the surface of the object.

In the case of one-dimensional barcodes, geometric code sections are alternately black and white vertical strokes, the width of which varies in accordance with the message, which is the code.

If dvuhjadernyh barcode, geometric code sections are squares, forming a regular lattice, the color of which is changed in accordance with the message, which is the code.

In the rest of the cha is t describe these geometric areas are referred to as elements.

The following steps detail the methods and devices for authentication of two-dimensional barcodes (also known as Data Matrix codes) and, more specifically, methods and devices for authentication two-dimensional bar codes printed with the inclusion of a digital authentication code, labeled by using lasers with variable power and lasers with a fixed capacity.

In relation to the inclusion of a digital authentication code in the two-dimensional barcode below with reference to figure 4 describes how to generate code Data Matrix containing the digital authentication code. It should be noted that the recommended setup for the generation of a digital authentication code, in particular the resolution in pixels per 1 inch and the type of item (i.e., shape and/or size of the elements forming the digital authentication code), determined in advance for printing (paper, ink, printer, document), for example using a known method.

For the method of determining the optimum settings for imaging structures authentication there is an optimal level of destruction that ensures the easy separation of different seals patterns authentication single source. Thus, if the level of destruction when printing is very low, such as 1 or 2% (1 or 2% of the elements of the structure broken Atara or pixels read correctly on the basis of ideal grip), different print one structure identifier are very close to each other, and their reliable identification is difficult, unless there is a very accurate capture and/or algorithm is very accurate analysis. Similarly, when the level of destruction is very high, for example 45 or 50% (45 or 50% of the elements of the structure ID or pixels read correctly on the basis of a perfect capture, and 50% means that there is no statistical correlation between the read matrix and the matrix of source), printed patterns ID almost indistinguishable from each other. In practice, the optimal level of destruction is approximately 25%, and preferably to provide approximately this level, if conditions allow it. In fact, under the assumption that modification when printing and, therefore, destruction are essentially probabilistic values for the level of destruction equal to 25%, the probability that each of the dots printed patterns ID differs from the other points of the printed structures of the identifier is the maximum.

Below is the second analysis of the magnitudes of the errors that are required when forming the image to be printed in the print media.

It should be recalled, h is about digital authentication codes are composed of various elements, taking discrete values. In the case of binary values of the elements can be represented in black (printed) or white (not printed). At the time of registration is determined by the magnitude of the error, which corresponds to the number of elements that contain an incorrect value. It should be noted that the magnitude of the error is directly related to the ratio signal/noise.

To determine, as may be generated dot patterns with varying characteristics, allowing to optimize the registration of copies, the following describes a model based on theory of decision making. Characteristics measured on the images (or points) are represented by signals. To simplify the analysis it is assumed, consisting in the fact that before printing digital signals imoet binary values corresponding to the characteristics that can have binary values (for example, two measurement points, two points and so on). This hypothesis is confirmed by the fact that most printing processes reproduce the binary image. Obviously, the conclusions of the analysis can be extended to more complicated cases, in particular with multiple possible values for point features. Print dot structures with changing characteristic is simulated by adding Gaussian noise. It is also assumed that copies of shotable which are stated using the same printing process, so print copies are also simulated by adding Gaussian noise with the same energy. In addition, the manufacturer of the counterfeit, which receives the signal before printing copies, forced to reproduce a binary signal by performing an evaluation of the initial value, which minimizes the probability of error.

This model directly corresponds to the dot structures with varying characteristics, the size of the dots which can be 1×1 (in pixels) or 1×2 (in pixels) (in print, for example, in the case of 2400 dots per inch) and for which the manufacturer of the counterfeit must choose one of the sizes of the dots in the image reproduced on the basis of scanning in accordance with the measured grayscale or estimated surface points. The model is also consistent with point structures with varying characteristics, in which the position change, for example, by 1 pixel.

From this model are derived optimal Registrar and statistical distribution of the values of the Registrar and the values of the parameters, leading to maximum efficiency when you register your copy.

