Tire sidewall arc identification arrangement verification system based on identity authentication

Abstract

The present application relates to the field of information security and identity authentication, and discloses a tire sidewall arc mark arrangement verification system based on identity authentication. The arrangement feature extraction node collects tire sidewall arc mark images, corrects macro arc distortion through a curved surface unfolding model to obtain a two-dimensional plane unfolding image, extracts a micro arrangement offset matrix of the mark edge node in the arc coordinate system, inputs the micro arrangement offset matrix into a cryptography hash function to generate a physical feature digest and transmit the same, the identity authentication server performs a bitwise exclusive or comparison between the physical feature digest and a pre-stored reference digest to calculate a hamming distance, and outputs a verification pass signal if the hamming distance is less than a preset verification threshold. The present application converts tire molding micro distortion into a cryptography verification element, establishes a unique binding between a physical entity and a digital identity, prevents plain text information from being copied, and blocks the cloning and counterfeiting path of entity identity.

Description

Technical Field

[0001] This invention relates to the field of information security and identity verification technology, and discloses a tire sidewall arc marking layout verification system based on identity authentication. Background Technology

[0002] In existing technologies, identity verification for tire sidewall markings typically employs plaintext information comparison. The specific process involves: acquiring an image of the marking on the tire sidewall surface using an image acquisition device; extracting the plaintext serial number or digital signature carried by the marking using optical character recognition (OCR) or QR code decoding technology; then transmitting the extracted plaintext data to an authentication server; the authentication server performs string matching between the received plaintext data and pre-stored valid credentials in its database; if a match is found, the tire entity is deemed legitimate.

[0003] The aforementioned existing technologies rely on plaintext information carried by the identifier for identity verification. Plaintext information is easily scanned optically and physically copied, making it impossible to establish a unique binding relationship between the physical entity and the digital identity, which makes the tire's physical identity easy to clone and forge. Summary of the Invention

[0004] The purpose of this invention is to provide a tire sidewall arc marking layout verification system based on identity authentication, which can solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the technical invention adopted by this invention is as follows: The tire sidewall arc marking layout verification system based on identity authentication includes: layout feature extraction nodes and identity authentication server; The arrangement feature extraction node acquires the tire side arc mark image, and corrects the macro arc distortion in the tire side arc mark image through the surface unfolding model to obtain a two-dimensional plane unfolded image. Extract the microscopic arrangement offset matrix of the marked edge nodes in the circular coordinate system from the two-dimensional planar unfolded image; The microscopic arrangement offset matrix is ​​input into a cryptographic hash function to generate a physical feature digest, and the physical feature digest is transmitted to the identity authentication server; The identity authentication server performs a bitwise XOR comparison operation between the received physical feature digest and the pre-stored reference digest, calculates the Hamming distance of the bit differences in the XOR result, and outputs an identity verification pass signal if the Hamming distance is less than a preset verification threshold.

[0006] Preferably, the process of obtaining a two-dimensional planar unfolded image by correcting the macroscopic arc distortion in the tire sidewall arc marking image through the surface unfolding model by the arrangement feature extraction node includes: identifying the arc positioning baseline in the tire sidewall arc marking image, and fitting the three-dimensional cylindrical parametric equation of the tire sidewall based on the arc positioning baseline. The inverse mapping relationship between image pixels from the two-dimensional imaging plane to the tire side surface is established based on the three-dimensional cylindrical parametric equation. Based on the inverse mapping relationship, bilinear interpolation resampling is performed on the tire sidewall arc marking image to transform the pixels in the curved coordinate system to a planar rectangular coordinate system with arc length and radial distance as axes, thereby generating the two-dimensional planar unfolded image that eliminates curvature deformation.

[0007] Preferably, the process of the arrangement feature extraction node extracting the micro-arrangement offset matrix of the marker edge nodes in the two-dimensional planar unfolded image in the circular arc coordinate system includes: performing sub-pixel edge detection on the two-dimensional planar unfolded image to locate the discrete coordinate point set of the marker contour; The discrete coordinate point set is fitted with an elliptic curve, and the endpoints of the major and minor axes of the ellipse are extracted as feature anchor points. Calculate the tangential offset component and normal offset component of each of the aforementioned feature anchor points relative to the vertices of the theoretically arranged mesh; The tangential offset components and normal offset components of all feature anchor points are matrix-recombined according to the topological order of the theoretically arranged mesh to generate the microscopic arrangement offset matrix.

[0008] Preferably, the process by which the arrangement feature extraction node inputs the micro-arrangement offset matrix into a cryptographic hash function to generate a physical feature digest includes: performing normalization processing on the micro-arrangement offset matrix to map continuous offset values ​​to discrete integer values ​​with a fixed bit depth; The discrete integer values ​​are expanded into a one-dimensional bit stream in a row-column traversal order; The head of the one-dimensional bitstream is spliced ​​with the position salt value corresponding to the physical location to form an enhanced bitstream; The enhanced bitstream is input into the cryptographic hash function, and a fixed-length hash value is output as the physical feature digest.

[0009] Preferably, the process of the identity authentication server performing a bitwise XOR comparison operation between the received physical feature digest and the pre-stored benchmark digest, and calculating the Hamming distance of the bit difference in the XOR result, includes: retrieving the benchmark digest associated with the unique identifier of the tire to be verified from the secure storage area of ​​the identity authentication server. Perform a bitwise logical XOR operation between the physical feature summary and the baseline summary to generate an XOR difference sequence; The total number of bits with a logical value of true in the XOR difference sequence is counted, and the total number of bits is identified as the Hamming distance. The Hamming distance characterizes the degree of fluctuation deviation between physical entity features and registration features.

[0010] Preferably, if the Hamming distance is less than a preset verification threshold, the process of outputting an identity verification pass signal includes: acquiring the current physical environment parameters collected by the environmental temperature and humidity sensor, dynamically correcting the initial verification threshold based on the current physical environment parameters, and obtaining the preset verification threshold. Construct comparator logic to numerically compare the Hamming distance with the preset verification threshold; When the Hamming distance is determined to be less than the preset verification threshold, the security register is triggered to write the verification success status bit. Read the verification success status bit, generate the authentication pass signal containing a timestamp and the unique identifier of the tire to be verified, and push the authentication pass signal to the application layer interface.

