Method for personalizing a security device, personalized security device, identity document containing such a device, and method for authenticating such a device

Abstract

The invention relates to a method for personalizing a security device comprising at least one uniqueness parameter (8) in a predetermined object geometric space (E), said method comprising the steps of: - identifying the at least one uniqueness parameter (8) of said security device; - determining a location of the at least one uniqueness parameter (8) in the object geometric space (E) of said security device; - generating a location of at least one micro-pattern (9) in an image geometric space (E') of the security device by a mathematical function, from said location of the at least one uniqueness parameter (8), said object geometric space (E) being determined such that the mathematical function is injective on this space; and - marking the at least one micro-pattern (9) of said location generated by the mathematical function in said security device. (Figure 2)

Description

Description Title of the invention: Method for personalizing a security device, personalized security device, identity document incorporating such a device, and method for authenticating such a device device

[0001] = The invention relates to a method of personalizing a security device. The invention also relates to a personalized security device as well as a authentication method for such a device. It also concerns a document identity documents that include such a device.

[0002] = A safety device here refers to an element that one wants to be able to verify, at any At the moment, authenticity is required to grant rights to the holder of such a device.

[0003] An identity document here refers, for example, to an identification card, often including, but not necessarily, a microcircuit, such as a circuit board bank account, identity card, driver's license; it could also be a passport, usually in booklet format with several pages, including one page of The data, legible to the naked eye, is linked by a hinge attached to the booklet to the others. pages.

[0004] — A security device generally includes an identity photograph of its holder, and / or alphanumeric data specific to the holder, a per- character These different data helps make each security device unique, depending on its holder.

[0005] — To demonstrate the authenticity of a security device, for example during a For identity verification, the security device is, for example, compared with a copy digital copy of it, the copy having been acquired before sending the device to its holder. This copy forms a digital code that can be used to track the evolution of the system and detecting fraud or falsification, for example through to image processing techniques based on comparison and correlation pondance.

[0006] — A practical example is to perform a high-resolution scan of the security device before sending it to the holder and storing characteristics uniqueness in a database. A matching result between the copy digital and a scan of the device after sending it to its owner thus provides an in- training on its authenticity.

