A PRODUCT COMPRISING A MARKING PATTERN ON A SURFACE AREA FOR IDENTIFYING THE PRODUCT

Abstract

 A marking pattern over a product surface area is proposed to facilitate a traceability of the product. Preferably, the material shape (as may be prepared by embossing or engraving or etching, for instance with a laser) or the printing structure (based on ink or varnish deposit) of an area (120) of the product surface is modified by a combinatorial marking pattern as a selection of base patterns (131, 132, 133) with different geometrical arrangements relative to the product surface, such that the product surface marking does not visually disturb the product surface artwork or marketing design, while each geometrically arranged base pattern remains individually detectable by a marking pattern sensor independently from the other geometrically arranged base patterns out of the combined marking pattern on the product surface shape or structure.

Description

Field

[0001] This application relates to the field of product surface marking technologies for traceability and anti-counterfeiting purpose.

Background

[0002] In order to determine whether product items are genuine, various anti-counterfeiting technologies may be used (Anti counterfeiting technology guide, European Union Intellectual Property Office, 2021). In general, these technologies add an element onto the item that is difficult to duplicate, or copy, or they characterize a specific physical or chemical feature of the item, similar to a fingerprint of the item. The challenge may be either technical, for instance on the reproduction of holograms, or require products which are not readily available on the market, such as rare isotopes or special inks. In general, anti-counterfeiting features may be classified as overt technologies (visible, or more generally perceptible by the end user with his own body senses, without the need for specific detection equipment, such as for instance holograms on banknotes) or covert technologies (invisible/imperceptible, but detectable with a dedicated equipment). In general, covert technologies are preferred for branded products because they do not interfere with the product surface artistic and/or marketing designs. Example of covert technologies include:

• Product markings: Technologies such as digital watermarks have been designed to better prevent the counterfeiting of product packages and security documents by electronic and digital means. As a widely deployed example of such a technology, the AlpVision Cryptoglyph exists in two flavors, either as a random or pseudo-random pattern of microdots printed with visible ink (WO0225599, WO04028140), or as a distributed cloud of micro-holes in the varnish layer (WO06087351). The distribution of microdots or micro-holes can be controlled with a secret cryptographic key. Authentication can be performed using conventional imaging devices, such as smartphones or off-the-shelf office scanners, in combination with dedicated signal processing software Product markings comprise digital markings, chemical markings, hologram and the like.
• Surface fingerprints: These technologies do not add any security element on the product, but rather use existing, intrinsic microscopic surface characteristics. For instance, a matte surface of a plastic injected product is an ideal candidate for the fingerprint solution. An image of the surface may be acquired during production and then compared to a later image capture from the product under inspection. Authentication can be performed using conventional imaging devices, such as smartphones or off-the-shelf office scanners, in combination with dedicated signal processing software. Examples of fingerprinting technologies are described for instance in US10332247.

[0003] The authentication of an object for any of the above technologies typically comprises a step of identifying, with a detector comprising a sensor adapted to the particular authentication technology employed for this object, whether the authentication technology can be retrieved from inspecting the object. Preferably, the detector employs a camera for imaging the object surface. In the past two decades, digital detection technologies have emerged which have facilitated the automation of this process and its generalization to non-specialized personnel, possibly also the general public, thanks to the use of digital signal processing algorithms embedded into software applications either embedded into the detector equipment (e.g., a smartphone) or executed on a computer in communication with the detector equipment (e.g., a scanner) through a communication network.

[0004] These digital authentication detection methods generally employ the following steps:

1) capture, with a sensor, a digital signal representation of the object to be authenticated (for instance, an image of an area on the object surface by a camera sensor);

2) process, with computer-implemented signal processing algorithms, the digital signal representation to characterize the object genuineness - for instance, by measuring a difference, a distance or a signal-to-noise ratio (SNR) between the captured digital signal representation and a template digital signal representation of a reference genuine object; or by extracting mathematical features from the digital signal representation which can be used by a classifier to discriminate between a fake and a genuine object. - for instance, feature points or feature vectors;

3) classify (with computer-implemented authentication classification or decision-making algorithms) the object as genuine or fake depending on the measurement (for instance, using a predetermined threshold or using a machine learning classifier). In general, these prior art product markings are not individualized. They often use a single anti-counterfeiting pattern marking for all the manufactured products of a given product series. This does not enable to individually trace a single manufactured product along its whole lifecycle, from factory to destruction or recycling. In particular, it is not possible with these prior art covert marking technologies to identify grey market illegal product distribution circuits.