[0011] Preferably, the process of performing bilinear interpolation resampling on the tire sidewall arc marking image according to the inverse mapping relationship includes: traversing the target pixel coordinates in the Cartesian coordinate system and calculating the floating-point source coordinates corresponding to the target pixel coordinates in the tire sidewall arc marking image through the inverse mapping relationship; Locate the four integer pixels adjacent to the floating-point source coordinates; Obtain the pixel grayscale values ​​of the four integer pixels; Using the spatial distance between the floating-point source coordinates and the four integer pixels as weights, a weighted summation operation is performed on the pixel grayscale values, and the result of the weighted summation operation is assigned to the target pixel coordinates until the full-image pixel mapping is completed.

[0012] Preferably, the process of calculating the tangential offset component and normal offset component of each of the feature anchor points relative to the vertices of the theoretically arranged mesh includes: A local two-dimensional orthogonal coordinate system is constructed with the vertices of the theoretically arranged grid as the origin, wherein the horizontal axis is parallel to the tangent direction of the tire circumference and the vertical axis is parallel to the radial normal direction of the tire. Calculate the horizontal and vertical projection distances of the feature anchor points in the local two-dimensional orthogonal coordinate system; The horizontal axis projection distance is determined as the tangential offset component, and the vertical axis projection distance is determined as the normal offset component; Smoothing filtering is performed on the tangential offset component and the normal offset component to eliminate high-frequency noise interference.

[0013] Preferably, the process of performing normalization processing on the micro-arrangement offset matrix to map continuous offset values ​​to discrete integer values ​​with a fixed bit depth includes: scanning the micro-arrangement offset matrix and extracting the absolute value of the global maximum offset and the absolute value of the global minimum offset. A linear mapping interval is constructed using the absolute value of the global maximum offset and the absolute value of the global minimum offset; Subtract the absolute value of the global minimum offset from each consecutive offset value in the micro-arrangement offset matrix, and then divide by the span of the linear mapping interval to obtain a normalized floating-point number. The normalized floating-point number is multiplied by a preset quantization factor and then rounded down to generate the discrete integer value with the fixed bit depth.

[0014] Preferably, the process of counting the total number of bits with logical values ​​of true in the XOR difference sequence and confirming the total number of bits as the Hamming distance includes: performing a bitwise accumulation operation on the XOR difference sequence to obtain the initial Hamming distance; The error correction coding verification matrix stored in the identity authentication server is invoked to perform adjoint calculation on the physical feature digest; If the result of the syndrome calculation is non-zero, it is confirmed that there is a channel transmission error in the physical feature digest, and the error bit is located based on the result of the syndrome calculation. A compensation deduction operation equal to the number of error bits is performed on the initial Hamming distance, and the value after compensation deduction is confirmed as the final Hamming distance.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The physical feature digest and the pre-stored baseline digest are compared using a bitwise XOR operation to calculate the Hamming distance. This invention corrects the macroscopic arc distortion of the tire sidewall arc marking image using a surface unfolding model, extracts the microscopic arrangement offset matrix of the marking edge nodes in the arc coordinate system, inputs the microscopic arrangement offset matrix into a cryptographic hash function to generate a physical feature digest, and the identity authentication server outputs an authentication pass signal based on the Hamming distance. The above process transforms the uncontrollable microscopic distortion generated during tire vulcanization into cryptographic verification elements, eliminates the exposure of plaintext information in the verification process, overcomes the defect that physical entities and digital identities cannot be uniquely bound, and blocks the path of cloning the tire entity identity by replicating plaintext information.

[0016] 2. In extracting microscopic arrangement offsets, the invention performs sub-pixel edge detection and elliptic curve fitting on the unfolded two-dimensional image, calculates the tangential and normal offset components of the feature anchor points relative to the vertices of the theoretical arrangement grid, and eliminates the interference of macroscopic deformation on microscopic feature extraction. In generating physical feature summaries, it splices and identifies the positional salt values ​​corresponding to the physical locations, enhancing the collision resistance of the physical feature summaries. It dynamically corrects the initial verification threshold using environmental temperature and humidity parameters, and uses error correction coding verification matrix to locate channel transmission errors for compensation and deduction when calculating Hamming distance, eliminating the interference of environmental fluctuations and transmission errors on the physical feature comparison results, ensuring the stability of arrangement feature extraction and the reliability of the identity verification system. Attached Figure Description

[0017] Figure 1 A flowchart of the overall workflow of the tire sidewall arc marking layout verification system based on identity authentication; Figure 2 Flowchart for macroscopic arc distortion correction of tire sidewall arc marking image; Figure 3 Flowchart for extracting the microscopic arrangement offset matrix; Figure 4 Flowchart for generating physical feature summaries; Figure 5 Flowchart for physical feature summary comparison and Hamming distance calculation; Figure 6 This is a flowchart for dynamic threshold correction and authentication signal output. Detailed Implementation

[0018] In one embodiment, the tire sidewall arc marking layout verification system based on identity authentication includes a layout feature extraction node and an identity authentication server. The layout feature extraction node is deployed in image acquisition terminal equipment at the end of the tire production line, in warehouse inspection stations, or at vehicle maintenance sites. The identity authentication server is deployed in a secure computing environment in the cloud or an enterprise's internal data center. Data transmission between the layout feature extraction node and the identity authentication server occurs via an encrypted communication link. This encrypted communication link is established using a transport layer security protocol, and the transmitted data is encrypted using a symmetric encryption algorithm.

[0019] The feature extraction nodes are arranged to acquire images of the tire sidewall arc markings using an industrial camera. The optical axis of the camera lens is perpendicular to the tire sidewall plane, and the acquisition area covers the entire marking area of ​​the tire sidewall. The feature extraction nodes then input the acquired sidewall arc marking images into a surface unfolding model. The surface unfolding model corrects the macroscopic arc distortion in the sidewall arc marking images, obtaining a two-dimensional planar unfolded image. Macroscopic arc distortion is due to the cylindrical structure of the tire sidewall, resulting in perspective distortion during planar imaging. This manifests as the marking characters curving in an arc along the circumference, with areas far from the center of the lens being compressed.