The following table shows the different variables.

Without diminishing the generality, the signal source is adopted uniform, s[i]:{+a,-a}, for i=0, 1,..., N-1, and a>0. Signal noise print obeys the Gaussian distribution N(0,σ²)

Hypotheses of the model are as follows:

The fact that minimize whether manufacturers of counterfeit probability of error can be easily controlled by restoring the signal to the nearest value ranging +and-and.

Therefore, the problem of registration includes the following two different hypotheses:

where N₀and H₁the hypothesis that the received signal represents respectively the original and the copy.

The probability that the manufacturer of counterfeit has correctly assessed an amount equal to:

where

Probability distribution for received signal such as listed below, and there is a combination of two Gaussian distributions in the hypothesis H₁.

Check whether a simple correlator optimal classification function. The criterion of Neyman-Pearson regarding registration allows you to decide what is over likelihood ratio threshold t:

The likelihood ratio is given by the expression:

Taking the logarithm, and the introduction of a new threshold level f allows to obtain the following:

Classification function is, therefore, a simple correlator T', the value of which should be less than the threshold t', in order to classify the signal as a copy.

Statistics T is defined for both hypotheses. We can assume that T' obeys the Gaussian distribution (for large values of N). Mathematical expectation and variance determined for both hypotheses:

The second member of variance for hypothesis H₁(a⁴Q(-a/σ)(1-Q(-a/σ))), can be excluded, if the copies correspond to the original. In practice, if the manufacturers of counterfeit minimize their work only by using one original for the manufacture of a large number of copies, the exclusion of that member advisable.

In the case when the variances are equal, the characteristic data logging can be represented by a coefficient of variation of d²that corresponds to the difference between the mathematical expectation of a function of T for the two hypotheses, normalized otnositelno variance T:

where γ=a²/σ²is the square root of the ratio signal/ noise.

Since the characteristic of the recording increases with increasing coefficient of variation, the task is to determine the value of γ, which gives the maximum of the expression (γ(1-Q(γ)))².

Figure 10 illustrates the value of the expression for equation (25) for a fixed value of N, normalized with respect to its optimal value and is obtained as a function of the value of γ. This can be interpreted as follows. The values of γ close to zero, correspond to very strong noise relative to the signal: when the noise is very large, the signal excessively deteriorates when the original print, and the manufacturer of the counterfeit introduces a number of valuation errors that are too small. On the contrary, for too large values of γ the signal is not sufficiently degraded, and in much of the cases, the manufacturer of the counterfeit does not introduce any estimated error. Between these two boundary cases mentioned expression passes through the optimal value for which the value is numerically evaluated as γ$$\approx$$0.752.

It is interesting to note that for this value the probability that the manufacturer of the counterfeit products do not have the proper size, approximately 22.6%.

In practice, this includes obtaining signal-to-noise ratio γ²as close as possible to 0.752²i.e. 0.565 during printing.

The following is an example for a better understanding of how to provide this value relationships. Assume that the pixel structure with the changing characteristic is formed with two possible sizes of points (expressed as number of pixels, and the size of the dots is nine pixels (for example, 3×3 pixels). It should be noted that the size of the dots can be measured using a large number of algorithms, such as local adaptive threshold representation for gray scale with counting the number of pixels below the threshold. The point of the nine pixels print a sufficient number of times. In the registered image measure mathematical expectation and standard deviation for the number of pixels of each point. Further assumed that the mathematical expectation is equal to 12 (average gain is 33%), and standard deviation equal to 4. This standard deviation corresponds to the value of σ that describes the noise in the formulas for the proposed model. Value, approximately equal to 3 is, therefore, the target value for the signal α to obtain the relations γ=0.75, which is very close to optimal. To obtain this led the ranks of the signal can be set two point size, for example, fifteen and six pixels.

Due to physical uncertainties inherent to image formation, at the step of forming the image, the digital authentication code is preferably match the magnitude of the error, which is greater than the first preset value and less than the second preset value. For example, the first preset value equal to 10%, and the second preset value is 35%.