[0007] — However, the above example as well as the other solutions proposed up to now present a number of drawbacks. Indeed, it is necessary to Performing high-quality scans of the security device is necessary to obtain a reliable comparison result. Scanning each security device leads to bulky storage and generally requires cloud computing. These solutions are therefore expensive and may not comply with personal data privacy laws. In some cases, the authentication of a security device, particularly the image processing required for this authentication, is carried out using standard image acquisition devices, such as smartphones. While the use of such devices is easy and inexpensive, it can be prone to errors due to the significant image processing required by state-of-the-art solutions. Document EP 3 459 757, for example, describes a solution based on guilloche patterns forming a network of wavy lines superimposed on an identity photograph and visible to the naked eye. This solution was developed to secure the alphanumeric data recorded on the security device by inserting variable guilloche patterns onto the identity photograph, encoding the alphanumeric data. However, such guilloché patterns are easily identifiable and can be reproduced and superimposed on another ID photograph. Furthermore, the ID photograph can be altered between the guilloché patterns through image processing or overprinting. This solution therefore does not prevent the falsification of an identity photograph on a security document by an imposter. Furthermore, the presence of guilloche patterns on the photograph can affect its visibility and make it more difficult to verify a security document during an identity check. In this context, the invention aims to resolve at least some of these drawbacks. The invention aims in particular, during the manufacture of a security device, to make it possible to obtain in a simple way a specific signature of each security device, the memorization of which does not involve the entry and storage of a large amount of data and the recognition of which at the time when one wishes to authenticate such a device does not involve a specific or complex device, therefore without requiring significant costs. The invention proposes, according to a first aspect, a method for personalizing a security device comprising at least one uniqueness parameter in a predetermined object geometric space, said method comprising the steps of: - identification of at least one uniqueness parameter of said security device; - determination of a location of at least one uniqueness parameter in the geometric space of the object of said security device; - generation of a location of at least one micro-motif in a geometric image space of said security device by an injective mathematical function, starting from said location of at least one uniqueness parameter; and - marking of at least one micro-pattern at said location generated by the injective mathematical function in the geometric image space of said security device. The location of at least one uniqueness parameter of the security device is thus linked to the location of at least one micro-motif by the mathematical function. Such a personalization process makes it possible to provide a specific signature by generating at least one micro-pattern whose location is reasonably easy to identify optically, while offering a very wide variety of possible reference configurations. Furthermore, it is possible to avoid adding a foreign element since the basic data, namely the at least one uniqueness parameter, is intrinsically present in the security device. The personalization process thus makes it possible to link this specific signature reversibly, or even uniquely, to at least a portion of the security device owner's data. In this case, the location of the at least one uniqueness parameter of the security device is linked to the location of the at least one micro-pattern by the injective mathematical function. By injective method, the present invention aims to define a function for which every element of the image geometric space has at most one preimage in the predetermined object geometric space. The preimage represents the location of at least one uniqueness parameter, and the image represents the location of at least one micro-motif. In other words, a function is said to be injective if in the object geometric space, two elements always have two distinct images under this function in the image geometric space. Thus, the sine function can be considered an injective function when defined on the interval [-5; Z], for example. [#] Furthermore, because at least one uniqueness parameter and at least one micro-pattern are visually identifiable, at least by a viewing device, on the personalized security device, there is no need to store information about them, such as their location in the security device. Indeed, at least one uniqueness parameter is subject to random phenomena during the manufacturing of security devices. These phenomena do not involve the addition of any foreign particles, but deviations from a reference configuration allow each device to be differentiated. These random phenomena in a manufacturing process are defined by the concept of a "physical unclonable function," abbreviated PUF (for "Physical unclonable function"). The greater the number of uniqueness parameters chosen for the security device, the higher the probability that another device will be unique. The security device exhibits the same uniqueness parameters at the same locations, which is low. Determining the location of at least one unique parameter can be done using known image processing techniques with relatively simple devices, such as smartphones. For example, the same device can be used to identify and locate at least one unique parameter on the security device. Given the observational accuracy of many current optical devices, which can reach 20 microns, it is possible to define a specific signature for each security device. The individualization of each security device is thus achieved in a much simpler way than in the state of the art where the specific signature of each security device implies the acquisition and storage of information specific to each security device. Thus, for a plurality of security devices, it is possible to store only the mathematical function. The injective