[0020] The arrangement feature extraction node extracts the microscopic arrangement offset matrix of the marked edge nodes in the two-dimensional planar unfolded image in the arc coordinate system. The microscopic arrangement offset matrix represents the random offset of the marked edge nodes relative to the theoretical design position during the tire vulcanization process. This offset is caused by a combination of factors such as mold manufacturing errors, uneven vulcanization temperature distribution, and inconsistent shrinkage of rubber materials, and is unique and cannot be replicated.

[0021] The arrangement feature extraction node inputs the micro-arrangement offset matrix into a cryptographic hash function to generate a physical feature digest. The cryptographic hash function uses a one-way hash function, which is collision-resistant and irreversible, capable of converting input data of arbitrary length into a fixed-length output hash value. The physical feature digest only contains a digital fingerprint identifying the micro-arrangement features and does not contain any plaintext identification information; therefore, the original micro-arrangement offset matrix cannot be derived from the physical feature digest.

[0022] The feature extraction nodes transmit the physical feature digest to the identity authentication server. The identity authentication server pre-stores baseline digests of all legitimate tires in a secure storage area. These baseline digests are generated after tire production through the same feature extraction and hash calculation process as the verification process, and are associated with the tire's unique identifier. The identity authentication server performs a bitwise XOR comparison between the received physical feature digest and the pre-stored baseline digest, calculating the Hamming distance of the bit differences in the XOR result. The Hamming distance represents the number of bits that differ in corresponding positions between two equal-length binary strings, quantifying the degree of difference between the physical feature digest and the baseline digest.

[0023] The authentication server compares the calculated Hamming distance with a preset verification threshold. If the Hamming distance is less than the preset threshold, it outputs an authentication success signal; if the Hamming distance is greater than or equal to the preset threshold, it outputs an authentication failure signal. The preset verification threshold is set based on the noise level of the feature extraction process, the degree of influence of environmental fluctuations, and the system's requirements for false acceptance and false rejection rates. A comparison of system module functions is shown in Table 1.

[0024]

[0025] In this embodiment, the layout feature extraction nodes and the identity authentication server work together to transform the uncontrollable microscopic distortions generated during the tire vulcanization process into cryptographic verification elements. The generation and comparison of physical feature summaries do not involve any plaintext identification information, eliminating the exposure of plaintext information in the verification process. The microscopic layout offset is generated by the combined effect of multiple random factors and cannot be copied by optical scanning or physical replication, establishing a unique binding relationship between the physical entity and the digital identity, blocking the path of cloning the tire entity's identity by replicating plaintext information.

[0026] In a preferred embodiment, the process of arranging feature extraction nodes to correct macroscopic arc distortion in the tire sidewall arc marking image using a surface unfolding model and obtaining a two-dimensional planar unfolded image includes: identifying the arc positioning baseline in the tire sidewall arc marking image; fitting the three-dimensional cylindrical parametric equation of the tire sidewall based on the arc positioning baseline; establishing an inverse mapping relationship between image pixels from the two-dimensional imaging plane to the tire sidewall surface based on the three-dimensional cylindrical parametric equation; performing bilinear interpolation resampling on the tire sidewall arc marking image according to the inverse mapping relationship, transforming the pixels in the surface coordinate system to a Cartesian coordinate system with arc length and radial distance as axes, and generating a two-dimensional planar unfolded image with curvature deformation eliminated.

[0027] The feature extraction node performs grayscale conversion and Gaussian filtering on the tire sidewall arc marking image to remove noise interference. The node then uses an edge detection algorithm, specifically the Canny edge detection algorithm, to extract edge features from the image. By setting high and low thresholds, it extracts strong and weak edges, retaining weak edges connected to strong edges and removing isolated weak edges. From the extracted edge features, the node identifies the arc positioning baseline. This baseline is a pre-designed circular reference ring on the tire sidewall, located inside or outside the marking area, and possesses continuous arc edge features.

[0028] The feature extraction nodes are arranged to extract discrete edge point sets on the circular arc positioning baseline. The least squares method is used to fit the discrete edge point sets to a circle, obtaining the center coordinates and radius of the circular arc positioning baseline on the two-dimensional imaging plane. Based on the camera's intrinsic and extrinsic parameter matrices, the center coordinates and radius on the two-dimensional imaging plane are converted into coordinates and radii in three-dimensional space, establishing the three-dimensional cylindrical parametric equations for the tire sidewall. The expression for the three-dimensional cylindrical parametric equations is:

[0029] in, Let be the radius of the cylinder on the tire sidewall. Angle in the circumferential direction. Let be the axial coordinates of the cylinder in three-dimensional space. , , These are Cartesian coordinates in three-dimensional space. The parameters of the three-dimensional cylindrical parametric equations are explained in Table 2.

[0030]

[0031] The feature extraction nodes are arranged according to the 3D cylindrical parametric equations and the camera's projection matrix to establish an inverse mapping relationship between image pixels from the 2D imaging plane to the tire sidewall surface. (Camera projection matrix...) A 3×4 matrix is ​​used to represent points in three-dimensional space. Points projected onto the two-dimensional imaging plane The formula for the projection relationship is:

[0032] in, As a scaling factor, , These are the pixel coordinates on the two-dimensional imaging plane.

[0033] The process of establishing the inverse mapping relationship is as follows: for any target pixel coordinate in the two-dimensional plane unfolded image... ,in Let the coordinates be the arc length. For radial distance coordinates, based on the relationship between arc length and angle Calculate the angle corresponding to the target pixel on the three-dimensional cylindrical surface. .Will and Substituting the values ​​into the three-dimensional cylindrical parametric equation, we obtain the coordinates of the target pixel in three-dimensional space. . 3D spatial coordinates By substituting the inverse operation of the camera projection matrix, the floating-point source coordinates of the target pixel in the tire sidewall arc marking image are calculated. .