Due to physical uncertainties inherent to image formation, at the step of image formation between the two groups of the same code with variable geometric areas, the digital authentication code is preferably match the variance, which is greater than the third preset value and less than the fourth preset value. For example, the third preset value is equal to 2%, and the fourth preset value is 45%.

Due to physical uncertainties inherent to image formation, at the step of forming the image, the digital authentication code is preferably compares the noise, for which the signal-to-noise representation of the digital authentication code is less than the toe which I preset value, and more than a sixth preset value.

In the first example, the fifth preset value of the ratio signal/noise is 0.05, and the sixth preset value is 2.63, resulting in the efficiency of registration copy is at least 25% of the optimum efficiency of Desk copies (obtained when the value is 0.56).

More preferably, the fifth preset value of the signal-to-noise ratio is 0.11, while the sixth preset value is 1.8, resulting in efficiency definition copy is at least 75% of the optimum efficiency of Desk copies (obtained when the value is equal to 0.56).

Even more preferably, the fifth preset value of the signal-to-noise ratio is 0.32, and the sixth preset value is 0.93, resulting in efficiency definition copy is at least 90% from the optimal efficiency of Desk copies (obtained when the value is 0.56).

The following describes a possible algorithm to optimize print settings:

- in step 720 take surface with some space available for the structure of the identifier, for example a square of size 1/6 inch;

- in step 721 to form a digital image of the structures of the identifier of the consumer digital dimensions, corresponding to the different possible permissions to print, for example one of the structures of the identifier has a size 66×66 pixels at 400 dpi, another structure identifier has a size of 100×100 pixels at 600 dpi, another structure identifier has a size 133×133 pixels at 800 dots per inch, and another structure identifier has a size of 200×200 pixels at 1200 dpi;

- in step 722 each structure identifier with a variety of digital print sizes several times, for example 100 times, with the appropriate resolution so that the print sizes corresponded to the available surface area;

- in step 723 each structure identifier capture several times for each type, for example 3 times;

- in step 724 for each structure identifier count output that is a function of the destruction of the digital authentication code in the step of forming the image, and generally unique for each generated image due to the random nature of each individual errors;

- in step 725 calculates the evaluation of similarity for all pairs of registered structures of ID with the same print resolution; and

- in step 726 carry out the method described with reference to measurement in accordance with the specified method of extraction of the total output of the data is x, it measures the rate of separation of output data for each resolution print and select the print resolution giving the maximum value for this index.

In one embodiment, the implementation matrix of protected information print resolution print and determines the resolution of the print, giving the magnitude of the error of 25%, as calculated by one of the described algorithms.

In one implementation options select the print resolution, which corresponds to the largest difference between the smallest value estimates calculated on the basis of comparison of the output data corresponding identical to the seals, and the maximum value estimates calculated on the basis of the comparison output corresponding to the different seals.

As shown in figure 4, in step 105 receive at least one message, at least one key, the physical size of the Data Matrix code and the print resolution.

Optionally, in step 110 based on the keys and messages to determine at least one of the encoded message for inclusion in the digital authentication code. In particular, messages from the digital authentication code may be associated with a message code Data Matrix, and one of these codes at least partially based on another to increase the reliability of authentication.

Step 11 on the basis of at least one message, adopted at step 105, form a Data Matrix code.

At step 120 to determine the number of black elements in the code, Data Matrix, and the structure detection means can be included or not included in the code Data Matrix.

In step 125, depending on the number of black elements specify the print resolution, the physical size and the number of elements in the digital authentication code.

In step 130, depending on the number of items, keys and messages define the values taken by each of the elements of the digital authentication code, using the algorithm of formation of the digital authentication code. It should be noted that the algorithms of generating the digital authentication code often contain steps encryption, encoding and reinstall elements (for example, rearrangements and/or substitutions).

