A holographic image anti-counterfeiting pattern laser processing method, storage medium and product

Abstract

The application provides a holographic image anti-counterfeiting pattern laser processing method, a storage medium and a product, a polarization angle sequence is determined through spectrum processing, a neural network is used to optimize microstructure arrangement, an encoding matrix is generated, hidden data is integrated to improve information dimension, then simulation calculation and iterative updating are used to enhance the anti-counterfeiting effect of the encoding matrix, resistance characteristics are extracted and robustness is verified, finally, material surface writing is guided to form a holographic image anti-counterfeiting pattern, and finally, a multi-dimensional verification protocol is combined to ensure the stability of the polarization response in different environments. The application significantly improves the cracking resistance and environmental adaptability of the anti-counterfeiting pattern through the deep combination of the polarization angle and the microstructure optimization.

Description

Technical Field

[0001] This invention belongs to the field of anti-counterfeiting and security technology, and more specifically relates to a laser processing method, storage medium and product for holographic anti-counterfeiting patterns. Background Technology

[0002] In the field of modern anti-counterfeiting technology, holographic imaging technology, due to its high security and difficulty in replication, has become an important means of protecting product authenticity and ensuring market order. Common anti-counterfeiting patterns on the market often rely on a single visual effect or simple structural design, lacking multi-layered information concealment capabilities. This limitation makes anti-counterfeiting labels easily cracked by high-level counterfeiting techniques, especially since the information presentation under different lighting conditions is insufficient, failing to meet the demands of high security. More importantly, existing methods have bottlenecks in the density and complexity of information encoding, unable to carry more hidden data within a limited space, thus limiting the depth and breadth of anti-counterfeiting effects.

[0003] Focusing on the technical challenges, the core issue lies in how to achieve the superposition and control of multiple information dimensions at a microscale. In particular, the precise control of the polarization state of light wave vibrations is a major difficulty. Without accurately dividing and dynamically switching different polarization angles, it is impossible to form diverse microstructures on or within the material surface, such as fine scratches or hole arrangements in different directions, thus limiting the information representation capability of the pattern under different observation conditions. This lack of controllability directly affects the precise design of the distance between microstructures. If the spatial interval during writing with different polarization states cannot be flexibly adjusted, it is difficult to integrate additional information encoding into the arrangement of the structures. When creating anti-counterfeiting patterns, if the distance between structures etched with one light direction and those etched with another light direction cannot be controlled to a specific value, it is impossible to hide more information through these distance differences, thus affecting the anti-counterfeiting effect.

[0004] Therefore, how to achieve precise division and dynamic switching of light polarization state at a microscale, and increase the information encoding dimension by accurately controlling the structural spacing, has become a key issue in improving the security of holographic anti-counterfeiting technology. Summary of the Invention

[0005] This invention proposes a laser processing method for holographic anti-counterfeiting patterns. First, the polarization angle sequence is determined through spectral processing, and the arrangement of microstructures is optimized using a neural network to generate an encoding matrix, incorporating hidden data to enhance information dimensionality. Subsequently, simulation calculations and iterative updates enhance the anti-counterfeiting effect of the encoding matrix, extracting resistance features and verifying their robustness. Finally, this guides the writing onto the material surface, forming a holographic anti-counterfeiting pattern. Finally, a multi-dimensional verification protocol is combined to ensure the stability of polarization response under different environments. This invention, through the deep integration of polarization angle and microstructure optimization, significantly improves the anti-counterfeiting pattern's resistance to cracking and its environmental adaptability.

[0006] The technical solution of this invention is implemented as follows: A laser processing method for holographic anti-counterfeiting patterns includes the following steps: Acquire initial light polarization state data, determine the preliminary range of polarization angles, and obtain a polarization angle sequence; A neural network algorithm is used to simulate and optimize the arrangement of microstructures. If the deviation of the structural spacing in the simulation results exceeds a preset threshold, the polarization angle sequence is adjusted to obtain optimized structural spacing data. A coding matrix is ​​generated for the multi-level information overlay process. Hidden data carrying elements are incorporated into the coding matrix, and the extended configuration of the information coding dimension is determined to obtain the extended coding matrix. The response of light interaction under different observation conditions is simulated and calculated. If the interaction response does not reach the anti-counterfeiting effect depth standard, the elements of the coding matrix are iteratively updated to obtain an enhanced coding matrix. Anti-counterfeiting resistance features are extracted from the enhanced coding matrix, the robustness of the features is verified, and the integrity index of the resistance mechanism is determined. Based on the integrity index, the writing process on the material surface is guided, and the resistance mechanism is integrated into the microstructure arrangement to obtain holographic anti-counterfeiting pattern data. The polarization response of the test pattern under different lighting conditions is evaluated. If the response deviation exceeds the preset threshold, the writing process parameters are adjusted to obtain optimized anti-counterfeiting pattern data. Based on the optimized anti-counterfeiting pattern data, a multi-dimensional anti-counterfeiting verification protocol is generated. The polarization angle sequence and resistance features are incorporated into the protocol to determine the extended configuration of the verification process.