mathematical function can, for example, be stored on a secure server and be accessible by a sender of the security device and / or a controlling entity. This mathematical function can be common to a plurality of security devices while having little risk of identically personalizing at least two devices for different holders. The injective mathematical function can be encrypted to increase the security of the personalized security device. Furthermore, in a case where a security device has a plurality of uniqueness parameters, the mathematical function can be configured to maximize the dispersion of the locations of the plurality of micro-patterns generated in the geometric image space of the security device. The mathematical function can be configured to generate a micro-pattern location distinct from the location of at least one uniqueness parameter. The predetermined object geometric space and the image geometric space can correspond to a surface of the security device. The geometric spaces of the object and image can be identical. Object and image geometric spaces can be at least partially superimposed. Object and image geometric spaces can be distinct. Furthermore, the geometric object and image spaces can be of any shape, arbitrary, or for example square, rectangular, triangular or even circular. In one particular case, the object geometric space includes at least one photograph of a security device holder, while the image geometric space is defined outside the photograph so as to avoid the generation of at least one micro-pattern on the photograph. According to another implementation method, the mathematical function can be bijective, when it is both injective and surjective. In other words, a bijective function is a function for which every element of its image geometric space has a unique preimage in the object geometric space, and vice versa. For example, two distinct micro-motifs are associated with two distinct uniqueness parameters, and vice versa. The marking step may include a sub-step of engraving and / or printing at least one micro-pattern into the security device. Such an engraving sub-step may, for example, be performed using a laser, while a printing step may be performed using any ink. The process may further include a parameterization step of the mathematical function in which at least one coefficient of said mathematical function is determined. According to an example, at least one coefficient can be determined from at least one characteristic data of the safety device. A characteristic data element can be data relating to the holder of the security device, such as a photograph, alphanumeric data, or a signature. This data can also be data relating to the issuer of the security device, which could be a bank in the case of a bank card, a security organization in the case of a badge, or a state in the case of a visa or passport. In this case, at least one characteristic data element could, for example, be alphanumeric data, a logo, or a guilloché pattern. At least one coefficient of the injective mathematical function can be determined based on at least one characteristic of the safety device, so the injective mathematical function can vary from one safety device to another. Thus, it is possible to store a common function for all safety devices, while parameterizing this function differently for each safety device. According to another example, at least one coefficient of said mathematical function may contain a cryptographic key. Such a cryptographic key can be configured to encrypt and decrypt the mathematical function and / or to define an additional degree of verification of the mathematical function. For example, the cryptographic key may contain at least one bit. In an example implementation where the security device has several uniqueness parameters associated with several micro-patterns, each micro-pattern can be associated with at least one bit. Identifying, determining or generating micro-patterns in a predetermined order allows for the definition of a sequence of bits forming an additional security code. For example, five micro-patterns can be associated respectively with one bit, together forming the sequence 10101. The 10101 suite can be stored on the security device, for example via an electronic chip integrated into the security device, and / or encoded on the security device, for example via a barcode or a QR code (from the English "Quick Response code"). Redundancy check elements, for example redundancy bits, can be added to the security code to verify its integrity. The injective mathematical function may include a coefficient determined from a characteristic data of the security device and a coefficient including a cryptographic key or part of a cryptographic key. The cryptographic key can be independent of the security device, that is, not be generated from data characteristic of the security device. The process may also include a step of defining a predetermined reference frame. Thus, the location of at least one uniqueness parameter in the object geometric space can be defined by coordinates in said predetermined frame and / or the location of at least one micro-motif in the image geometric space can be defined by coordinates in the predetermined frame. Thus, thanks to the simple identification of coordinates by optical measurement, it is possible to characterize a very large number of security devices. The predetermined reference point can be linked to the safety device and can be defined from at least one characteristic data of said safety device. In this way, the predetermined frame can depend on at least one characteristic datum and can therefore also satisfy the conditions of a non-clonable physical function (PUF). The coordinates of the at least one micro-motif thus generated in the predetermined frame can be specific to each security device. According to one variant, the coordinates of at least one uniqueness parameter can be defined in a first predetermined frame and the coordinates of the micro-patterns can be established in a second frame, different from the first. The invention also relates, according to a second aspect, to a personalized security device comprising at least one uniqueness parameter in a predetermined object geometric space and at least one micro-motif in an image geometric space, and at least one location of the at least one uniqueness