Method and system for determining genuine product by using xrf

Abstract

The present disclosure relates to a method and a system for determining whether a product is genuine by using XRF, and provides a method and a system for determining whether a product is genuine, in which: XRF data for the product is collected; multivariate statistical analysis is performed on the collected XRF data to generate a determination model enabling determination of whether a product is genuine; XRF data for a first product for which whether the product is genuine is to be determined is received; and whether the first product is genuine is determined on the basis of the XRF data for the first product.

Description

Authenticity determination method and system using XRF

[0001] The present disclosure relates to a method and system for determining authenticity using X-ray Fluorescence (XRF). More specifically, the present disclosure relates to a method and system for determining the authenticity of a product using multivariate statistical analysis based on XRF data.

[0002] So-called branded goods refer to products that have gained trust and popularity among consumers and traders from specific manufacturers or sellers. These branded goods provide high value to consumers based on quality, design, and reliability, thereby forming a consumer base that trusts the brand. However, branded goods may face the problem of counterfeit or illegal reproductions being distributed.

[0003] One existing method to address this issue is for manufacturers to issue certificates verifying authenticity. However, since certificates can also be forged, there is a problem in that authenticity cannot be trusted based solely on the certificate for products distributed through channels other than official networks. For instance, in the case of parallel imports or second-hand goods, the authenticity of a product cannot be confirmed based on the certificate alone, and the same problem is occurring in direct person-to-person transactions, which have been increasing recently.

[0004] Existing methods for authenticating products have primarily utilized methods such as measuring metal purity using electrical conductivity and eddy currents, or having experts visually examine the product's appearance—including stitching patterns other than metals—and compare whether the shapes are identical. However, these methods have limitations in that they cannot determine the detailed composition of the metal, make it difficult to distinguish sophisticated counterfeits from genuine products, and rely on the subjective judgment of experts to determine authenticity, which makes it difficult to achieve high reliability.

[0005] The present disclosure aims to solve such problems by providing a method and system for accurately and quickly determining authenticity using multivariate statistical analysis of XRF data.

[0006]

[0007] One embodiment of the present disclosure is intended to provide a method and system for determining whether a product is genuine.

[0008] In addition, one embodiment of the present disclosure is intended to provide a method and system for determining authenticity using XRF data of a product subject to authentication.

[0009] In addition, one embodiment of the present disclosure is intended to provide a method and system for determining whether a product is genuine using multivariate statistical analysis of XRF data.

[0010]

[0011] The purposes of the present disclosure are not limited to those mentioned above, and other purposes and advantages of the present disclosure not mentioned may be understood from the following description and will be more clearly understood by the embodiments of the present disclosure. Furthermore, it will be readily apparent that the purposes and advantages of the present disclosure can be realized by the means and combinations thereof set forth in the claims.

[0012]

[0013] One embodiment of the present disclosure aims to provide a method for determining whether a product is genuine using XRF.

[0014] A method for determining the authenticity of a product using XRF according to one embodiment of the present disclosure may include the steps of: collecting XRF data for the product; performing multivariate statistical analysis on the collected XRF data to generate a determination model capable of determining authenticity; receiving XRF data for a first product to be determined authenticity; and determining the authenticity of the first product based on the XRF data for the first product.

[0015] In one embodiment, XRF data for the first product may be obtained by measuring at least one metal part included in the first product.

[0016] In one embodiment, the XRF data for the first product may be measured using a portable X-ray fluorescence analyzer.

[0017] In one embodiment, the method may further include a judgment model capable of determining authenticity and a step of visualizing and providing XRF data for the first product.

[0018] In one embodiment, the determination model capable of determining authenticity may include at least one of a PCA determination model generated from the result of analyzing the collected XRF data by Principal Component Analysis (PCA) and a PLS-DA determination model generated from the result obtained by analyzing the XRF data of the product by PLS-DA.

[0019] In one embodiment, the PCA determination model may set genuine product areas and / or counterfeit product areas in a coordinate system (score plot) generated by performing principal component analysis (PCA) on the collected XRF data.

[0020] In one embodiment, the genuine area and / or the counterfeit area may be determined by grouping XRF data corresponding to the genuine and / or counterfeit products after plotting them on the coordinate system (score plot).

[0021] In one embodiment, the step of determining whether the first product is genuine may involve plotting XRF data for the first product on the coordinate system (score plot) and determining whether it is genuine based on whether it belongs to a genuine area or a counterfeit area.

[0022] In one embodiment, the PLS-DA judgment model may be a mathematical prediction model for an output variable generated by analyzing XRF data with PLS-DA, with the authenticity of the product as the output variable (Y) and the XRF data as the input variable (X).

[0023] In one embodiment, the step of determining whether the first product is genuine involves inputting XRF data for the first product into the mathematical prediction model as an input variable (X), and the predicted output variable ( It is possible to determine whether it is genuine based on ).

[0024] In one embodiment, the step of determining whether the first product is genuine may include a first step of determining whether it is genuine using a pre-stored ingredient table, and a second step of determining whether it is genuine using the PCA determination model and / or the PLS-DA determination model if it cannot be determined whether it is genuine in the first step.

[0025] In one embodiment, the pre-stored component table may include at least one of the type, content, and plating thickness of the constituent components.

[0026] In one embodiment, if it is confirmed that the similarity between the content of the components of the first product and the components of the component table exceeds a reference value, the first product can be determined to be genuine.