[0007] Preferably, initial light polarization state data is acquired, a preliminary range of polarization angles is determined, and a polarization angle sequence is obtained, including: The preset light source device acquires the initial light polarization state data, and uses the Stokes parameter solution model to process the initial light polarization state data to generate a Stokes vector matrix. The Stokes vector matrix is ​​then subjected to frequency domain transformation and feature extraction to obtain a spectral feature dataset. The gradient change value of the spectral feature dataset is calculated and the state change points are marked. The distribution interval of the state change points is used to determine the preliminary division range of the polarization angle. Based on the preliminary division range, the Stokes vector matrix is ​​segmented and mapped and calculated to obtain the polarization angle sequence.

[0008] Preferably, a neural network algorithm is used to simulate and optimize the arrangement of microstructures. If the deviation in the structural spacing in the simulation results exceeds a preset threshold, the polarization angle sequence is adjusted to obtain optimized structural spacing data, including: The polarization angle sequence is input into the convolutional neural network model to generate a simulated arrangement feature map, which contains spatial location information of small structures. The geometric centroid coordinate data of the simulated arrangement feature map are extracted to construct the structure spacing distribution matrix, which is calculated from the Euclidean distance between adjacent geometric centroid coordinates. The difference between the structural spacing distribution matrix and the standard design spacing template is calculated to generate a structural spacing deviation tensor. If the structural spacing deviation tensor exceeds a preset tolerance threshold, an angle correction gradient vector is calculated to update the polarization angle sequence until the deviation meets the condition, thereby obtaining optimized structural spacing data.

[0009] Preferably, an encoding matrix is ​​generated for the multi-level information overlay process, hidden data-carrying elements are incorporated into the encoding matrix, and the extended configuration of the information encoding dimension is determined to obtain the extended encoding matrix, including: An initial sparse matrix is ​​constructed based on optimized structural spacing data, which is obtained by transforming the basic topological mapping graph. An encoding matrix framework is generated in the multidimensional feature subspace corresponding to the initial sparse matrix, and hidden data is embedded in the low-rank region of the encoding matrix framework to form a hybrid encoding block. The extended configuration parameters of the information encoding dimension are determined according to the dimensional characteristics of the hybrid encoding block, and the tensor product operation is performed on the hybrid encoding block using the extended configuration parameters to obtain an extended encoding matrix containing multi-level superimposed information and hidden data.

[0010] Preferably, the response of light interaction under different observation conditions is simulated and calculated. If the interaction response does not reach the anti-counterfeiting depth standard, the elements of the encoding matrix are iteratively updated to obtain an enhanced encoding matrix, including: A ray interaction model for simulating light field behavior is constructed based on the extended encoding matrix, ray tracing is performed, and interactive response data formed by light rays on the surface of microstructures are collected. Calculate the deviation matrix between the interactive response data and the preset anti-counterfeiting effect depth standard. If the deviation matrix exceeds the threshold, calculate the gradient descent direction and matrix update step size to iteratively update the expanded encoding matrix to obtain the enhanced encoding matrix.

[0011] Preferably, anti-counterfeiting resistance features are extracted from the enhanced coding matrix, the robustness of the features is verified, and the integrity index of the resistance mechanism is determined. Based on the integrity index, the writing process on the material surface is guided, integrating the resistance mechanism into the microstructure arrangement to obtain holographic anti-counterfeiting pattern data, including: An enhanced coding matrix is ​​obtained and multi-scale singular value decomposition is performed. Singular value sequences are extracted to construct anti-counterfeiting and anti-cracking resistance feature vectors. The anti-counterfeiting and anti-cracking resistance feature vectors are mapped to a multi-dimensional spectral feature space and superimposed with nonlinear optical distortion noise to generate a test spectral dataset. The correlation decay curve between the test spectral dataset and the standard lossless spectral data is calculated. If the robustness quantization score determined by the correlation decay curve is lower than the preset defense threshold, the construction parameters of the anti-counterfeiting and anti-cracking resistance feature vectors are corrected. Based on the distribution density of the corrected anti-counterfeiting and anti-cracking resistance feature vectors in the multi-dimensional spectral feature space, the integrity index of the resistance mechanism is calculated. Integrity indicators are obtained and mapped to phase modulation control variables to obtain diffraction grating spatial frequency distribution data. A microstructure arrangement topology model is constructed, and the resistance mechanism is transformed into etching depth perturbation amount and superimposed on the microstructure arrangement topology model to generate three-dimensional surface relief structure data. The exposure energy density distribution matrix is ​​calculated, and the exposure energy density distribution matrix is ​​encoded to obtain holographic image anti-counterfeiting pattern data.

[0012] Preferably, the polarization response of the test pattern under different lighting conditions is evaluated. If the response deviation exceeds a preset threshold, the writing process parameters are adjusted to obtain optimized anti-counterfeiting pattern data, including: The holographic anti-counterfeiting pattern data is acquired and mapped onto a virtual optical surface model. A dynamic spectral illumination environment model containing incident light from multiple angles is constructed. The polarization response feature matrix is ​​extracted from the virtual optical surface model. The polarization response feature matrix is ​​compared with a standard reference spectral database to determine the response deviation value. If the response deviation value exceeds a preset threshold, the compensation correction amount for the writing process parameters is calculated. The compensation correction amount is then used to reconstruct the holographic image anti-counterfeiting pattern data to obtain optimized anti-counterfeiting pattern data.