In step 135 to form digital image Data Matrix code by introducing the values of the digital authentication code in the specified order (for example, from left to right, then top to bottom) in the pixels corresponding to the number of black elements.

In step 140, the object is printed or labeled with the development of digital image Data Matrix code that includes a digital authentication code.

Below is an example where the code 150 Data Matrix, shown in figure 1, generated according to a standard algorithm based on the message. Size code 150 is 26x26 ale is new, including patterns detection. This code contains 150 344 black item, including patterns detection.

In this example you want to print code 150 using printing devices, providing a resolution of 600 pixels per inch, with a coverage area of approximately 1 cm × 1 cm Size in pixels of the image, equivalent to 1 cm, is equal to 236×236 pixels, or 236 pixels for 26 elements, i.e. 9.07 pixels per 1 element (for each dimension). After rounding up to nine pixels for 1 item in each dimension resulting code size Data Matrix, equal to 234×234 pixels (as 26×9=234), i.e. there are 9×9=81 pixels per 1 element.

Since there are 344 black element, the digital authentication code generated in all black elements may have a 81×344=27864 pixel. Digital authentication code having a 1 bit by 1 pixel can therefore be formed in such a way as to have 27864 pixel. Digital authentication code is generated from the keys and messages according to known algorithms, and its value is entered in the black elements.

Figure 2 shows the code 160 Data Matrix according to the present invention, which is formed from the code 150 and in which the digital authentication code is distributed on all black items. Depending on the means of printing - ink, paper, or other media - it is possible that quality to the Yes Data Matrix is not sufficient for decoding, for example, due to the fact that it has "holes" in the elements. To address this in one implementation options of the selected subset of pixels of the black elements for storing the values of the digital authentication code. For example, selected any other pixel, and the non-selected pixels remain black, giving an average of 75% of the black pixels on 1 item. Therefore, the digital authentication code is 27864/2=13932 bit. Figure 3 shows such a Data Matrix code 170. A subset of the pixels black elements may also be selected based on the cryptographic key.

It is desirable that the readability of the code Data Matrix was not significantly affected by the changes. For example, the authors of the present invention printed codes 150, 160 and 170 with office laser printer with a resolution of 600 pixels per 1 inch (the size of 1 cm) and used a device verification barcode "TruCheck USB verifier" (registered trademark), which determines the type of code Data Matrix. Code 150 that performs the function of a reference sample obtained type 'A'and code 160 and 170 derived types "B" and "A" respectively. It should be noted that the code 160 contains more information for the authentication and/or contains a message)than code 170. Plainly may be an inverse relationship between the quality of the Data Matrix and the amount of information contained in the digital authentication code. Space, to the will to change the digital authentication code, in practice depends on the type that is acceptable for a particular application. In this case, if the type is acceptable (usually a type to a type "C" are acceptable), then, as a rule, choose code 160 that contains more information. If only acceptable type "A", then choose code 170. In other cases, the utilization of the selected item regulate to achieve the minimum desired type.

To increase the number of elements of the digital authentication code can be also used white (or not marked) areas. In this case it is necessary to maintain a low color density for the white area, so that decoding is not disturbed. For example, can be used 20% of the pixels in the white space, preferably not including the boundary of the site, which may come in contact with a black area. It should be noted that if the digital authentication code contains equiprobable binary value, then 10% of the pixels in the middle are black and slightly reduce the correlation coefficient of the element. In the previous example we used a 7×7 internal pixels of the white area, with 10 of them selected pseudo-random manner and contain the element of the digital authentication code. Items can only be placed in even-numbered columns and rows, for example to prevent them from touching. These elements can be clucene in the digital authentication code black elements, or they can be considered as one of the digital authentication code, providing another means for authentication. On figa and 9B presents an example code 180 Data Matrix for which white areas also contain elements of authentication.

It should be noted that the elements of the digital authentication code can also have a variable size, such as items of size 1×2 pixels and 1×1 pixels, more difficult to identify items by the manufacturer of the counterfeit trying to completely recreate the original digital authentication code.