[0005] In theory, it would be possible to produce an individual marking pattern for the product surface that enables to uniquely mark each product. However, for large series of manufactured products, this approach is not realistic, because of the cost of producing and detecting so many different product surface patterns.

[0006] There is therefore a need for improved, cost-effective, scalable product markings which can be individualized at the product manufacturing stage, which can be retrieved with conventional product marking optical detectors all along the product lifecycle, to facilitate the individual product traceability.

[0007] It is an object of the invention to provide product surface markings to facilitate the individual identification of a product without visually disturbing the product surface design, artwork, matte or shiny finish. It is a further object of the invention to provide product surface markings to facilitate the individual identification of a product out of a large number of a manufactured product series.

Summary

[0008] The present application is based on the finding that combinations of marking base patterns over a product surface can facilitate a traceability of the product. Preferably, the material shape (as may be prepared by embossing or engraving, for instance with a laser) or the printing structure (based on ink or varnish deposit) of an area of the product surface is enhanced by a plurality of marking patterns using a pre-defined geometrical arrangement relative to the product surface, such that the marked product surface shape does not visually disturb the product surface artwork or marketing design, while each of the combined marking patterns remain individually detectable by a marking pattern sensor independently from the other marking patterns out of the combined marking patterns on the product surface shape.

[0009] It is proposed a product comprising a surface area for identifying the product,

• wherein the surface area exhibits a marking pattern made of one or more geometrical arrangements of at least one base pattern of structural features,
• wherein each base pattern of structural features is arranged as a random or pseudo-random distribution of structural features having surface diffuse reflection, color and/or specular properties which are non-visually disturbing to the naked eye when looking at the product surface area,
• and wherein the geometrical arrangements are defined along an x-axis, a y-axis, a z-axis, or a combination thereof, relative to the surface area, the random or pseudo-random distribution of structural features of each geometrically arranged base pattern over the surface being detectable independently from the random or pseudo-random distribution of structural features of the other geometrically arranged base patterns by sensing a surface area with an optical detector.

Description of the figures

[0010] The present application will be better understood thanks to the attached figures in which :

• Figure 1) is an abstract representation of a product surface shape modified with an overlay of marking patterns (z-axis arrangement).
• Figure 2) is an abstract representation of a product surface shape modified with a juxtaposition of marking patterns (x-axis and y-axis arrangement)
• Figure 3) is an abstract representation of a product surface shape modified with an overlay of translated marking patterns (combination of x-axis/y-axis and z-axis arrangement).
• Figure 4a) and 4b) show an abstract representation of a possible translated marking pattern adapted not to lose the structural elements from the initial pattern which have been moved beyond the image pattern borders after translation.
• Figure 5) shows a) a reference pattern of pseudo-random, low density distribution of dot structural features; b) the resulting overlay pattern obtained by translating the reference pattern of Figure 5a) with 10 different translation vectors and overlaying the resulting 10 translated marking patterns; and c) the 10 cross-correlation peaks of the reference pattern of Figure 5a) compared with its 10 translated overlayed patterns of Figure 5b).
• Figure 6) shows a digital pattern of a product combinatorial marking in accordance with a proposed embodiment.