[0013] Preferably, based on the optimized anti-counterfeiting pattern data, a multi-dimensional anti-counterfeiting verification protocol is generated, incorporating polarization angle sequences and resistance features into the protocol, and determining the extended configuration of the verification process, including: Optimized anti-counterfeiting pattern data is obtained and the phase distribution of the holographic grating is analyzed. High-frequency diffraction feature points are extracted as basic encryption nodes to construct a polarization response mapping model to calculate the polarization angle sequence. Weak segments of the polarization angle sequence in the virtual optical noise environment are identified and optical resistance feature masks are generated. The optical resistance feature masks and polarization angle sequences are superimposed to construct a multi-dimensional anti-counterfeiting verification protocol. Polarization modulation parameters are extracted according to the protocol to determine the extended configuration of the verification process.

[0014] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the method described thereon.

[0015] In another aspect, the present invention also provides a computer program product, comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method described herein.

[0016] The beneficial effects of this invention are as follows: Addressing the problems of large response deviations in anti-counterfeiting patterns under different lighting conditions, insufficient robustness of the encoding matrix, and poor scalability of multi-dimensional verification protocols, this invention achieves a complete process from initial light polarization state data analysis to the generation of optimized anti-counterfeiting pattern data by integrating spectral processing algorithms, neural network optimization, and dynamic spectral verification. Through a deep integration of polarization angle and microstructure optimization, this invention significantly improves the anti-counterfeiting pattern's resistance to cracking and its environmental adaptability, providing an efficient solution for information security in complex scenarios. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] A laser processing method for holographic anti-counterfeiting patterns includes the following steps: S101: Obtain initial light polarization state data, determine the preliminary division range of polarization angles, and obtain the polarization angle sequence.

[0021] The preset light source device acquires the initial light polarization state data, and uses the Stokes parameter solution model to process the initial light polarization state data to generate a Stokes vector matrix. The Stokes vector matrix is ​​then subjected to frequency domain transformation and feature extraction to obtain a spectral feature dataset. The gradient change value of the spectral feature dataset is calculated and the state change points are marked. The distribution interval of the state change points is used to determine the preliminary division range of the polarization angle. Based on the preliminary division range, the Stokes vector matrix is ​​segmented and mapped and calculated to obtain the polarization angle sequence.

[0022] Initial light polarization state data is obtained from a pre-set laser source device, such as linearly polarized light output from a He-Ne laser with a center wavelength of 632.8 nm after passing through a polarizer. The detector samples the original Stokes vector data: S0=1.000, S1=0.982, S2=0.034, S3=0.008. This data is analyzed using a spectral processing algorithm. First, the degree of polarization P=√(S1²+S2²+S3²) / S0=0.983, confirming that the light is close to a fully polarized state. Then, the polarization azimuth angle ψ=(1 / 2)•arctan(S2 / S1)=0.0173 radians, approximately equal to 0.99°, is calculated. Simultaneously, the ellipticity angle χ=(1 / 2)•arcsin(S3 / S0)=0.004 radians, indicating that the ellipticity is extremely small. Finally, using the polarization azimuth angle ψ as the core parameter, a preliminary division is made, dividing the region from -90° to +90°. The polarization angle range is divided into 37 intervals with a step size of 5°, namely [-90°, -85°], [-85°, -80°]...[+85°, +90°], and the currently calculated 0.99° is mapped to the interval [+0°, +5°]. Then, based on the light intensity distribution characteristics and comparison with the historical sample database, when ψ falls into [+0°, +5°], the representative value of the polarization angle sequence of the light is further refined to +3°, and the sequence is recorded as {…, -2°, +3°, +8°, ​​...}, which is used for subsequent polarization compensation or adaptive adjustment of the optical path.

[0023] S102, a neural network algorithm is used to simulate and optimize the arrangement of microstructures. If the deviation of the structural spacing in the simulation results exceeds the preset threshold, the polarization angle sequence is adjusted to obtain optimized structural spacing data.

[0024] The polarization angle sequence is input into a convolutional neural network model to generate a simulated arrangement feature map, which contains spatial location information of minute structures. The geometric centroid coordinates are extracted from the simulated arrangement feature map to construct a structural spacing distribution matrix, which is calculated from the Euclidean distance between adjacent geometric centroid coordinates. The difference between the structural spacing distribution matrix and the standard design spacing template is calculated to generate a structural spacing deviation tensor. If the structural spacing deviation tensor exceeds a preset tolerance threshold, the angle correction gradient vector is calculated to update the polarization angle sequence until the deviation meets the condition, thereby obtaining optimized structural spacing data.