parameter and at least one location of the at least one micro-motif being linked by a mathematical function injective. Such a device is obtained, for example, through a customization process that includes all or part of the characteristics described above. This type of customized security device thus offers advantages similar to those described above in relation to the process. At least one micro-pattern may include at least one dot engraved and / or printed in the personalized security device. At least one micro-motif may have a characteristic dimension of less than 200 µm, or even less than 50 µm, or even less than 5 µm. The image geometric space can be configured to be distinct from at least one predetermined area, such as a photograph of a device holder. A predetermined area can, for example, be defined so as not to alter the visibility of certain characteristic data, or so as to avoid interference with at least one uniqueness parameter. A predetermined area can also be defined so as to include characteristic data in order to reduce the visibility of at least one micro-pattern to the naked eye. The invention also relates, according to a third aspect, to an identity document comprising an identity photograph and / or alphanumeric data relating to a holder of said identity document and a personalized security device according to the second aspect, the identity photograph comprising at least one uniqueness parameter. As an example, one side and / or the other side of the identity document may contain the geometric object and image spaces of the security device. As another example, one side of an identity document may contain either the object geometric space or the image geometric space of the security device, while the other side may contain the object geometric space or the image geometric space of the security device. These examples are particularly relevant when the identity document is an identity card, a bank card, or any other document with at least two opposite sides. In yet another example, the first page of an identity document may contain the object geometry, and the second page may contain the image geometry of the security device. This example is particularly relevant when the identity document is a passport or any other identity document comprising at least two pages. The invention also relates, according to a fourth aspect, to a method for authenticating a security device personalized according to the second aspect of The invention, said authentication method comprising the steps of: - identification of at least one uniqueness parameter of said personalized security device; - determination of a location for at least one identified uniqueness parameter; - generation of a check location of at least one micro-check pattern by the injective mathematical function, from the location of at least one identified uniqueness parameter; - identification of at least one micro-pattern in said personalized security device; - determination of a location of at least one identified micro-motif; - comparison of the location of the verification of at least one micro-pattern with the location of the at least one micro-pattern identified in said personalized security device; and - authentication of said personalized security device in case of correspondence between the location of verification of at least one micro-pattern of verification and the location of at least one micro-pattern identified in said personalized security device. Therefore, it is possible to verify the authenticity of a personalized security device according to the invention, by comparing the location of at least one micro-pattern identified on the personalized security device with the verification location of at least one micro-pattern generated by the injective mathematical function from the identified location of at least one uniqueness parameter. A match between locations can be considered when the difference between locations is less than a predefined tolerance zone. This tolerance zone may be intended to account for inaccuracies related to certain steps in the personalization or authentication processes, such as the image acquisition, location determination, or even micro-pattern marking stages. When the personalized security device has a plurality of uniqueness parameters and associated micro-patterns, it may be possible to detect if only part of the security device has been tampered with. In the case where a coefficient of the mathematical function has a cryptographic key, it is possible to find this cryptographic key by iterating from the location of at least one uniqueness parameter and the mathematical function until a match with a micro-pattern is found. The cryptographic key found or the security code can be verified with information that can be stored on the security device, for example on a chip or encoded with a barcode or QR code in the security device. It is therefore possible to verify the integrity of the mathematical function by comparing the cryptographic key (or security code) with the information stored and / or encoded on the security device. The invention, according to an exemplary embodiment, will be better understood and its advantages will become more apparent upon reading the following detailed description, given by way of example and in no way limiting, with reference to the attached drawings in which: -Fig.1 represents a personalized security device according to the invention, comprising characteristic data and micro-patterns; - [Fig.2] represents a non-personalized security device containing characteristic data before a micro-pattern marking step; - [Fig.3] represents an example of the application of an authentication process for a personalized security device a) comprising micro-patterns and an identity photograph, and a falsified version b) comprising micro-patterns in identical locations and a different identity photograph than the authentic device; - [Fig.4] schematically represents a method of personalizing a security device according to an example of implementation of the invention; - [Fig.5] schematically represents a method of authentication of a personalized security device according to an example of implementation of the invention. Identical elements represented in the aforementioned figures are identified by identical numerical references. Figure 1 represents a personalized security