[0027] In one embodiment, the second step may include a second-1 step using the PCA judgment model and a second-2 step using the PLS-DA judgment model if the authenticity cannot be determined in the second-1 step.

[0028] According to one embodiment of the present disclosure, a program stored on a computer-readable recording medium may be provided to execute a method for determining the authenticity of a product using XRF.

[0029] One embodiment of the present disclosure aims to provide a system for determining the authenticity of a product using XRF.

[0030] A system for determining the authenticity of a product using XRF according to one embodiment of the present disclosure comprises an X-ray fluorescence analyzer, a memory, and at least one processor that executes one or more instructions stored in the memory. The at least one processor collects XRF data for the product, performs multivariate statistical analysis on the collected XRF data to generate a determination model capable of determining authenticity, receives XRF data for a first product whose authenticity is to be determined, and determines the authenticity of the first product based on the XRF data for the first product.

[0031] In one embodiment, the X-ray fluorescence analyzer may be portable.

[0032] In one embodiment, an output unit indicating the result of determining authenticity may be further included.

[0033] In one embodiment, the result of determining whether the first product is genuine can be represented by visualizing a determination model capable of determining the genuineness and XRF data for the first product.

[0034] In one embodiment, the determination model capable of determining authenticity may include at least one of a PCA determination model generated from the result of analyzing the collected XRF data by Principal Component Analysis (PCA) and a PLS-DA determination model generated from the result obtained by analyzing the XRF data of the product by PLS-DA.

[0035] In one embodiment, the principal component of the PCA decision model is Z = a1X1 + a2X2...a m X m Expressed as, and [X 1, X 2, ..., X m ] is input data obtained by sampling m from the XRF spectrum, where m is a natural number greater than or equal to 3 and less than or equal to 2048, and a i (Here, 1≤i≤m) can be greater than or equal to -1 and less than or equal to 1.

[0036] In one embodiment, the PLS-DA decision model is Y = w1X1 + w2X2...w n X n It is expressed as, where Y is an output variable for determining authenticity, and [X 1, X 2, ..., X n ] is input data obtained by sampling n samples from the XRF spectrum, where n is a natural number up to 2048, and w i (where 1≤i≤n) can be greater than or equal to -0.000001 and less than or equal to 0.000002.

[0037]

[0038] According to one embodiment of the present disclosure, there is an effect of providing a method and system for determining the authenticity of a product that can be used in cases where it is difficult to determine authenticity solely through visual inspection or comparison of component contents.

[0039] In addition, according to one embodiment of the present disclosure, there is an effect of accurately determining whether a product is genuine through multivariate statistical analysis of XRF data.

[0040] In addition, according to one embodiment of the present disclosure, trading can be promoted by increasing the credibility of the secondary distribution market, including the second-hand market, by facilitating the trading of genuine products.

[0041]

[0042] In addition to the above, the specific effects of the present disclosure are described together with the specific details for implementing the disclosure below.

[0043]

[0044] FIG. 1 is a diagram showing XRF data that can be used to determine authenticity in one embodiment of the present disclosure.

[0045] FIG. 2 is a flowchart of a method for determining whether a product is genuine or not according to one embodiment of the present disclosure.

[0046] FIG. 3 is a schematic diagram of a step for determining whether a product is genuine or not according to one embodiment of the present disclosure.

[0047] FIG. 4 is a graph comparing the main components of genuine and counterfeit products according to one embodiment of the present disclosure.

[0048] FIG. 5 is an example diagram showing the result of determining whether a product is genuine using a PCA judgment model according to one embodiment of the present disclosure.

[0049] FIG. 6 is an example diagram showing the result of determining whether a product is genuine using a PLS-DA judgment model according to one embodiment of the present disclosure.

[0050] FIG. 7 is a diagram showing the configuration of a system for determining whether a product is genuine or not according to one embodiment of the present disclosure.

[0051] FIG. 8 is an example of a screen showing observation of a measurement area using an X-ray fluorescence analyzer in a product authenticity determination system according to one embodiment of the present disclosure.

[0052]

[0053] To clarify the technical concept of the present disclosure, embodiments of the present disclosure will be described in detail with reference to the attached drawings. In describing the present disclosure, detailed descriptions of related known functions or components will be omitted if it is determined that such detailed descriptions would unnecessarily obscure the essence of the present disclosure. Components having substantially the same functional configuration among the drawings have been assigned the same reference numerals and symbols as much as possible, even if they are shown in different drawings. For convenience of explanation, devices and methods will be described together where necessary. Each operation of the present disclosure does not necessarily have to be performed in the order described and may be performed in parallel, selectively, or individually.

[0054] The terms used in the embodiments of this disclosure have been selected to be as widely used and general as possible, taking into account the functions of this disclosure; however, these terms may vary depending on the intent of those skilled in the art, case law, the emergence of new technologies, etc. Additionally, in specific cases, terms have been selected at the applicant's discretion, and in such cases, their meanings will be described in detail in the description of the relevant embodiments. Therefore, terms used in this specification should be defined not merely by their names, but based on their meanings and the overall content of this disclosure.

[0055] Throughout this disclosure, singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms such as "comprising" or "having" are intended to specify the existence of features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. That is, throughout this disclosure, when a part is described as "comprising" a certain component, it means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0056] Expressions such as "at least one" modify the entire list of components and do not modify the components of the list individually. For example, "at least one of A, B, and C" and "at least one of A, B, or C" refer to only A, only B, only C, both A and B, both B and C, both A and C, all of A, B, and C, or any combination thereof.