[0025] Based on the existing polarization angle sequence, the system first builds a convolutional neural network model using a deep learning framework to simulate the arrangement characteristics of microstructures. Assuming the input polarization angle sequence is {-5°, 0°, +5°, +10°}, the network model uses these angle values ​​as input features, combined with a pre-trained weight matrix, and predicts the initial spacing distribution of the microstructures through three convolutional layers and two fully connected layers. The simulated result is a spacing value array {2.3nm, 2.5nm, 2.8nm, 3.1nm}. Subsequently, the system automatically calculates the deviation of these spacing values, setting a preset threshold of 0.5nm. Analysis reveals that the difference between the maximum spacing of 3.1nm and the minimum spacing of 2.3nm in the array is 0.8nm, exceeding the threshold and triggering an adjustment mechanism. Next, the system invokes an optimization algorithm, using gradient descent to iteratively adjust the polarization angle sequence. The initial learning rate is 0.01. After five iterations, the sequence is updated to {-4°, +1°, +6°, +9°}, and then re-input into the neural network model for simulation, yielding a new spacing array {2.4nm, 2.6nm, 2.7nm, 2.9nm}. At this point, the difference between the maximum and minimum spacing is 0.5nm, meeting the threshold requirement. Finally, the system stores the optimized spacing data in the database and links it to the subsequent optical device design module, forming a data loop. The entire process is driven by automated scripts, with parameter calculations and model training completed through a cloud computing platform, ensuring the real-time performance and accuracy of data processing.

[0026] S103, generate an encoding matrix for the multi-level information overlay process, incorporate hidden data carrying elements into the encoding matrix, determine the extended configuration of the information encoding dimension, and obtain the extended encoding matrix.

[0027] An initial sparse matrix is ​​constructed based on optimized structural spacing data, which is obtained by transforming the basic topological mapping graph. An encoding matrix framework is generated in the multidimensional feature subspace corresponding to the initial sparse matrix, and hidden data is embedded in the low-rank region of the encoding matrix framework to form a hybrid encoding block. The extended configuration parameters of the information encoding dimension are determined according to the dimensional characteristics of the hybrid encoding block, and the tensor product operation is performed on the hybrid encoding block using the extended configuration parameters to obtain an extended encoding matrix containing multi-level superimposed information and hidden data.

[0028] Based on optimized structural spacing data, the system first reads the stored spacing array {2.2nm, 2.5nm, 2.7nm, 3.0nm}, constructs an initial encoding matrix through a matrix generation algorithm, assuming the matrix dimension is 4x4, and the element values ​​are linearly mapped from the spacing data to weight values ​​between 0 and 1, resulting in matrix elements ranging from 0.22 to 0.30. Subsequently, the system calls a pseudo-random number generator, combined with a hash algorithm, to embed the hidden data carrying elements into the diagonal positions of the matrix. Specifically, the hidden data "1011" is mapped to a weight adjustment value of ±0.05 in binary encoding. After adjustment, the diagonal elements are {0.27, 0.30, 0.25, 0.35}. Next, the system analyzes the expansion requirements of the information encoding dimension. Based on the matrix eigenvalue decomposition algorithm, the eigenvalue distribution of the current matrix is ​​calculated. It is found that the maximum eigenvalue is 1.2, the minimum is 0.8, and the difference of 0.4 is less than the preset expansion threshold of 0.5, triggering the dimension expansion mechanism. The system automatically expands the matrix dimension from 4x4 to 6x6. The new elements are filled using an interpolation algorithm, resulting in a new matrix element range between 0.20 and 0.38. At the same time, the eigenvalues ​​are recalculated, with a maximum value of 1.5, a minimum value of 0.9, and a difference of 0.6, which meets the expansion conditions. Finally, the system stores the expanded encoding matrix in a distributed file system and associates it with the subsequent information decoding module, forming a closed loop of data flow. The entire process is automated script scheduling and uses a parallel computing framework to complete matrix operations and data embedding, ensuring high processing efficiency.

[0029] S104 simulates the response of light interaction under different observation conditions. If the interaction response does not meet the anti-counterfeiting depth standard, the elements of the encoding matrix are iteratively updated to obtain an enhanced encoding matrix.

[0030] A ray interaction model for simulating light field behavior is constructed based on the extended encoding matrix. Ray tracing is performed to collect interactive response data formed by light on the surface of the microstructure. The deviation matrix between the interactive response data and the preset anti-counterfeiting effect depth standard is calculated. If the deviation matrix exceeds the threshold, the gradient descent direction and matrix update step size are calculated to iteratively update the extended encoding matrix to obtain an enhanced encoding matrix.

[0031] After obtaining the expanded encoding matrix, the system immediately loads the 6×6 matrix to the high-performance computing node. Using the Monte Carlo ray tracing algorithm combined with Fresnel equations, it simulates the interaction behavior of incident light under different wavelengths and incident angles. Specifically, the incident light wavelength range is set to 380nm to 780nm with a step size of 10nm, and the incident angle increases by 5° from 0° to 75°. Simultaneously, considering that the polarization states are 50% s-polarization and 50% p-polarization, each ray is tracked for a maximum of 8 scattering and reflection events. The intensity, phase, and polarization changes after each interaction are recorded, accumulating approximately 240 million parallel simulations of ray paths to obtain the original response spectrum curve and angle-intensity distribution map. Subsequently, the system extracts key anti-counterfeiting features, including a reflectivity peak of 0.78 at a wavelength of 550nm at a 45° incident angle, a full width at half maximum (FWHM) of 32nm, and an average peak shift of 18nm at different viewing angles. These are compared one by one with the preset anti-counterfeiting depth standard, revealing that the reflectivity peak of 0.78 is below the target threshold. If the value is 0.85 and the peak shift is 18nm, which is less than the required value of 22nm, the current interactive response is deemed substandard. The iterative update module is triggered. The system uses gradient descent optimization combined with the L-BFGS algorithm, using the weighted error of the two indicators as the loss function. Perturbations are applied to the non-zero elements in the encoding matrix. The initial learning rate is 0.008, and the momentum coefficient is set to 0.92. After 47 iterations, the overall adjustment range of the matrix elements is controlled within ±0.042. After the update, the elements near the main diagonal of the matrix change to {0.31, 0.36, 0.28, 0.39, 0.24, 0.33}, with corresponding minor adjustments at other positions. The same light interaction simulation is re-executed, resulting in a new peak reflectance of 0.87 and a peak shift of 24nm, both exceeding the anti-counterfeiting depth standard. The system determines the enhancement is successful, overwrites and saves the enhanced encoding matrix to the original storage path, and generates a version identifier associated with the anti-counterfeiting verification database for subsequent optical reading devices to call and compare, forming a complete closed-loop processing flow.

