FR3084651A1 - METHOD FOR SECURING A BOTTLE AND CHECKING THE BOTTLE. - Google Patents

Abstract

The invention relates to a device for securing a bottle (100) comprising: • a container (10), • a closing device (20) closing said container (10), and • at least one first marking element, integral with the container (10); • at least one second marking element, integral with the closure device; the method comprising: • an enrollment step consisting in producing a source image of the bottle (100), characterized in that it further comprises • a processing step consisting in: ○ defining on the source image a first set remarkable points, and selecting at least a first remarkable point (31), corresponding to a first marking element; ○ defining a second set of remarkable points on the source image, and selecting at least one second remarkable point (32), corresponding to a second marking element; and ○ define the position of each selected remarkable point (32).

Description

Illllllllllllllllllllllllllllllllllllllllllll

METHOD FOR SECURING A BOTTLE AND CHECKING THE BOTTLE.

FIELD OF THE INVENTION

The present invention relates to the field of securing the distribution of products placed in containers.

In this field, the present invention aims to protect at least one of:

the container, ie the closable container (bottle, flask ...), the contents of the closable container;

in particular the improper re-use of the container or content.

It helps protect against undue re-use of the container or content.

The present invention applies to any closable container; whether in the luxury sector, for example in perfumery, in the field of wines and spirits, chemicals, pharmaceuticals, cosmetics, etc.

The content is for example in the form of liquid, gas, gel, cream, powder, granules, in powder form, etc.

Within the meaning of the present invention, by “bottle”, or indistinctly bottle, means:

• a closable container, • a device for closing said container, and • the content of said container, • the container being closed by said closure device.

For the sake of brevity, only a non-limiting example concerning wines and spirits will be described here.

Indeed, it is known that some fraudulent users produce 2 bottles from an authentic bottle, possibly legally purchased:

a first bottle whose container (bottle) is counterfeit but the content (liquid) is authentic, and a second bottle whose container (bottle) is authentic but the content (liquid) is counterfeit.

Some containers are also coated with a secondary packaging, for example in the form of a heat-shrinkable sleeve or in quasi-contact with the bottle which surrounds all or part of the container.

The present invention aims to reduce the risk of such improper use.

Different solutions exist on the market, for example by printing security labels (2D codes, holograms, etc.), affixing seals on the cap, implementing double identifiers, etc.

The present invention aims to improve existing solutions by avoiding the use of a security label.

SUMMARY OF THE INVENTION

More specifically, according to a first of its objects, the invention relates to a method for securing a bottle comprising:

• a container, • a device for closing said container closing said container, • a first set of at least one marking element, integral with the container;

• a second set of at least one marking element, integral with the closure device;

the process comprising:

• an enrollment step consisting in producing, by means of shooting comprising an optical sensor, a source image of the bottle.

It is essentially characterized in that it further comprises • a step of processing the source image consisting of:

defining on the source image a first set of remarkable points, and selecting at least a first remarkable point among the first set of remarkable points, each first remarkable point corresponding to a respective first marking element of the first set of marking elements ;

define on the source image a second set of remarkable points, and select at least one second remarkable point among the second set of remarkable points, each second remarkable point corresponding to a respective second marking element of the second set of marking elements ; and o define the position of each selected remarkable point.

The position of each selected remarkable point can be an absolute position or a relative position.

It can be provided that the first set of remarkable points and the second set of remarkable points are selected so that the two marking elements corresponding to said remarkable points can both enter simultaneously the field of observation of an optical sensor.

It can be expected that precise enrollment conditions are known (distance and position of the bottle relative to the sensor, lighting system, light intensity, etc.).

It can be provided that the two marking elements, corresponding respectively to the first set of remarkable points and to the second set of remarkable points, are chosen so that the opening of the bottle causes the destruction or the displacement of at least 1 'one of the remarkable points.

We can also provide a step consisting in calculating, and recording in the database, at least one of the metrics from:

• the distance between the first remarkable point and the second selected remarkable point, and • a function of the distance between the first remarkable point and the selected second remarkable point, in particular a hash function of said distance.

It is also possible to provide a step consisting in securing a security code to the bottle; the security code optionally comprising a mathematical function of the position of the first remarkable point and of the position of the second remarkable point selected.