[0057] Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but the components are not limited by the terms. The terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present disclosure, the first component may be named the second component, and similarly, the second component may be named the first component.

[0058] Additionally, terms such as "...part," "...module," etc., as described in this disclosure refer to a unit that processes at least one function or operation, and may be implemented in hardware or software, or a combination of hardware and software.

[0059] Throughout the entire disclosure, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" with other elements interposed between them. Furthermore, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0060] Throughout this disclosure, the expression “configured to” may be replaced, depending on the context, with, for example, “suitable for,” “having the capacity to,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured to” may not necessarily mean only “specifically designed to” in hardware. Instead, in some situations, the expression “system configured to” may mean that the system is “capable of” in conjunction with other devices or components. For example, the phrase “processor configured to perform A, B, and C” may mean a dedicated processor for performing the said operations (e.g., an embedded processor) or a generic-purpose processor (e.g., a CPU or an application processor) capable of performing said operations by executing one or more software programs stored in memory.

[0061]

[0062] The present disclosure relates to a method and system for determining authenticity using XRF. More specifically, the present disclosure relates to a method and system for determining whether a product is authentic using multivariate statistical analysis of XRF data.

[0063] Throughout this disclosure, Multivariate Statistical Analysis refers to a technique for statistically analyzing complex data in which multiple variables influence each other, and signifies an analytical method used to identify relationships between various variables or to perform tasks such as prediction and classification. Multivariate Statistical Analysis can be used as a useful tool for transforming data into an intuitive form, and can visualize high-dimensional data by converting it into low-dimensional data through dimensionality reduction, or make the data easier to understand. Furthermore, Multivariate Statistical Analysis has the advantage of improving the performance of prediction models compared to considering only a single variable by identifying complex relationships that arise from the interaction of multiple variables.

[0064] For example, multivariate statistical analysis includes Principal Components Analysis (PCA), Principal Least Square (PLS) regression, Principal Least Square - Discrimination Analysis (PLS-DA), Factor Analysis, Cluster Analysis, Discriminant Analysis, k-Means Clustering, Multiple Regression Analysis, Logistic Regression Analysis, Multiple Correlation Analysis, Analysis of Covariance (ANCOVA), Structural Equation Modeling (SEM), Path Analysis, Multivariate Time Series Analysis, Independent Component Analysis (ICA), Multidimensional Scaling (MDS), Association Analysis, and Chi-square. It may be at least one of the analyses (Chi-Square Analysis).

[0065] A method and system for determining the authenticity of a product according to one embodiment of the present disclosure can determine whether a product is authentic based on the results of multivariate statistical analysis of XRF data regarding genuine and counterfeit products of a product whose authenticity is to be determined.

[0066]

[0067] FIG. 1 is a diagram showing XRF data that can be used to determine authenticity in one embodiment of the present disclosure.

[0068] Throughout this disclosure, XRF data refers to data obtained using X-ray fluorescence (XRF) analysis (or fluorescence X-ray analysis). X-ray fluorescence analysis is a non-destructive method used to analyze the composition of a substance, which is a technique that measures the fluorescence X-rays emitted from a sample by irradiating the sample with X-rays. Data obtained from this X-ray fluorescence analysis can be expressed as the intensity (strength) of the fluorescence X-rays for a specific energy (KeV).

[0069] In X-ray fluorescence analysis, the energy of the fluorescent X-rays emitted from a sample has a unique value for each element, so the elements contained in the sample can be identified through the measured energy. For example, since iron (Fe) has an energy of approximately 6.4 KeV and copper (Cu) has an energy of approximately 8.0 KeV, the elements contained in the sample can be identified by utilizing whether fluorescent X-rays with the energy corresponding to the element are detected. Furthermore, since the intensity of fluorescent X-rays emitted at a specific energy is proportional to the concentration of the corresponding element, a higher intensity indicates a higher content of that element in the sample; by utilizing this, data regarding the mass percentage (%) or concentration (ppm or mg / kg) of each element can be obtained. Additionally, an XRF spectrum can be obtained by plotting the data obtained from X-ray fluorescence analysis with energy (KeV) on the X-axis and the intensity of the fluorescent X-rays on the Y-axis.

[0070] In the present disclosure, XRF data is data obtained by applying X-ray fluorescence analysis to a sample. For example, XRF data may include a list of elements contained in the sample obtained using X-ray fluorescence analysis, the content of elements contained in the sample, and an XRF spectrum (or data obtained by sampling the XRF spectrum).

[0071] Generally, since counterfeit products do not use the exact same components as genuine products, the constituent materials of the parts used in counterfeits differ from those used in genuine products; therefore, the authenticity of the product to be authenticated can be verified by comparing the XRF data of the counterfeit and the genuine product. In other words, the XRF data obtained from X-ray fluorescence analysis of a counterfeit product differs from the XRF data obtained from a genuine product, and by comparing the XRF data of the product to be authenticated with that of the genuine product, it is possible to confirm whether the product is genuine.