Detailed description

[0011] The terms "product" or "item" or "manufactured item" (used interchangeably) refer to a manufactured or an artisanal product. Examples of products include, but are not limited to a luxury product, a pharmaceutical product, a cosmetics product, a food product, a tobacco product, a consumer electronics product, a sports product, a spare part product (e.g. automotive), a biomedical product, a coin, a bullion, a jewel, an artistic product, a banknote, a security document, a collectible card product, a precious metal, a watch, a leather product such as a bag. A product may also be a part of a manufactured object which is attached to or associated with this object, such as a component, a tag, a label, or a package. Preferably, a product comprises at least one surface characterized by its shape and/or its structure which may be modified with a pattern by using various manufacturing or printing processes known in the art. Examples of modifications comprises changes of the diffuse properties, surface color and/or specular properties. Examples of product surfaces include, but are not limited to, an engraved surface, an embossed surface, a debossed surface, a metallized surface, a varnished surface, or a printed surface. The product surface may comprise a metallic area, a plastic area, a printed area, a painted area, or a natural material area such as stone, precious metal, vegetal fiber, wood, or leather. The product may be part of a batch of similarly manufactured products. The "marking number" is the number serving to parametrize the generation of the "marking pattern". The marking number can be a serial number (SN) of the product (i.e. an individual or unique marking number), a number identifying a batch of products, or a signature or a copy of other data displayed on the product such the serial number, the batch number, the expiration date, a QR code or a Barcode. The shape and/or structure of the product surface may be modified with the marking pattern, encoding a marking number which can be used to match a product number or a product identifier.

[0012] The term "marking" refers to the result of a physical modification of a surface condition. Marking may be applied at the end of a manufacturing process to change the visual appearance or certain optical properties of a surface area relative to its surrounding areas, for instance a surface area on a product, or a surface area on a product part in association with a product. Examples of physical modifications comprise modifications of the surface shape or structure by processes such as for instance engraving, etching, embossing, debossing, stamping, printing, varnishing, and the like. Examples of visual appearance or optical properties for covert markings comprise surface diffuse reflection, color and/or specular properties which are non-visually disturbing to the naked eye when looking at the product surface design. The "marking" can be part of the manufacture process, e.g. the mold or the die from which the product is produced can comprise the marking pattern which is transferred to the surface area of the product.

[0013] The term "structural feature" refers to an elementary component inherent to the marking process. A structural feature physically modifies the shape or structure of a surface area in the marking process.

[0014] In some embodiments, a structural feature may correspond to a dot and a structural feature value may correspond to a colour code (e.g. for printing), a depth or thickness of the dot (e.g. for engraving, embossing, debossing, or varnishing), or a size of the dot. Other embodiments are also possible, for instance in embossing or debossing markings a structural feature may be an embossing or debossing elementary shape of a predefined geometry. In general, structural features are chosen so that their marking onto the surface causes changes to the surface diffuse reflection, color and/or specular properties which are visually non-disturbing to the end user, but can be detected with an optical detector.

[0015] The term "pattern" refers to a one-dimensional (1D), a two-dimensional (2D) or a three-dimensional representation (3D) of the distribution of structural features along a line, within an area, or throughout a volume in the process of marking a product surface. A "base pattern" refers to a pre-defined distribution of structural features as may be represented in a computer-implemented form as a vector (1D), a matrix (2D) or a tensor (3D) of structural feature values. A "marking pattern" or a "combinatorial marking pattern" refers to a product surface marking which is made of a combination of one or more geometrical arrangements of one or more base patterns. A "digital pattern" may be produced as a data representation of a vector, a matrix or a tensor by a computer-implemented method such as a data encoding algorithm. Examples of 2D patterns suitable for use in a dot marking process comprise "image bitmaps", wherein each pixel represents a dot and the bitmap is used to guide the pattern marking process.

[0016] The term "density" refers to the ratio of the number of actual structural features relative to the number of possible structural features within a pattern (e.g. ratio of active dots over the number of pixel positions in a bitmap).

[0017] Figure 1 shows an abstract representation of a first possible embodiment in which the shape or the structure of a surface area 120 of a product 100 is modified by using an overlay of a plurality of base patterns 131, 132, 133 selected in a database 110 of P pre-defined patterns. In a preferred embodiment, a database 110 of pre-defined base patterns is established wherein each base pattern is indexed as 0, 1, 2, 3..., P-1. The total number P of pre-defined base patterns in the database 110 is chosen such that the combination of a subset of k base patterns in the set of P pre-defined base patterns enables to identify, from the indices of the k base patterns in the set, a unique number in the range from 1 to N, where N is the number of products to be individually tracked out of the batch of products. As will be apparent to those skilled in the art of combinatorics, this corresponds to the binomial coefficient. The P and k values should be chosen such that their binomial coefficient

is larger than or equal to the number N of products to be identified. Each individual product may then be identified by a marking number derived from a subset of k indices, for instance using a lookup table, a combinatorial numbering system such as combinadics, enumerative combinatorics, or other indexing encoding methods.