We can also provide for an enrollment step, preferably implemented before the bottle is marketed, and consisting of:

• send, to a database located on an optionally secure server, optionally in an encrypted manner, at least one from the source image and a set of at least one of the metrics calculated for said source image, and • assign a identifier in the database for each source image or for each set of at least one of the metrics calculated for said source image, said identifier possibly being one of the calculated metrics or a serial number.

According to another of its objects, the invention relates to a method for verifying a secure bottle according to the invention, the method comprising steps consisting in:

• obtain a set of at least one test image of the bottle by an optical sensor, • process at least a subset of the test image with the same algorithm as that used for processing the source image, • define on the test image, a first set of remarkable points, and selecting at least a first remarkable point from the first set of remarkable points, each first remarkable point corresponding to a respective first marking element of the first set of marking elements;

• define on the test image a second set of remarkable points, and select at least one second remarkable point among the second set of remarkable points, each second remarkable point corresponding to a respective second marking element of the second set of marking elements ;

• define the position of each selected remarkable point; then • calculate at least one of the metrics from:

• the distance between the first remarkable point and the second remarkable point, and • a function of the distance between the first remarkable point and the second remarkable point, in particular a hash function of said distance; and • compare at least one of the metrics calculated on the test image with at least one of the metrics calculated on the source image, the method optionally further comprising an optical recognition step of the first set of at least one marking element (12) and the second set of at least one marking element (21), implemented by an artificial intelligence system, the method comprising in a learning step, supervised or unsupervised.

Thus, for the step of processing the test image, provision may be made to use only a variable subset of the remarkable points of the test image.

There can be a step of straightening the test image.

There may be a step of recording a video sequence of the bottle by the optical sensor; the test image being one of the frames of the video sequence.

There may be a step of recording a video sequence of the bottle by the optical sensor; the test image being an image reconstructed from several frames, so as to improve the quality of the test image.

It is possible to provide for a step of recording a video sequence of the bottle by the optical sensor from which a 3D test representation is calculated.

According to another of its objects, the invention also relates to a computer program comprising program code instructions for the execution of the steps of the verification method according to the invention, when said program is executed on a computer.

Other characteristics and advantages of the present invention will appear more clearly on reading the following description given by way of illustrative and nonlimiting example and made with reference to the appended figures.

DESCRIPTION OF THE DRAWINGS FIG. 1 illustrates an embodiment of a bottle of Champagne, FIG. 2 illustrates an embodiment of a bottle of wine, FIG. 3 illustrates an embodiment of a source image according to the invention of a label for marking the bottle in FIG. 1, and FIG. 4 illustrates an embodiment of a source image of an enlargement of the area A in dotted lines in FIG. 1.

DETAILED DESCRIPTION

By “bottle 100” we mean:

• a container 10, for example for a fluid, and in particular a liquid, • a device 20 for closing said container 10, and • the contents of said container 10, said container 10 being closed by said closing device 20.

The closure device 20 is for example in the form of a stopper, capsule, etc. It may include a wire cap, that is to say a wire frame of which some caps are capped, for example for bottles of champagne (Figure 1), sparkling wine, cider or beer.

The container 10, see also FIG. 2, comprises for example at least one of the elements from:

a neck, typically comprising:

o a ring, for example a ring, cord, full or square, and o a collar, for example straight or swollen, a shoulder, for example drooping, rounded or straight, a body, also called bole or belly, for example straight , conical, cylindrical, bulging, etc., a heel, a bottom, for example straight or pitted.

A marking element is an identification or decoration element, that is to say a graphic element observable by an optical sensor.

The bottle 100 comprises a first set of at least one marking element 12, integral with the container 10.

Typically, a first marking element can be obtained by machining, etching or molding of said container 10, by deposition on the container 10 or even by bonding to the container 10. For example, laser marking or jet marking can be provided. ink.

For example, it can be provided that a first marking element is understood by, or consists of, at least one of:

a marking label 11, stuck on at least one of the neck, the shoulder and the body of the container 10, a molding of the container 10 representing a predetermined pattern (not illustrated), a decoration, in particular deposited by screen printing on the body of the container 10 (not shown).

It is also expected that the bottle 100 further comprises a second set of at least one marking element 21, integral with the closure device 20.

For example, it can be provided that a second marking element is included with, or consists of, at least one of:

an optionally heat-shrinkable cap or sleeve, glued or deposited by overcapping, a wire cap affixed to a closure device 20, for example a capsule or a cap, of the closure device 20, an element of decoration of said closure device 20, a traceability or authentication label, preferably inviolable (self-destructive), preferably affixed astride said closure device 20 and the container 10, etc.

