Multi-parameter variable-sequence three-dimensional encrypted binary anti-counterfeit printing method

Abstract

The invention relates to the technical field of anti-counterfeit printing, in particular to a multi-parameter variable-sequence three-dimensional encrypted binary anti-counterfeit printing method, which comprises the following steps of: analyzing pixel information of an original printing image, and decomposing the pixel information into three independent dimensional components of space, brightness and texture; each component is mapped with an encryption parameter library formed by associated dynamically adjustable encryption elements, and differential processing of position disorder, brightness migration and texture weaving is performed on each component; and synchronizing each component and calculating a global synchronization factor, updating the encryption parameter library by using the factor in a collaborative variable sequence manner, and carrying out secondary encryption on each component based on the updated parameter library to generate a final anti-counterfeiting coding sequence. According to the method, encryption pertinence, diversity and concealment are enhanced, reverse cracking is avoided, uniqueness and non-replicability of anti-counterfeiting codes are improved, anti-counterfeiting capability is enhanced, and the method is adaptive to anti-counterfeiting printing of high-end products.

Description

Technical Field

[0001] This invention relates to the field of anti-counterfeiting printing technology, and in particular to a multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method. Background Technology

[0002] Currently used binary anti-counterfeiting printing methods mostly process the pixel information of the original printed image as a whole, achieving encryption through single-dimensional pixel position disorder, brightness adjustment, or simple texture overlay. The encryption parameters are mostly fixed values, resulting in an encryption parameter library lacking dynamic adjustment capabilities. The encrypted elements are independent of each other, with no collaborative relationship. This type of technology is widely used for anti-counterfeiting in various commodity packaging, certificates, and tickets, relying on simple parameter mapping and a single encryption process to achieve basic anti-counterfeiting effects.

[0003] Existing technical solutions have shortcomings. The overall encryption method cannot take into account the feature differences of different dimensions of pixel information, resulting in insufficient encryption targeting. The fixed encryption parameter library is unable to cope with complex counterfeiting methods and is easily cracked by reverse engineering. At the same time, the encryption process is mostly a one-time, single-dimensional process without dynamic parameter updates and multi-stage collaborative mechanisms. This results in limited concealment and anti-counterfeiting capabilities of the anti-counterfeiting code, which cannot meet the high anti-counterfeiting security requirements of high-end products. Therefore, an encryption method is needed that can achieve precise encryption in different dimensions, dynamically update encryption parameters, and coordinate all stages. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method, comprising: The pixel information of the original printed image is analyzed and decomposed into independent spatial dimension components, luminance dimension components, and texture dimension components. The spatial dimension component, the brightness dimension component, and the texture dimension component are respectively mapped using an encrypted parameter library, which consists of a set of interconnected and dynamically adjustable encrypted elements. For the spatial dimension component, a spatial perturbation seed is generated based on the elements in its encryption parameter library, and the spatial perturbation seed is used to induce the pixel positions to generate a preliminary position disorder. For the brightness dimension component, a brightness modulation waveform is generated based on the elements in its encrypted parameter library, and the brightness modulation waveform is used to drive the brightness value to cyclically migrate. For the texture dimension component, a texture fusion template is generated based on the elements in its encrypted parameter library, and the texture fusion template is used to guide the texture features to be cross-woven. The components that have completed the initial position disordering, cyclic migration and cross weaving processes are synchronized, and the global synchronization factor is calculated based on the synchronization state of the components. Based on the global synchronization factor, a cooperative reordering operation is performed on all encrypted elements in the encrypted parameter library to update the encrypted parameter library of each component. Based on the updated encryption parameter library, a new round of perturbation, modulation, and weaving is performed on the spatial dimension component, the brightness dimension component, and the texture dimension component to generate the final anti-counterfeiting encoding sequence.

[0006] As a further aspect of the present invention, the step of parsing the pixel information of the original printed image and decomposing the pixel information into independent spatial dimension components, luminance dimension components, and texture dimension components includes: Obtain the pixel matrix of the original printed image, wherein each pixel in the pixel matrix contains coordinates, brightness value and neighborhood texture descriptor; Extract the coordinate set of all pixels from the pixel matrix to form a spatial dimension component, which is used to characterize the geometric distribution information of the pixels; The set of brightness values ​​of all pixels is extracted from the pixel matrix to form a brightness dimension component, which is used to characterize the optical intensity information of the image. Based on the neighborhood texture descriptor of each pixel, the texture vector of each pixel is generated by calculating the consistency of gradient directions within the neighborhood. The set of texture vectors of all pixels constitutes the texture dimension component, which is used to characterize the structural information of the image surface.