[0034] The feature extraction nodes are arranged to perform bilinear interpolation resampling on the tire sidewall arc marking image based on the inverse mapping relationship. All target pixel coordinates in the Cartesian coordinate system are traversed, and the floating-point source coordinates corresponding to each target pixel coordinate are calculated using the inverse mapping relationship. The four integer pixels adjacent to the floating-point source coordinates are then located. , , , ,in , , This indicates the floor function.

[0035] Get the pixel grayscale values ​​of four integer pixels. , , , Using the spatial distances between the floating-point source coordinates and four integer pixel points as weights, a weighted summation operation is performed on the pixel grayscale values. The formula for bilinear interpolation is:

[0036] in, The fractional part of the floating-point source coordinates in the horizontal direction. This represents the decimal part of the floating-point source coordinates in the vertical direction.

[0037] The result of the weighted summation operation is assigned to the target pixel coordinates. This process continues until full pixel mapping is completed, generating a two-dimensional planar unfolded image free from curvature distortion. Bilinear interpolation resampling effectively avoids aliasing and blurring during pixel mapping, preserving image smoothness and detail.

[0038] In this embodiment, by identifying the arc positioning baseline and fitting the three-dimensional cylindrical parametric equation, a precise inverse mapping relationship is established. A bilinear interpolation resampling method is then used for pixel transformation, effectively correcting macroscopic arc distortion in the tire sidewall arc marking image. In the corrected two-dimensional planar unfolded image, the marking characters are restored to a standard rectangular shape, eliminating circumferential curvature and radial perspective compression, providing an accurate image basis for subsequent microscopic arrangement offset extraction.

[0039] In a preferred embodiment, the process of extracting the microscopic arrangement offset matrix of the marker edge nodes in the circular coordinate system in the two-dimensional planar unfolded image by arranging feature extraction nodes includes: performing sub-pixel edge detection on the two-dimensional planar unfolded image to locate the discrete coordinate point set of the marker contour; fitting the discrete coordinate point set to an elliptic curve to extract the endpoints of the major and minor axes of the ellipse as feature anchor points; calculating the tangential offset component and normal offset component of each feature anchor point relative to the vertices of the theoretical arrangement grid; and reorganizing the tangential offset component and normal offset component of all feature anchor points according to the topological order of the theoretical arrangement grid to generate the microscopic arrangement offset matrix.

[0040] Feature extraction nodes are arranged to perform sub-pixel edge detection on the unfolded 2D image. The sub-pixel edge detection employs a Zernike moment-based algorithm. This algorithm calculates the Zernike moments of the image and leverages their rotation invariance and orthogonality to improve edge localization accuracy to the sub-pixel level. The sub-pixel edge detection process is as follows: the unfolded 2D image is divided into multiple equally sized sub-windows, and the Zernike moment of each sub-window is calculated; the direction and sub-pixel position of the edges within each sub-window are calculated based on the Zernike moments; and the sub-pixel edge points of all sub-windows are connected to form a discrete coordinate point set identifying the contour.

[0041] The feature extraction nodes are arranged to fit an elliptic curve to the discrete coordinate point set. The elliptic curve fitting uses the least squares method. The general formula for an ellipse is:

[0042] in, , , , , , Let be the coefficients of the equation of the ellipse, and satisfy . For each point in the discrete coordinate point set Substituting into the general equation of an ellipse, we obtain the error formula as follows:

[0043] By minimizing the sum of squares of all errors Solve for the coefficients of the equation of the ellipse. , , , , , Calculate the coordinates of the center of the ellipse based on the coefficients of the ellipse equation. Long half shaft short half shaft and rotation angle The four endpoints of the major and minor axes of the ellipse are extracted as feature anchor points. The coordinates of the endpoints of the major axis are... The coordinates of the minor axis endpoints are .

[0044] The feature extraction nodes are arranged to calculate the tangential and normal offset components of each feature anchor point relative to the vertices of the theoretical layout mesh. The theoretical layout mesh is generated based on the design drawings of the marking, and the mesh vertices correspond to the feature anchor point positions of the marking contour under ideal conditions. A local two-dimensional orthogonal coordinate system is constructed with the vertices of the theoretical layout mesh as the origin, where the horizontal axis is parallel to the tangent direction of the tire circumference, and the vertical axis is parallel to the radial normal direction of the tire.

[0045] Calculate the horizontal and vertical projection distances of the feature anchor points in a local two-dimensional orthogonal coordinate system. Let the global coordinates of the vertices of the theoretically arranged mesh be... The global coordinates of the feature anchor point are The horizontal unit vector of the local coordinate system is The vertical axis unit vector is ,in The angle between the circumferential tangent direction at the vertex of the theoretically arranged mesh and the positive direction of the global x-axis.

[0046] The displacement vector of the feature anchor point relative to the vertices of the theoretically arranged mesh is Tangential offset component The normal offset component is the projection of the displacement vector onto the unit vector on the horizontal axis. The projection of the displacement vector onto the unit vector on the vertical axis is calculated using the following formula:

[0047] The tangential and normal offset components are smoothed using a Gaussian filter with a 3×3 window size and a Gaussian kernel standard deviation of 1.0. An example of the feature anchor point offset component calculation is shown in Table 3.

[0048]

[0049] The feature extraction nodes are arranged by rearranging the tangential and normal offset components of all feature anchor points into a matrix according to the topological order of the theoretical mesh. The theoretical mesh is... The grid structure, corresponding to There are 10 identifier characters, each corresponding to 4 feature anchor points. The micro-arrangement offset matrix is ​​as follows: The matrix contains odd-numbered rows storing tangential offset components and even-numbered rows storing normal offset components. Each pair of columns corresponds to two feature anchor points of an identifier character.

