A three-dimensional code element label with self-anti-counterfeiting function, a generation method and a verification method thereof

Abstract

A three-dimensional code label with self anti-fake function, comprising the following parts: a medium carrier formed by standard printing process, the medium carrier surface randomly having at least three positioning patterns in regular geometric shapes, the center coordinates of multiple positioning patterns being able to be constructed into a verification area in square structure, the verification area being uniformly divided into multiple sub-areas; a three-dimensional code pattern formed by randomly fixing colored particles mixed with glue on the surface of the medium carrier by letterpress printing process; a verification code area including three rows of verification code encoding: the first row of verification code containing more than 13 bits of information encoding; the second row of verification code containing more than 21 bits of check encoding, multiple check encodings including total particle number encoding in the verification area, particle number encoding in each sub-area, anti-fake verification encoding; the third row of verification code containing more than 25 bits of coordinate encoding.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional code labels, and in particular to a three-dimensional code element label with a self-anti-counterfeiting function, a generation method thereof, and a verification method thereof. Background Art

[0002] Anti-counterfeiting technologies fall into seven categories: material, printing, laser, digital, electronic identification, texture, and security thread digital information. The current mainstream approach involves attaching a QR code to a product, which consumers scan to authenticate the product and obtain traceability information. Since the QR code algorithm is open source, counterfeiters can directly generate the code. However, for QR codes that use encryption algorithms, counterfeiters can use instruments (such as QR code copiers and high-definition cameras) to capture the printed QR code and obtain an electronic image of the code. They can then use image correction software to make adjustments to the electronic image and then print the QR code image using the same material, resulting in a counterfeit QR code.

[0003] In the prior art, a method for generating a 3D code, such as that disclosed in application number 2021108820314, includes the following steps: a code generator obtains product information and a background image; the code generator generates a 2D code matrix based on the product information; the code generator fuses the background image and the 2D code matrix to obtain a 3D code; the code generator selects a plurality of first specific regions and a plurality of second specific regions on the 3D code; and the color values ​​of each first specific region and each second specific region are recorded. The 3D code is then associated with the product, and the 3D code is used for product anti-counterfeiting and traceability.

[0004] First, while the aforementioned 3D code labels can integrate three elements of information, they essentially still use QR code technology, simply adding a product background image. This merely increases the difficulty of anti-counterfeiting, but their effectiveness is limited. Second, label generation requires a product background image, making the label highly dependent on the product. This not only increases the cost of label production but also significantly limits the scope of the label itself. Summary of the Invention

[0005] In order to solve the above problems, the present invention provides a three-dimensional code element label with self-anti-counterfeiting function, a generation method and a verification method thereof:

[0006] A three-dimensional code element label with a self-anti-counterfeiting function, comprising the following parts:

[0007] A medium carrier formed by a standard printing process, wherein the surface of the medium carrier is randomly provided with at least three regular geometrically shaped positioning patterns, wherein the center coordinates of the plurality of positioning patterns can be used to form a square-shaped verification area, wherein the verification area is evenly divided into a plurality of sub-areas;

[0008] a three-dimensional code pattern formed by colored particles randomly fixed on the surface of the medium carrier after mixing with glue by the letterpress printing process, the three-dimensional code pattern having a three-dimensional concave-convex touch, the colored particles randomly distributed in the verification area, the colored particles having a particle size of 0.1-5.5 mm;

[0009] a verification code area including three rows of verification code encoding:

[0010] a first row of verification code containing 13-bit or more information encoding, the multiple information encoding including one or more of enterprise code, production time or production serial number;

[0011] a second row of verification code containing 21-bit or more check encoding, the multiple check encoding including total number of particles in the verification area encoding, particle number in each sub-area encoding, and anti-counterfeiting verification encoding generated according to existing encoding by encryption algorithm;

[0012] a third row of verification code containing 25-bit or more coordinate encoding, the multiple coordinate encoding including a first field value and center coordinates of four or more specific particles, the first field value used to determine the sub-area where the multiple specific particles are located, the specific particles being four or more, and the coordinate information extracted and encoded by Hough transform algorithm;

[0013] the verification code encoding displayed in plaintext or ciphertext by encryption algorithm.

