Double-layer anti-counterfeit label system and method based on artificial intelligence

Abstract

The invention relates to the technical field of anti-counterfeiting, in particular to a double-layer anti-counterfeiting label system and method based on artificial intelligence, and the system comprises an image acquisition module, a color block analysis module, a particle distribution module and an anti-counterfeiting recognition module. The method comprises the following steps: acquiring an initial anti-counterfeiting image containing an anti-counterfeiting label according to user requirements, and splitting the initial anti-counterfeiting image into a color block frame image, a particle distribution image and a two-dimensional code image; analyzing the color block frame image to obtain an edge feature of the color block frame image, and performing feature matching on the edge feature of the color block frame image and the two-dimensional code image to obtain an edge matching feature; performing matching analysis on metal particle distribution in the particle distribution image to obtain a particle distribution state of the metal particles; performing feature matching on the particle distribution state and the two-dimensional code image to obtain a particle state vector; verifying the initial anti-counterfeiting image based on the edge matching feature and the particle state vector; and the identification precision and reliability of the anti-counterfeit label are improved.

Description

Technical Field

[0001] The present invention relates to the field of anti-counterfeiting technology, and in particular to a double-layer anti-counterfeiting label system and method based on artificial intelligence. Background Art

[0002] In the field of anti-counterfeiting technology, traditional anti-counterfeiting methods often rely on a single anti-counterfeiting method, such as single QR code verification, special ink printing, etc. Although these methods have improved the anti-counterfeiting ability of products to a certain extent, there is still a risk of being imitated or cracked. With the development of artificial intelligence technology, anti-counterfeiting technology has also begun to develop in the direction of intelligence and multi-layering. However, most of the existing intelligent anti-counterfeiting systems only focus on a single level of anti-counterfeiting information, such as only verifying the QR code, or only analyzing special textures, and lack comprehensive anti-counterfeiting verification methods.

[0003] For example, Chinese patent publication No. CN114580597A discloses a method for producing and re-labeling NFC anti-counterfeiting labels, the method comprising: receiving a first login instruction of a first production equipment group; performing a first equipment login check; if the check is successful, performing production task allocation to generate a first production task list and issuing it; receiving a first bad label statistical list upon completing the production task; if the first bad label statistical list is not empty, receiving a second login instruction of the first re-labeling equipment group; performing a second equipment login check; if the check is successful, performing re-labeling task allocation to generate a first re-labeling task list and issuing it; receiving a second bad label statistical list upon completing the re-labeling task; if the first bad label statistical list is empty, the first production task list is used to form a first production task backup record and save it; if it is not empty, the first production task list and the second bad label statistical list are used to form a first production task backup record and save it.

[0004] For example, Chinese patent publication number CN116227524A discloses a label-based anti-counterfeiting system. The method includes: the generator of the anti-counterfeiting label obtains an initial anti-counterfeiting image containing a graphic code for anti-counterfeiting, and splits the initial anti-counterfeiting image into multiple sub-images; the multiple sub-images are divided into multiple times and printed on the label substrate to generate a printed anti-counterfeiting image corresponding to the graphic code, and there is a random offset between the printing positions of the multiple sub-images on the label substrate; the verifier of the anti-counterfeiting label obtains the anti-counterfeiting image to be verified collected by the user through the terminal; obtains the printed anti-counterfeiting image corresponding to the graphic code contained in the anti-counterfeiting image to be verified; and verifies the authenticity of the anti-counterfeiting image to be verified based on the image similarity between the anti-counterfeiting image to be verified and the printed anti-counterfeiting image.

[0005] The prior art describes that when processing anti-counterfeiting labels, the anti-counterfeiting labels can be set through NFC, and that the offset between multiple sub-images can be used for anti-counterfeiting; however, these anti-counterfeiting means have the problem of a single setting means, which is easy to be imitated and cracked. By setting up multi-level comprehensive verification of the content on the anti-counterfeiting label, the recognition accuracy and reliability of the anti-counterfeiting label can be improved. Summary of the invention

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a double-layer anti-counterfeiting label system based on artificial intelligence, including: an image acquisition module, used to collect the initial anti-counterfeiting image containing the anti-counterfeiting label according to user needs, and split the initial anti-counterfeiting image into a color block border image, a particle distribution image and a QR code image.

[0007] The color block analysis module is used to analyze the color block border image, determine the edge features of each sub-image in the color block border image, combine the edge features of each sub-image to obtain the edge features of the color block border image, and perform feature matching between the edge features of the color block border image and the QR code image to obtain the edge matching features of the color block border image.

[0008] The particle distribution module is used to match and analyze the metal particle distribution in the particle distribution image to obtain the particle distribution state of the metal particles; the particle distribution state is feature matched with the two-dimensional code image to obtain the particle state vector.