[0007] As a further aspect of the present invention, the spatial dimension component, the brightness dimension component, and the texture dimension component are respectively mapped using an encrypted parameter library, including: A spatial parameter library is established for the spatial dimension components. The spatial parameter library contains a series of variable displacement vectors, rotation angles, and scaling factors. There is a non-linear coupling relationship between the displacement vectors, rotation angles, and scaling factors. A luminance parameter library is established for the luminance dimension components. The luminance parameter library contains a series of variable waveform functions, phase offsets, and amplitude coefficients. There are modulation constraint relationships among the waveform functions, phase offsets, and amplitude coefficients. A texture parameter library is established for the texture dimension components. The texture parameter library contains a series of variable fusion kernels, weaving rules and replacement probabilities. There are weaving dependencies among the fusion kernels, weaving rules and replacement probabilities. After initializing the encrypted parameter library, a cross-dimensional association mapping table is established between the spatial parameter library, the brightness parameter library, and the texture parameter library. The cross-dimensional association mapping table is used to record the chain effect weights of changes in elements in one parameter library on elements in the other two parameter libraries.

[0008] As a further aspect of the present invention, for the spatial dimension component, a spatial perturbation seed is generated based on elements in its encryption parameter library, and the spatial perturbation seed is used to induce a preliminary position disorder of pixel positions, including: From the spatial parameter library corresponding to the spatial dimension components, a set of displacement vectors and rotation angles are dynamically selected as basic perturbation elements based on the coordinate values ​​of the current pixel. The basic perturbation element is input into a chaotic iteration function, which uses the state value of the previous pixel after processing as the initial input to iteratively generate the spatial perturbation seed specific to the current pixel. The coordinates of the current pixel are nonlinearly transformed using the spatial perturbation seed. The transformation rule includes finding the corresponding target coordinates in a preset coordinate permutation table using the spatial perturbation seed as an index. Traverse all pixels and perform coordinate transformations sequentially according to the spatial perturbation seed. The set of all transformed coordinates constitutes the initial position disorder.

[0009] As a further aspect of the present invention, for the luminance dimension component, a luminance modulation waveform is generated based on elements in its encrypted parameter library, and the luminance value is driven to cyclically migrate using the luminance modulation waveform, including: Extract a waveform function from the brightness parameter library corresponding to the brightness dimension component, and dynamically calculate the phase offset and amplitude coefficient of the waveform function based on the statistical brightness characteristics of the pixels in the current processing batch. By combining the waveform function, the phase offset, and the amplitude coefficient, a complete brightness modulation waveform is synthesized. The original luminance value sequence to be processed is convolved with the luminance modulation waveform, and the convolution result generates a set of luminance shift values. The brightness migration amounts are sequentially superimposed onto the corresponding original brightness values. If the superposition result exceeds the brightness representation range, the value is folded back from the range boundary to complete the cyclic migration of brightness values ​​and generate a sequence of migrated brightness values.

[0010] As a further aspect of the present invention, for the texture dimension component, a texture fusion template is generated based on elements in its encrypted parameter library, and the texture fusion template is used to guide the cross-weaving of texture features, including: From the texture parameter library corresponding to the texture dimension components, the fusion kernel and weaving rules are adaptively selected according to the roughness level of the image partition; The selected fusion kernel is applied to the neighborhood texture vector of the current pixel to generate an intermediate texture feature template; According to the weaving rules, key feature lines are extracted from the intermediate texture feature template, and the key feature lines are cross-connected and woven with the texture feature lines of adjacent partitions to form a new texture network; Based on the replacement probabilities in the texture parameter library, some nodes in the new texture network are replaced with preset texture fragments from the parameter library to complete the cross weaving and generate a set of woven texture vectors.