[0032] S105, extract anti-counterfeiting and anti-counterfeiting resistance features from the enhanced coding matrix, verify the robustness of the features, and determine the integrity index of the resistance mechanism; guide the writing process on the material surface according to the integrity index, integrate the resistance mechanism into the microstructure arrangement, and obtain holographic image anti-counterfeiting pattern data.

[0033] An enhanced coding matrix is ​​obtained and multi-scale singular value decomposition is performed. Singular value sequences are extracted to construct anti-counterfeiting and anti-cracking resistance feature vectors. The anti-counterfeiting and anti-cracking resistance feature vectors are mapped to a multi-dimensional spectral feature space and superimposed with nonlinear optical distortion noise to generate a test spectral dataset. The correlation decay curve between the test spectral dataset and the standard lossless spectral data is calculated. If the robustness quantization score determined by the correlation decay curve is lower than the preset defense threshold, the construction parameters of the anti-counterfeiting and anti-cracking resistance feature vectors are corrected. Based on the distribution density of the corrected anti-counterfeiting and anti-cracking resistance feature vectors in the multi-dimensional spectral feature space, the integrity index of the resistance mechanism is calculated. Integrity indicators are obtained and mapped to phase modulation control variables to obtain diffraction grating spatial frequency distribution data. A microstructure arrangement topology model is constructed, and the resistance mechanism is transformed into etching depth perturbation amount and superimposed on the microstructure arrangement topology model to generate three-dimensional surface relief structure data. The exposure energy density distribution matrix is ​​calculated, and the exposure energy density distribution matrix is ​​encoded to obtain holographic image anti-counterfeiting pattern data.

[0034] The system first inputs a 6×6 enhanced matrix into the feature extraction module. The matrix element sequence is processed using Fourier transform combined with wavelet decomposition to separate high-frequency components as potential resistance features. This yields two sets of significant periodic signals: a main frequency peak at 0.42 periods / element and a secondary peak at 0.19 periods / element. Simultaneously, the singular value decomposition of the matrix is ​​calculated, extracting the first three singular values ​​as 12.47, 8.92, and 5.61 as structural stability indicators. Subsequently, a spectral processing algorithm is used to verify the robustness of these features. Specifically, a simulated attack dataset is loaded, including 7200 perturbation samples with Gaussian noise (standard deviation 0.015 to 0.085), uniform pruning resulting in 15% to 40% element loss, affine transformation distortion angles of ±12°, and compression quality factors of 30 to 85. The feature vectors of each sample are recalculated, and their cosine similarity is calculated with the original feature vectors. Statistical results show that the average similarity remains above 0.963 within a noise standard deviation of 0.065. When 30% of the data is lost during cropping, the similarity drops to 0.874 but remains above the warning threshold of 0.82. With a compression quality factor of 40, the similarity is 0.918. Further, a resistance mechanism integrity index is constructed. A weighted fusion method is used to normalize four sub-indicators: periodic signal retention rate (0.941), maximum relative change rate of singular values ​​(0.073), minimum similarity under comprehensive attacks (0.874), and feature distinguishability (KL divergence compared to a random matrix) (3.28). Weights of 0.38, 0.24, 0.22, and 0.16 are assigned, respectively. The final integrity score is calculated to be 0.906, exceeding the preset security threshold of 0.88. This indicates a complete resistance mechanism with high anti-counterfeiting and anti-cracking resistance. The system automatically stores this integrity index and core feature vector in the encrypted metadata area, synchronously associated with the enhanced matrix version, facilitating rapid comparison and confirmation by subsequent anti-counterfeiting verification devices within 0.3 seconds.