[0009] The anti-counterfeiting recognition module is used to verify the initial anti-counterfeiting image based on edge matching features and particle state vectors, determine the recognition method of the initial anti-counterfeiting image under different user requirements, and complete the anti-counterfeiting image recognition.

[0010] A double-layer anti-counterfeiting label method based on artificial intelligence includes: S1, collecting an initial anti-counterfeiting image containing an anti-counterfeiting label through an image acquisition module, and splitting it into a color block border image, a particle distribution image and a QR code image.

[0011] S2, analyzing the color block border image, determining the edge features of each sub-image, and performing feature matching between the edge features of each sub-image and the two-dimensional code image to obtain edge matching features.

[0012] S3, performing matching analysis on the metal particle distribution in the particle distribution image, obtaining the particle distribution state, and performing feature matching between the particle distribution state and the two-dimensional code image to obtain a particle state vector.

[0013] S4, based on edge matching features and particle state vectors, verifies the initial anti-counterfeiting image, determines the recognition method according to user needs, and completes the anti-counterfeiting image recognition.

[0014] The beneficial effects of the present invention are as follows: 1. The present invention uses a system to collect an initial anti-counterfeiting image containing an anti-counterfeiting label through an image acquisition module, and splits it into a color block frame image, a particle distribution image, and a two-dimensional code image; the color block frame and metal particles in the initial anti-counterfeiting image are analyzed respectively, and the correlation of the corresponding data bound to the data in the two-dimensional code image can be obtained, which provides a corresponding data basis for the subsequent specific authentication of the anti-counterfeiting label. At the same time, the method of setting an anti-counterfeiting mark on the current anti-counterfeiting label through metal particles and color block frames can reduce the problem of a single setting means of anti-counterfeiting means, making the anti-counterfeiting label difficult to be imitated and cracked.

[0015] Second, the present invention analyzes the color block border image, determines the edge features of each sub-image, and matches these edge features with the two-dimensional code image to obtain edge matching features; this step enhances the reliability of anti-counterfeiting verification by comparing the association between the color block border image and the two-dimensional code image. At the same time, the color block border is generated in an irregular manner, which can improve the setting effect of the anti-counterfeiting label and further reduce the risk of being imitated or cracked.

[0016] 3. The present invention obtains the particle distribution state by matching and analyzing the metal particle distribution in the particle distribution image, and performs feature matching of this state with the two-dimensional code image to obtain a particle state vector. This step utilizes the randomness and non-replicability of the metal particle distribution to further improve the accuracy of anti-counterfeiting verification. At the same time, the vector after the metal particle matching is combined with the matching content of the color block border to complete the anti-counterfeiting image recognition. This step realizes the comprehensive verification of the anti-counterfeiting label by comprehensively considering multiple levels of anti-counterfeiting information. It can also determine the identification method according to user needs, provide different verification modes such as fast verification, detailed analysis, batch processing, etc., to meet the needs of different users. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0018] Figure 1 It is a label schematic diagram of a double-layer anti-counterfeiting label system based on artificial intelligence.

[0019] Figure 2 It is a system schematic diagram of a double-layer anti-counterfeiting label system based on artificial intelligence.

[0020] Figure 3 It is a flow chart of the color block analysis module of a double-layer anti-counterfeiting label system based on artificial intelligence.

[0021] Figure 4 It is a flow chart of the particle distribution module of a double-layer anti-counterfeiting label system based on artificial intelligence.

[0022] Figure 5 It is a flow chart of a double-layer anti-counterfeiting label method based on artificial intelligence. DETAILED DESCRIPTION

[0023] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. If no specific techniques or conditions are specified in the embodiments, the techniques or conditions described in the literature in the art or the product specifications are used.

[0024] See also Figure 1 , Figure 2 , a double-layer anti-counterfeiting label system based on artificial intelligence, including: an image acquisition module, a color block analysis module, a particle distribution module and an anti-counterfeiting identification module; wherein the output end of the image acquisition module is connected to the color block analysis module, the output end of the color block analysis module is connected to the particle distribution module, and the output end of the particle distribution module is connected to the anti-counterfeiting identification module.

[0025] The image acquisition module is used to collect the initial anti-counterfeiting image containing the anti-counterfeiting label according to user needs, and split the initial anti-counterfeiting image into a color block border image, a particle distribution image and a QR code image.

[0026] The color block analysis module is used to analyze the color block border image, determine the edge features of each sub-image in the color block border image, combine the edge features of each sub-image to obtain the edge features of the color block border image, and perform feature matching between the edge features of the color block border image and the QR code image to obtain the edge matching features of the color block border image.