Preferably, a step of optical recognition of the first set of at least one marking element and of the second set of at least one marking element is provided.

In one embodiment, it is provided that said optical recognition step is implemented by an artificial intelligence system, the method comprising in a learning step, optionally supervised.

Enrollment

First of all, an enrollment step is provided which consists in particular of producing a digital image of the bottle 100, called “source image” and sending the source image to a secure database, located on a secure server, preferably of encrypted way.

More specifically, enrollment consists of:

produce a source image by an optical sensor, for example a digital camera, a camera, in particular placed on a production line, the optical sensor being possibly integrated, in particular with a communicating object, implementing a processing step of the 'source image to calculate a footprint, and save the footprint, and optionally the source image, in a database.

To have a complete reference of the bottle 100, it is possible to provide for producing several source images, by photography or video frames, possibly each source image corresponding to a respective relative position of the shooting means and of the bottle 100, and possibly of keep a 3D representation of the bottle 100.

Preferably, the source image is produced in situ, that is to say on the production site of the bottle 100.

The enrollment step is preferably implemented before the marketing of the bottle 100.

After the marketing of the bottle 100, it can be authenticated by a verification process described later.

First, a source image processing step is provided. The source image processing step can be implemented in situ, that is to say on computing means linked by computer to the shooting means, or implemented on the secure server.

A first step in the source image processing step is to define a set of remarkable points on the source image.

Remarkable points

A remarkable point is defined as a point of the source image, that is to say a pixel or a set of adjacent pixels two by two, for which the contrast gradient, according to a predefined direction and distance, is greater at a predefined threshold value.

A remarkable point therefore corresponds to a particular point of the bottle 100, the particular point having an optical characteristic on the source image. A marking element of the bottle 100 has a non-uniform optical characteristic which generates at least one remarkable point on the source image. A remarkable point on the source image therefore corresponds to all or part of a marking element of the bottle 100.

For example, for a container 10 whose mold includes a relief defining a pattern, a remarkable point can be defined on the image of said pattern of the container 10.

For a container 10 comprising at least one label 11, a remarkable point can be defined by virtue of an alphanumeric or kanji character of the label 11, the shape or edge of the label 11, or even an element of the decoration of label 11.

A remarkable point can also be purely graphic, such as for example a point of a watermark, a point of the decoration, of the motif or of a photograph of the label 11, etc.

If the label 11 includes a photograph of a person, a biometric recognition step can be provided to define a remarkable point.

More generally, remarkable points can be graphic elements located in an environment with specific physical or mathematical characteristics (in image processing), such as for example graphic elements around which an intense or zero gradient is formed, or which respond to image processing criteria such as Stephen-Harris detectors. By “intense” gradient is meant a gradient whose value is greater than a threshold value.

So we get:

a first set of remarkable points, each first remarkable point 31 corresponding to a respective first marking element of the first set of marking elements 12, integral with the container 10 (illustrated by dotted circles in FIG. 3); and a second set of remarkable points, each second remarkable point 32 corresponding to a respective second marking element of the second set of marking elements 21, integral with the closure device 20 (illustrated by dotted circles in FIG. 4).

It can be foreseen that the step consisting in defining a first set of remarkable points 31 and a second set of remarkable points 32 is implemented by learning, optionally supervised, the artificial intelligence system.

Selection

We then plan to select at least a first remarkable point 31 and a second remarkable point 32 from the set of defined remarkable points, such as:

the first remarkable point 31 is selected from the first set of remarkable points, and the second remarkable point 32 is selected from the second set of remarkable points.

We can predict that the set of first remarkable points is predetermined, and identical for a set of bottles.

For example, provision may be made to select a first remarkable point 31 determined by the very structure of the container 10, for example the seam, a relief of the container 10, the possible notch of the container 10, etc. Thus, while a visible ad hoc security device, for example a security label, could attract the attention of a counterfeiter, a structural element of the bottle 100 necessarily present is advantageously used here for the security of the bottle 100: nothing distinguishes a bottle 100 thus secured from an ordinary bottle 100.

It is also possible to provide for selecting a first remarkable point 31 by its position relative to a structural element of the container 10; for example the identification or decoration element closest to the seam, the closest to a predetermined relief of the container 10, the closest to the possible notch of the container 10, etc.