Below with reference to figure 5 shows an example of the authentication algorithm code, Data Matrix, used in conjunction with a digital authentication code.

In step 205, the image containing code, Data Matrix and, therefore, the digital authentication code, receive from the device to capture images, such as using a scanner. In parallel, decoding codes, accept settings read digital authentication code (for example, the size in pixels of each element) and the level of decision-making.

In step 210 decode the message code, Data Matrix, specified using the medium shapes and colors of square elements.

In step 215 determines whether the read message Data Matrix code, for example in accordance with the enabled detection code and error correction. If the relationships read correctly, code Data Matrix is considered as inauthentic, and the user display the following: "the Code is not authentic". If the message is read correctly, at step 220 form the original image code Data Matrix.

Then at step 225 determines the number of black elements and determine according to the parameters read digital authentication code the number of elements of the digital authentication code.

In step 230, depending on the original image code Data Matrix and the number of elements of the digital authentication code to determine the position (in pixels) in the image of each element of the digital authentication code. In step 235 the value associated with the value of each element of the digital authentication code, for example, the gray level of the pixel extracted from the image. This gives the vector data representing the printed digital authentication code and having the place of destruction.

In step 240 decrypts at least one digital message authentication code using a key decryption (if the digital authentication code is encrypted.

In step 245 determines the assessment, describing the magnitude of the destruction of the digital authentication code. As assessment is used, for example, the percentage of correct bits, the correlation coefficient between the original digital authentication code is eficacia and digital authentication code, measured on the basis of recorded images, etc

If the digital message authentication code and Data Matrix code correlated in a time when they are formed, then the correlation if necessary check in step 250; otherwise, the user is given the message: "Unauthenticated code Data Matrix".

Finally, in step 255 the measured score is compared with predetermined limit or threshold level of decision-making". If it is higher, for example, due to the low magnitude of the error or a high degree of correlation, the user is presented with the message "authenticated Data Matrix". Otherwise, display the message "Data Matrix is not authenticated". In some cases, the display read every message.

In some embodiments of the bits of the digital authentication code are reserved for synchronization essentially known manner.

Note that the same code Data Matrix can be repeatedly printed or plotted by marking with static printing (offset printing, printing aniline dyes and so on) or may vary with each printing using digital printing.

In the second variant of implementation for the integration of digital authentication code implementing a system for laser marking and microdata. Some marking system for code, Data Matrix, cast the tis laser or micro system, can't work with large images, as can be seen from the foregoing. For example, the marking image size 236×236 pixels, as in the example above, it takes too long time and too slows down the pace of the production line or generates large Data Matrix code. Described below private embodiment is intended to eliminate these drawbacks.

Several methods of implementing a Data Matrix code is possible with the use of laser:

the laser pulse can create each element,

- multiple shared laser pulses can create each element,

element or group of elements can be vectorized and engraved by means of a continuous laser exposure.

In addition, some laser marking systems can locally modify the following differentiating features:

- power laser;

the polarization of the laser;

- focusing a laser beam on the surface are marked;

- micropositioning;

- the direction or order in which the points are marked with a laser; and

the shape of the wavefront.

Similarly, some of the ways to run the code Data Matrix can be realized by using microdata:

- mikrovolovka can create every single element; or

- many micrographic can in order to create each element.

In a similar manner can be used the possibility of marking systems microgravity locally modify the following differentiating features:

- power;

- micropositioning;

- orientation mikrovolovka; and/or

- the form of mikrovolovka.

For optimizing the timing of the production code, Data Matrix, consisting of various elements, (in accordance with the controlled settings of the laser device or devices microvolume), the elements are divided into subsets or classes. Each subset of elements, the differentiating characteristics of which are the same form, usually within one pass of the tool. Due to this configuration, the differential impact when marking, change only once for each subset of items instead of individual changes for each item. Each setting that can be changed locally and which has a significant impact on the generated code Data Matrix, can be used to store information. For example, if the laser power has two levels, 1-bit information can be stored by changing the thickness or other properties. In some embodiments of the changes locally configurable settings can be used in combination.