[0035] Guided by the integrity index of 0.906, the system initiates the material surface writing control process. First, the holographic anti-counterfeiting pattern data is decomposed into a 512×512 pixel phase distribution map, which is then optimized using the Gerchberg-Saxton iterative method. The number of iterations is set to 180, and the initial random phase error is controlled within 0.05 radians. Finally, a phase modulation template with a mean square error of 0.0072 is obtained through convergence. Subsequently, according to the resistance mechanism requirements, the key high-frequency periodic components (frequency range of 0.35 to 0.48 cycles / pixel) in the phase template are mapped to the microstructure depth modulation. A three-layer nanostructure arrangement is generated using a layered photolithography strategy. The first layer has a depth modulation amplitude of 68nm corresponding to the main periodic signal, the second layer has a depth of 42nm with superimposed secondary modulation to enhance scattering characteristics, and the third layer uses 15nm random dithering. Unpredictable perturbations are introduced into the layer to enhance the resistance to replication. The system then calculates the spatial autocorrelation function of the microstructure, verifying that the peak width is controlled within 1.8 pixels. At the same time, the number of connected components after local binarization is 2847, and the average domain area is 3.14 pixels², ensuring that the pattern has sufficient complexity and information entropy (the calculated Shannon entropy value is 7.82 bits / pixel). Finally, a holographic reconstruction image is generated under the conditions of an incident angle of ±18° and a wavelength of 532nm using a ray tracing simulation algorithm. The analysis shows that the peak signal-to-noise ratio of the reconstruction reaches 41.6dB, and the sidelobe suppression ratio is better than -29.4dB, confirming that the microstructure arrangement has been effectively integrated into the resistance mechanism, forming holographic anti-counterfeiting pattern data with high security and optical verifiability. The written parameters and integrity indicators are encapsulated together in the blockchain anchor record to achieve tamper-proof traceability.

[0036] S106, test the polarization response of the pattern under different lighting conditions. If the response deviation exceeds the preset threshold, adjust the writing process parameters to obtain optimized anti-counterfeiting pattern data.

[0037] Holographic anti-counterfeiting pattern data is acquired and mapped onto a virtual optical surface model. A dynamic spectral illumination environment model containing incident light from multiple angles is constructed. The polarization response feature matrix is ​​extracted from the virtual optical surface model. The polarization response feature matrix is ​​compared with a standard reference spectral database to determine the response deviation value. If the response deviation value exceeds a preset threshold, the compensation correction amount of the writing process parameters is calculated. The compensation correction amount is used to reconstruct the holographic anti-counterfeiting pattern data to obtain optimized anti-counterfeiting pattern data.

[0038] The system first uses holographic anti-counterfeiting pattern data and calls a dynamic spectral verification algorithm to automatically test the polarization response of the pattern under various lighting conditions. Specifically, a polarization decomposition model is used, such as decomposing the incident light into transverse electric field and transverse magnetic field components. The test lighting environment parameters are set as wavelength range 450-650nm, incident angle range ±25°, and scanning is performed in 0.1° steps to calculate the eigenvalue distribution of the polarization response matrix. The maximum eigenvalue deviation is recorded as 0.032, and the average deviation is 0.015. Then, the response data is compared with a preset threshold of 0.025. The system automatically determines that the deviation exceeds the threshold at some angles, triggering the parameter adjustment process and entering the writing process optimization stage. In the optimization process, the system adjusts the writing parameters through an adaptive feedback algorithm, specifically increasing the laser writing power from the initial 3.5mW to 3.8mW and shortening the pulse duration from 2.4ns to 2.1ns to improve the microstructure accuracy of the pattern surface. Next, the anti-counterfeiting pattern data is regenerated. The system uses an improved Fourier transform algorithm to perform frequency domain analysis on the new pattern, extracting spectral features and ensuring that the proportion of high-frequency components increases from 42.3% to 46.7%. The dynamic spectral verification algorithm is run again to calculate the polarization response deviation, yielding a new deviation value of 0.019, below the preset threshold, confirming the optimization's effectiveness. Finally, the system associates and stores the adjusted parameters with the newly generated anti-counterfeiting pattern data, forming a traceable optimization record. This ensures that subsequent verification processes can call the latest data to complete consistency checks. The entire process forms a closed-loop logic through algorithm-driven and data analysis, ensuring the stability of the anti-counterfeiting pattern data under different environments.

[0039] S107. Based on the optimized anti-counterfeiting pattern data, a multi-dimensional anti-counterfeiting verification protocol is generated, incorporating polarization angle sequences and resistance features into the protocol, and determining the extended configuration of the verification process.

[0040] Optimized anti-counterfeiting pattern data is obtained and the phase distribution of the holographic grating is analyzed. High-frequency diffraction feature points are extracted as basic encryption nodes to construct a polarization response mapping model to calculate the polarization angle sequence. Weak segments of the polarization angle sequence in the virtual optical noise environment are identified and optical resistance feature masks are generated. The optical resistance feature masks and polarization angle sequences are superimposed to construct a multi-dimensional anti-counterfeiting verification protocol. Polarization modulation parameters are extracted according to the protocol to determine the extended configuration of the verification process.

[0041] Based on optimized anti-counterfeiting pattern data, the system automatically generates a multi-dimensional anti-counterfeiting verification protocol. Specifically, it constructs a fusion model of polarization angle sequence and resistance features. First, the system extracts the polarization angle data of the anti-counterfeiting pattern. For example, setting the angle scanning range to 0° to 180°, sampling is performed in 0.5° steps to generate a polarization angle sequence containing 360 data points. Then, the sequence is encoded using a vector quantization algorithm, and the polarization intensity value corresponding to each angle is calculated, resulting in an average intensity deviation of 0.012 and a maximum deviation of 0.028. This data is then embedded into the core fields of the verification protocol. Next, for the construction of resistance features, the system uses an entropy-based feature extraction algorithm to analyze the signal attenuation characteristics of the anti-counterfeiting pattern under different environments. The test environment noise range is set to -30dB to -10dB, and the signal entropy distribution is calculated. The average entropy value is recorded as 3.45, and the maximum entropy value is 3.82. By associating and mapping the entropy features with the polarization angle sequence, a resistance feature matrix is ​​formed, ensuring that the verification protocol has anti-interference capabilities. Then, the system designs extended configurations for the verification process, calling a dynamic weight allocation algorithm to automatically adjust the verification priority based on environmental variables. The initial value for polarization verification weight is set to 0.6, and the resistance feature weight to 0.4. When noise exceeds -15dB, the weights are automatically adjusted to 0.5 and 0.5 respectively to balance verification accuracy. Simultaneously, the system generates extended configuration scripts, including parameters such as a 500ms verification time window and a 10kHz data sampling rate, ensuring the verification process adapts to various application scenarios. The entire process is executed automatically through the algorithm, forming a closed-loop verification logic between data. The mapping relationship between the polarization angle sequence and the resistance feature matrix is ​​stored in an encrypted database, providing data support for subsequent business operations such as batch verification or cross-platform verification, ensuring the stability and scalability of the verification protocol.