It is also possible to plan to select a first remarkable point 31 determined by the very structure of an identification or decoration element, for example a predetermined corner of a label 11.

Similarly, it can be provided that the second remarkable point 32 selected from the second set of remarkable points is predetermined, and identical for a set of bottles.

For example, provision may be made to select a second remarkable point 32 as a function of the very structure of the closure device 20, for example a pattern on a capsule, a wire cap, etc.

In one embodiment, it is provided that the closure device, in particular a sleeve, comprises a variable image diffractive device (not illustrated), better known by its acronym DOVID for "Diffractive Optical Variable Image Device" in English, in especially a hologram or a holographic track. In a first variant, the DOVID is positioned identically for all the bottles in a batch of bottles. In a second variant, the DOVID is variably positioned between the bottles of a batch of bottles. The second remarkable point 32 can be an optical element of DOVID. Alternatively or in combination, the second remarkable point 32 can be selected as a function of its position relative to a set of at least one predetermined optical element of the DOVID, or alternatively selected as a function of its relative position with respect to a predetermined element of the form of DOVID. Thus, whatever the variant, the second remarkable point 32 can be selected.

One can also provide for selecting a second remarkable point 32 determined by its relative position with respect to the first remarkable point 31; for example the remarkable point 31 of the closure device 20 closest to the top right corner of the label 11 affixed to the body or the neck of the container 10, etc.

It is also possible to provide for the second remarkable point 32 to have a variable or even random position relative to the position of the first remarkable point 31.

Thus, the position of the second remarkable point 32 can be defined:

either relatively; that is to say with respect to the position of the first remarkable point 31, said position of the first remarkable point 31 being variable from one container 10 to another;

either absolutely; that is to say relative to the position of a structural element of the container 10, said position of the first remarkable point 31 being invariable from one container 10 to another.

Preferably, the first remarkable point 31 and the second remarkable point 32 of the bottle 100 are selected so that the two marking elements corresponding to said remarkable points can both enter simultaneously the field of observation of an optical control sensor .

Advantageously, the two marking elements corresponding to said remarkable points are chosen so that the opening of the bottle 100 causes the destruction or the displacement of at least one of the two marking elements, therefore of their relative position.

We can predict that the step of selecting:

at least a first remarkable point 31 among the first set of remarkable points and at least a second remarkable point 31 among the second set of remarkable points, is implemented by learning, optionally supervised, of the artificial intelligence system.

The advantage of learning, optionally supervised, the artificial intelligence system is to reduce the processing time to go from one bottle to another or from a first type of bottle to a second type of bottle.

attributes

From the set of remarkable points selected, a step is planned which consists in calculating a set of attributes which contribute to the identifiable, even unique, character of the image.

The attributes include a set of metrics, that is, for example, a set of distances or angles between certain remarkable points.

The attributes include, for example, the coordinates of remarkable points (relative to a predetermined reference point), the distances between certain remarkable points, values of contrast gradient around the remarkable points, etc.

The position of a predetermined remarkable point 31 of the image can also be an attribute, for example a remarkable point 31 can be:

- an edge of the image,

- a benchmark calculated biometrically,

- a label edge 11,

- a decorative edge,

- etc.

We plan to calculate, and record in the secure database, at least one of the metrics from:

the distance, in particular Euclidean and / or cylindrical, between the first remarkable point 31 and the second remarkable point 32, and a function of the distance between the first remarkable point 31 and the second remarkable point 32, in particular a hash function of said distance.

For a photograph of a person or an animal, attributes can be for example usual elements of biometrics such as distance ratios between the positions of the eyes, the nose, the commissure of the lips or the center of the mouth. or muzzle, or angles between these same elements, etc.

For a photograph or a graphic representation, for example of a castle, attributes can be for example ratios of distance between predetermined elements, for example the positions of windows, towers, flags, etc.

The attributes can be calculated using known software or software libraries.

We can also use the algorithm called SIFT for “Scale Invariant Feature Transform” in English, or the algorithm called SURF for “Speeded Up Robust Features” in English, both of which are local descriptors which consist, first, detecting a certain number of remarkable points in the image, in order to then calculate a descriptor locally describing the image around each remarkable point. The quality of the descriptor is measured by its robustness to the possible changes that an image can undergo, for example a change of scale and a rotation.