In one and the options for implementing the configuration can be changed locally and quasi-continuous, and can be defined for an arbitrary number of levels for storing information. For example, the implement 10 levels for a given configuration (for example, change the color or size) instead of the two described above. In another implementation, the value of this parameter may even change continuously.

However, in the examples of implementation, it is preferable to use only two levels of values for the signal source. In this case, can be used for optimum signal-to-noise ratio equal to 0.56, with maximizing the effectiveness of your copy. To ensure this value relations determine the noise properties of the channel, usually using the print pair material/tools, characterizing the distribution of the measured signal during image capture. In the case of two energy levels, the statistical distribution of the magnitude of the impact are examined to determine the two energy levels sufficiently close for the consideration of the partial superposition of their distributions, with the goal of providing above the ideal signal-to-noise ratio. In particular, if the standard deviation is 0.01 mm²for the laser power with specified values of influence 0.10 mm²choose power levels, providing the average magnitude of the impact, similar to the values 0.1075 mm²and 0.0925 mm². Indeed, BL is through this receive signal-to-noise ratio, equal to 0.0075²/0.01²=0.5625, which is very close to theoretical optimal value.

Thus, if the size of the Data Matrix, equal to 26×26 items, use 344 laser exposure. Digital authentication code formed in the same manner as described above, and change through the points with dimensions 0.1075 mm²and 0.0925 mm².

As stated above, check the cause if the changes made to the destruction of the Data Matrix code that is incompatible with the terms of the application, for example, type "C" is the minimum possible for the Data Matrix.

In some cases it is impossible to change for more information. For example, the means of printing or marking can only be used to label or not to label the element (a single binary labeling). There are also cases when additional identification information does not provide high security against copying. For example, tool marking, providing pre-determined levels of labeling, which are stored separately from the others and clearly visible in the marked image, allow you to enter additional information, but in principle still possible to make an identical copy of it.

In these cases, it may be used residual noise when printing or marking, and it can function as a digital authentication code is. In fact, regardless of the type of printing, at a certain scale or resolution, there are some defects. For example, if the laser effect can in principle leaves round or elliptical affecting the spot, at sufficiently high resolutions form of influencing spots usually are not quite regular. This is evident even if you use an inkjet printer system. Even at higher resolutions there are irregularities in the depth of the acting point etc.

The irregularity of the mark can be registered, measured and used to generate a digital authentication code. Then the digital authentication code may be stored in a database or saved as a two-dimensional barcode. The latter approach, however, is not very desirable, because the marking of the second code is an expensive operation and requires space on the document, which is usually contrary to the desired effect. On the other hand, the digital authentication code may be stored in Association with the message code, Data Matrix, which in the case of its uniqueness (which is preferable) allows the authentication step to make a simple verification by comparing measurements of the defect marking with digital authentication code. Many measurements of defects is possible, for example, you can measure the average color or Owen gray color for each item in the code, Data Matrix, may be determined by the contour of the item can be measured the distance between the center of gravity and the external circuit at different angles, etc. point of impact may overlap. In this case, can be established limit value for the distance to the external circuit.

For the considered channel print "average" the labeling of a particular two-dimensional barcode can be modeled based on the average size of the points of impact and, if necessary, on the basis of possible interactions, when the point of impact related to each other. The image is evaluated using a simulation of the registered image can be subtracted, and the result has less redundancy, which increases the signal-to-noise ratio and at the same time, the effectiveness of the registration.

Below with reference to Fig.6 shows an example algorithm for recording defect marking barcode.

In step 305 accept the recorded image containing the barcode.

In step 310 decodes the message barcode, expressed through the medium of shapes and colors of square elements.

In step 315 message-based compute the ID.

In step 320 is measured characteristics of the defect marking barcode.