[0050] In this embodiment, sub-pixel edge detection improves the positioning accuracy of the marker contour to the sub-pixel level, enabling the capture of minute microscopic distortions generated during tire vulcanization. Elliptic curve fitting is used to extract feature anchor points, reducing the impact of edge noise on feature point positioning. By constructing a local two-dimensional orthogonal coordinate system to calculate the tangential and normal offset components, the offset of the feature anchor points can be decomposed into two orthogonal directions related to the tire geometry, eliminating the interference of macroscopic deformation on microscopic feature extraction.

[0051] In a preferred embodiment, the process of the arrangement feature extraction node inputting the micro-arrangement offset matrix into a cryptographic hash function to generate a physical feature digest includes: performing normalization processing on the micro-arrangement offset matrix to map continuous offset values ​​to discrete integer values ​​with a fixed bit depth; expanding the discrete integer values ​​into a one-dimensional bit stream according to the row and column traversal order; concatenating the position salt value corresponding to the physical position to the beginning of the one-dimensional bit stream to form an enhanced bit stream; and inputting the enhanced bit stream into the cryptographic hash function to output a hash value of fixed length as the physical feature digest.

[0052] The node scan of the arrangement feature extraction node is used to extract the micro-arrangement offset matrix and extract the absolute value of the global maximum offset. absolute value of global minimum offset A linear mapping interval is constructed using the absolute values ​​of the global maximum and minimum offsets. The values ​​of each consecutive offset in the micro-arrangement offset matrix. Subtract the absolute value of the global minimum offset, then divide by the span of the linear mapping interval to obtain the normalized floating-point number. The calculation formula is:

[0053] Normalized floating-point numbers and preset quantization factors Multiply and perform floor function to generate discrete integer values ​​with a fixed bit depth. The calculation formula is:

[0054] Among them, the preset quantization factor Based on fixed bit depth Sure, For example, when the fixed bit depth is 8 bits, The range of discrete integer values ​​is .

[0055] The feature extraction nodes are arranged to expand discrete integer values ​​into a one-dimensional bitstream in row-column traversal order. The traversal order starts from the first row and first column, traversing all columns of the first row, then all columns of the second row, and so on, until the entire matrix has been traversed. Each discrete integer value is converted into a binary representation of its corresponding bit depth, with the most significant bit first and the least significant bit last. All binary representations are then concatenated sequentially to form a one-dimensional bitstream.

[0056] The layout feature extraction nodes are appended to the header of the one-dimensional bitstream along with a location salt value corresponding to the physical location. The location salt value is a fixed-length binary string generated by hashing the tire's unique identifier, the location number on the tire sidewall, and the production batch number. The location salt value is 128 bits long and is concatenated with the one-dimensional bitstream to form an enhanced bitstream. The location salt value enhances the collision resistance of the physical feature summary; even if different tires have partially similar microscopic layout offset matrices, the generated physical feature summaries will differ significantly due to the different location salt values.

[0057] The arrangement of feature extraction nodes enhances the bitstream input to a cryptographic hash function, which employs the SHA-256 algorithm. The SHA-256 algorithm converts input data of arbitrary length into a fixed-length hash value of 256 bits, exhibiting collision resistance and irreversibility. The SHA-256 algorithm's processing includes five steps: message padding, message block division, hash value initialization, compression function processing, and hash value output. The final 256-bit hash value is the physical feature digest.

[0058] In a preferred embodiment, the process of the identity authentication server performing a bitwise XOR comparison operation between the received physical feature digest and the pre-stored baseline digest, and calculating the Hamming distance of the bit differences in the XOR result, includes: retrieving the baseline digest associated with the unique identifier of the tire to be verified from the secure storage area of ​​the identity authentication server; performing a bitwise logical XOR operation between the physical feature digest and the baseline digest to generate an XOR difference sequence; counting the total number of bits with logical values ​​of true in the XOR difference sequence, and confirming the total number of bits as the Hamming distance; the Hamming distance characterizes the degree of fluctuation deviation between the physical entity features and the registration features.

[0059] The authentication server receives the physical feature digest and the unique identifier of the tire to be verified from the feature extraction nodes. Based on the unique identifier of the tire to be verified, the authentication server queries and retrieves the corresponding baseline digest from the secure storage area. The secure storage area is protected by a hardware security module, and the baseline digest is stored in encrypted form, only to be decrypted and loaded into memory when the verification operation is performed.

[0060] The authentication server performs a bitwise logical XOR operation between the physical feature digest and the reference digest. Let the physical feature digest be... The benchmark abstract is ,in The length of the digest in bits. and For the first The bit value of each bit takes the value of either 0 or 1. XOR difference sequence The calculation formula is:

[0061] in, Represents the logical XOR operation, when and At the same time, ;when and When they are the same, .

[0062] The authentication server counts the total number of bits with logically true values ​​in the XOR difference sequence and uses this total number of bits as the Hamming distance. The calculation formula is:

[0063] In a preferred embodiment, the process of outputting an authentication pass signal if the Hamming distance is less than a preset verification threshold includes: acquiring the current physical environment parameters collected by an environmental temperature and humidity sensor; dynamically correcting the initial verification threshold based on the current physical environment parameters to obtain a preset verification threshold; constructing comparator logic to compare the Hamming distance with the preset verification threshold; triggering the security register to write a verification success status bit when it is determined that the Hamming distance is less than the preset verification threshold; reading the verification success status bit, generating an authentication pass signal containing a timestamp and the unique identifier of the tire to be verified, and pushing the authentication pass signal to the application layer interface.

[0064] The identity authentication server obtains current physical environment parameters from environmental temperature and humidity sensors. These sensors are deployed near the feature extraction nodes to collect real-time temperature data at the verification site. and relative humidity The identity authentication server sets the initial verification threshold based on the current physical environment parameters. Dynamic correction is performed to obtain the preset verification threshold. The corrected formula is:

[0065] in, The standard temperature is 25℃. The standard relative humidity is 50%. This is a temperature correction factor. The humidity correction coefficient is used. The temperature and humidity correction coefficients were experimentally determined to characterize the impact of environmental temperature and humidity changes on the feature extraction results. A comparison of environmental parameters and validation threshold correction coefficients is shown in Table 4.