device 1 according to the invention. Such a personalized security device 1 is intended to authenticate an individual, in particular its holder, and thus allow verification of that individual's identity. The personalized security device 1 is specifically designed to be unique, so as to avoid any confusion between several individuals. For this purpose, the personalized security device 1 includes a security device 10 represented for example in [Fig.2], which includes at least one characteristic data linked to at least one characteristic specific to the holder or issuer of the security device 10. In the embodiment shown in [Fig.2], the security device 10 includes several characteristic data specific to its holder, including an identity photograph 2, alphanumeric data 3 and a signature 4. The security device 10 also includes at least one characteristic of the issuer of the security device 1, for example, a bank in the case of a bank card, a security organization in the case of a badge, or a State in the case of a visa or passport. This at least one characteristic of the issuer is presented here in the form of alphanumeric data 5, a logo 6, and guilloché patterns 7. At least one characteristic data point may be presented in other forms such as a drawing. This at least one characteristic data is present on the security device 10 before personalization according to the invention and is linked to random characteristics caused during its manufacture. Therefore, such a security device 10 has a uniqueness parameter 8 which here includes at least one of such characteristic data. By definition, the security device 10 includes a predetermined object geometric space E comprising at least one uniqueness parameter 8. The geometric object space E corresponds here to the identity photograph 2 of the security device 10. The uniqueness parameter 8 is used to define a specific signature to distinguish a security device among a multitude of customized or non-customized security devices. Indeed, during the manufacturing cycle, small variations may occur, for example, variation in the position of a biometric point or the position of a logo, emblem or design, or any other characteristic data. It has emerged that these variations can be measured in a reasonably easy manner, so that the conditions for a non-clonable physical function (PUF) are therefore satisfied. Such a uniqueness parameter & is for example identified on figures 1 and 2 by a hatched circle and corresponds to a biometric point on the identity photograph 2. In this case, several uniqueness parameters 8 are identified and correspond to different biometric points such as the pupils, nose and corners of the mouth on the identity photograph 2 of the holder of the security device 10. The uniqueness parameters 8 identified by hatched circles represent here only a selection from the set of uniqueness parameters of the security device 10. Each uniqueness parameter 8 includes, for example, at least one alphanumeric data point, a logo, a design or a coat of arms, or a particular point of these elements. More generally, a uniqueness parameter 8 can include any characteristic data of the security device 10 that is capable of satisfying the conditions of a non-clonable physical function (PUF). It follows that the position of a uniqueness parameter 8 can constitute uniqueness information that can be used on the one hand for the authentication of a security device, and on the other hand for the differentiation between several of these security devices. The personalized security device 1 shown in [Fig.!] includes for this purpose the security device 10 shown in [Fig.2] as well as at least one micro-motif 9 identified by a cross. By definition, the security device 10 further includes an image geometric space E' in which the location of at least one micro-motif 9 is generated. The geometric image space E' corresponds here to the represented face of the safety device 10. Each micro-pattern 9 is designed to be invisible to the naked eye. However, at least one micro-pattern 9 is adapted to be detectable by a device such as a smartphone, from an image that can also be obtained via such a device. At least one micro-pattern 9 comprises, for example, a point or point cloud that can have a maximum dimension on the order of a micrometer. For example, a micro-pattern 9 can have dimensions ranging from 1 µm to 500 µm. In particular, the dimensions of a micro-pattern 9 can range from 50 µm to 200 µm for authentication with a smartphone or scanner and from 5 µm to 50 µm for laboratory authentication with a high-resolution scanner. In the described implementation example, each micro-pattern 9 is laser-engraved on the safety device 1. A micro-pattern 9 can also be produced from various processes, such as printing, allowing their marking on a safety device, for example. The micro-pattern 9 can include, for example, at least one point, or a cloud of points, or any type of pattern such as a spiral, a circle, a triangle, a square, etc. The micro-motif 9 is linked to at least one uniqueness parameter 8 via a mathematical function which is injective here. Furthermore, the micro-pattern 9 is configured to encode a uniqueness parameter 8 and thus authenticate the personalized security device 1. Because it is invisible to the naked eye, detecting and / or reproducing a micro-pattern 9 is difficult for an imposter. Furthermore, reproducing a micro-pattern 9 is even more difficult without knowledge of the mathematical function and at least one uniqueness parameter 8 chosen to generate it. The location of at least one micro-pattern 9 in the image geometric space E” of the personalized security device 1 corresponds to a result of the mathematical function applied to a location of the uniqueness parameter 8 in the object geometric space E. For example, the injective mathematical function can be a sine function taken on an interval { = 5, 5] “(ai At least one micro-pattern 9 is marked here towards the periphery of the personalized security device 1, so as not to interfere with at least one characteristic data item such as the identity photograph 2, for example. The image geometric space E' can be defined in such a way that the location of the micro-pattern 9 does not interfere with other data such as characteristic data, for example. The personalized security device 1 of [Fig.1] is obtained from a security