In step 325, if necessary, subtract the average value of characteristics for a set of Data Matrix codes bil is istics of, measured for the considered code Data Matrix.

In step 330 characteristics expressed quantitatively, if necessary, tighten and define the vector of characteristic data representing defects.

At step 335 the characteristic vector of data store in the database associated with the message ID.

Below with reference to Fig.7 shows an example algorithm for authentication two-dimensional barcode based on the measurement marking defects when printing.

In step 405 take the recorded image containing the barcode.

In step 410 decodes the message barcode, expressed through the medium of shapes and colors of square elements. If the message cannot be decoded, for the user, display: "Naschityvaetsya barcode.

Otherwise, at step 415 message-based compute the ID.

In step 420 from the database, get the data vector corresponding to the identifier, and the threshold level of decision-making associated with this vector characteristic data. If the database does not contain the identifier for user display: Barcode not authenticated".

Otherwise, at step 425 measure the characteristics of the defect marking barcode.

At step 430 the average characteristics for multiple barcodes if necessary the cost deducted from their characteristics, measured for the considered code.

In step 435 characteristics expressed quantitatively, if necessary, tighten and define the vector of characteristic data representing defects.

At step 440 the extracted data vector is compared with the vector data obtained from the database, and calculate a similarity metric, called evaluation.

At step 445 the measured score is compared with a threshold level of decision-making. If the score is greater than this threshold level, the user displays the message "Data Matrix Code authenticated". Otherwise, the user displays the message "Data Matrix Code is not authenticated". Each read the message, if necessary, to display to the user.

On Fig presents local terminal 505, equipped with a printer 510, means 535 for image capture, the two sensors 540, 545 and means 515 for access to the network 520 connected to the server 525. Server 525 is equipped with a base 530 of the data.

As a local terminal 505 is used, for example, a typical computer. It is installed on the production or manufacturing line 550 of various objects, for example on the packaging line. Line 550 contains, for example, an unloading mechanism for flat objects (not shown) and a conveyor (not shown)driving processed one after the other about the projects.

The sensor 540 is installed on the production line 550 before the optical field sensor 535 image, when viewed in the direction of travel, and is designed to register the arrival of the processed object. As sensor 540 is used, for example, an optical element containing a transmitter and receiver of the light rays. The sensor 545 is placed on line 550 and determines the rate of interest on this line. For example, the sensor 545 is connected to a programmable logic controller (not shown)that controls the operation of the line 550 to the base for moving objects, for example a conveyor belt. Local terminal 505 operates essentially in a known manner by printing objects using the printer 510, for example by inkjet printing or laser marking. Means 515 for access to the network 520 represents, for example, the modem known type for access to the network 520, such as the Internet.

As a means 535 to capture images used, for example, a digital camera, a linear sensor or an industrial camera.

As the server 525 used server known type. In the 530 data saved at least a list of object identifiers and data vectors defect associated with these objects, defined in accordance with the proposed method. In the 530 is preferably stored in combination with the ID manually is where it will each object identifier object type and the location of the geometric code implemented according to the present invention for this type of object and the identifier of the service provider performing the manufacture or processing.

Terminal 505 contains a program that when executed implements the steps of the proposed method. Terminal 525 contains a program that when executed implements the steps of the method of storing vector data of the defect and restore method.

In some embodiments of the terminal 505 does not contain any special software, and uses a web browser and a web service stored on the server 525.

1. The method of generating code for authentication with geometric areas, the shape and/or color which can be changed in accordance with the message, characterized in that it includes:
- the step that generate code with variable geometric areas, with the formation of geometric sections in accordance with the message;
- the step that generates a digital authentication code with providing digital values, and
- step on which form the image areas specified geometric code, containing part of the digital authentication code at least in some of its geometrical areas and/or at least one gap between the geometric areas, and generate destruction of the digital code autentificat and random.

2. The method according to claim 1, characterized in that, due to physical uncertainties inherent to image formation, at the step of forming the image specify the magnitude of the error display of the digital authentication code, which is greater than the first preset value and less than the second preset value.