[0072] Referring to Figure 1, the XRF spectrum for a part of the genuine product differs from that of the same part of the counterfeit product in terms of the intensity of X-ray fluorescence (Y-axis) according to XRF energy (X-axis). That is, since the genuine product consists of parts with different material components than the counterfeit product, the XRF spectra obtained by applying X-ray fluorescence analysis to the same parts of the genuine and counterfeit products, respectively, show different intensities of X-ray fluorescence according to energy.

[0073] In one embodiment, if there is a significant difference in the major constituent materials between the genuine product and the counterfeit product, the genuine product and the counterfeit product can be distinguished by checking the presence and / or content of the said material in the product to be examined from XRF data. For example, if a part of the genuine product is a precious metal alloy containing gold (Au, 62.45%) and nickel (Ni, 11.67%) as major constituents, and the counterfeit product has a lower content of gold (Au) or nickel (Ni) in the same part, the authenticity can be determined simply by checking the presence and content of gold (Au) and nickel (Ni) in the XRF data for the same part of the product to be examined. In one embodiment, the content of the constituent components for a part of the product to be examined can be stored in a component table, and the authenticity can be determined based on this.

[0074] However, when there are no significant differences in the major constituent materials between genuine and counterfeit products, authenticity cannot be determined solely by the presence or content of these materials (identified using XRF data). For example, if a component of a genuine product is an alloy metal rather than a precious metal, a counterfeit product may also use a similar alloy metal; in this case, since there is no significant difference in the major constituent materials or their content, authenticity cannot be verified by comparing these factors. In such cases, to determine authenticity, one must compare the content of multiple constituent materials overall, rather than checking the presence and / or content of a few specific materials. However, not only is it difficult to compare multiple factors simultaneously, but it is also challenging to establish criteria for determining authenticity based on the results of such comparisons.

[0075] In one embodiment of the present disclosure, in order to distinguish between genuine and counterfeit products even in such cases, XRF data may be analyzed using a multivariate statistical analysis method to derive a determination model capable of determining authenticity. In one embodiment, the determination model capable of determining authenticity may include criteria for distinguishing between genuine and counterfeit products. For example, the determination model capable of determining authenticity may be a dataset capable of distinguishing authenticity. As another example, the determination model capable of determining authenticity may be represented as a region of a coordinate system capable of distinguishing authenticity. As another example, the determination model capable of determining authenticity may be a mathematical prediction model for authenticity and may include weights and / or coefficients used in the mathematical model.

[0076] In one embodiment, XRF data analyzed using a multivariate statistical analysis method may be the intensities of X-ray fluorescence at energy levels corresponding to predetermined materials. For example, since the authenticity of metal parts from luxury goods manufacturer Company A can be distinguished based on the composition of titanium (Ti), chromium (Cr), manganese (Mn), cobalt (Co), iron (Fe), nickel (Ni), copper (Cu), and zinc (Zn), the intensities of X-ray fluorescence measured at energy levels corresponding to the materials can be analyzed as XRF data using a multivariate statistical analysis method.

[0077] In one embodiment, the XRF data to which multivariate statistical analysis is applied may include the intensities of X-ray fluorescence sampled in a predetermined manner from the XRF spectrum. For example, multivariate statistical analysis may be applied to the intensities of n X-ray fluorescence sampled at predetermined intervals from the XRF spectrum as XRF data. Here, n may be a predetermined value as a natural number greater than or equal to 1.

[0078] In one embodiment, a judgment model for determining authenticity can be derived from XRF data collected for genuine and counterfeit products using Principal Components Analysis (PCA). Principal Components Analysis is a method for reducing the dimensionality of high-dimensional data containing multiple variables; it transforms data into a lower dimension by generating new axes (PC axes, singular vectors, or eigenvectors) based on the correlations of variables while preserving data variance as much as possible. In one embodiment, if the XRF data of the product to be authenticated is plotted on a score plot, which is a coordinate system based on the new axes (PC axes) for dimensionality reduction derived by performing Principal Components Analysis on the XRF data, the authenticity of the product can be determined based on this. For example, the judgment model derived by Principal Components Analysis may include information regarding the regions representing genuine and counterfeit products in the score plot.

[0079] In one embodiment, a judgment model for determining authenticity can be derived from XRF data collected for genuine and counterfeit products using the Principal Least Square Discrimination Analysis (PLS-DA) method. PLS-DA is a method for finding the relationship between input data X and output variable Y (class label or label value). After identifying latent variables that can be commonly well explained by both input data X and output variable Y, a judgment model capable of effectively classifying data can be constructed based on these variables. In one embodiment, XRF data is input as input data X, and PLS-DA analysis can be performed by setting the output variable Y to 1 if the corresponding XRF data pertains to a genuine product and to 0 if it pertains to a counterfeit product. As a result of the PLS-DA analysis, the PLS variable t i The weight w corresponding to iand regression variable b i If we obtain , we can use it to obtain a judgment model that can predict the output variable Y for XRF data of the product to be evaluated.

[0080]

[0081] FIG. 2 is a flowchart of a method for determining whether a product is genuine or not according to one embodiment of the present disclosure.

[0082] A method according to one embodiment of the present disclosure may include the step (S100) of collecting product identification information and XRF data corresponding to the product. In one embodiment, the product identification information may include information for identifying the product, such as the product name, product model name, product engraving number, product serial number, etc. In one embodiment, the product identification information may include information regarding whether the product is genuine or counterfeit. In one embodiment, the XRF data corresponding to the product may be stored by measuring the constituent components and / or XRF fluorescence spectrum of at least a part of the product using an X-ray fluorescence analyzer.