device 10 shown in [Fig.2] to which is applied a personalization process described below according to an example of implementation of the invention. Such a personalization process includes a step of determining the location of the uniqueness parameter 8 in the pre-determined object geometric space E of the security device 10 in order to determine a micro-pattern location. To identify the uniqueness parameter 8, an image of the security device 10 is acquired. The uniqueness parameter 8 can be identified using recognition algorithms when an image of the security device 1 is acquired. In the present implementation example, five uniqueness parameters 8 are identified in the object geometric space E in the security device 10 and correspond to particular biometric points of the identity photograph 2. In practice, the same characteristic data can be identified from one security device to another. In order to determine the location of at least one uniqueness parameter 8 identified in the geometric object space E of the safety device 10, a predetermined frame 11 is defined here from the image of the safety device 10 represented in [Fig.2]. The predetermined coordinate system 11 is linked to the security device 10 and is formed by two axes, X and Y. Specifically, the X-axis is defined from the alphanumeric data 5 adjacent to the logo 6, while the Y-axis is defined from an edge of the identification photograph 2 of the security device 10. The characteristic data used to define the coordinate system are chosen here so that the predetermined coordinate system 11 is orthogonal. Any type of coordinate system can be considered, such as a normalized or even orthonormal coordinate system, and the predetermined coordinate system 11 can be defined from any other characteristic data of the security device 10. In an embodiment not shown, the predetermined reference mark 11 can also be defined from one or more of the edges of the safety device 10. In this way, coordinates of at least one uniqueness parameter 8 identified in the object geometric space E by the predetermined frame 11 are determined and used to generate a location for at least one micro-motif 9 in the geometric space image E' security device 10. Depending on the predetermined frame type 11, at least one uniqueness parameter 8 can be located by Cartesian, polar, cylindrical or spherical coordinates. The coordinates of at least one uniqueness parameter 8 in the predetermined frame 11 are obtained here directly after image processing since each uniqueness parameter 8 corresponds to a point. Alternatively, the coordinates can be subjected to further processing, in particular when at least one uniqueness parameter 8 corresponds not to a point, but to a gap, for example to the distance between the eyes. The coordinates of at least one uniqueness parameter 8 identified in the object geometric space E with the predetermined frame 11 are stored temporarily, for example in a device external to the security device 10, in order to be used as an input parameter of a mathematical function configured to generate coordinates of one or more micro-patterns, these coordinates being uniquely linked to the coordinates of at least one uniqueness parameter 8. Once the coordinates of a micro-pattern 9 are generated, it is no longer necessary to store the coordinates of at least one uniqueness parameter 8 because they can be determined as described previously from the security device 10. In particular, the coordinates of at least one micro-motif 9 are generated by the injective mathematical function from the coordinates of at least one uniqueness parameter 8 identified in the predetermined frame 11. The coordinates of at least one micro-motif 9 generated by the injective mathematical function are established from the predetermined frame 11. Alternatively, the coordinates of at least one micro-motif can be established in another frame. In the implementation example shown, the mathematical function is adapted to generate the coordinates of a micro-pattern 9 from the coordinates of a uniqueness parameter 8 so that the number of micro-patterns 9 is equal to the number of uniqueness parameters 8 identified. Alternatively, the injective mathematical function can be adapted to generate the coordinates of a micro-motif 9 from the coordinates of at least two uniqueness parameters 8 identified in the predetermined frame 11, so that the number of micro-motifs 9 is less than the number of uniqueness parameters 8 identified. In order to authenticate a personalized security device 1 according to the invention, an image is acquired, for example during an identity check, in order to identify at least one uniqueness parameter 8 and to determine its coordinates in the predetermined frame 11. This acquisition step is identical to the acquisition step of the personalization process for the security device 10 described previously. The coordinates of at least one micro-motif 9 in the image geometric space E' of the security device 1 are also determined in a manner analogous to the personalization process. Therefore, determining the coordinates from the image of the personalized security device 1 also does not require significant image processing. We then generate so-called verification coordinates of at least one micro-verification pattern 12 ([Fig.3]), with the injective mathematical function, from the coordinates of at least one uniqueness parameter 8 identified on the image of the personalized security device 1, in the predetermined frame 11. The coordinates of at least one micro-pattern are compared to the verification coordinates of at least one verification micro-pattern 12, calculated via the injective function from the coordinates of at least one uniqueness parameter 8 in order to establish or not a correspondence between these coordinates. The personalized security device 1 is considered authentic when the coordinates match, within a tolerance zone. Conversely, security device 1 is declared non-authentic when at least one of the coordinates does not match, within a tolerance zone. In practice, the tolerance zone is defined for the coordinates of at least one uniqueness parameter 8 and / or at least one micro-pattern 9 and allows to take into account inaccuracies related for example to the acquisition of the image, the determination of the predetermined reference frame 11, the determination of