[0027] The particle distribution module is used to match and analyze the metal particle distribution in the particle distribution image to obtain the particle distribution state of the metal particles; the particle distribution state is feature matched with the two-dimensional code image to obtain the particle state vector.

[0028] The anti-counterfeiting recognition module is used to verify the initial anti-counterfeiting image based on edge matching features and particle state vectors, determine the recognition method of the initial anti-counterfeiting image under different user requirements, and complete the anti-counterfeiting image recognition.

[0029] The initial anti-counterfeiting image can be collected through image acquisition and recognition such as a camera, and at the same time, according to the distribution of each part in the initial anti-counterfeiting image, the color block border image and particle distribution image used to identify the authenticity of the current anti-counterfeiting label are extracted; the color block border image and the particle distribution image will be distributed on the left side of the initial anti-counterfeiting image, and the right side of the initial anti-counterfeiting image will contain a QR code and a verification code, which is convenient for rapid identification when authenticating the anti-counterfeiting label. For the color border image, a border image randomly generated by multiple color blocks will be used to encrypt the content corresponding to the current anti-counterfeiting label, and then the distribution form of metal particles will be used as the particle distribution image, and the encrypted information will be represented in the data form of the particle distribution. When any content of the anti-counterfeiting label in the color block border image and the particle distribution image does not correspond, the current anti-counterfeiting label will be considered to be false.

[0030] When performing normal QR code recognition on the identified QR code image, the features existing in the QR code image will be matched with the color block border and the metal particle distribution to further verify whether the irregularly distributed metal particles and color blocks conform to the pre-set rules, thereby completing the identification setting of the anti-counterfeiting label.

[0031] When using an anti-counterfeiting label, a unique and irregular color block is used on one side to form a border. The edges and corresponding areas of each color block need to be identified on this border to verify whether the color block is the content bound to the QR code. When binding, a hash value will be set for the color block border image. Then, when the QR code is identified, it will check whether the image is consistent with the image corresponding to the pre-set QR code to find out whether the border of the current initial anti-counterfeiting image is a genuine product.

[0032] In one embodiment of the present invention, Figure 3 As shown, the implementation method of the edge feature of the color block border image also includes: preprocessing the color block border image and converting the color block border image into a grayscale histogram. The grayscale histogram is to convert the color image into a grayscale image, and filter the grayscale image to reduce the impact of noise on subsequent steps. Common filtering methods include Gaussian filtering, median filtering, etc.

[0033] According to the distribution position of each color block in the color block border image, the color block border area is divided into sub-images corresponding to each color block in turn; and the edge feature of each sub-image is identified.

[0034] The implementation method of identifying the edge features of each sub-image includes: performing contour detection on the color blocks in each sub-image, and setting contour pixel points corresponding to each color block according to the minimum circumscribed rectangle of the contour detection.

[0035] The contour pixels are combined into an edge coordinate sequence, and the shape descriptor and Fourier descriptor of the contour pixels are extracted. The edge coordinate sequence, shape descriptor and Fourier descriptor are mapped to each other and output as edge features of each sub-image.

[0036] The shape descriptor refers to the Hu moment (7-dimensional invariant moment). The Hu moment is a shape descriptor used for image recognition and shape analysis. It is based on the invariant moment theory. The invariant moment is a description of the shape features of an image and can remain invariant when the image is translated, rotated, and scaled. The Hu moment includes seven standardized invariant moments, which are all functions of the geometric moment and are invariant to translation, rotation, and scaling.

[0037] The Hu moment calculates the geometric moment of the image based on the contour information, that is, it is calculated based on the distance between the contour pixels, and then normalizes the geometric moment and calculates seven invariant moments to obtain the shape descriptor.

[0038] The Fourier descriptor uses Fourier transform to apply discrete Fourier transform to the contour point set to obtain the representation of each point in the frequency domain, and extracts the amplitude and phase of the first 20 low-frequency components from the frequency domain data as the Fourier descriptor. This is used to describe the contours of the current color blocks in the corresponding area. It should be noted that each color block here represents a sub-image composed of multiple irregular shapes of the same color, and the information of each color block is identified in this sub-image.

[0039] According to the edge features of each sub-image, the position and area of ​​each color block in each sub-image are calculated; and according to the position and area of ​​each color block, the color block adjacency relationship between the color blocks is constructed, and the color block adjacency relationship describes the adjacency relationship between the color blocks. At this time, the position can use the color block border image that is not divided into sub-images as the coordinate system, such as using the lower left corner of the color block border image as the coordinate origin and the two adjacent sides as the coordinate axes to quantify the position of each color block in all sub-images, and then identify its area according to the pixels contained in each color block.

[0040] According to the color block adjacency relationship between the color blocks, the edge features of each sub-image are spliced ​​and output as the edge features of the color block border image.