[0066]

[0067] The authentication server constructs a comparator logic to compare the calculated Hamming distance with a preset verification threshold. When the Hamming distance is determined to be less than the preset verification threshold, a successful verification status bit is written to the security register. The security register is a non-volatile register, and write operations can only be performed under the control of the authentication server's secure kernel to prevent malicious tampering with the verification result.

[0068] The authentication server reads the successful authentication status bit from the security register and generates an authentication pass signal containing a timestamp and the unique identifier of the tire to be verified. The timestamp uses Coordinated Universal Time (UTC) and is accurate to the second. The authentication server pushes the authentication pass signal to the application layer interface, which can be a tire production management system, a warehouse management system, or a vehicle maintenance management system.

[0069] refer to Figure 4 In a preferred embodiment, the process of performing bilinear interpolation resampling on the tire sidewall arc marking image based on the inverse mapping relationship includes: traversing the target pixel coordinates in the Cartesian coordinate system, calculating the floating-point source coordinates corresponding to the target pixel coordinates in the tire sidewall arc marking image through the inverse mapping relationship; locating four integer pixels adjacent to the floating-point source coordinates; obtaining the pixel grayscale values ​​of the four integer pixels; performing a weighted summation operation on the pixel grayscale values ​​using the spatial distance between the floating-point source coordinates and the four integer pixels as weights, and assigning the weighted summation result to the target pixel coordinates, until the full-image pixel mapping is completed.

[0070] The dimensions of the Cartesian coordinate system are pre-calculated for the arrangement of feature extraction nodes. The horizontal axis of the Cartesian coordinate system is the arc length coordinate, and its range is... The vertical axis represents the radial distance coordinate, with a range of [missing information]. ,in The inner radius of the identified area. The outer radius of the marked area. The resolution of the Cartesian coordinate system is determined based on the resolution of the tire sidewall arc marking image to ensure that the unfolded image has sufficient detail.

[0071] Arrange feature extraction nodes to traverse the coordinates of all target pixels in the Cartesian coordinate system. ,in Let the coordinates be the arc length. These are radial distance coordinates. For each target pixel coordinate, the corresponding floating-point source coordinates are calculated using an inverse mapping relationship. The calculation process for the reverse mapping relationship has been described above and will not be repeated here.

[0072] The feature extraction nodes are arranged to locate four adjacent integer pixels from the floating-point source coordinates. These four integer pixels form a square, and the floating-point source coordinates are located inside this square. The pixel grayscale values ​​of these four integer pixels are then obtained, with the range of pixel grayscale values ​​being... .

[0073] Arrange feature extraction nodes and calculate the spatial distance weights between the floating-point source coordinates and four integer pixel points. Horizontal weights. The horizontal distance between the floating-point source coordinates and the integer pixel point to the left, and the vertical weight. This represents the vertical distance between the floating-point source coordinates and the integer pixel point above. The weights of the four integer pixels are as follows: , , and The sum of the weights is 1.

[0074] The feature extraction nodes are arranged to perform a weighted summation operation on the grayscale values ​​of four integer pixels, and the result of the weighted summation operation is assigned to the target pixel coordinates. The result of the weighted summation operation is a floating-point number, which is converted to an integer grayscale value by rounding. The above process is repeated until the mapping of all target pixels in the Cartesian coordinate system is completed, generating a two-dimensional planar unfolded image.

[0075] In a preferred embodiment, the process of calculating the tangential offset component and normal offset component of each feature anchor point relative to the vertices of the theoretically arranged grid includes: constructing a local two-dimensional orthogonal coordinate system with the vertices of the theoretically arranged grid as the origin, wherein the horizontal axis is parallel to the tangent direction of the tire circumference and the vertical axis is parallel to the radial normal direction of the tire; calculating the horizontal axis projection distance and the vertical axis projection distance of the feature anchor point in the local two-dimensional orthogonal coordinate system; determining the horizontal axis projection distance as the tangential offset component and the vertical axis projection distance as the normal offset component; and performing smoothing filtering on the tangential offset component and the normal offset component to eliminate high-frequency noise interference.

[0076] The feature extraction nodes calculate the circumferential tangent direction at the vertex of the theoretical layout mesh based on its position. The vertices of the theoretical layout mesh are located on the cylindrical surface of the tire sidewall, and the circumferential tangent direction is perpendicular to the line connecting that vertex and the tire center. Let the global coordinates of the tire center be... The global coordinates of the theoretically arranged mesh vertices are Then the vector connecting the vertex and the center of the tire is The unit vector along the tangent direction of the circle. The unit vector is the vector formed by rotating the connecting vector 90 degrees counterclockwise. The calculation formula is:

[0077] in, The magnitude of the connecting vector, i.e., the cylinder radius of the tire sidewall. The unit vector in the radial normal direction. The unit vector of the connecting vector is calculated using the following formula:

[0078] Using the vertices of the theoretically arranged mesh as the origin of the coordinate system, The horizontal axis is the unit vector, with Let the vertical axis be a unit vector, and construct a local two-dimensional orthogonal coordinate system.

[0079] The feature extraction nodes are arranged to calculate the horizontal and vertical projection distances of the feature anchor points in a local two-dimensional orthogonal coordinate system. The global coordinates of the feature anchor points are... The displacement vector relative to the vertices of the theoretically arranged mesh is The horizontal projection distance is the dot product of the displacement vector and the horizontal unit vector, and the vertical projection distance is the dot product of the displacement vector and the vertical unit vector.

[0080] The horizontal axis projection distance is defined as the tangential offset component, which represents the offset of the feature anchor point along the tangential direction of the tire circumference; the vertical axis projection distance is defined as the normal offset component, which represents the offset of the feature anchor point along the radial normal direction of the tire.

[0081] The feature extraction nodes are arranged to perform smoothing filtering on the tangential and normal offset components to eliminate high-frequency noise interference. Gaussian filtering is used for smoothing, and the kernel function of Gaussian filtering is:

[0082] in, The standard deviation of the Gaussian kernel controls the smoothness of the filter. The tangential and normal offset component matrices are convolved with the Gaussian kernel to obtain the smoothed offset component matrices.