For the SIFT algorithm, described in particular in the publication D. Lowe. Object recognition from local scale-invariant features. IEEE International Conference on Computer Vision, pages 1150-1157, 1999, the detection of points is based on the differences of the Gaussians (DoG) obtained by calculating the difference between each pair of images smoothed by a Gaussian filter, by varying to each time the sigma parameter (ie the standard deviation) of the filter. DoGs can be calculated for different scale levels allowing the introduction of the concept of scale space. The detection of potential areas of points of interest / remarkable points is carried out by searching for the extrema according to the plane of the dimension of the image (x, y) and the plane of the scale factor. Then a filtering step is necessary to remove the irrelevant points, by eliminating for example the points whose contrast is too weak.

The calculation of the SIFT descriptor is carried out on an area around each point of interest for example of 16x16 pixels, subdivided into 4x4 areas of 4x4 pixels. On each of the 16 zones, a histogram of the orientations of the gradient based on 8 intervals is then calculated. The concatenation of the 16 histograms gives a descriptor vector of 128 values.

For the SURF algorithm, described in particular in the publication H. Bay, T. Tuylelaars, and L. Van Gool. Surf: Speeded up robust features. European Conference on Computer Vision, pages 404-417, 2006, the method consists in using the determinant of the Hessian matrix, in calculating an approximation of the second derivatives of the Gaussians of the image by means of filters at different scales using masks of different sizes (e.g. 9x9, 15x15, 21x21, ...). For the calculation of the orientation of the points and the descriptors around the points, the principle is based on the sums of the responses of the horizontal and vertical Haar wavelets as well as their standards. The circular description area is further divided into 16 regions. A wavelet analysis is performed on each region in order to construct the final descriptor. The latter is made up of the sum of the gradients in x and y as well as the sum of their respective norms for all 16 regions. The descriptor vector is thus made up of 64 values which represent properties extracted both in normal space and in that of the scales of magnitude.

Preferably, there is provided a step consisting in classifying the attributes according to an order of priority of likelihood, which makes it possible for example to select only the most effective, in particular for detecting possible manipulation of the image.

We can plan to reject an attribute, that is to say not to select it, if it has a metric greater or less than a predetermined value, or a range of predetermined values. For example the distance between a label edge 11 and a predetermined inscription, between a graphic element and a text element, etc. This in particular limits the risk of false labels.

We can therefore provide a step consisting in selecting all or part of the calculated attributes.

Preferably, provision is made to select a number of attributes greater than a threshold value recorded in a memory. The more numerous and different the metrics (number of attributes) from each other, the lower the confusion rate.

The enrollment step also includes assigning an identifier to each source image saved in the database.

The bottle identifier 100 can be assigned when the source image is saved in the secure database. The identifier can be one of the calculated metrics. Preferably, however, each bottle 100 itself comprises a unique identifier, for example a serial number.

It is also possible to provide a step consisting in securing a security code, for example a 1D bar code or a 2D bar code, to the bottle 100, for example by marking said security code on an ad hoc label stuck on the bottle 100, for example by laser engraving or ablation, or printing of said security code on a preexisting label 11 of the bottle 100 in which case the code is entered just after the step of calculating one of the metrics.

For example, the security code comprises a mathematical function of the position of the first remarkable point 31 and of the position of the second remarkable point 32, for at least two remarkable points. We can notably foresee that the security code is one of the calculated metrics, which makes it possible to personalize the security code per bottle 100.

The advantage of such a security code lies in particular in the possibility of its offline control.

The security code may be the identifier of the bottle 100, or separate from it.

It is also possible, as an alternative or in combination, for the bottle 100, in particular the container 10, to include a set of bitters, possibly confused with the bitters of a 2D bar code.

Digital footprint

The set of selected attributes defines a digital imprint of the image.

The digital footprint of the image can be calculated on the secure server, or calculated by a programmable computer, or programmed, in computer connection with the optical sensor.

Once the attributes have been selected, the digital fingerprint of the image can then be saved in a memory. In this case, the digital fingerprint is stored as a data vector in temporary memory. Typically, the data vector comprises the values of the selected attributes, juxtaposed two by two.

A security advantage in calculating the digital imprint of the image, in particular of the reference image, is then that only the person responsible for applying the sleeve to the bottle 100 is the holder of the imprint of the bottle. 100, as ready to go in the distribution circuit.

The digital imprint of the image, in particular of the reference image, subsequently makes it possible to verify the authenticity of the bottle 100, typically after it has been put on the market.