[0041] When performing feature stitching, the raster scanning and marking methods are used to connect the edge features of the sub-image through the color block adjacency relationship of multiple color blocks to obtain the topological structure between the color blocks, and according to the topological structure of each color block, the adjacent parts of the edge features are spliced ​​and combined to obtain the edge features that describe the current color block border image.

[0042] At the same time, the edge combination features will also be matched with the relevant data on the QR code. For example, the size and position of each irregular color block in the edge combination features can be verified according to the extended data contained in the QR code to complete further matching of the QR code, ultimately improving the overall recognition speed and anti-counterfeiting effect.

[0043] Therefore, the implementation method of the edge matching feature includes: determining the relative distance and distribution sequence of each sub-image in the color block border image based on the edge features of the color block border image, using the QR code image to obtain the preset edge features, matching the relative distance and distribution sequence of each sub-image with the preset edge features, and determining the relative relationship and distribution quantity of multiple sub-images after matching.

[0044] The relative distance indicates the relative distance between each sub-image and the relative distance between each color block in the sub-image. The distribution composition of each color block indicates the type, quantity and arrangement of the color blocks in the sub-image, indicating the sequence of the corresponding order of the color blocks. Since each color block represents a corresponding graphic under a color, and the graphic is randomly generated in an irregular manner, the type of the graphic will be described according to the edge or shape represented by the graphic, and then the distribution sequence of each sub-image will be generated according to whether there are multiple of the same type and how the graphics are sorted on the border.

[0045] At this time, when outputting the relevant content of the color block border image, a border composed of irregular color blocks will be used to set an identification code corresponding to the QR code image, and generate an initially set preset edge feature. Then, the collected color block border image is compared with the preset edge feature to understand whether the difference between the collected image and the initially set image meets the normal specifications, and to understand the type, quantity and arrangement of each color block at this time to describe the color block border image. Generally, when generating a border composed of irregular color blocks, a fixed-length hash value will be formed according to the distribution ratio of these data and the form of adjacent color blocks. After the color block border image is collected, the corresponding content on this image is compared with the hash value of this description to determine the difference between the current image and the standard image.

[0046] Afterwards, according to the relative distance and distribution sequence of each sub-image, it will be described whether some color blocks appear in the corresponding positions of this sub-image in the image corresponding to the preset edge feature, and whether there are adjacent and symmetrical relationships. The number of these corresponding relationships will be identified to obtain the distribution number of sub-images. This distribution number indicates what kind of matching relationship exists between the current color block border image and the preset edge feature, and the existence of the relationship and the corresponding number are recorded.

[0047] The implementation method of determining the relative relationship and distribution quantity of multiple sub-images after matching includes: using a feature matching algorithm to calculate the similarity between the relative distance and distribution sequence of each sub-image during matching and the preset edge feature, and the calculated similarity uses Euclidean distance and Hamming distance to describe the similarity between these features. If the similarity of the relative distance and the distribution sequence is described at this time, the cosine similarity is used to calculate the similarity under the relative distance, and the consistency of the sequence is used to describe the similarity under the type, quantity and arrangement of the color blocks, that is, the value of the consistent and similar characters occupying the entire sequence length, to describe the similarity that the distribution sequence can correspond to, and the position of this part of the corresponding data on the sub-image and the preset edge feature can also be used to calculate the sum of squares of the difference to be used as the matching result described at this time. The matching results of each sub-image and the preset edge feature are obtained, and the relative relationship of each sub-image, such as adjacent, symmetrical, etc., is constructed using breadth-first search according to the matching results of each sub-image and the preset edge feature; the breadth-first search is to obtain a relationship diagram that can describe the features on all color blocks by continuously combining and connecting the results of whether there is an adjacent and symmetrical relationship between the corresponding color blocks in each sub-image and the preset edge feature. The data of different relative relationships on the relationship graph are measured to obtain the distribution quantity of the sub-images. Then, the data in the matching process can be classified according to these relative relationships and distribution quantities to identify whether there are corresponding errors.

[0048] The relative relationship and distribution quantity of the sub-images are used to perform consistency verification, identify the relevant particle size differences and area ratio differences of the sub-images, and obtain the degree of difference of the sub-images.

[0049] For sub-images, the relevant granularity difference represents the average value of the pixel difference between the corresponding distribution numbers of color blocks in the corresponding relative relationship of the sub-images; the area ratio difference is also the average value of the difference ratio of the areas contained in its color blocks. The values ​​corresponding to the relevant granularity difference and the area ratio difference are taken as the degree of difference of the sub-images.

[0050] It is determined whether the sub-image with the largest difference is a normal change. If it is a normal change, an anti-counterfeiting mark is set for the color block border image, and the distribution sequences of each sub-image are combined and output as the edge matching features of the color block border image.