[0083] In a preferred embodiment, the process of normalizing the micro-layout offset matrix to map continuous offset values ​​to discrete integer values ​​with a fixed bit depth includes: scanning the micro-layout offset matrix and extracting the absolute values ​​of the global maximum and minimum offsets; constructing a linear mapping interval using the absolute values ​​of the global maximum and minimum offsets; subtracting the absolute value of the global minimum offset from each continuous offset value in the micro-layout offset matrix and dividing by the span of the linear mapping interval to obtain a normalized floating-point number; multiplying the normalized floating-point number by a preset quantization factor and performing a rounding operation to generate discrete integer values ​​with a fixed bit depth.

[0084] The node scans all elements of the micro-layout offset matrix to extract the global maximum absolute value of the offset. and the absolute value of the global minimum offset The absolute value of the global maximum offset is the maximum value of the absolute values ​​of all elements in the matrix, and the absolute value of the global minimum offset is the minimum value of the absolute values ​​of all elements in the matrix.

[0085] Arrange feature extraction nodes to construct linear mapping intervals The span of the linear mapping interval is Each element in the micro-arrangement offset matrix Mapping to interval Within, we obtain normalized floating-point numbers. Normalization eliminates the differences in offset values ​​between different batches of tires, providing a unified standard for subsequent quantification.

[0086] Arrange the feature extraction nodes by normalizing the floating-point number and the preset quantization factor. Multiply and perform rounding to the nearest integer to generate discrete integer values ​​with a fixed bit depth. Preset quantization factor The required bit depth is determined by the desired quantization depth. Higher bit depths result in higher quantization accuracy and richer information in the generated physical feature summary, but also higher computational complexity. In practical applications, 8-bit or 16-bit quantization depths are typically chosen.

[0087] In a preferred embodiment, the process of counting the total number of bits with logically true values ​​in the XOR difference sequence and confirming the total number of bits as the Hamming distance includes: performing a bitwise accumulation operation on the XOR difference sequence to obtain the initial Hamming distance; calling the error correction coding check matrix stored in the identity authentication server to perform a synodal calculation on the physical feature digest; if the synodal calculation result is non-zero, it is confirmed that there is a channel transmission error in the physical feature digest, and the error bits are located according to the synodal calculation result; performing a compensation deduction operation equal to the number of error bits on the initial Hamming distance, and confirming the value after compensation deduction as the final Hamming distance.

[0088] The authentication server performs a bitwise accumulation operation on the XOR difference sequence, and the accumulation result is the initial Hamming distance. Bitwise accumulation can be implemented using hardware circuitry, resulting in high computational efficiency.

[0089] The authentication server calls the error correction encoding verification matrix stored in the secure storage area. The error correction coding uses Hamming code, and the parity-check matrix... for Matrix, where The bit length of the physical feature digest. This refers to the number of check bits. (For the physical feature summary) Performing the adjoint calculation, adjoint The calculation formula is:

[0090] in, Let be the transpose of the physical feature summary, and all operations are performed over the binary field GF(2). If the adjoint equation calculation results... If the vector is zero, it indicates that no errors occurred during the transmission of the physical feature digest, and the initial Hamming distance is the final Hamming distance. If the result of the syndrome calculation... A non-zero vector indicates that a bit error occurred during the transmission of the physical feature digest. According to the syndrome... In the verification matrix The position in the equation can be used to locate the position of the error bit. For example, if the syntactic equation... Equal to the parity check matrix The The column indicates the first of the physical feature summaries. A bit error occurred.

[0091] The identity authentication server counts the number of error bits. Perform a compensation deduction operation on the initial Hamming distance to obtain the final Hamming distance. The calculation formula is:

[0092] The compensation deduction operation can eliminate the impact of channel transmission errors on the Hamming distance calculation results and improve the accuracy of identity verification.

[0093] In this embodiment, by performing normalization and quantization on the micro-arrangement offset matrix, continuous offset values ​​are converted into discrete integer values ​​with a fixed bit depth, facilitating subsequent hash calculations. The collision resistance of the physical feature digest is enhanced by using a salt value at the concatenation position of the one-dimensional bitstream header. Dynamic correction of the initial verification threshold using environmental temperature and humidity parameters adapts to fluctuations in feature extraction under different environmental conditions. By utilizing the error-correcting coding parity matrix to locate and compensate for channel transmission errors, interference from transmission errors on the physical feature comparison results is eliminated, ensuring the reliability of the identity verification system.

Claims

1. A tire sidewall arc marking layout verification system based on identity authentication, characterized in that, include: Arrange feature extraction nodes and identity authentication servers; The arrangement feature extraction node acquires the tire side arc mark image, and corrects the macro arc distortion in the tire side arc mark image through the surface unfolding model to obtain a two-dimensional plane unfolded image. Extract the microscopic arrangement offset matrix of the marked edge nodes in the circular coordinate system from the two-dimensional planar unfolded image; The microscopic arrangement offset matrix is ​​input into a cryptographic hash function to generate a physical feature digest, and the physical feature digest is transmitted to the identity authentication server; The identity authentication server performs a bitwise XOR comparison operation between the received physical feature digest and the pre-stored reference digest, calculates the Hamming distance of the bit differences in the XOR result, and outputs an identity verification pass signal if the Hamming distance is less than a preset verification threshold.

2. The tire sidewall arc marking layout verification system based on identity authentication according to claim 1, characterized in that, The process of obtaining a two-dimensional planar unfolded image by correcting the macroscopic arc distortion in the tire sidewall arc marking image through the surface unfolding model by the arrangement feature extraction node includes: identifying the arc positioning baseline in the tire sidewall arc marking image, and fitting the three-dimensional cylindrical parametric equation of the tire sidewall based on the arc positioning baseline. The inverse mapping relationship between image pixels from the two-dimensional imaging plane to the tire side surface is established based on the three-dimensional cylindrical parametric equation. Based on the inverse mapping relationship, bilinear interpolation resampling is performed on the tire sidewall arc marking image to transform the pixels in the curved coordinate system to a planar rectangular coordinate system with arc length and radial distance as axes, thereby generating the two-dimensional planar unfolded image that eliminates curvature deformation.