the coordinates of at least one uniqueness parameter 8 and at least one micro-pattern 9, or even to the marking of at least one micro-pattern 9 on the safety device 10. One such example of the application of the authentication process is shown in [Fig. 3], in which a first personalized security device 1, represented by Figure 3a, includes an identity photograph 2 of a first holder. This security device 1 also includes alphanumeric data relating to its holder 3 and its issuer 5, as well as a signature 4, a logo 6 and a plurality of micro-patterns 9 having coordinates corresponding to a result of an injective mathematical function applied to the coordinates of uniqueness parameters 8 in a pre-determined coordinate system 11, which corresponds to one or more biometric points of the identity photograph 2. Figure 3, and in particular Figure 3b, also depicts a second personalized security device l' which features the same alphanumeric data 3 and 5, the same logo 6, and the same guilloché patterns 7 as the first personalized security device 1. However, the second personalized security device l' features an identification photograph 2' which differs from the identification photograph 2 of the first personalized security device 1 and for this purpose presents uniqueness parameters 8" different from the uniqueness parameters 8 of the first personalized security device 1. It is assumed here that the identity photographs 2 and 2' of the first and second personalized security devices 1 and 1' are positioned identically, so that the predetermined marker 11 is identical for each of these personalized security devices. Such an example is likely to occur when an identity photograph is altered by an imposter, for example. In the present implementation example, although the coordinates of the uniqueness parameters 8' are different in the predetermined frame 11, the same micro-patterns 9 and 9' have been marked in the first and second custom security devices let 1". According to the authentication process described above, the uniqueness parameters 8 and 8' and the micro-patterns 9 and 9' are identified and their coordinates are respectively determined in the predetermined frame 11. From the theoretical coordinates of the uniqueness parameters 8 and 8' in the predetermined frame 11, verification coordinates of verification micro-patterns 12 are generated by the injective mathematical function. In the case of the first personalized security device | (a) the verification coordinates of micro-patterns of verification 12, generated by the injective mathematical function from the coordinates of the uniqueness parameters 8 in the pre-determined frame 11, have a correspondence with the coordinates of the micro-patterns 9. Therefore, the first personalized security device 1 is recognized as original. In the case of the second security device l' (b), verification coordinates of micro-patterns 12', generated by the injective mathematical function from the uniqueness parameters 8' of the identity photograph 2" do not correspond with the coordinates of the micro-patterns 9. Therefore, the security device 1' is not recognized as original. The non-match can be established when the coordinates of the identified micro-patterns 9 and the verification coordinates of the verification micro-patterns 12 differ by a distance at least equal to the tolerance zone. Figure 4 schematically illustrates an example of the implementation of a method for securing a safety device 10 according to the invention. According to this example, the process involves: - a step S1 of acquiring an image of the security device 10 as illustrated in [Fig.2]: - a step S2 of defining a geometric object space E; - a step S3 of identification of at least one uniqueness parameter 8 of the safety device 10 in the geometric object space E; - a step S4 of defining a predetermined frame 11; - a step S5 of determining the coordinates of at least one uniqueness parameter 8 in a predetermined frame 11 on the safety device 10; - a step S6 of generating coordinates of at least one micro-motif in a geometric image space E' of the security device 10 by an injective mathematical function, from the coordinates of at least one uniqueness parameter 8 in the predetermined frame 11; - a step S7 of marking at least one micro-pattern at a location defined by the coordinates generated by the mathematical function on the security device 10. A security device 1 such as represented in [Fig.1] is obtained for example as a result of such a securing process. Figure 5 schematically illustrates an example of the implementation of a method for authenticating a personalized security device 1 according to the invention, as shown in Figure 1, for example. According to this example, the method comprises: - an S8 step for acquiring an image of the personalized security device;1 - a step S9 of identification of at least one uniqueness parameter 8 in the object geometric space E of the personalized security device 1; - a step S10 of determining the coordinates of at least one uniqueness parameter 8 in the predetermined frame 11; - a step S11 of generating verification coordinates of at least one micro- verification motif 12 in the predetermined frame | | by the injective mathematical function, from the coordinates of at least one identified uniqueness parameter; - an S12 identification step of at least one micro-pattern in the personalized security device 1; - a step S13 of determining the coordinates of at least one micro-motif identified in the personalized security device, in the pre-determined reference frame 11; - a step S14 of comparing the verification coordinates of at least one micro-pattern of verification 12 and the coordinates of at least one micro-pattern identified in the personalized security device 1; and - a step S15 of authenticating said personalized security device 1 in case of correspondence between the verification coordinates of at least one micro-pattern of verification 12 and the coordinates of at least one micro-pattern identified in the personalized security device 1. The match is established when the verification coordinates of at least a micro-verification pattern 12 belong to a predetermined tolerance zone, while non-matching is established when the verification coordinates of at least one micro-verification pattern 12 differ by a distance greater than the tolerance zone.