[0051] The edge matching features of the final output will indicate under what circumstances the current color block border image differs from the preset edge features, and the specific value of the difference, so as to increase the difficulty of counterfeiting the current color block border image. As for judging whether the sub-image with the largest degree of difference is a normal change, when the error in the degree of difference exceeds 0.1%, it means that the currently identified color block border image has an abnormality and is not a normal change; when the error value is less than 0.1%, it means that the currently identified color block border image is an image that normally matches the QR code and is a normally produced product.

[0052] At this time, 0.1% means that the color block border image currently being identified should have a very small error with the content bound to the QR code. This 0.1% error may be caused by the way the image is processed after it is taken. Only when the error is so small that it can be ignored can the color block border be considered normal, thereby completing the anti-counterfeiting identification.

[0053] When the difference is normal after matching, the sequence of specific distribution of color blocks in the color block border image will be selected for output to assist the metal particles in completing the double-layer authentication of the anti-counterfeiting image.

[0054] In one embodiment of the present invention, the particle distribution module is mainly used to identify the distribution state of metal particles in the particle distribution image. After obtaining a coordinate sequence, the random auxiliary index of each position and the combined index of each position are determined to describe the anti-counterfeiting information represented by the metal particles.

[0055] like Figure 4 As shown, the matching analysis of the distribution of metal particles in the particle distribution image includes: according to the position of the metal particles in the particle distribution image, the coordinates corresponding to each metal particle are identified to generate a particle coordinate sequence; the particle distribution image represents an image of irregular distribution of metal particles in a specific area, and the distribution state of the metal particles is used as an anti-counterfeiting method for the anti-counterfeiting label. And the metal particles are of the same size, and in the specific area, the lower left corner is used as the starting point, and the area where the metal particles are located is divided into multiple rectangles of the same size, and the coordinates corresponding to each metal particle are identified, and the coordinates of these metal particles are input into the coordinate sequence to obtain the coordinates corresponding to the metal particles.

[0056] Record the distance and angle between each metal particle, convert the particle coordinate sequence into a topological structure matrix, and connect each metal particle in the topological structure matrix to output the particle distribution state. The topological structure matrix is ​​mainly used to combine the data of the position, distance, angle and adjacency relationship of each metal particle to describe the specific distribution of the current metal particles, so as to facilitate the subsequent analysis of the actual distribution of metal particles and the corresponding data of the QR code image to determine whether the distribution of metal particles meets the preset rules.

[0057] The implementation method of feature matching the particle distribution state with the two-dimensional code image includes: based on the distance between each metal particle in the particle distribution state, the particle distribution state is associated with the data corresponding to the two-dimensional code image to generate a relational data set of the particle distribution state; the relational data set is to associate the topological structure matrix corresponding to the particle distribution state with the corresponding data set in the two-dimensional code image to determine whether the current distribution of the metal particles is correct, and the data corresponding to the two-dimensional code image is the image of the metal particle distribution, which is used to verify the distribution of the metal particles in the current particle distribution image.

[0058] The position index of the particle distribution state is obtained from the relational data set, and the position index is combined to obtain each position combination index. The position index represents the position of the metal particles in the current particle distribution state and the position of the corresponding data in the QR code image. Each position index represents a numerical pair that associates the coordinates of these two positions; each position combination index is a statistical analysis of these data. For example, each position index contains the physical coordinates of the metal particles and the logical coordinates of the data block corresponding to the QR code. After that, for each position index, the deviation distance and deviation angle of the two coordinates corresponding to the position index are calculated as the deviation distance and deviation angle of each position index, and these data are combined into each position combination index.

[0059] The particle distribution state is classified using the combined index of each position, and the average deviation distance and average deviation angle of the position index under each classification are calculated.

[0060] According to the average deviation distance and average deviation angle of the position index under each category, a position code is set for each category; according to the dimension corresponding to each category, the position codes are combined into a particle distribution vector.

[0061] When classifying, the combined index of each position is used to classify the particle distribution state. The implementation method includes: multi-dimensional hierarchical clustering of the particle distribution state, and sequentially performing clustering of spatial position, distribution mode and deviation angle association.

[0062] The current metal particle distribution area is cut into multiple grid areas of the same size. There are metal particles in each grid area. The average deviation distance and average deviation angle of the corresponding position index of the metal particles in each grid area are calculated; this is used as the clustering result of the spatial position.

[0063] The grid area is converted into a density histogram according to the density of metal particles, and KL divergence clustering is performed using the density histogram. The KL divergence corresponding to the probability distribution of the values ​​of metal particles in the grid area and the area where the metal particles are distributed is calculated, and the metal particles in each grid area are clustered according to the value of KL divergence to obtain multiple clusters. The average deviation distance and average deviation angle of the corresponding position index of the metal particles in each cluster are output to obtain the clustering result of the distribution pattern.