[0011] As a further aspect of the present invention, the step of synchronizing the components that have completed the initial position disordering, cyclic migration, and cross-weaving processes, and calculating the global synchronization factor based on the synchronization state of the components, includes: Record the processing completion times of the initial position disorder, the migrated brightness value sequence, and the woven texture vector set, respectively, and calculate the maximum time difference between the three times; The initial position disorder, the migrated brightness value sequence, and the woven texture vector set are sampled, and the state association entropy of the initial position disorder, the migrated brightness value sequence, and the woven texture vector set at the same sampling point is calculated. The state association entropy is used to quantify the degree of coordination of the changes in the three component data. The maximum time difference and the state association entropy are input into a synchronization factor calculation model. The synchronization factor calculation model outputs a scalar value as the global synchronization factor. The larger the global synchronization factor, the more the processing of the three components needs to be adjusted to achieve synchronization.

[0012] As a further aspect of the present invention, a cooperative reordering operation is performed on all encrypted elements in the encrypted parameter library according to the global synchronization factor, thereby updating the encrypted parameter library of each component, including: The global synchronization factor is compared with a preset synchronization threshold. If the global synchronization factor is greater than the synchronization threshold, a cooperative reordering operation is triggered. In the cooperative reordering operation, the variation range benchmark of elements in the spatial parameter library, brightness parameter library, and texture parameter library is first determined according to the magnitude of the global synchronization factor. Guided by the aforementioned change range benchmark, when adjusting an element in the spatial parameter library by traversing the cross-dimensional correlation mapping table, the values ​​of the affected elements in the brightness parameter library and texture parameter library are simultaneously calculated and pre-adjusted based on the chain influence weight recorded in the cross-dimensional correlation mapping table. After completing one round of traversal adjustment based on association mapping, an updated spatial parameter library, an updated brightness parameter library, and an updated texture parameter library are generated.

[0013] As a further aspect of the present invention, the step of performing a new round of perturbation, modulation, and weaving on the spatial dimension component, the luminance dimension component, and the texture dimension component based on the updated encryption parameter library, and generating the final anti-counterfeiting encoding sequence, includes: The initial position disorder is used as a new spatial dimension component input. Using the updated spatial parameter library, the generation of spatial perturbation seeds and the position disorder induction process are repeated to generate position disorder. The migrated luminance value sequence is used as the input of the new luminance dimension component. Using the updated luminance parameter library, the generation and cyclic migration process of the luminance modulation waveform is repeated to generate a luminance sequence. The woven texture vector set is used as the input of new texture dimension components. The updated texture parameter library is used to repeat the generation and cross-weaving process of the texture fusion template to generate a texture network. The position disorder, the brightness sequence, and the texture network are recombined according to the original pixel correspondence to generate a final pixel matrix containing three-dimensional encryption information. The final pixel matrix is ​​converted into binary code to obtain the final anti-counterfeiting code sequence.

[0014] As a further aspect of the present invention, it also includes: extracting feature fingerprints from the final anti-counterfeiting code sequence, compressing the feature fingerprints into an identification code, and embedding the identification code into the printed product, specifically including: The final anti-counterfeiting encoding sequence is subjected to block hash calculation to obtain a set of hash values, which are then used as the original feature fingerprint. Extract the key parameter trajectory generated during the encryption process. The key parameter trajectory includes the evolution path of the spatial perturbation seed, the morphological sequence of the brightness modulation waveform, and the change record of the texture fusion template. The original feature fingerprint is mixed with the key parameter trajectory and then fed into a compression encoder; The compression encoder uses a combination of run-length encoding and differential encoding to compress the mixed data and output a compact bit stream, which is the identifier encoding. During the printing plate-making stage, the identification code is embedded into a designated area or the full-size image of the printed product in the form of a micro-dot matrix or spectral watermark.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: The pixel information of the original printed image is decomposed into independent spatial dimension components, brightness dimension components, and texture dimension components. Each of these three components is mapped one-to-one with a set of interconnected and dynamically adjustable encryption elements, forming an encryption parameter library. Compared to the conventional method of uniformly mapping encryption parameters to the entire printed image, the dimensional parameter mapping can accurately match the feature differences of each dimension, making the encryption process of each dimension more targeted. The correlation between each encryption element can prevent the leakage of encryption parameters in a single dimension from causing the overall encryption system to fail. At the same time, the dynamically adjustable encryption elements can flexibly adapt to the characteristics of different printed images, enhancing the diversity and concealment of encryption and reducing the probability of being cracked and counterfeited.