[0014] Preferably, the positioning pattern is any one of a circle, a triangle, a square or a rectangle, when the positioning pattern is a circle, the diameter of the circle is greater than one time of the average diameter of the colored particles; when the positioning pattern is a polygon, the minimum side length of the polygon is greater than one time of the average diameter of the colored particles.

[0015] Preferably, the first check code of the second row of verification code is a division code used to determine the type of equal division of the verification area, the division code being greater than zero. The division code is non-0-bit, when indicating non-0, the placeholder meaning is reserved, indicating 1-F, a total of 15 types. The division code determines the type of equal division of the verification area by its own table, wherein the content of the table can be set by the user as needed. For example, when the division code is 1, it indicates that the square verification area is divided into 9 equal sub-areas;

[0016] total number of particles in the verification area encoding, 2-bit or more, used to indicate the total number of colored particles in the verification area;

[0017] Region particle quantity encoding, 18 bits or more, representing the particle quantity in each sub-region of the verification region division, the particle quantity being encoded in decimal or hexadecimal;

[0018] Anti-counterfeiting verification code, 4 bits or more, generated by extracting part of the hash value through hash operation on the aforementioned code.

[0019] Preferably, the third line of the verification code includes a first field value and the center coordinates of four or more specific particles, the first field value coordinates using 16 hexadecimal, the first field value range is 0-F, the first field value is used to determine the sub-region of the plurality of specific particles by the provided lookup table 1, the lookup table 1 is set by the user, the lookup table 1 contains the first field value and the sub-region label of the plurality of specific particles corresponding to the first field value; the center coordinates of each specific particle are represented by 6 hexadecimal numbers, the first 3 bits are the horizontal coordinates, and the last 3 bits are the vertical coordinates, the selection rule of the specific particle is: according to the X-axis coordinate from left to right, Y-axis from top to bottom, when the serial number is odd, odd+1, then divided by 2; when the serial number is even, directly divided by 2, the obtained number is the serial number of the specific particle center point to be taken, and the coordinates of the specific particle center point are taken out and coded into the coding table.

[0020] Preferably, the Hough transform algorithm includes the following steps:

[0021] (1) Perform grayscale, Gaussian blur and Canny edge detection processing on the verification region image;

[0022] (2) Detect the center coordinates and radius of the colored particles by Hough gradient method, and screen the particles that meet the preset radius range;

[0023] (3) Convert the detection results to the original image coordinate system, and record the particle quantity and position.

[0024] Preferably, the encryption algorithm uses symmetric encryption, asymmetric encryption, hybrid encryption system, national encryption algorithm or hash algorithm to generate ciphertext encoding.

[0025] Preferably, the label is applied to product origin identification, anti-counterfeiting authentication, meta-universe scene mapping, or unique coding binding scene with one-dimensional code and two-dimensional code.

[0026] A method for generating the three-dimensional code label, comprising the following steps:

[0027] S1: Prepare a medium carrier, form a medium carrier substrate by printing process, and print a positioning pattern on the surface of the medium carrier substrate to delimit a verification region;

[0028] S2: randomly distributing colored particles on the surface of the medium carrier by a relief printing process to form a three-dimensional pattern with a concave-convex tactile feel;

[0029] S3: using a Hough transform algorithm to extract the quantity, position and center coordinate information of the colored particles in the three-dimensional pattern;

[0030] S4: Generate a verification code according to the extracted information, the verification code including an information code, a check code and a coordinate code, and print the verification code on the medium carrier.

[0031] A method for verifying the three-dimensional code element tag comprises the following steps:

[0032] S1: Acquire the label image through machine vision recognition technology, identify the positioning pattern and delineate the verification area;

[0033] S2: Identify the information code, check code and coordinate code in the verification area, and parse the particle quantity, position and center coordinate information of the three-dimensional pattern contained in the verification code;

[0034] S3: Using the Hough transform algorithm to extract the number, position and center coordinate information of the particles in the three-dimensional pattern within the verification area;

[0035] S4: The extracted information is compared with the data parsed from the verification code for verification. If they are consistent, it is judged to be true, otherwise it is judged to be false.