[0064] According to the deviation angle of the position index, a deviation angle association matrix is ​​generated for each grid area, and a principal component analysis is performed on the deviation angle association matrix of the adjacent grid area. The average deviation distance and average deviation angle of the position index corresponding to the output data of the principal component analysis are calculated as the clustering result of the deviation angle association. When the deviation angle association matrix is ​​generated, according to the deviation angle existing in the position index, a corresponding deviation angle association matrix is ​​generated for the metal particles existing in the current grid area, and then the principal component analysis is performed on the deviation angle association matrix in the adjacent grid area. The data with the highest output contribution under the principal component analysis are used as the clustering result corresponding to the adjacent grid area, and the deviation angle and deviation distance on the corresponding position index are averaged to obtain the clustering result under the deviation angle association. At the same time, when performing principal component analysis and outputting corresponding data, the output will be based on the cumulative variance contribution rate of the output features, that is, the ratio of the sum of the features of the output data to the sum of all eigenvalues. At this time, 70% will be set as the threshold of the cumulative variance contribution rate to select the deviation angle association matrix of adjacent grid areas when the principal component analysis is used to convert it into the corresponding eigenvalue. Finally, multiple cluster-like data combinations are obtained, and the deviation distances and deviation angles corresponding to these data combinations are averaged to obtain the analysis results of the corresponding data at this time.

[0065] Position codes are set for the average deviation distance and average deviation angle of the position index corresponding to the clustered data in turn, and the position codes are combined according to the clustering dimension to obtain a particle distribution vector. The particle distribution vector represents the combination of corresponding data after clustering in multiple dimensions. Finally, these values ​​can be converted into a specific hash value to represent the result of metal particle identification.

[0066] At the same time, the implementation method of the particle distribution module also includes: identifying the texture of each metal particle in the particle distribution image, obtaining the etching characteristics of each metal particle under the particle distribution state, and the etching characteristics represent the texture of the metal particle surface after laser etching, and then performing hierarchical analysis according to the etching characteristics and metal particle distribution, respectively performing distribution layer verification and etching layer analysis, and completing the analysis of the particle distribution state based on the analysis results of the two levels. Since the texture corresponding to laser etching is mostly fine content, an optical microscope or an electron microscope can be used to magnify and observe the metal particles at this time, and the traces of laser etching can be clearly seen.

[0067] This processing method is generally used when suppliers and distributors check product anti-counterfeiting labels, and is not applicable to the camera usage scenario described above. The above description can be a scenario in which users use mobile phones or other devices to scan QR codes and color block border images, as well as particle distribution images after purchasing products to complete anti-counterfeiting label verification.

[0068] The particle distribution state is analyzed hierarchically according to the etching characteristics of each metal particle, and the distribution layer analysis result and the etching layer analysis result are determined in turn.

[0069] The distribution layer analysis results represent the relevant values ​​of the particle distribution vector calculated using the position index, and the etching layer analysis results identify the holographic image on the metal particles and determine whether the holographic image is consistent with the data features bound to the QR code to complete the identification results of the etching layer.

[0070] For example, the etching layer analysis result is represented by matching the etching features with the image features in the QR code image binding data, calculating the similarity of the etching features, and when the similarity of the etching features is greater than 95%, the current etching layer analysis result is considered normal. To calculate the similarity, the etching features can be represented by a grayscale histogram, and the pixel value of the etching feature can be compared with the value of the corresponding pixel in the database, and the cosine similarity calculation method can be used to complete the feature matching.

[0071] According to the analysis results of the distribution layer and the etching layer, the particle state vector is constructed.

[0072] At this time, when using two levels to construct the particle distribution state vector, since the distribution layer analysis result will indicate the process of constructing the particle state vector using the position index, the etching layer analysis result will be used to perform data identification on the content in the distribution layer analysis result. For example, when there is an inconsistency between the etching layer analysis result and the data bound to the QR code, the metal particles corresponding to the inconsistency are identified, and the metal particles at the corresponding position are recorded, and the corresponding image data is uploaded to facilitate subsequent analysis of the reasons for the inconsistency in the current anti-counterfeiting label during the etching analysis; if they are consistent, the current metal particle analysis result is considered normal, and the content of the distribution layer analysis result is used to construct the particle state vector.

[0073] In one embodiment of the present invention, after completing the identification of the color block border and metal particles on the anti-counterfeiting label, it is possible to directly determine whether the current anti-counterfeiting label is valid based on the two identification results. At the same time, since different users authenticate different contents of the anti-counterfeiting label, the verification method of the anti-counterfeiting label can be adjusted.