[0016] First, the three components undergo differential processing: spatial perturbation seed-induced initial pixel position disorder, brightness modulation waveform-driven cyclic migration of brightness values, and texture fusion template-guided cross-weaving of texture features. Then, based on the synchronization state of each component after processing, a global synchronization factor is calculated. This factor is used to perform a cooperative reordering operation on all encrypted elements in the encryption parameter library to update the library. Finally, based on the updated encryption parameter library, perturbation, modulation, and weaving operations are performed on the three components again, forming a closed-loop encryption process. Compared to conventional single-time, single-dimensional encryption, the dynamically updated encryption parameter library causes the encryption logic to constantly change, making it impossible for attackers to reverse-engineer using fixed encryption rules. The cooperative reordering of each component ensures the consistency and integrity of the encryption process, making the generated anti-counterfeiting code unique and uncopyable, further strengthening the security and anti-counterfeiting capabilities of the encryption. Attached Figure Description

[0017] Figure 1 This is a flowchart of a multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method according to the present invention; Figure 2 A flowchart for decomposing pixel information into components; Figure 3 A flowchart for generating preliminary position disorder for spatial dimension component perturbations; Figure 4 This is a trend chart of global synchronization factor changes; Figure 5 A trend chart showing the variation of eigenvalues ​​of each dimension component in 3D encryption. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1 The process involves parsing the pixel information of the original printed image and decomposing it into independent spatial, luminance, and texture components. Each of these components is then mapped using an encrypted parameter library, which consists of a set of dynamically adjustable, interconnected encrypted elements. For the spatial component, a spatial perturbation seed is generated based on the elements in the encrypted parameter library, and this seed is used to induce a preliminary disorder of pixel positions. For the luminance component, a luminance modulation waveform is generated based on the elements in the encrypted parameter library, and this waveform is used to drive a cyclic luminance value. Migration; for the texture dimension component, a texture fusion template is generated based on the elements in its encryption parameter library, and the texture fusion template is used to guide the texture features to perform cross weaving; the components that have completed the initial position disordering, cyclic migration and cross weaving processing are synchronized, and a global synchronization factor is calculated based on the synchronization state of the components; according to the global synchronization factor, a cooperative reordering operation is performed on all encryption elements in the encryption parameter library, thereby updating the encryption parameter library of each component; based on the updated encryption parameter library, a new round of perturbation, modulation and weaving is performed on the spatial dimension component, the brightness dimension component and the texture dimension component, and the final anti-counterfeiting encoding sequence is generated.

[0021] In one embodiment of the present invention, see [reference] Figure 2The process involves obtaining the pixel matrix of the original printed image. Each pixel in the pixel matrix contains coordinates, a brightness value, and a neighborhood texture descriptor. The neighborhood texture descriptor can be a vector containing gradient magnitudes in eight directions. The set of coordinates of all pixels is extracted from the pixel matrix to form a spatial dimension component, which characterizes the geometric distribution information of the pixels. The set of brightness values ​​of all pixels is also extracted from the pixel matrix to form a brightness dimension component, which characterizes the optical intensity information of the image. Based on the neighborhood texture descriptor of each pixel, a texture vector for each pixel is generated by calculating the consistency of gradient directions within the neighborhood. The set of texture vectors of all pixels constitutes the texture dimension component. The texture vector generation process involves quantizing the main peak of the neighborhood gradient direction histogram. In a specific calculation, the consistency index... It can be done through the formula:

[0022] in: Indicates the first Texture orientation consistency index per pixel Each of the eight preset gradient directions represents the sum of the gradient magnitudes of the pixel, and the texture dimension component is used to characterize the structural information of the image surface.