3. The tire sidewall arc marking layout verification system based on identity authentication according to claim 1, characterized in that, The process of the arrangement feature extraction node extracting the microscopic arrangement offset matrix of the marker edge nodes in the two-dimensional planar unfolded image in the circular arc coordinate system includes: performing sub-pixel edge detection on the two-dimensional planar unfolded image and locating the discrete coordinate point set of the marker contour; The discrete coordinate point set is fitted with an elliptic curve, and the endpoints of the major and minor axes of the ellipse are extracted as feature anchor points. Calculate the tangential offset component and normal offset component of each of the aforementioned feature anchor points relative to the vertices of the theoretically arranged mesh; The tangential offset components and normal offset components of all feature anchor points are matrix-recombined according to the topological order of the theoretically arranged mesh to generate the microscopic arrangement offset matrix.

4. The tire sidewall arc marking layout verification system based on identity authentication according to claim 1, characterized in that, The process by which the arrangement feature extraction node inputs the micro-arrangement offset matrix into the cryptographic hash function to generate a physical feature digest includes: performing normalization processing on the micro-arrangement offset matrix to map continuous offset values ​​to discrete integer values ​​with a fixed bit depth. The discrete integer values ​​are expanded into a one-dimensional bit stream in a row-column traversal order; The head of the one-dimensional bitstream is spliced ​​with the position salt value corresponding to the physical location to form an enhanced bitstream; The enhanced bitstream is input into the cryptographic hash function, and a fixed-length hash value is output as the physical feature digest.

5. The tire sidewall arc marking layout verification system based on identity authentication according to claim 1, characterized in that, The process of the identity authentication server performing a bitwise XOR comparison operation between the received physical feature digest and the pre-stored benchmark digest, and calculating the Hamming distance of the bit difference in the XOR result, includes: retrieving the benchmark digest associated with the unique identifier of the tire to be verified from the secure storage area of ​​the identity authentication server. Perform a bitwise logical XOR operation between the physical feature summary and the baseline summary to generate an XOR difference sequence; The total number of bits with a logical value of true in the XOR difference sequence is counted, and the total number of bits is identified as the Hamming distance. The Hamming distance characterizes the degree of fluctuation deviation between physical entity features and registration features.

6. The tire sidewall arc marking layout verification system based on identity authentication according to claim 1, characterized in that, If the Hamming distance is less than the preset verification threshold, the process of outputting the identity verification pass signal includes: acquiring the current physical environment parameters collected by the environmental temperature and humidity sensor, dynamically correcting the initial verification threshold based on the current physical environment parameters, and obtaining the preset verification threshold. Construct comparator logic to numerically compare the Hamming distance with the preset verification threshold; When the Hamming distance is determined to be less than the preset verification threshold, the security register is triggered to write the verification success status bit. Read the verification success status bit, generate the authentication pass signal containing a timestamp and the unique identifier of the tire to be verified, and push the authentication pass signal to the application layer interface.

7. The tire sidewall arc marking layout verification system based on identity authentication according to claim 2, characterized in that, The process of performing bilinear interpolation resampling on the tire sidewall arc marking image based on the inverse mapping relationship includes: traversing the target pixel coordinates in the Cartesian coordinate system and calculating the floating-point source coordinates corresponding to the target pixel coordinates in the tire sidewall arc marking image through the inverse mapping relationship; Locate the four integer pixels adjacent to the floating-point source coordinates; Obtain the pixel grayscale values ​​of the four integer pixels; Using the spatial distance between the floating-point source coordinates and the four integer pixels as weights, a weighted summation operation is performed on the pixel grayscale values, and the result of the weighted summation operation is assigned to the target pixel coordinates until the full-image pixel mapping is completed.

8. The tire sidewall arc marking layout verification system based on identity authentication according to claim 3, characterized in that, The process of calculating the tangential offset component and normal offset component of each feature anchor point relative to the vertices of the theoretically arranged mesh includes: A local two-dimensional orthogonal coordinate system is constructed with the vertices of the theoretically arranged grid as the origin, wherein the horizontal axis is parallel to the tangent direction of the tire circumference and the vertical axis is parallel to the radial normal direction of the tire. Calculate the horizontal and vertical projection distances of the feature anchor points in the local two-dimensional orthogonal coordinate system; The horizontal axis projection distance is determined as the tangential offset component, and the vertical axis projection distance is determined as the normal offset component; Smoothing filtering is performed on the tangential offset component and the normal offset component to eliminate high-frequency noise interference.

9. The tire sidewall arc marking layout verification system based on identity authentication according to claim 4, characterized in that, The process of performing normalization on the micro-arrangement offset matrix and mapping continuous offset values ​​to discrete integer values ​​with a fixed bit depth includes: scanning the micro-arrangement offset matrix and extracting the absolute value of the global maximum offset and the absolute value of the global minimum offset. A linear mapping interval is constructed using the absolute value of the global maximum offset and the absolute value of the global minimum offset; Subtract the absolute value of the global minimum offset from each consecutive offset value in the micro-arrangement offset matrix, and then divide by the span of the linear mapping interval to obtain a normalized floating-point number. The normalized floating-point number is multiplied by a preset quantization factor and then rounded down to generate the discrete integer value with the fixed bit depth.

10. The tire sidewall arc marking layout verification system based on identity authentication according to claim 5, characterized in that, The process of counting the total number of bits with logical values ​​of true in the XOR difference sequence and confirming the total number of bits as the Hamming distance includes: performing a bitwise accumulation operation on the XOR difference sequence to obtain the initial Hamming distance; The error correction coding verification matrix stored in the identity authentication server is invoked to perform adjoint calculation on the physical feature digest; If the result of the syndrome calculation is non-zero, it is confirmed that there is a channel transmission error in the physical feature digest, and the error bit is located based on the result of the syndrome calculation. A compensation deduction operation equal to the number of error bits is performed on the initial Hamming distance, and the value after compensation deduction is confirmed as the final Hamming distance.

Images