[0036] Preferably, the comparison and verification includes:

[0037] In the networked mode, the extracted label images are compared with the original generated images in the database;

[0038] In offline mode without a network environment, the particle quantity and location information are inferred by verifying the code and compared with the current image extraction value.

[0039] The three-dimensional code element label generation and verification method with anti-counterfeiting function provided by the present invention has the following significant beneficial effects compared with existing one-dimensional codes, two-dimensional codes and ordinary anti-counterfeiting labels: the present invention achieves an anti-counterfeiting effect that is extremely difficult to imitate, quickly identified, and widely applicable through a comprehensive anti-counterfeiting solution of physical randomness (three-dimensional concave-convex particles) + high-precision machine vision + high-strength encryption (SM4 national secret algorithm and Hough transform algorithm) + multi-mode verification (online / offline / touch). It can be widely used in luxury goods anti-counterfeiting, food traceability, drug authenticity, metaverse product authentication and other fields, and has extremely high market application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a schematic diagram of the three-dimensional code element label structure.

[0041] Figure 2 This is a diagram of the verification area of ​​the meta tag.

[0042] Figure 3 This is a diagram illustrating the encoding of the meta tag.

[0043] Figure 4 Schematic diagram of Hough transform algorithm application. DETAILED DESCRIPTION

[0044] The following is combined with Figure 1-2 The present invention is further illustrated with examples.

[0045] A three-dimensional code element label with a self-anti-counterfeiting function, comprising the following parts:

[0046] A medium carrier, the medium carrier is formed by a standard printing process, and the surface of the medium carrier randomly has at least three positioning patterns with regular geometric shapes, and the center coordinates of multiple positioning patterns can be constructed into a verification area with a square structure. The positioning pattern is any one of a circle, a triangle, a square or a rectangle. When the positioning pattern is a circle, its diameter is more than twice the average diameter of the colored particles; when the positioning pattern is a polygon, its minimum side length is more than twice the average diameter of the colored particles. In this embodiment, a circle is taken as an example, and the positioning pattern is a positioning circle. The positioning circle is printed on the medium carrier in a standard process of inkjet or printing to form a standard. When printing, the diameter of the circle needs to be more than twice the average particle diameter to facilitate machine reading and differentiation / partitioning; according to the center point of each positioning circle, a rectangular area with a predetermined length and width is constructed as the verification area for the next step, and then the verification area is evenly divided into multiple sub-areas.

[0047] A three-dimensional code pattern is formed by randomly fixing colored particles on the surface of the medium carrier through a letterpress printing process and mixing them with glue. The three-dimensional code pattern has a three-dimensional concave and convex tactile feel. The colored particles are randomly distributed in the verification area, and the particle size of the colored particles is 0.1 to 5.5 mm.

[0048] by Figure 3 For example, the verification code area includes three lines of verification code:

[0049] The first line of verification code contains 21 bits of information code, which is a plain text code. The information code can be expanded. The 21 bits of information code include the enterprise code, production time, and production serial number in sequence. The enterprise code can be composed of the original 13-digit one-dimensional barcode of the product, or it can be determined by the enterprise itself when using it. The production time is a 6-digit code, and the production serial number is a 2-digit code. Figure 3The first line of verification code for a certain company may be: 48970313524**-250308-03, which represents the product code of the third production batch of this company on March 8, 2025.

[0050] The second line of verification code contains a 25-digit verification code, which consists of the following parts in order:

[0051] The first check code is used to determine the division code of the verification area, and the division code is greater than zero. The division code is a non-zero bit. When it indicates that it is not 0, the placeholder is reserved, indicating a total of 15 types from 1 to F. The division code determines the verification area division type through its own table, and the table content can be set according to user needs. Figure 3 As shown, when the division code is 1, it means that the square verification area is divided into 9 equally divided sub-areas.