[0074] For example, ordinary users can directly scan the QR code and the corresponding color block border for verification, or in addition to verifying the color block border, they also need to verify the distribution of metal particles. In this case, they can directly use a normal mobile phone device to verify through specific software; if it is a manufacturer or dealer who verifies, they will not only directly verify the color block border and metal particles, but also verify the texture on the metal particles, and use the results of the three verifications to determine whether the anti-counterfeiting label is valid.

[0075] Therefore, the implementation methods of the anti-counterfeiting identification module include: obtaining the user's identification requirements, such as quick verification, detailed analysis, batch processing, etc., and determining the identification method of the current initial anti-counterfeiting image according to the user's identification requirements; for quick verification requirements, the matching steps and output results can be simplified; for detailed analysis requirements, detailed matching reports and particle status information can be provided; for batch processing requirements, the processing flow can be optimized to improve efficiency.

[0076] According to the recognition method of the current initial anti-counterfeiting image, the initial anti-counterfeiting image is verified in turn using edge matching features and particle state vectors.

[0077] For example, when fast verification is required, verification can be completed only through the output content of the edge matching feature. When the edge matching feature is obtained, it will be determined whether the color block border is compared with the data bound to the QR code image, that is, the edge matching feature will contain a successful match mark. When fast verification is required, verification can be completed only in this way.

[0078] Detailed analysis will use edge matching features and particle state vectors simultaneously to complete the preliminary anti-counterfeiting image verification. For the particle state vector, a set of position-coded data will be output, or converted into a position code of a hash value. This value can be directly compared with the data bound to the QR code. If they are consistent, the preliminary anti-counterfeiting image verification is completed.

[0079] If it is batch processing, for a batch of anti-counterfeiting labels to be identified, the identification mark of the etching feature will be searched in the particle state vector, and the edge matching feature of the color block border will be used to find the deficiencies in the production of anti-counterfeiting labels in the current batch, so as to complete the comprehensive authentication of the anti-counterfeiting labels.

[0080] like Figure 5 As shown, the present invention also provides a double-layer anti-counterfeiting label method based on artificial intelligence, including: S1, collecting an initial anti-counterfeiting image containing an anti-counterfeiting label through an image acquisition module, and splitting it into a color block border image, a particle distribution image and a QR code image.

[0081] S2, analyzing the color block border image, determining the edge features of each sub-image, and performing feature matching between the edge features of each sub-image and the two-dimensional code image to obtain edge matching features.

[0082] S3, performing matching analysis on the metal particle distribution in the particle distribution image, obtaining the particle distribution state, and performing feature matching between the particle distribution state and the two-dimensional code image to obtain a particle state vector.

[0083] S4, based on edge matching features and particle state vectors, verifies the initial anti-counterfeiting image, determines the recognition method according to user needs, and completes the anti-counterfeiting image recognition.

[0084] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.

Claims

1. A double-layer anti-counterfeiting label system based on artificial intelligence, characterized in that: include: An image acquisition module is used to collect an initial anti-counterfeiting image including an anti-counterfeiting label according to user needs, and split the initial anti-counterfeiting image into a color block border image, a particle distribution image and a QR code image; The color block analysis module is used to analyze the color block border image, determine the edge features of each sub-image in the color block border image, combine the edge features of each sub-image to obtain the edge features of the color block border image, perform feature matching on the edge features of the color block border image and the two-dimensional code image, and obtain the edge matching features of the color block border image; The particle distribution module is used to match and analyze the metal particle distribution in the particle distribution image to obtain the particle distribution state of the metal particles; perform feature matching between the particle distribution state and the two-dimensional code image to obtain a particle state vector; The anti-counterfeiting recognition module is used to verify the initial anti-counterfeiting image based on edge matching features and particle state vectors, determine the recognition method of the initial anti-counterfeiting image under different user requirements, and complete the anti-counterfeiting image recognition.

2. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 1, characterized in that: The implementation methods of edge features of color block border images include: According to the distribution positions of the color blocks in the color block border image, the color block border area is divided into sub-images corresponding to the color blocks in sequence; and the edge features of each sub-image are identified; According to the edge features of each sub-image, the position and area of ​​each color block in each sub-image are calculated; and according to the position and area of ​​each color block, the color block adjacency relationship between the color blocks is constructed; According to the color block adjacency relationship between the color blocks, the edge features of each sub-image are spliced ​​and output as the edge features of the color block border image.

3. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 2 is characterized in that: The implementation methods of identifying the edge features of each sub-image include: Perform contour detection on the color blocks in each sub-image, and set the contour pixel points corresponding to each color block according to the minimum circumscribed rectangle of the contour detection; The contour pixels are combined into an edge coordinate sequence, and the shape descriptor and Fourier descriptor of the contour pixels are extracted. The edge coordinate sequence, shape descriptor and Fourier descriptor are mapped to each other and output as edge features of each sub-image.

4. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 1, characterized in that: The implementation methods of edge matching features include: Based on the edge features of the color block border image, the relative distance and distribution sequence of each sub-image in the color block border image are determined, the preset edge features are obtained using the QR code image, the relative distance and distribution sequence of each sub-image are matched with the preset edge features, and the relative relationship and distribution quantity of multiple sub-images after matching are determined; Using the relative relationship and distribution number of sub-images, consistency verification is performed to identify the relevant particle size differences and area ratio differences of the sub-images, and obtain the degree of difference of the sub-images; It is determined whether the sub-image with the largest difference is a normal change. If it is a normal change, an anti-counterfeiting mark is set for the color block border image, and the distribution sequences of each sub-image are combined and output as the edge matching features of the color block border image.

5. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 1, characterized in that: The matching analysis of the metal particle distribution in the particle distribution image can be realized by: According to the position of the metal particles in the particle distribution image, the coordinates corresponding to each metal particle are identified to generate a particle coordinate sequence; The distance and angle between each metal particle are recorded, the particle coordinate sequence is converted into a topological structure matrix, and each metal particle in the topological structure matrix is ​​connected and output as a particle distribution state.

6. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 1, characterized in that: The implementation methods of feature matching between particle distribution state and QR code image include: Based on the distance between each metal particle in the particle distribution state, the particle distribution state is associated with the data corresponding to the two-dimensional code image to generate a relational data set of the particle distribution state; Obtaining position indexes of particle distribution states from the relational data set, combining the position indexes to obtain position combination indexes; Use the combined index of each position to classify the particle distribution state, and calculate the average deviation distance and average deviation angle of the position index under each classification; According to the average deviation distance and average deviation angle of the position index under each category, a position code is set for each category; according to the dimension corresponding to each category, the position codes are combined into a particle distribution vector.

7. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 6, characterized in that: The implementation methods for classifying the particle distribution state include: Multi-dimensional hierarchical clustering of particle distribution states is performed, and spatial position, distribution pattern and deviation angle correlation clustering are performed in turn; The current metal particle distribution area is cut into multiple grid areas of the same size, each of which contains metal particles. The average deviation distance and average deviation angle of the corresponding position index of the metal particles in each grid area are calculated as the clustering result of the spatial position; The grid area is converted into a density histogram according to the density of the metal particles, and the density histogram is used for KL divergence clustering. The KL divergence corresponding to the probability distribution of the values ​​of the metal particles in the grid area and the area where the metal particles are distributed is calculated, and the metal particles in each grid area are clustered according to the value of the KL divergence to obtain multiple clusters. The average deviation distance and average deviation angle of the corresponding position index of the metal particles in each cluster are output to obtain the clustering result of the distribution pattern; According to the deviation angle of the position index, the deviation angle association matrix of each grid area is generated, and the principal component analysis is performed on the deviation angle association matrix of adjacent grid areas. The average deviation distance and average deviation angle of the position index corresponding to the principal component analysis output data are calculated as the clustering result of the deviation angle association.

8. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 1, characterized in that: The particle distribution module can also be implemented by: Identify the texture of each metal particle in the particle distribution image to obtain the etching characteristics of each metal particle under the particle distribution state; Performing a hierarchical analysis on the particle distribution state according to the etching characteristics of each metal particle, and determining the distribution layer analysis result and the etching layer analysis result in turn; According to the analysis results of the distribution layer and the etching layer, the particle state vector is constructed.

9. The double-layer anti-counterfeiting label system based on artificial intelligence according to claim 1, characterized in that: The implementation methods of the anti-counterfeiting identification module include: Obtaining the user's identification requirements, and determining the identification method of the current initial anti-counterfeiting image according to the user's identification requirements; According to the recognition method of the current initial anti-counterfeiting image, the initial anti-counterfeiting image is verified in turn using edge matching features and particle state vectors.

10. A double-layer anti-counterfeiting label method based on artificial intelligence, characterized in that: include: S1, collecting an initial anti-counterfeiting image including an anti-counterfeiting label through an image acquisition module, and splitting it into a color block border image, a particle distribution image and a QR code image; S2, analyzing the color block border image, determining the edge features of each sub-image, and matching the edge features of each sub-image with the two-dimensional code image to obtain edge matching features; S3, performing matching analysis on the metal particle distribution in the particle distribution image to obtain the particle distribution state, and performing feature matching between the particle distribution state and the two-dimensional code image to obtain a particle state vector; S4, based on edge matching features and particle state vectors, verifies the initial anti-counterfeiting image, determines the recognition method according to user needs, and completes the anti-counterfeiting image recognition.

Images