[0083] In one embodiment, the method may include a step (S200) of performing multivariate statistical analysis on collected XRF data to derive and store a determination model capable of determining authenticity. For example, the determination model capable of determining authenticity may relate to areas corresponding to genuine and / or counterfeit products in a coordinate system (score plot) using PC axes (or singular vector axes) obtained by performing Principal Component Analysis (PCA) on the collected XRF data. As another example, the result obtained by performing PCA on the collected XRF data may be represented as a singular vector in a graph, and a loading value may be extracted. Here, the loading value is information containing the metal component and its content that must be considered most important for determining authenticity among the total metal material components appearing in the XRF data in the model for determining authenticity.

[0084] In one embodiment, the PCA decision model is an equation for the principal components Z = a1X1 + a2X2… a m X m It can be expressed as, and [X 1, X 2,..., X m ] is XRF data obtained by sampling m from the XRF spectrum, and [a 1, a 2,..., a m ] is a loading value indicating the degree to which each component of the XRF data influences the principal components. Here, m is a natural number greater than or equal to 3 and less than or equal to 2048, and a i (where 1 ≤ i ≤ m) can have a value greater than or equal to -1 and less than or equal to 1. For example, [X 1, X 2,..., X m If ] corresponds to major metallic components such as nickel, copper, or gold, the loading value [a 1, a 2,..., a m] can be interpreted as a value indicating the degree to which each major metallic component contributes to the main component (PC).

[0085] In one embodiment, a determination model capable of determining authenticity can be expressed by a mathematical model obtained using PLS-DA analysis. For example, the PLS-DA determination model is Y = w1X1 + w2X2… w n X n It can be expressed as, where Y is an output variable for determining authenticity, and [X 1, X 2,..., X n ] is input data obtained by sampling n samples (where n ≤ 2048) from the XRF spectrum. Also, [w 1, w 2,..., w n ] can be obtained as weights for each input data using PLS-DA analysis. For example, using the Nonlinear Iterative Partial Least Squares (NIPALS) algorithm, weights can be calculated using PLS variables, their corresponding weights, and regression variables.

[0086] In one embodiment, the method may include a step (S300) of receiving XRF data for a first product to be determined as genuine. In one embodiment, the XRF data for the first product may be measured using an X-ray fluorescence analyzer. For example, the X-ray fluorescence analyzer may be portable.

[0087] In one embodiment, the method may include a step (S400) of determining whether the first product is genuine using XRF data for the first product. In one embodiment, the XRF data for the first product may be applied to a dataset capable of determining genuineness derived in step S200 to determine whether the first product is genuine. For example, the determination of whether the first product is genuine may be provided as 'genuine' or 'counterfeit'. As another example, the determination of whether the first product is genuine may be provided as '00% probability of being genuine'.

[0088] In one embodiment, the method may include a step (S500) of visualizing and providing data related to determining the authenticity of the first product. For example, XRF data for the first product and a determination model capable of determining authenticity (derived through multivariate statistical analysis in step S200) may be visualized and provided.

[0089]

[0090] FIG. 3 is a schematic diagram of a step for determining whether a product is genuine or not according to one embodiment of the present disclosure.

[0091] In one embodiment, the step (S400) of determining whether a product is genuine or not may include a first step (S410) using a pre-stored component table. Details related to this will be described later with reference to FIG. 4.

[0092] In one embodiment, if it is impossible to determine authenticity by the first step (S410), the authenticity of the product under appraisal can be determined by using a determination model for authenticity derived by a multivariate analysis method in the second step (S450, S470). First, as a second step (S450), the authenticity is determined using a PCA determination model derived by Principal Component Analysis (PCA). If it is impossible to determine authenticity using the PCA determination model, the authenticity can be determined using a PLS-DA determination model derived by PLS-DA analysis in the second step (S450). Since the determination model in the second step allows for the analysis of differences in overall composition ratios, the authenticity can be determined even in cases where it is impossible to determine authenticity in the first step due to a lack of significant difference in the content of major components.

[0093]

[0094] FIG. 4 is a graph comparing the main components of genuine and counterfeit products according to one embodiment of the present disclosure.

[0095] Referring to Figure 4, it can be seen that the main components and content of genuine and counterfeit products are different. In this case, the genuine and counterfeit products can be distinguished using a component table for the genuine product.

[0096] A component table according to one embodiment of the present disclosure may be stored to correspond to product identification information. For example, the component table may include one or more of the type, content, and plating thickness of the constituent components. In one embodiment, the component table may include a pre-set reference value regarding the similarity of the type, content, and / or plating thickness of the constituent components. In one embodiment, the pre-set reference value may be set differently for each product.

[0097] In one embodiment, the first step of using a pre-stored ingredient table can analyze the constituent components of the product using XRF data measured for the first product and compare them with a preset reference value to determine whether the first product is genuine.

[0098] In one embodiment, XRF data of a first product to be determined as genuine is analyzed and the component content of a component table is compared; if it is confirmed that the similarity of a specific component (corresponding to a genuine product) exceeds a preset threshold, the first product can be determined to be genuine. In another embodiment, if it is confirmed that the similarity of a specific component (corresponding to a counterfeit product) does not exceed a preset threshold, the first product can be determined to be genuine.