[0052] The 2nd and 3rd check codes are the total number of particles in the verification area. The total number of particles in the verification area has 2 bits and is used to represent the total number of colored particles in the verification area. The lower limit is 000 and the upper limit is 255. It is calculated in hexadecimal code. Figure 3 As shown in the figure, the 1st to 3rd check code is 17E vertically, indicating that the verification area image is divided into 9 sub-areas, and the number of particles recognized by the program is 7E in hexadecimal and 126 in decimal.

[0053] The regional particle number code has 18 bits because this embodiment has 9 sub-regions, indicating the number of particles in each of the 9 sub-regions divided into the verification region. Figure 3 As shown in the figure, the regional particle number code is 1412090C0C110D0E0B, which means that the number of particles identified by the algorithm in each of the 9 sub-regions is 14, 12, 09, 0C, 0C, 11, 0D, 0E, and 0B in hexadecimal code.

[0054] The 4-digit security verification code is generated by performing a standard hash operation on the preceding coded digits. The first and last three digits of the hash value (the first, second, and third digits from the right) are used to form the 4-digit security verification code area. The preceding coded digits described in this application can be either an information code, a verification code, or even a coordinate code, and can be customized by the user.

[0055] The third line of verification code contains a coordinate code of more than 25 bits. The multiple coordinate codes include the first bit field value and the center coordinates of more than 4 specific particles. The first bit field value is used to determine the sub-area where the multiple specific particles are located. There are more than 4 specific particles. The coordinate information is extracted and encoded using the Hough transform algorithm.

[0056] The third row of the verification code includes a first bit field value and the center coordinates of four or more specific particles. The first bit field value coordinates adopt 16 hexadecimal, and the first bit field value ranges from 0 to F. The first bit field value is used to determine the sub-region of the plurality of specific particles by the provided lookup table 1, which is set by the user. The lookup table 1 contains the first bit field value and the sub-region label of the plurality of specific particles corresponding to the first bit field value. The center coordinates of each specific particle are represented by 6 hexadecimal numbers, of which the first 3 are the horizontal coordinates and the last 3 are the vertical coordinates. The selection rule of the specific particles is as follows: the sequence number is marked from left to right along the X-axis and from top to bottom along the Y-axis. When the sequence number is odd, the odd number plus 1 and then divided by 2. When the sequence number is even, it is directly divided by 2, and the obtained number is the sequence number of the specific particle center point. The coordinates of the specific particle center point are taken out and coded into the coding table. For example, when the first bit field value is 1, the coordinates of the specific particle center points in the four sub-regional grids with labels 1, 2, 7 and 9 need to be taken out. The lower left corner is the origin coordinate, the lower left corner is 000, and each 6 bits is a group of coordinate values, a total of 4 groups. The first three bits of each group are the horizontal coordinates, and the last three bits are the vertical coordinates. The value range of the horizontal and vertical coordinates is 000-FFF.

[0057] The above verification code coding forms a string of numbers and English letters (including uppercase and lowercase) through a complete encryption algorithm and operation rules. The embodiment has 21-bit information coding, 21-bit verification coding and 25-bit coordinate coding to finally form a ciphertext / plaintext coding of not less than 50 bits, which is printed on the medium carrier synchronously, including but not limited to being printed on the upper, lower, left and right of the pattern or the pattern itself. The technical solution can not only change all data from 0-F to ciphertext through the self-set lookup table 2, but also can convert the above plaintext to ciphertext by using the national standard SM4 encryption algorithm. The embodiment does not convert ciphertext by using the lookup table 2, and the lookup table 2 is only an example of primary encryption.

[0058] As shown in Figure 4 The Hough transform algorithm of the embodiment includes the following steps:

[0059] (1) Perform grayscale, Gaussian blur and Canny edge detection processing on the verification region image;

[0060] (2) Detect the center coordinates and radius of the colored particles by the Hough gradient method, and screen the particles that meet the preset radius range;

[0061] (3) Convert the detection result to the original image coordinate system, and record the number and position of the particles.