3. The method according to claim 1 or 2, characterized in that, due to physical uncertainties inherent to image formation, at the step of image formation between the two groups of the same code with variable geometric areas set the deviation display digital authentication code, which is larger than the third preset value and less than the fourth preset value.

4. The method according to claim 1, characterized in that, due to physical uncertainties inherent to image formation, at the step of forming the image set the noise relating to the display of the digital authentication code, and the signal-to-noise representation of the digital authentication code is less than a fifth preset value.

5. The method according to claim 1, characterized in that it further includes:
- the step that determines the conditions for the formation of the specified image; and
- step depending on the condition the th formation of the image determine the physical characteristics of the elements at least part of the digital authentication code.

6. The method according to claim 1, characterized in that the specified step of generating code with variable geometric areas specified variable geometric areas are essentially parallel rectangular strips whose width and/or the distance between them is changed in accordance with the specified message.

7. The method according to claim 1, characterized in that the specified step of generating code with variable geometric areas specified variable geometric areas represent an array of square plots, color, and/or at least one dimension which is changed in accordance with the message.

8. The method in accordance with claim 1, characterized in that at this step of forming the image code with geometric areas containing at least some of the geometric areas of the digital authentication code, as a digital authentication code using the deviation of at least one dimension changeable geometric sections.

9. The method according to claim 1, characterized in that at this step of forming an image of the specified code with geometric areas containing, at least in some of its geometric sections, part of the digital authentication code, each part of the digital authentication code, introduced the geometrical area code with variable geometric areas, has the shape of a rectangular distribution of elements whose size is at least one order of magnitude smaller than the specified geometric sections, and part of these elements has a color different from the color of the specified geometric sections.

10. The method according to claim 1, characterized in that each geometric area containing part of the digital authentication code, the area of these elements is less than a quarter of the area of the geometric segment.

11. The method according to claim 1, characterized in that it further includes a step of encoding information in a specified digital code authentication.

12. The method according to claim 11, characterized in that the above information is based on the message and/or message based on the specified information.

13. The method according to claim 11 or 12, characterized in that said information is a measure of the destruction of the digital authentication code due to physical uncertainties that affect the image at the specified step of forming the image.

14. The method in accordance with claim 1, characterized in that it further includes a step in which measure the destruction of the digital authentication code generated at the specified step of forming the image.

15. The method according to 14, characterized in that at this step the dimension of destruction used registration codes errors, is built in the digital authentication code.

16. The method according to claim 1, characterized in that it includes a step in which you determine the output image data is formed, dependent on the destruction of the digital authentication code at the specified step of forming the image.

17. The forming device code for the authentication code with geometric areas, the shape and/or color of which is changed in accordance with the message, characterized in that it comprises:
the means of generating the specified code with variable geometric areas in accordance with the message, to ensure geometric areas;
- means the generation of a digital authentication code, to provide digital values; and
- the means through which is formed the image areas specified code with geometric areas containing at least some of its geometrical areas and/or at least one gap between the geometric areas of the digital authentication code, generated by the destruction of the digital authentication code at random.

18. The authentication code with geometric areas of variable shape and/or color, presented in the form of a recorded image, characterized in that it includes:
- the step that reads the message in the middle fo the max and colors of geometric sections;
- step on which measure the level of destruction of the digital authentication code, presents at least in some geographical areas specified code with geometric areas; and
- the step that determines the authenticity of the specified code with geometric areas on the basis of at least the specified level of destruction.

19. Device authentication code with geometric areas having shapes and/or colors present in the form of a registered image, characterized in that it comprises:
- readers of the message contained in the middle of all colors and shapes of geometric sections;
- a means of measuring the level of destruction of the digital authentication code, presents at least some of the geometric sections of the code with geometric areas; and
- means for determining the authenticity code with geometric areas to determine the authenticity of the specified code with variable geometric areas on the basis of at least the specified level of destruction.