[0018] Preferably, the P pre-defined base patterns are prepared with a low density of a random or pseudo-random distribution of structural elements suitable for marking the product surface without being visually disturbing when the product user is looking at the product surface design. In particular, the density of structural elements shall be low enough to avoid significant overlap between any k separate base patterns in the set of P pre-defined base patterns. This preserves the ability to individually discriminate, at retrieval stage, k base patterns out of a combined marking pattern formed by overlaying the k patterns and used as the marking pattern at manufacturing stage. A statistical method (such as for instance methods in relation with the Cryptoglyph technology from AlpVision, but other embodiments are also possible), an empirical method (such as using a trial and error test with different patterns) or a combination thereof may be used to this end in a preparatory stage to populate the database 110 with pre-defined based patterns.

[0019] In a possible embodiment, base patterns are chosen such as they are orthogonal to each other, to facilitate their retrieval in the detection process. In another possible embodiment, base patterns are chosen such as they are partially orthogonal. Other embodiments are also possible with non-orthogonal patterns, for instance by combining them with further geometrical transforms in the marking process to facilitate their separation at retrieval time. It is also possible to design base patterns such that the structural feature satisfies specific layout conditions, for instance minimum distance between two structural features.

[0020] When the product 100 is produced with an individual marking number, it is possible to select, with a computer-implemented method, a subset of k base patterns 131, 132, 133 out of the database 110 such that their individual indices enable to retrieve the product individual marking number.

[0021] In the first proposed embodiment of Figure 1, the k base patterns 131, 132, 133 corresponding to indices 2,4 and 9 in the database 110 can be superposed to produce a combinatorial marking pattern on the surface area 120. The superposition of the k base patterns 131, 132, 133 into the combinatorial marking pattern virtually corresponds to a simple "stacking" geometrical arrangement along a z-axis perpendicular to the surface area. At manufacturing stage, depending on the available surface shape or structure modification technology for marking the product 100 along the production chain, in a possible embodiment the subset of k base patterns 131, 132, 133 may be overlaid into a digital pattern (not represented) by a computer-implemented method, and the resulting digital pattern is then used to guide the surface shape or structure modification at manufacturing stage (virtual stacking). In another possible embodiment physical stacking , the subset of k base patterns 131, 132, 133 may be directly used to serially overlay the k patterns modifications each from a pattern-dedicated physical marking method, for instance by selecting k base patterns 131, 132, 133 out of P possible patterns in a sequential marking process (e.g. rolling stamp or multiple die stamps) (physical stacking). Both the virtual stacking and the physical stacking overlay marking embodiments enable to produce a combinatorial marking pattern on the product surface area 120 which is a mixture of k base patterns 131, 132, 133.

[0022] At verification stage, an optical detection device 150 is used to capture one or more images from the combinatorial marking pattern over the product surface area 120, and a computer-implemented method is used to individually extract the k base patterns from the one or more images. The computer-implemented method can then retrieve the index of each of the k base patterns in the database (indices 2, 4 and 9 in the example of Figure 1) and decode accordingly the individual marking number of the product based on the retrieved indices.

[0023] Figure 2 shows an abstract representation of a second possible embodiment in which the shape or the structure of a surface area 120 of a product 100 is modified by using a marking pattern made of a plurality of geometrically arranged base patterns, for instance an array of juxtaposed base patterns over the product surface area 120. In a preferred embodiment, a database 110 of pre-defined base patterns is established wherein each base pattern is indexed as 0, 1, 2, 3..., P-1. The total number P of pre-defined base patterns in the database 110 is chosen such that the ordered arrangements of a subset of k base patterns in the set of P base patterns enables to identify, from the indices of the k base patterns in the set, a unique number i in the range from 1 to N, where N is the number of products to be individually tracked out of the batch of products. As will be apparent to those skilled in the art of combinatorics, this corresponds to the number of permutations of k in P. The P and k values should be chosen such that their number of ordered arrangements

is at least equal to the number N of products to be identified. Each product may then be identified by a marking number derived from a subset of k indices and their respective orders, for instance using a lookup table, a combinatorial numbering system such as combinadics, enumerative combinatorics, or other indexing encoding methods.