[0062] The following is a part of the program for extracting the center coordinates of small particles:

[0063] import cv2

[0064] import numpy as np

[0065] def draw_rectangle_and_extract_circles(image_path, points):

[0066] # Read the image

[0067] image = cv2.imread(image_path)

[0068] # Make sure the input points are a list of four points

[0069] if len(points) != 4:

[0070] raise ValueError("Please provide the coordinates of four points.")

[0071] #points = [(49, 63), (323, 67), (43, 353), (317, 357)]

[0072] # Get the coordinates of the upper left and lower right corners of the rectangle

[0073] top_left = (min(points[0][0], points[1][0]), min(points[0][1],points[1][1]))

[0074] bottom_right = (max(points[2][0], points[3][0]), max(points[2][1],points[3][1]))

[0075] # Draw a rectangle

[0076] cv2.rectangle(image, top_left, bottom_right, (0, 255, 0), 2)

[0077] # Extract rectangular area

[0078] roi = image[top_left[1]:bottom_right[1], top_left[0]:bottom_right[0]]

[0079] # Convert to grayscale image

[0080] gray_roi = cv2.cvtColor(roi, cv2.COLOR_BGR2GRAY)

[0081] # Blur the grayscale image to reduce noise

[0082] gray_roi = cv2.GaussianBlur(gray_roi, (5, 5), 0)

[0083] # Use Canny edge detection

[0084] edges = cv2.Canny(gray_roi, 50, 150)

[0085] Morphological Operations: Dilation

[0086] kernel = np.ones((3, 3), np.uint8)

[0087] dilated_edges = cv2.dilate(edges, kernel, iterations=1)

[0088] # Create a mask

[0089] mask = np.zeros(gray_roi.shape, dtype=np.uint8)

[0090] cv2.rectangle(mask, (0, 0), (gray_roi.shape[1], gray_roi.shape[0]),(255), thickness=-1)

[0091] # Apply the mask

[0092] masked_edges = cv2.bitwise_and(dilated_edges, dilated_edges, mask=mask)

[0093] # Display edge image

[0094] cv2.imshow('Masked Edges', masked_edges)

[0095] # Detect circles using Hough circle transform

[0096] circles = cv2.HoughCircles(masked_edges, cv2.HOUGH_GRADIENT, dp=1,minDist=10,

[0097] param1=30, param2=10, minRadius=1, maxRadius=15)

[0098] #Store the coordinates of the circle center that meet the conditions

[0099] circle_centers = []

[0100] if circles is not None:

[0101] circles = np.uint16(np.around(circles))

[0102] for i in circles[0, :]:

[0103] radius = i[2]

[0104] if radius>1: # Only extract circles with a radius greater than 1

[0105] # Calculate the coordinates of the center of the circle in the original image

[0106] center = (i[0] + top_left[0], i[1] + top_left[1]) # Convert to original image coordinates

[0107] circle_centers.append(center)

[0108] # Mark the center of the circle on the image

[0109] cv2.circle(image, center, 10, (255, 0, 0), -1) # Draw the center of the circle with a radius of 10 and a blue color

[0110] # Display the result image

[0111] cv2.imshow('Result', image)

[0112] cv2.waitKey(0)

[0113] cv2.destroyAllWindows()

[0114] # Example call

[0115] image_path = 'C:\\Users\\w\\Desktop\\tu1.jpg'

[0116] points = [(69, 83), (303, 87), (63, 333), (297, 337)]

[0117] 1. The basic idea of ​​Hough transform

[0118] The Hough transform is a feature extraction technique used to detect specific shapes (such as lines and circles) in images. For circle detection, the Hough transform maps a circle in image space to a point in parameter space, defined by the circle's center coordinates (a, b) and radius rr. Specifically, a circle in image space corresponds to a point in parameter space, and multiple points in image space that lie on the same circle will intersect at the same point in parameter space.

[0119] 2. Construction of parameter space

[0120] In the Hough transform, the parameter space is typically a three-dimensional space, representing the abscissa aa of the circle's center, the ordinate bb, and the radius rr. For each edge point in the image, the Hough transform iterates over all possible circle centers and radii and accumulates the corresponding points in the parameter space. If multiple edge points lie on the same circle, their accumulator values ​​in the parameter space increase significantly, forming a peak.