Verification

To ensure that there has been no undue re-use of the container 10 or of the contents, a verification step can be provided, typically after the bottle 100 has been placed on the market.

The verification step comprises a step consisting in obtaining a set of at least one image, called the test image, of the bottle 100 by an optical control sensor.

The optical control sensor can be that of any type of communicating object, telephone, tablet, digital camera, etc., capable of communicating by computer with the secure database.

The test image can be produced by the same means (optical sensor, computer program) as those used for the reference image, or by completely different means, for example the optical sensor of a communicating object (mobile phone , smartphone, tablet, etc.), in particular the communicating object of a buyer.

For example, the set of at least one test image of the bottle 100 is a video sequence.

For a given bottle 100, a plurality of test images can be provided, in which each test image corresponds to a respective angular position of the optical control sensor relative to the bottle 100.

Each test image comprises at least part of the closure device 20 of the bottle 100 and all or part of the container, in particular the neck.

We can provide a cropping step, consisting in selecting a part of the test image, which increases the calculation speed. In this case, we plan to select a predetermined subset of the test image whose shape and dimensions are predetermined.

A test image processing step is provided, consisting of processing the test image with the same algorithm as that used for processing the source image, and calculating at least one of the metrics from:

the distance, in particular Euclidean and / or cylindrical, between the first remarkable point 31 and the second remarkable point 32, and a function of the distance between the first remarkable point 31 and the second remarkable point 32, in particular a hash function of said distance.

The digital footprint of each test image is thus calculated identically, preferably with the same computer program, or at least the same algorithms, as that used for the calculation of the digital footprint of the source image.

For example, the computer program is stored on a remote server where the test image and the source image are sent. Alternatively, the same computer program can be downloaded to a first programmable computer for calculating the digital footprint of the source image, and to a second programmable computer for calculating the digital footprint of the test image in order to select the same attributes.

The test image processing step can be implemented in situ, that is to say on computing means linked by computer to the optical control sensor, or implemented on the secure server.

We can then provide a comparison step consisting in comparing the imprint of the test image and the imprint of the source image.

If at least one attribute of the fingerprint of the test image is identical to an attribute of the fingerprint of the source image, then the bottle 100 has not been opened.

Conversely, if the bottle 100 has been opened, then the device 20 for closing the container 10 has been detached from the container 10. Consequently, the relative position of the first marking element and of the second marking element will have varied, therefore the respective remarkable points will have different positions, so the imprint of the test image will be different from the imprint of the source image.

Thus during the acquisition of the authentication zone on the bottle 100, one can foresee several possible situations, and for example send a specific message whose content relates to one of the following situations:

Situation 1: the imprint of the test image is identical to the imprint of the source image; bottle 100 is authentic; container 10 has probably not been opened;

Situation 2: the imprint of the test image is different from the imprint of the source image; then the closing device 20 has probably been manipulated (replaced, removed or the like). The bottle 100 does not comply with its placing in the distribution circuit, there is a risk that the bottle 100 has been opened / handled, there is a risk of fraud.

There may be a step of straightening the test image before the comparison step. For example, it can be provided that the bottle 100 includes bitters, or that the contours of the label, the shape and dimensions of which are known, are used as bitters for the righting calculations. When the bottle 100 comprises a 2D bar code, the bitters of said bar code can also be advantageously used for the step of straightening the test image. The rectification step of the test image makes it possible to correct the possible variation in angular position of the bottle 100 and of the optical sensor between the source image and the test image.

In the absence of pre-existing bitters, one can use some of the remarkable points corresponding to one of the first marking elements whose position is known, for example:

- corners of a label, stuck on at least one of the neck, the shoulder and the body of the container 10,

a molding of the container 10 representing a predetermined pattern,

- elements of a decoration, in particular deposited by screen printing on the body of the container 10.

The invention is not limited to the embodiments previously described.

We can for example provide:

a step of calculating and recording the result of a hash function of the fingerprint of the reference image, a step of calculating and recording the result of the same hash function of the footprint of the image test, then a step of comparing the results of these two hash functions, which adds an additional level of security.

Thanks to the present invention, a bottle 100 can then have an opening indicator.

During the inspection at the end of the bottling line or during an in-store inspection by an approved inspector, all the elements necessary for a complete inspection are visible / accessible.