[0024] When the product 100 is produced with an individual marking number, it is possible to select, with a computer-implemented method, with the marking number as parameter, a subset of k base patterns out of the database 110 and to juxtapose them to produce an overlaid marking pattern on the surface area 120 so that the indices of the k base patterns enable to retrieve the individual marking number. Depending on the manufacturer surface shape or structure modification technology for the product 100, in a possible embodiment the subset of k base patterns 131, 132, 133 may be virtually juxtaposed into a digital pattern (not represented) by a computer-implemented method, and the resulting array pattern is then used to guide the modification on the product surface area 120 at manufacturing stage. In another possible embodiment, the subset of k base patterns may be directly used to physically juxtapose, one by one, the k base patterns modifications each from a pattern-dedicated physical marking method, for instance by selecting k patterns out of N possible patterns in a serial marking process and positioning them at a different position over the surface area 120 with an automated positioning mechanism (e.g robot arm, x-y table, conveyer belt, etc). Both the virtual and physical juxtaposition embodiments enable to produce a combinatorial marking pattern on the product surface area 120 which is a mixture of k base patterns 131, 132, 133.

[0025] The order of the patterns into the array may for instance be defined from bottom to top, from left to right with reference to a horizontal x-axis and a vertical y-axis relative to the position of the surface area 120. In the example of Figure 2, the pattern of index 9 is embedded at array position (x=0, y=0), the pattern of index 0 at array position (x=1, y=0), the pattern of index 4 at array position (x=0, y=1) and the pattern of array 2 at array position (x=1, y=1). At verification stage, an optical detection device 150 is used to capture one or more images from the product surface pattern 120 and a computer-implemented method is used to individually extract the k patterns as found onto the product surface area 120 according to the array order. The computer-implemented method can then retrieve the ordered indices of the patterns in the database (9, 0, 4, 2 in the example of Figure 2) and decode accordingly the individual marking number of the product based on the retrieved ordered indices. Other embodiments for ordering are also possible.

[0026] Figure 3 shows an abstract representation of another possible embodiment for producing a combinatorial marking pattern in which the shape or the structure of a surface area 120 of a product 100 is modified by using a selection of geometrical arrangements of a single predefined base pattern. Indeed, it was observed that multiple possibilities of geometrical arrangements may also provide enough indices to uniquely encode an individual product marking number while keeping the geometrical arrangements of the base pattern independently retrievable from each other at detection stage. Geometrical arrangements may comprise geometrical transforms such as rotations, translations, scaling, warping, mirroring and/or a combination thereof. Geometrical arrangements may also be specialized on sophisticated base patterns serving purposes such as increasing encoding space, introducing resilience to geometrical transformations, or allowing for the identification of rotations, translations, and scale. Such sophistication may consist for instance in adding shifted, rotated, or even mirrored versions of the base pattern. It is also possible to generate a base pattern with inherent geometrical properties to facilitate its retrieval with an optical sensor operating under different angles relative to the product surface area, for instance symmetries, such as rotational and/or central symmetries.

[0027] In the example of Figure 3), the base pattern is translated 3 times to produce 3 different geometrical arrangements of the base patterns, indexed by the translation coordinates ((0,0),{3,1},{0,1}) along a x-axis and y-axis reference relative to the surface area 120. At detection time, the marking number of the product can be derived from the indices of the actual retrieved geometrical arrangements coordinates as a subset selected out of a set of multiple possible geometrical arrangements of a base pattern.

[0028] In the example of Figure 4a), a geometrical arrangement of index {2,5} is illustrated, corresponding to a translation of D*2 pixels (where D is a parameter defining the minimum number of pixels between 2 translated base patterns in the marking pattern bitmap) along a horizontal axis (x-axis) and a translation of D*5 pixels along a vertical axis (y-axis). In this illustration, possible translations can be of 0, 1, 2, 3, 4, 5 or 6 steps of D pixels along each axis, which enables to select j possible geometrical placements into a set of Q=7^∗7=49 possible transformed marking patterns. The total number P of pre-defined base patterns in the database 110 is chosen such that the combined encoding of the indices of a subset of k base patterns in the set of P base patterns and of the j sets of coordinates selected out of Q possible geometrical transforms enables to identify a marking number in the range from 1 to N, where N is the number of products to be individually tracked out of the batch of products.