[0121] 3. Accumulator and Voting Mechanism

[0122] The Hough transform uses an accumulator array to count the votes for each possible circle. For each edge point in the image, the Hough transform calculates all possible circle centers and radii and adds 1 to the corresponding position in the accumulator. Ultimately, the parameters (a, b, r) corresponding to the point with the largest value in the accumulator are the parameters of the circle in the image.

[0123] 4. Optimization of Hough Gradient Method

[0124] To improve computational efficiency, OpenCV uses the Hough gradient method to optimize the standard Hough circle transform. The Hough gradient method first calculates the image gradient using Canny edge detection and the Sobel operator. It then draws lines along the gradient and counts possible circle centers. Next, it determines the radius by calculating the distance from the edge point to the circle center. This method reduces computational effort and improves detection speed.

[0125] A method for generating the three-dimensional code element label includes the following steps:

[0126] S1: preparing a medium carrier, forming a medium carrier substrate through a printing process, and printing a positioning pattern on the substrate surface of the medium carrier to demarcate a verification area;

[0127] S2: randomly distributing colored particles on the surface of the medium carrier by a relief printing process to form a three-dimensional pattern with a concave-convex tactile feel;

[0128] S3: using a Hough transform algorithm to extract the quantity, position and center coordinate information of the colored particles in the three-dimensional pattern;

[0129] S4: Generate a verification code according to the extracted information, the verification code including an information code, a check code and a coordinate code, and print the verification code on the medium carrier.

[0130] A method for verifying the three-dimensional code element tag comprises the following steps:

[0131] S1: Acquire the label image through machine vision recognition technology, identify the positioning pattern and delineate the verification area;

[0132] S2: Identify the information code, check code and coordinate code in the verification area, and parse the particle quantity, position and center coordinate information of the three-dimensional pattern contained in the verification code;

[0133] S3: Using the Hough transform algorithm to extract the number, position and center coordinate information of the particles in the three-dimensional pattern within the verification area;

[0134] S4: The extracted information is compared with the data parsed from the verification code for verification. If they are consistent, it is judged to be true, otherwise it is judged to be false.

[0135] The comparison and verification includes:

[0136] In the networked mode, the extracted label images are compared with the original generated images in the database;

[0137] In offline mode, the particle quantity and location information are inferred through the verification code and compared with the current image extraction value.

[0138] A method for verifying a three-dimensional code element label comprises the following steps: (1) obtaining a label image through machine vision recognition technology, identifying the positioning pattern and demarcating a verification area; (2) identifying the information code, check code and coordinate code of the verification code area, and parsing the number, position and center coordinate information of the particles of the three-dimensional pattern contained in the code; (3) extracting the number, position and center coordinate information of the particles of the three-dimensional pattern in the verification area by using the Hough transform algorithm; (4) comparing and verifying the extracted information with the data parsed from the verification code; if they are consistent, it is judged to be true, otherwise it is judged to be false.

[0139] The comparison and verification of this 3D code supports online or offline verification. During online verification, after reading / taking a photo of the label, the photo that needs to be compared and read can also be transmitted to the background, and the pattern is read in reverse / forward. At the same time, the label position and other information in the photo are parsed and compared with the verification code. Then, the photo when the label was generated is queried for comparison to determine the authenticity of the label and return the identification result. During offline verification without a network environment, only the photo is parsed, the verification code is checked, and the number of particles contained in the label pattern and the location information of specific particles are deduced from the verification code. The information is then compared with the pattern to determine the authenticity of the label. The three-dimensional characteristics of the pattern can be verified by touch or binocular camera, thereby preventing the possibility of scanning or printing counterfeit patterns.