[0099] In one embodiment, if it is confirmed that the similarity of a specific component (corresponding to a genuine product) does not exceed a preset threshold, the first product may be determined to be a counterfeit. In another embodiment, if it is confirmed that the similarity of a specific component (corresponding to a counterfeit product) exceeds a preset threshold, the first product may be determined to be a counterfeit.

[0100] In one embodiment, XRF data may be measured for at least one metal part included in the product. The metal part may be a logo, a connecting ring, a string, etc. For example, if the metal parts included in the product are a 'logo' and a 'string' respectively, the inspector may measure XRF data for each of the product's 'logo' and 'string'.

[0101] In one embodiment, when the metal parts included in the product are gold-plated and silver-plated, the authenticity can be determined based on the content of gold (Au), silver (Ag), platinum (Pd), rubidium (Rh), iridium (Is), osdium (Os), lithium (Li), ruthenium (Rth), nickel (Ni), etc.

[0102] In one embodiment, in the case of an alloy included in a product, the authenticity can be determined based on the content of copper (Cu), zinc (Zn), iron (Fe), chromium (Cr), etc.

[0103]

[0104] FIG. 5 is an example diagram showing the result of determining whether a product is genuine using a PCA judgment model according to one embodiment of the present disclosure.

[0105] In one embodiment of the present disclosure, multivariate statistical analysis of XRF data may utilize Principal Component Analysis (PCA). Principal Component Analysis (PCA) is a method for efficiently converting high-dimensional data into low-dimensional data while minimizing information loss, and can reduce the dimensionality of the data to low-dimensional data by identifying principal components that maximize the correlation and variance between input variables.

[0106] In one embodiment, to apply Principal Component Analysis (PCA), the covariance matrix of the XRF data is calculated, and the eigenvalues ​​and eigenvectors are obtained by decomposing the covariance matrix. The eigenvector with the largest eigenvalue is defined as the principal component axis (first axis, PC1 or singular value) corresponding to the principal component. A second axis (PC2) that is orthogonal to the first axis (PC1) and maximizes the variance can be found, and a third axis (PC3) that is orthogonal to both the first and second axes and maximizes the variance can be found. In this way, as many axes as the required dimensions (number of features) can be found.

[0107] In one embodiment, loading values ​​extracted by the optimal axis calculated by Principal Component Analysis (PCA) are expressed, and the loading values ​​can be selected as latent variable values ​​that best explain the constituent material. Referring to FIG. 5(a), the loading values ​​of copper (Cu), iron (Fe), and chromium (Cr) are large, so they can be determined as major constituents representing the characteristics of the material under examination. In one embodiment, the larger the loading value derived by the optimal axis using Principal Component Analysis, the greater the influence on the determination of authenticity, and the authenticity can be determined using the constituents with large loading values. For example, after storing a component table for copper (Cu), iron (Fe), and chromium (Cr) with large loading values ​​in FIG. 5(a), the authenticity can be determined by comparing the content of the corresponding material.

[0108] In one embodiment, XRF data can be displayed on a score plot defined by principal component axes obtained by principal component analysis (PCA). When XRF data for genuine products and XRF data for counterfeit products form clusters in the score plot, the areas corresponding to these clusters can be determined as corresponding to genuine products or counterfeit products. For example, a score plot containing genuine or counterfeit product areas can be utilized as a judgment model. Referring to FIG. 5(b), since XRF data for counterfeit products forms a cluster in the lower left part of the score plot, this area marked with a red circle in FIG. 5(b) can be designated as the counterfeit product area, and the remaining area as the genuine product area to determine whether it is a genuine product.

[0109] In one embodiment, XRF data of a first product to be determined as genuine can be displayed on a score plot such as FIG. 5(b), and the product can be determined as genuine or counterfeit based on whether it belongs to a counterfeit area or a genuine area. In one embodiment, the genuineness of the product in the score plot can be determined by the distance from the counterfeit area or the genuine area. For example, the Euclidean distance method can be used to determine whether the product is counterfeit or genuine.

[0110]

[0111] FIG. 6 is an example diagram showing the result of determining whether a product is genuine using a PLS-DA judgment model according to one embodiment of the present disclosure.

[0112] According to one embodiment of the present disclosure, multivariate statistical analysis of XRF data may utilize Partial Least Squares (PLS) analysis or Principal Least Squares-Discrimination Analysis (PLS-DA). PLS or PLS-DA is a method for finding the relationship between an input variable X and an output variable Y (class label). After finding latent variables (regression coefficients) that can effectively explain both the input variable X and the output variable Y, a judgment model capable of effectively classifying data can be constructed based on these latent variables. In one embodiment, referring to FIG. 6(a), XRF data is input as the input variable X, and PLS or PLS-DA analysis can be performed by setting the output variable Y to 1 if the XRF data pertains to a genuine product and to 0 if it pertains to a counterfeit product. As a result of the PLS or PLS-DA analysis, the PLS variable t i The weight w corresponding to i and regression variable b iIf we obtain this, we can use it to obtain a PLS judgment model that can predict the output variable Y for XRF data of the product to be evaluated.