[0029] In the embodiment illustrated on Figure 3, the number of base patterns in the database is only 1 and the indexing encoding solely depends on the numbers of its actual and possible geometrical transforms, so as to accelerate the matching process at retrieval time. In this particular case where P=1, with j sets of geometrical placements out of Q, the number of combinations is the coefficient

.

[0030] More generally, the exemplary embodiments of Figure 1, Figure 2 and Figure 3 may be combined to produce a combinatorial marking pattern made of one or more geometrical arrangements of at least one base pattern of structural features. This may be particularly advantageous in the case of a very large number of products. Indeed, increasing the number P of base pattern references as well as the number k of patterns to embed into a combinatorial marking pattern directly impacts the number of pattern matching detection operations and thus the computational efficiency of the detection stage . On the other hand, increasing the number of possible geometrical arrangements Q of a single base pattern is inherently limited by the size available for the combinatorial marking on the surface area 120 and the constraint to keep them individually separable at detection stage.

[0031] The individual product marking number combinatorial encoding method may thus jointly employ a first selection of k values among P possibilities for the pre-defined reference base patterns and a second set of j values among Q possibilities for their possible geometrical arrangements, for instance using a lookup table, a combinatorial numbering system such as combinadics, enumerative combinatorics, or other indexing encoding methods. As will be apparent to those skilled in the art of digital encoding, this may even comprise the addition of parity checks or error correction codes to increase the robustness of the encoding method to the noise induced by the physical marking process.

[0032] In the case where the same base pattern can be selected multiple times and Q = j (all the Q geometrical transforms are used), this corresponds to position encoding where each geometrical transformation identifies a unique position, similar to digits for numbers. In this case we have for instance, for P=10 and Q=j=49 a decimal encoding system which can encode N=10⁴⁹ product and a binary encoding with P=2 and N=2⁴⁹.

[0033] Each individual product may then be identified from the combined encoding of the indices of a subset of k base patterns in the set of P patterns and of the geometrical transform coordinates of j geometrical arrangements of each base pattern in the set of Q possible geometrical arrangements of a base pattern.

[0034] As can be seen in Figure 4a), the geometrical transform of a base pattern may move some of its structural features beyond the borders of the product surface area 120 available space. To overcome this limitation without losing the information of those structural features in the random or pseudo-random pattern retrieval, the geometrical arrangement may employ pattern folding or wrap-around of the outside areas back into the regular base pattern area, for instance using the reconstruction scheme as illustrated on Figure 4b). As will be apparent to those skilled in the art of image processing, this folded marking of products is particularly well suited when the geometrical transform is a translation along the x-axis and/or y-axis relative to the product surface are 120, and when the retrieval process uses cross-correlation to match to a base pattern reference, as the extraction of the cross-correlation calculated peaks enable to retrieve the number of translation steps along the x-axis as well as the y-axis (and thus the translation coordinates of the corresponding geometrical arrangements of the base pattern). In the example of Figure 4b), the dashed square represents a translated base pattern and the parts exiting the base window represented by the black line are wrapped around. This wrap around has no impact on the cross-correlation signal of the translated base pattern with the original base pattern as recorded into the database 110 if the cross-correlation is computed using the Fourier transform.

[0035] Figure 5 shows 5a) a base pattern; 5b) the combinatorial marking pattern formed by overlaying 10 translated and wrapped around versions of the base pattern using the following translations (10 out of 49 possibilities), with following 10 vectors noted {dx,dy} for each translation { {0,0}, {0,1}, {0,4}, {3,1}, {3,6}, {4,2}, {5,0}, {5,1}, {5,6}, {6,6} }; and 5c) the cross-correlation of the base pattern and of the combined pattern showing 10 peaks corresponding to each translation vector.