[0140] The three-dimensional code element label first uses randomly printed colored small particles to form a scattered concave and convex circular pattern. According to the center points of the pre-printed positioning circles, a coordinate map of the verification area is formed. Then, an optimized Hough transform algorithm is used to detect the contour of the circle. Using machine vision recognition technology, the number of colored small particles in the verification area and the coordinates of the center points of each small circle are identified and recorded. By formulating coding rules, the above information is encoded and encrypted and printed on the second and third lines of the verification code area. During verification, machine vision recognition technology is first used to identify the code in the second and third lines of the verification code area. This code then parses the information contained within it, including the number of small colored particles and the location of specific particles within the verification area. The same optimized Hough transform algorithm for detecting circular contours is then used to extract the number of small particles and the coordinates of the center points of each small dot within the concave-convex circular pattern. These extracted values ​​are then compared with the values ​​parsed from the code in the second and third lines. If the comparison is true, the label (pattern + verification code) is considered to be consistent with the original production and has not been counterfeited. If the comparison is false, the label is considered to be counterfeit and inconsistent with the original production. The 3D code element label exploits the randomness and extremely difficult-to-reproduce nature of the colored particle distribution pattern when printed, using the number and location coordinates of the particles to achieve self-counterfeiting.

[0141] The application scenarios of this 3D code element tag include the following:

[0142] Origin, packaging, time, etc. Scene identification. Such as the proof of the goods bought by the sea buyer, the label of the country of origin of foreign fruits, beef, red wine, etc.

[0143] Identify the WYSIWYG scene. Such as the consistency proof of used luxury goods inspection products and shipped products, the origin of specialty agricultural products, and the processing (picking) time proof;

[0144] Identify the manufacturer's promotion of its own characteristic brand products;

[0145] Identify the meta-universe application scene, the meta-universe scene;

[0146] Identify existing one-dimensional barcodes on goods to ensure that the actual goods and the code correspond one-to-one, forming a single anti-counterfeiting traceability code;

[0147] Identify the applicable scene of the existing two-dimensional code, which corresponds to the two-dimensional code one-to-one, and performs anti-counterfeiting traceability.

[0148] Compared with the prior art, the present application has the following advantages:

[0149] 1. Very high anti-counterfeiting security

[0150] Physical anti-counterfeiting: random distribution of colored particles forms a three-dimensional pattern with concave-convex tactile sensation, and the particle distribution of each label is unique and cannot be replicated, which cannot be copied or printed.

[0151] Encrypted anti-counterfeiting: the particle position information is accurately extracted by an optimized Hough transform algorithm, and a high-strength verification code is generated by combining the SM4 national encryption algorithm to ensure that the code cannot be tampered with or forged.

[0152] Multi-verification mechanism: supports network verification + offline verification + tactile verification, forming a multi-level anti-counterfeiting system.

[0153] 2. High-precision machine recognition capability

[0154] The detection accuracy of the particle center is optimized by the improved Hough gradient method, combined with Canny edge detection + Sobel gradient calculation to improve the recognition accuracy of small particles. The verification area is divided into grids, and the number and coordinates of particles in each partition are recorded, so that the fault tolerance of machine vision recognition is higher, and the misjudgment is reduced.

[0155] 3. Flexible application adaptability

[0156] Compatible with multiple media: can be printed on paper, plastic, metal, textiles and other carriers to meet different industry needs.

[0157] Supports multiple verification methods: Online verification: Suitable for high-value goods (such as luxury goods, imported food), can be combined with blockchain technology to ensure that data cannot be tampered with. Offline verification: Suitable for environments without a network (such as remote areas), verification can be completed by simply taking a photo. Tactile verification: Users can detect the concave and convex touch through touch or binocular camera to quickly distinguish the authenticity. Obviously, the above embodiments of the present invention are only examples to illustrate the present invention, and are not limitations on the implementation methods of the present invention. For ordinary technicians in the field, other different forms of changes or modifications can be made on the basis of the above description. It is not necessary and impossible to give exhaustive examples of all implementation methods here. However, these obvious changes or modifications derived from the essential spirit of the present invention still fall within the scope of protection of the present invention.