[0113] FIG. 6(b) is an example of the result of analyzing XRF data using a PLS-DA judgment model. Referring to FIG. 6(b), if the XRF data for a counterfeit product has a value close to 0 and the XRF data for a genuine product has a value close to 1 according to the PLS-DA judgment model, the authenticity can be determined using this. For example, by applying the XRF data of a first product to be determined as genuine to the PLS-DA judgment model, if the value of the output variable (Y) is predicted to be close to 1, it can be determined as a genuine product. If the value of the output variable (Y) is output as a value close to 0, it can be determined as a counterfeit product. In one embodiment, whether it is a genuine product or a counterfeit product can be determined according to a predetermined threshold value. For example, using 0.5 as the threshold value, if it is greater than or equal to that value, it can be determined as a genuine product, and if it is less than that value, it can be determined as a counterfeit product.

[0114]

[0115] FIG. 7 is a diagram showing the configuration of a system for determining whether a product is genuine or not according to one embodiment of the present disclosure.

[0116] Referring to FIG. 7, a system (10) according to one embodiment of the present disclosure may include an X-ray fluorescence analyzer (100), a control unit (200), and an output unit (300).

[0117] In one embodiment, the X-ray fluorescence analyzer (100) can emit X-rays to a sample and detect X-ray fluorescence emitted from the sample. For example, the X-ray fluorescence analyzer (100) may include an X-ray tube (not shown) for X-ray emission. In one embodiment, the X-ray fluorescence analyzer (100) may be portable and battery-operated. The portable X-ray fluorescence analyzer (100) can freely select the measurement area and can measure even a narrow area of ​​3 mm or less, so it can measure at least one measurement area included in each product without obstruction.

[0118] In one embodiment, the control unit (200) may be composed of a computing device including at least one memory (210) and at least one processor (220) that executes one or more instructions stored in the memory. In one embodiment, the control unit (200) may be implemented as a single device or a single computing device. In another embodiment, the control unit (200) may be implemented by including a plurality of computing devices.

[0119] In one embodiment, various types of data, such as programs and files, such as applications, may be installed and stored in the memory (210). In one embodiment, the processor (220) may access and use the data stored in the memory (210) or store new data in the memory (210). For example, the memory (210) may store programs for processing and control of the control unit (200), reference data, various updateable storage data, etc. Additionally, the memory (210) may store reference template images or various data required for motion recognition.

[0120] In one embodiment, the memory (210) can store XRF data and a program for performing, processing, and controlling multivariate statistical analysis thereon.

[0121] In one embodiment, the processor (220) controls the overall operation of the product authenticity determination system and may include at least one processor such as a CPU, GPU, etc. For example, the processor (220) may execute a program stored in memory (210), read a stored file, or store a new file.

[0122] In one embodiment, the control unit (200) can collect XRF data, perform multivariate statistical analysis to extract and store a dataset capable of determining authenticity, and operate as a system for determining authenticity of the first product accordingly.

[0123] In one embodiment, the control unit (200) can receive XRF data detected by the X-ray fluorescence analyzer (100). Additionally, the control unit (200) can determine whether the product is genuine and, along with the result, visualize the XRF data for the first product and the dataset capable of determining the genuineness and transmit it to the output unit (300).

[0124]

[0125] FIG. 8 is an example of a screen showing observation of a measurement area using an X-ray fluorescence analyzer in a product authenticity determination system according to one embodiment of the present disclosure.

[0126] In one embodiment, the XRF measurement method may vary depending on whether the metal part of the product is gold-plated, silver-plated, or an alloy. In the case of gold-plated and silver-plated products, the measurement of precious metals may be selected in the X-ray fluorescence analyzer (100), and in the case of alloys, the measurement of alloys may be selected. The X-ray fluorescence analyzer (100) may measure at a low voltage in the case of precious metals and at a relatively high voltage in the case of alloys.

[0127] In one embodiment, after turning on the power of the X-ray fluorescence analyzer (100), the operation to check whether the hardware is operating normally by checking the probe and the measurement window, etc., may be performed first. In addition, the measurement area may be observed through a visible light camera installed on the X-ray fluorescence analyzer (100), and XRF data may be measured by making close contact with the surface of the measurement area. In one embodiment, the constituent components of the measurement area and the XRF spectrum may be stored as measurement results.

[0128] In one embodiment, the X-ray fluorescence analyzer (100) may include an X-ray tube that outputs energy up to 40 kV, and rubidium (Rh) may be used as the target metal used when emitting X-rays from the X-ray tube. As the X-ray fluorescence analyzer (100), a Silicon Drift Detector (SDD) capable of counting X-rays at a rate of about 10 times higher per second than a conventional PIN detector may be used.

[0129]

[0130] One embodiment of the present disclosure may also be implemented in the form of a recording medium comprising computer-executable instructions, such as program modules executed by a computer. A computer-readable medium may be any available medium accessible by a computer and includes both volatile and non-volatile media, and both removable and non-removable media. Additionally, a computer-readable medium may include both computer storage media and communication media. A computer storage medium includes both volatile and non-volatile, removable and non-removable media implemented by any method or technique for storing information, such as computer-readable instructions, data structures, program modules, or other data. A communication medium typically includes computer-readable instructions, data structures, or program modules and includes any information transmission medium.

[0131] The foregoing description of the present disclosure is for illustrative purposes only, and those skilled in the art will understand that modifications can be easily made to other specific forms without altering the technical spirit or essential features of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may likewise be implemented in a combined form.

[0132] The scope of the present disclosure is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present disclosure.