Thus, thanks to the present invention, the implementation of securing a bottle 100 is facilitated by:

a single control gesture (shooting a test image), with limited dimensions the control zone (maximum dimensions which are those of a bottle 100);

unique to the control area, regardless of the condition of the bottle 100.

The present invention is particularly simple to implement: there is nothing to add to an existing capping process. In addition, the associated costs are very low since a single shot may suffice.

In addition, thanks to the invention, a bottle can be secured without additional ad hoc element, in particular such as a security label.

Nomenclature

100 Bottle 10 Container 11 Container marking label 12 First set of at least one marking element, integral with the container 20 Closing element 21 Second set of at least one marking element, integral with the closure device 31 First remarkable point 32 Second remarkable point

Claims (10)

1. Method for securing a bottle (100) comprising: • a container (10), • a device (20) for closing said container (10) closing said container (10), • a first set of at least one marking element (12), integral with the container (10); • a second set of at least one marking element (21), integral with the closure device (20); the process comprising: • an enrollment step consisting in producing, by means of shooting comprising an optical sensor, a source image of the bottle (100), characterized in that it further comprises • a step of image processing source consisting of: defining on the source image a first set of remarkable points, and selecting at least a first remarkable point (31) from the first set of remarkable points, each first remarkable point (31) corresponding to a respective first marking element of the first set of marking elements; defining on the source image a second set of remarkable points, and selecting at least one second remarkable point (32) from the second set of remarkable points, each second remarkable point (32) corresponding to a respective second marking element of the second set of marking elements; and o define the position of each selected remarkable point (31, 32). 2. Method according to claim 1, in which the first set of remarkable points (31) and the second set of remarkable points (32) are selected so that the first marking element and the second marking element respectively corresponding to said first and second sets of remarkable points can simultaneously enter the field of observation of an optical sensor. 3. Method according to any one of the preceding claims, in which the first and the second marking element, corresponding respectively to the first set of remarkable points and to the second set of remarkable points, are chosen so that the opening of the bottle (100) causes the destruction or displacement of at least one of the remarkable points (31, 32). 4. Method according to any one of the preceding claims, comprising a step consisting in calculating, and recording in the database, at least one of the metrics from: • the distance between the first remarkable point (31) and the second remarkable point (32) selected, and • a function of the distance between the first remarkable point (31) and the second remarkable point (32) selected, in particular a function hash of said distance. 5. Method according to any one of the preceding claims, further comprising a step consisting in securing a security code to the bottle (100); the security code optionally comprising a mathematical function of the position of the first remarkable point (31) and of the position of the second remarkable point (32) selected. 6. Method according to any one of the preceding claims, comprising an enrollment step, preferably implemented before the marketing of the bottle (100), and consisting in: • send, optionally encrypted, the source image to a database located on an optionally secure server, • assign an identifier to each source image saved in the database, said identifier being one of the calculated metrics or a serial number. 7. Method for verifying a secure bottle (100) according to any one of the preceding claims, comprising a verification step, comprising steps consisting in: • obtain a set of at least one test image of the bottle (100) by an optical sensor, • process at least part of the test image with the same algorithm as that used for processing the source image, and • define on the test image a first set of remarkable points, and select at least a first remarkable point (31) from the first set of remarkable points, each first remarkable point (31) corresponding to a respective first marking element of the first set of marking elements (12); • define on the test image a second set of remarkable points, and select at least one second remarkable point (32) from the second set of remarkable points, each second remarkable point (32) corresponding to a respective second marking element of the second set of marking elements (21); and • define the position of each selected remarkable point (31, 32); then • calculate at least one of the metrics from: • the distance between the first remarkable point (31) and the second remarkable point (32), and • a function of the distance between the first remarkable point (31) and the second remarkable point (32), in particular a hash function of said distance; and • compare at least one of the metrics calculated on the test image with at least one of the metrics calculated on the source image, the method optionally further comprising an optical recognition step of the first set of at least one marking element (12) and the second set of at least one marking element (21), implemented by an artificial intelligence system, the method comprising in a learning step, supervised or unsupervised. 8. The method of claim 7, comprising a step of straightening the test image. 9. Method according to any one of claims 7 or 8, comprising a step of recording a video sequence of the bottle (100) by the optical sensor; the test image being a frame of the video sequence. 10. Computer program comprising program code instructions for executing the steps of the method according to any one of claims 7 to 9, when said program is executed on a computer.