Claims

1. A three-dimensional code element label with self-anti-counterfeiting function, characterized in that: Includes the following sections: A medium carrier formed by a standard printing process, wherein the surface of the medium carrier is randomly provided with at least three regular geometrically shaped positioning patterns, wherein the center coordinates of the plurality of positioning patterns can be used to form a square-shaped verification area, wherein the verification area is evenly divided into a plurality of sub-areas; A three-dimensional code pattern, wherein the three-dimensional code pattern is formed by mixing colored particles with glue through a relief printing process and then randomly fixed on the surface of the medium carrier, the three-dimensional code pattern having a three-dimensional concave-convex tactile feel, and the colored particles are randomly distributed in the verification area, and the particle size of the colored particles is 0.1 to 5.5 mm; The verification code area includes three lines of verification code: The first line of verification code contains information codes of more than 13 digits, wherein the information codes include one or more of the company code, production time, or production serial number; The second line of verification code contains a verification code of more than 21 digits. The verification codes include the total number of particles in the verification area, the number of particles in each sub-area, and an anti-counterfeiting verification code. The anti-counterfeiting verification code is generated using an encryption algorithm based on the existing code; The third line of verification code includes a coordinate code of more than 25 digits, wherein the plurality of coordinate codes include a first field value and the center coordinates of four or more specific particles, wherein the first field value is used to determine the sub-regions where the plurality of specific particles are located, and the coordinate codes are extracted and encoded using a Hough transform algorithm; The verification code is displayed in plain text or in cipher text through an encryption algorithm.

2. The three-dimensional code element label with self-anti-counterfeiting function according to claim 1, characterized in that: The positioning pattern is any one of a circle, a triangle, a square or a rectangle. When the positioning pattern is a circle, its diameter is more than twice the average diameter of the colored particles; when the positioning pattern is a polygon, its minimum side length is more than twice the average diameter of the colored particles.

3. The three-dimensional code element label with self-anti-counterfeiting function according to claim 1, characterized in that: The Hough transform algorithm comprises the following steps: (1) Perform grayscale conversion, Gaussian blurring, and Canny edge detection on the verification area image; (2) Detect the center coordinates and radius of the colored particles using the Hough gradient method, and select particles that meet the preset radius range; (3) Convert the detection results into the original image coordinate system and record the number and position of particles.

4. The three-dimensional code element label with self-anti-counterfeiting function according to claim 1, characterized in that: The encryption algorithm uses symmetric encryption, asymmetric encryption, a hybrid encryption system, a national secret algorithm or a hash algorithm to generate ciphertext code.

5. The three-dimensional code element label with self-anti-counterfeiting function according to claim 1, characterized in that: The label is used for product origin identification, anti-counterfeiting authentication, metaverse scene mapping, or unique coding binding with one-dimensional code or two-dimensional code.

6. A method for generating a three-dimensional code element label according to any one of claims 1 to 5, comprising the following steps: S1: preparing a medium carrier, forming a medium carrier substrate through a printing process, and printing a positioning pattern on the substrate surface of the medium carrier to demarcate a verification area; S2: randomly distributing colored particles on the surface of the medium carrier by a relief printing process to form a three-dimensional pattern with a concave-convex tactile feel; S3: using a Hough transform algorithm to extract the quantity, position and center coordinate information of the colored particles in the three-dimensional pattern; S4: Generate a verification code according to the extracted information, the verification code including an information code, a check code and a coordinate code, and print the verification code on the medium carrier.

7. A method for verifying a three-dimensional code element tag according to any one of claims 1 to 5, comprising the following steps: S1: Acquire the label image through machine vision recognition technology, identify the positioning pattern and delineate the verification area; S2: Identify the information code, verification code and coordinate code in the verification code area, and parse the particle quantity, position and center coordinate information of the three-dimensional pattern contained in the verification code; S3: Using the Hough transform algorithm to extract the number, position and center coordinate information of the particles in the three-dimensional pattern within the verification area; S4: The extracted information is compared with the data parsed from the verification code for verification. If they are consistent, it is judged to be true, otherwise it is judged to be false.

8. The method for verifying a three-dimensional code element tag according to claim 7, characterized in that: The comparison and verification includes: In the networked mode, the extracted label images are compared with the original generated images in the database; In offline mode without a network environment, the particle quantity and location information are inferred through the verification code and compared with the current image extraction value.

Images