Claims

Demands

1. Method for personalizing a security device (10) comprising at least one uniqueness parameter (8) in an object geometric space (E) predetermined, said process comprising the steps of: - identification (S2) of at least one uniqueness parameter (8) of said safety device (10); - determination (S4) of a location of at least one parameter of uniqueness (8) in the object geometric space (E) of said device security (10); - generation (S6) of a location of at least one micro-pattern (9) in a geometric image space (E') of the safety device (10) by a injective mathematical function, from said location of the other minus a uniqueness parameter (8); and - marking (S7) of at least one micro-pattern (9) said location generated by the injective mathematical function in geo- space metric image of said security device (10).

2. The method according to claim |, characterized in that it comprises in in addition to a parameterization step of the mathematical function in which at least one coefficient of said mathematical function is determined from at least one characteristic data point of the device security (10).

3. Method according to claim 1 or 2, characterized in that it comprises in addition, a parameterization step of the mathematical function in which at least one coefficient of said mathematical function is determined from a cryptographic key.

4. A method according to any one of claims 1 to 3, characterized in which also includes a step of defining a pre-reference point determined (11), in that the location of at least one parameter uniqueness in the object geometric space (E) is defined by co- ordered in said predetermined frame (11) and in that the location of at least one micro-motif (9) in the geo- space image metric (E”) is defined by coordinates in said frame predetermined (11).

5. Method according to claim 4, characterized in that said marker pre- determined (11) is linked to the safety device (10) and defined from at least one characteristic data point of said security device (10).

6. A customized security device comprising at least one parameter of uniqueness (8) in a predetermined object geometric space (E) and at less a micro-motif (9) in an image geometric space (E”), and at at least one location of at least one uniqueness parameter (8) and at at least one location of at least one micro-motif (9) being connected by an injective mathematical function.

7. A customized safety device according to claim 6, characterized in that at least one micro-motif (9) comprises at least one point engraved and / or printed in said personalized security device (1).

8. Personalized security device according to any one of the claims indications 6 or 7, characterized in that at least one micro-motif (9) has a characteristic dimension of less than 200 µm, or even 50 µm.

9. Custom safety device according to any one of the claims indications 6 to 8, characterized in that the geometric image space (E') is configured to be distinct from at least one predetermined area.

10. Identity document bearing a photograph (2) and / or alphanumeric data (3) relating to a holder of said identity document and a personalised security device (1) according to any one of claims 6 to 9, said identity photograph (2) comprising at least one uniqueness parameter (8).

11. Method for authenticating a personalized security device according to any one of claims 6 to 9, said process authentication including at least the following steps: - identification (S9) of at least one uniqueness parameter (8) of said personalized security device (1); - determination (S10) of the location of at least one parameter uniqueness (8) identified; - generation (S11) of a check location of at least one micro-verification pattern (12) by the injective mathematical function, from the location of at least one uniqueness parameter (8) identified; - identification (S12) of at least one micro-motif (9) in the device of personalized security (1); - determination (S13) of a location of at least one micro-motif (9) identified; - comparison (S14) of the verification location of at least one micro-pattern of verification (12) with the location of the at least a micro-pattern (9) identified in said personalized security device (1); and - authentication (S15) of said personalized security device (1) in case of correspondence between the location of verification of at least one micro-pattern of verification (12), and the location of at least one micro-pattern (9) identified in said personalized security device (1).