[0036] It will be apparent to those skilled in the art of covert technologies that the resulting modification can remain visually non-disturbing for the end-user of the product when looking at the product design, as the number k of base patterns to be combined in the surface marking pattern can be chosen low enough to maintain the property of low-density, random or pseudo-random structural features distribution in the combined marking pattern without significant difference from the properties of each of the k individual patterns. Furthermore, the structural features for pattern marking may be chosen such that they are inherently imperceptible to the human eye for instance through changes in size and/or contrast. This is significant compared to a naive method of product serial number tracking. To individually track 1 million of manufactured products, it is possible to produce and retrieve individual markings at a much lower cost than 1 million individual markings and retrievals, for example:

• Example 1: Product surface overlays of 4 patterns out of a database of 72 patterns; the retrieval process only requires 72 matching comparisons;
• Example 2: Product surface overlays of 3 patterns out of a database of 183 patterns; the retrieval process only requires 183 matching comparisons;
• Example 3: Product surface juxtaposition of 4 patterns out of a database of 34 patterns; the retrieval process requires only 34 matching comparisons.

[0037] For larger sets of objects:

• Example 4: Product surface overlays of 10 patterns out of a database of 40 patterns; the retrieval process only requires 40 matching comparisons; this combination enables to index up to 847 million products.
• Example 5: Product surface overlay of the same pattern with 10 different translations out of 49 possible translations; the retrieval process requires only 1 matching comparison, corresponding to over 8 billion combinations.

[0038] Figure 6) shows the digital bitmap corresponding to a combinatorial marking pattern of low density adapted from a set of geometrically arranged Cryptoglyph base patterns. This combinatorial marking pattern is suitable for printing yellow dots or varnish dots on a 4cm*4cm surface area at serialization stage, at the end of a product package manufacturing process.

[0039] The proposed method produces, with one of more processors of a computer-implemented digital marking pattern processing system, a marking pattern from a marking number through the steps of:

• Acquiring predefined parameters comprising the total number of base patterns in a database 110 of base patterns, the number of possible geometrical arrangements of a base pattern, and/or the maximum number of combinable base patterns to produce a combinatorial marking pattern made of a subset of one or more geometrical arrangements of at least one base pattern selected in a subset of base patterns;
• Calculating the indices of the subset of base patterns and/or the indices of the subset of geometrical arrangements of each base pattern as a bijective function of the product marking number;
• Acquiring the base patterns identified by the calculated indices of the subset of base patterns;
• Composing a combinatorial marking pattern by arranging the subset of base patterns at the geometrical positions identified by the calculated indices of the geometrical arrangements for each base pattern;
• Marking a product surface with the combinatorial marking pattern.

[0040] Preferably, the geometrical arrangements are defined and indexed along an x-axis, a y-axis, a z-axis, or a combination thereof, relative to a product surface area suitable for the product marking process, such that each geometrical arrangement of a base pattern over the product surface remains detectable independently from the other geometrical arrangements of base patterns in the combinatorial marking pattern.

[0041] Exemplary embodiments:

• In the embodiment of Figure 1, only one geometrical arrangement is predefined as an overlay of base patterns over the z-axis, and the base patterns are chosen in a database of base patterns which remain detectable from each other once overlaid into a combinatorial marking pattern (e.g. orthogonal patterns).
• In the embodiment of Figure 2, 4 geometrical arrangements are predefined as 4 possible positions in an array of 2*2 base patterns (x-axis coordinate and y-axis coordinate as geometrical arrangement indices) and the arrangement order (not just the combination) of a maximum number of 4 selected base pattern indices may be further used to encode the product marking number.
• In the embodiment of Figure 3, only one reference base pattern is used (subset of 1 base pattern in a database of 1 single base pattern) and 3 geometrical arrangements may be chosen out of 16 possible translations (4 possible indices along x-axis * 4 possible indices along the 4 y-axis) to produce the combinatorial marking pattern.

[0042] The proposed method retrieves from the combinatorial marking image, with one of more processors of a computer-implemented digital marking pattern detection system, the product marking number through the steps of:

• Matching the combinatorial marking image with one or more base patterns registered in the base pattern database 110 to identify the indices of the subset of base patterns which have been embedded into the combinatorial marking and the indices of the geometrical arrangements of each base pattern; (e.g. translation coordinates);
• Calculating, from the retrieved base pattern indices and/or the geometrical arrangement indices, the product marking number.