[0042] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method.

[0043] In another aspect, the present invention also proposes a computer program product, comprising a computer program, characterized in that the computer program implements the above-described method when executed by a processor.

[0044] In particular, according to some embodiments of this disclosure, the processes described above can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined above in the methods of some embodiments of this disclosure.

[0045] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a task data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated task data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0046] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol, such as HTTP (Hypertext Transfer Protocol), and can interconnect with digital task data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0047] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: determine the network connection status of the switch production line management application in response to detecting a query operation on a production collaboration document in the switch production line management application; replace the webpage entry information corresponding to the production collaboration document with target entry file information and load target webpage resource information in response to determining that the network connection status of the switch production line management application indicates an offline state, so as to display the webpage of the production collaboration document offline in the switch production line management application, wherein the target entry file information is the file information of the entry file corresponding to the webpage of the production collaboration document downloaded in advance, and the target webpage resource information is the resource information corresponding to the webpage stored locally; in response to determining that the network connection status of the switch production line management application indicates an online state and that the webpage resource information corresponding to the production collaboration document is not stored locally, download the webpage resource information of the webpage from the production line document server, wherein the webpage resource information includes an entry file and resource information; display the webpage of the production collaboration document in the switch production line management application according to the webpage resource information, and store the webpage resource information in a local database.

[0048] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including product-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A laser processing method of a holographic image anti-counterfeiting pattern, characterized in that, Includes the following steps: Acquire initial light polarization state data, determine the preliminary range of polarization angles, and obtain a polarization angle sequence; A neural network algorithm is used to simulate and optimize the arrangement of microstructures. If the deviation of the structural spacing in the simulation results exceeds a preset threshold, the polarization angle sequence is adjusted to obtain optimized structural spacing data. A coding matrix is ​​generated for the multi-level information overlay process. Hidden data carrying elements are incorporated into the coding matrix, and the extended configuration of the information coding dimension is determined to obtain the extended coding matrix. The response of light interaction under different observation conditions is simulated and calculated. If the interaction response does not reach the anti-counterfeiting effect depth standard, the elements of the coding matrix are iteratively updated to obtain an enhanced coding matrix. Anti-counterfeiting resistance features are extracted from the enhanced coding matrix, the robustness of the features is verified, and the integrity index of the resistance mechanism is determined. Based on the integrity index, the writing process on the material surface is guided, and the resistance mechanism is integrated into the microstructure arrangement to obtain holographic anti-counterfeiting pattern data. The polarization response of the test pattern under different lighting conditions is evaluated. If the response deviation exceeds the preset threshold, the writing process parameters are adjusted to obtain optimized anti-counterfeiting pattern data. Based on the optimized anti-counterfeiting pattern data, a multi-dimensional anti-counterfeiting verification protocol is generated. The polarization angle sequence and resistance features are incorporated into the protocol to determine the extended configuration of the verification process.

2. The laser processing method of the holographic image anti-counterfeit pattern according to claim 1, characterized in that, Acquire initial light polarization state data, determine the preliminary range of polarization angles, and obtain a polarization angle sequence, including: The preset light source device acquires the initial light polarization state data, and uses the Stokes parameter solution model to process the initial light polarization state data to generate a Stokes vector matrix. The Stokes vector matrix is ​​then subjected to frequency domain transformation and feature extraction to obtain a spectral feature dataset. The gradient change value of the spectral feature dataset is calculated and the state change points are marked. The distribution interval of the state change points is used to determine the preliminary division range of the polarization angle. Based on the preliminary division range, the Stokes vector matrix is ​​segmented and mapped and calculated to obtain the polarization angle sequence.

3. The laser processing method of the holographic image security pattern according to claim 1, wherein, A neural network algorithm is used to simulate and optimize the arrangement of microstructures. If the deviation in the structural spacing in the simulation results exceeds a preset threshold, the polarization angle sequence is adjusted to obtain optimized structural spacing data, including: The polarization angle sequence is input into the convolutional neural network model to generate a simulated arrangement feature map, which contains spatial location information of small structures. The geometric centroid coordinate data of the simulated arrangement feature map are extracted to construct the structure spacing distribution matrix, which is calculated from the Euclidean distance between adjacent geometric centroid coordinates. The difference between the structural spacing distribution matrix and the standard design spacing template is calculated to generate a structural spacing deviation tensor. If the structural spacing deviation tensor exceeds a preset tolerance threshold, the angle correction gradient vector is calculated to update the polarization angle sequence until the deviation meets the condition, thereby obtaining optimized structural spacing data.