[0133]

Claims

1. Regarding the method for determining whether a product is genuine, A step of collecting XRF data for the above product; A step of generating a judgment model capable of determining authenticity by performing multivariate statistical analysis on the collected XRF data; A step of receiving XRF data for a first product to be determined as genuine; and A step of determining whether the first product is genuine based on XRF data for the first product; comprising Method for determining the authenticity of a product.

2. In Paragraph 1, XRF data for the first product is obtained by measuring at least one metal part included in the first product. Method for determining the authenticity of a product.

3. In Paragraph 1, The XRF data for the first product above is measured using a portable X-ray fluorescence analyzer, Method for determining the authenticity of a product.

4. In Paragraph 1, The method further comprises the step of providing a judgment model capable of determining authenticity and visualizing XRF data for the first product. Method for determining the authenticity of a product.

5. In Paragraph 1, The judgment model capable of determining the above-mentioned authenticity is, A PCA judgment model generated from the result of analyzing the collected XRF data by Principal Component Analysis (PCA) and a PLS-DA judgment model generated from the result obtained by analyzing the XRF data of the product by PLS-DA, comprising at least one of Method for determining the authenticity of a product.

6. In Paragraph 5, The above PCA decision model is, A genuine region and / or counterfeit region established in a coordinate system (score plot) generated by performing Principal Component Analysis (PCA) on the above collected XRF data, Method for determining the authenticity of a product.

7. In Paragraph 6, The above genuine area and / or the above counterfeit area, XRF data corresponding to the above genuine product and / or the above counterfeit product are determined by grouping them after plotting them in the above coordinate system (score plot), Method for determining the authenticity of a product.

8. In Paragraph 6, The step of determining whether the first product is genuine is: Displaying XRF data for the first product above on the score plot above and determining whether it is genuine based on whether it belongs to a genuine area or a counterfeit area, Method for determining the authenticity of a product.

9. In Paragraph 5, The above PLS-DA judgment model is, A mathematical prediction model for an output variable, generated by analyzing XRF data using PLS-DA with the authenticity of the above product as the output variable (Y) and the XRF data as the input variable (X). Method for determining the authenticity of a product.

10. In Paragraph 9, The step of determining whether the first product is genuine is: XRF data for the first product is input into the mathematical prediction model as an input variable (X), and the predicted output variable ( Determining authenticity based on ), Method for determining the authenticity of a product.

11. In Paragraph 5, The step of determining whether the first product is genuine is: A first step of determining authenticity using a pre-stored ingredient table; and If the authenticity cannot be determined in the first step above, a second step of determining the authenticity using the PCA determination model and / or the PLS-DA determination model; comprising Method for determining the authenticity of a product.

12. In Paragraph 11, The above-mentioned pre-stored component table includes at least one of the type, content, and plating thickness of the constituent components, Method for determining the authenticity of a product.

13. In Paragraph 12, If it is confirmed that the similarity between the content of the constituent components of the first product and the constituent components of the ingredient table exceeds a standard value, the first product is determined to be genuine. Method for determining the authenticity of a product.

14. In Paragraph 11, The above second step is, Step 2-1 using the above PCA decision model; and A 2-2 step utilizing the PLS-DA judgment model when the authenticity cannot be determined in the 2-1 step above; comprising Method for determining the authenticity of a product.

15. A program stored on a computer-readable recording medium to execute the method of any one of paragraphs 1 through 14 on a computer.

16. In a system for determining the authenticity of a product, X-ray fluorescence analyzer; Memory; and It includes at least one processor that executes one or more instructions stored in the memory, and The above at least one processor is, Collect XRF data for the above product, and By performing multivariate statistical analysis on the collected XRF data above, a judgment model capable of determining authenticity is generated, and XRF data for the first product to be determined as genuine is received, and Determining whether the first product is genuine based on XRF data for the first product, Product authenticity determination system.

17. In Paragraph 16, The above X-ray fluorescence analyzer is portable, Product authenticity determination system.

18. In Paragraph 16, Further comprising: an output unit indicating the result of determining whether the first product is genuine or not; Product authenticity determination system.

19. In Paragraph 18, The result of determining whether the above-mentioned first product is genuine is, A determination model capable of determining the authenticity of the above product and a visualization of XRF data for the above first product Product authenticity determination system.

20. In Paragraph 16, The judgment model capable of determining the above-mentioned authenticity is, A PCA judgment model generated from the result of analyzing the collected XRF data by Principal Component Analysis (PCA) and a PLS-DA judgment model generated from the result obtained by analyzing the XRF data of the product by PLS-DA, comprising at least one of Product authenticity determination system.

21. In Paragraph 20, The principal components of the above PCA decision model are Z = a1X1 + a2X2...a m X m It is expressed as, [X 1, X 2, ..., X m ] is input data obtained by sampling m samples from the XRF spectrum, and Here, m is a natural number greater than or equal to 3 and less than or equal to 2048, and a i (where 1 ≤ i ≤ m) is greater than or equal to -1 and less than or equal to 1, Product authenticity determination system.

22. In Paragraph 20, The above PLS-DA decision model is Y = w1X1 + w2X2...w n X n It is expressed as, Y is an output variable for determining authenticity, and [X 1, X 2, ..., X n ] is input data obtained by sampling n samples from the XRF spectrum, and Here, n is a natural number at most 2048, and w i (where 1 ≤ i ≤ n) is greater than or equal to -0.000001 and less than or equal to 0.000002, Product authenticity determination system.

Images