Claims

1. A product comprising a marking pattern on a surface area for identifying the product,

- wherein the surface area exhibits the marking pattern made of one or more geometrical arrangements of at least one base pattern of structural features,

- wherein each base pattern of structural features is arranged as a random or pseudo-random distribution of structural features having surface diffuse reflection, color and/or specular properties which are non-visually disturbing to the naked eye when looking at the product surface area,

- and wherein the geometrical arrangements are defined along an x-axis, a y-axis, a z-axis, or a combination thereof, relative to the surface area, the random or pseudo-random distribution of structural features of each geometrically arranged base pattern over the surface being detectable independently from the random or pseudo-random distribution of structural features of the other geometrically arranged base patterns by sensing a surface area with an optical detector.

2. The product of claim 1, wherein each base pattern is selected from a database of pre-defined base patterns.

3. The product of claims 1 or 2, wherein geometrical arrangements are translations to produce an array of side-by-side base patterns juxtaposed to form the marking pattern along the x-axis and/or along the y-axis.

4. The product of claims 1, 2 or 3, wherein geometrical arrangements are produced by overlaying base patterns along the z-axis perpendicular to the surface area.

5. The product of any of the preceding claims, wherein the product is identifiable by extracting from the applied marking pattern the coordinates of the geometrical arrangements and/or the references to the corresponding base patterns.

6. The product of any of the preceding claims, wherein the marking pattern is applied on the surface area by embossing, debossing, die-stamping, water jetting, engraving, or by depositing ink dots or varnish droplets.

7. The product of any of the preceding claims, wherein the product is a manufactured product, a package of a manufactured product or a label of manufactured product.

8. The product of claim 7, wherein the product is a luxury product, a pharmaceutical product, a cosmetics product, a food product, a tobacco product, a consumer electronics product, a sports product, a spare part product (e.g. automotive), a biomedical product, a coin, a bullion, a jewel, an artistic product, a banknote, a security document, a collectible card product, a precious metal, a watch, a leather product such as a bag.

9. The product of any of the preceding claims, wherein the product surface comprises a metallic area, a plastic area, a printed area, a painted area, or a natural material area such as stone, precious metal, vegetal fiber, wood, or leather.

10. A method of marking a product with a marking number, comprising:

- Acquiring predefined parameters comprising the total number of base patterns in a database 110 of base patterns, the number of possible geometrical arrangements of a base pattern, and/or the maximum number of combinable base patterns to produce a combinatorial marking pattern made of a subset of one or more geometrical arrangements of at least one base pattern selected in a subset of base patterns;

- Calculating the indices of the subset of base patterns and/or the indices of the subset of geometrical arrangements of each base pattern as a bijective function of the marking number;

- Acquiring the base patterns identified by the calculated indices of the subset of base patterns;

- Composing a combinatorial marking pattern by arranging the subset of base patterns at the geometrical positions identified by the calculated indices of the geometrical arrangements for each base pattern;

- Marking a product surface with the combinatorial marking pattern.

11. The method of claim 10, wherein the number of geometrical arrangements is one and wherein the base patterns from the subset of base patterns are overlaid to compose a combinatorial marking.

12. The method of claim 10, wherein the number of base patterns is one and wherein the geometrical arrangements are translations of the base pattern.

13. The method of claim 10, wherein a subset of base patterns is organized into an array to compose a combinatorial marking and wherein the geometrical arrangement indices comprise the x-axis and/or the y-axis coordinates of each base pattern in the array.

14. A method of detecting a marking number on surface area of a product, comprising:

- Capturing an image of the product surface area of the product of any of the claims 1 to 9;

- Matching the captured image with one or more base patterns registered in a base pattern reference database to identify the indices of the subset of base patterns which have been embedded into the combinatorial marking and the indices of the geometrical arrangements of each base pattern;

- Calculating, from the retrieved base pattern indices and/or the geometrical arrangement indices, the marking number.

15. The method of claim 14, characterized in that matching comprises a cross-correlation signal processing step and identifying the geometrical arrangement indices comprises identifying translation coordinates from the cross-correlation signal peaks.

Drawing