4. The method of claim 1, wherein the holographic image security pattern is a laser- processed image. A coding matrix is ​​generated for the multi-level information overlay process. Hidden data-carrying elements are incorporated into the coding matrix, and the extended configuration of the information coding dimension is determined to obtain the extended coding matrix, including: An initial sparse matrix is ​​constructed based on optimized structural spacing data, which is obtained by transforming the basic topological mapping graph. An encoding matrix framework is generated in the multidimensional feature subspace corresponding to the initial sparse matrix, and hidden data is embedded in the low-rank region of the encoding matrix framework to form a hybrid encoding block. The extended configuration parameters of the information encoding dimension are determined according to the dimensional characteristics of the hybrid encoding block, and the tensor product operation is performed on the hybrid encoding block using the extended configuration parameters to obtain an extended encoding matrix containing multi-level superimposed information and hidden data.

5. The method of claim 1, wherein the holographic image security pattern is a laser- processed image of a hologram. The response of light interaction under different observation conditions is simulated and calculated. If the interaction response does not meet the anti-counterfeiting depth standard, the elements of the encoding matrix are iteratively updated to obtain an enhanced encoding matrix, including: A ray interaction model for simulating light field behavior is constructed based on the extended encoding matrix, ray tracing is performed, and interactive response data formed by light rays on the surface of microstructures are collected. Calculate the deviation matrix between the interactive response data and the preset anti-counterfeiting effect depth standard. If the deviation matrix exceeds the threshold, calculate the gradient descent direction and matrix update step size to iteratively update the expanded encoding matrix to obtain the enhanced encoding matrix.

6. The method of claim 1, wherein the holographic image security pattern is a holographic image of a person. Anti-counterfeiting and anti-cracking resistance features are extracted from the enhanced coding matrix, the robustness of the features is verified, and the integrity index of the resistance mechanism is determined. Guided by integrity indicators, the writing process on the material surface incorporates resistance mechanisms into the arrangement of microstructures, resulting in holographic anti-counterfeiting pattern data, including: An enhanced coding matrix is ​​obtained and multi-scale singular value decomposition is performed. Singular value sequences are extracted to construct anti-counterfeiting and anti-cracking resistance feature vectors. The anti-counterfeiting and anti-cracking resistance feature vectors are mapped to a multi-dimensional spectral feature space and superimposed with nonlinear optical distortion noise to generate a test spectral dataset. The correlation decay curve between the test spectral dataset and the standard lossless spectral data is calculated. If the robustness quantization score determined by the correlation decay curve is lower than the preset defense threshold, the construction parameters of the anti-counterfeiting and anti-cracking resistance feature vectors are corrected. Based on the distribution density of the corrected anti-counterfeiting and anti-cracking resistance feature vectors in the multi-dimensional spectral feature space, the integrity index of the resistance mechanism is calculated. Integrity indicators are obtained and mapped to phase modulation control variables to obtain diffraction grating spatial frequency distribution data. A microstructure arrangement topology model is constructed, and the resistance mechanism is transformed into etching depth perturbation amount and superimposed on the microstructure arrangement topology model to generate three-dimensional surface relief structure data. The exposure energy density distribution matrix is ​​calculated, and the exposure energy density distribution matrix is ​​encoded to obtain holographic image anti-counterfeiting pattern data.

7. The method of claim 1, wherein the holographic image security pattern is a laser- processed image of a hologram. The polarization response of the test pattern under different lighting conditions is evaluated. If the response deviation exceeds a preset threshold, the writing process parameters are adjusted to obtain optimized anti-counterfeiting pattern data, including: The holographic anti-counterfeiting pattern data is acquired and mapped onto a virtual optical surface model. A dynamic spectral illumination environment model containing incident light from multiple angles is constructed. The polarization response feature matrix is ​​extracted from the virtual optical surface model. The polarization response feature matrix is ​​compared with a standard reference spectral database to determine the response deviation value. If the response deviation value exceeds a preset threshold, the compensation correction amount for the writing process parameters is calculated. The compensation correction amount is then used to reconstruct the holographic image anti-counterfeiting pattern data to obtain optimized anti-counterfeiting pattern data.

8. The method of claim 1, wherein the holographic image security pattern is a laser- processed image. Based on the optimized anti-counterfeiting pattern data, a multi-dimensional anti-counterfeiting verification protocol is generated. Polarization angle sequences and resistance features are incorporated into the protocol, and the extended configuration of the verification process is determined, including: Optimized anti-counterfeiting pattern data is obtained and the phase distribution of the holographic grating is analyzed. High-frequency diffraction feature points are extracted as basic encryption nodes to construct a polarization response mapping model to calculate the polarization angle sequence. Weak segments of the polarization angle sequence in the virtual optical noise environment are identified and optical resistance feature masks are generated. The optical resistance feature masks and polarization angle sequences are superimposed to construct a multi-dimensional anti-counterfeiting verification protocol. Polarization modulation parameters are extracted according to the protocol to determine the extended configuration of the verification process.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 8.

10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 8.

Images