[0023] In some embodiments, a spatial parameter library is established for the spatial dimension components. This library contains a series of variable displacement vectors, rotation angles, and scaling factors. The component values ​​of the displacement vectors and rotation angles can be predefined within a numerical range. The scaling factors are defined as a series of positive real numbers. A non-linear coupling relationship exists between the displacement vectors, rotation angles, and scaling factors. This relationship manifests as follows: when a displacement vector is selected from the library, it is associated with a set of candidate rotation angles and scaling factors according to a built-in mapping function. A brightness parameter library is established for the brightness dimension components. This library contains a series of variable waveform functions, phase offsets, and amplitude coefficients. The waveform functions include sine functions and sawtooth wave functions. The phase offsets and amplitude coefficients are defined as adjustable parameters. A modulation constraint relationship exists between the waveform functions, phase offsets, and amplitude coefficients. This constraint relationship is manifested as follows: for a selected waveform function, the range of available combinations of phase offsets and amplitude coefficients is limited by a constraint table. A texture parameter library is established for the texture dimension components. The texture parameter library contains a series of variable fusion kernels, weaving rules, and replacement probabilities. The fusion kernel can be a Gaussian kernel or a Laplacian kernel of different sizes. The weaving rules define the specific pattern of texture feature line cross-connection. The replacement probability is a value between 0 and 1. There is a weaving dependency relationship between the fusion kernel, weaving rules, and replacement probability. This dependency relationship is manifested in that the selected fusion kernel type will affect the set of available weaving rules, and the selection of weaving rules will further determine the recommended value range of the replacement probability.

[0024] It is understandable that after initializing the encrypted parameter library, a cross-dimensional association mapping table is established between the spatial parameter library, the brightness parameter library, and the texture parameter library. This cross-dimensional association mapping table records the cascading impact weights of changes in elements within one parameter library on elements in the other two parameter libraries. For example, in the cross-dimensional association mapping table, when the X component of a displacement vector in the spatial parameter library changes, this event will be associated with the need to adjust the weight values ​​of the amplitude coefficients of certain waveform functions in the brightness parameter library, and simultaneously with the need to adjust the weight values ​​of the size parameters of a certain type of fusion kernel in the texture parameter library. The cross-dimensional association mapping table is stored in matrix form. The row index of the matrix corresponds to the specific element identifier in one parameter library, and the column index corresponds to the element identifier in another parameter library. The values ​​within the matrix cells are the cascading impact weights: a positive weight indicates a positive association adjustment, a negative weight indicates a negative association adjustment, and a weight of zero indicates no direct impact.

[0025] In one embodiment of the present invention, see [reference] Figure 3From the spatial parameter library corresponding to the spatial dimension components, a set of displacement vectors and rotation angles are dynamically selected as basic perturbation elements based on the coordinates of the current pixel. The selection rule can be based on the parity of the sum of the horizontal and vertical coordinates, indexing different preset combinations in the spatial parameter library. These basic perturbation elements are input into a chaotic iteration function, which uses the state value of the previous pixel after processing as its initial input to iteratively generate a spatial perturbation seed specific to the current pixel. The chaotic iteration function can employ a logistic mapping, and its expression is:

[0026] in: Indicates the generation of the first The first pixel spatial perturbation seed required Intermediate value of the next iteration The initial value for this iteration is the final result of generating the spatial perturbation seed for the previous pixel. This is a control parameter calculated from the basic perturbation elements (the magnitude of the displacement vector and the rotation angle). The coordinates of the current pixel are nonlinearly transformed using the spatial perturbation seed. The transformation rule includes searching for the corresponding target coordinates in a predefined coordinate permutation table, which is a predefined lookup table that maps seed values ​​to new coordinates in the image plane. All pixels are traversed, and the coordinate transformation is performed sequentially according to the spatial perturbation seed. The set of all transformed coordinates constitutes the initial position disorder.

[0027] In some embodiments, a waveform function is extracted from a luminance parameter library corresponding to the luminance dimension components, and the phase offset and amplitude coefficient of the waveform function are dynamically calculated based on the statistical luminance characteristics of the pixels in the current processing batch. The statistical luminance characteristics can be the average and standard deviation of the luminance values ​​of the pixels in the current batch. The waveform function, the phase offset, and the amplitude coefficient are combined to synthesize a complete luminance modulation waveform. The synthesis method is to use the phase offset as the horizontal translation amount of the waveform function and the amplitude coefficient as the vertical scaling factor of the waveform function. The original luminance value sequence to be processed is convolved with the luminance modulation waveform. The convolution result generates a set of luminance transfer amounts, and the length of the convolution kernel is consistent with the period length of the luminance modulation waveform. The luminance transfer amounts are sequentially superimposed on the corresponding original luminance values. If the superposition result exceeds the luminance representation range, it is folded back from the range boundary. Taking a 256-level grayscale image as an example, when the superposition result is greater than 255, the result value is subtracted by 256; when the superposition result is less than 0, the result value is added by 256, completing the cyclic migration of luminance values ​​and generating a migrated luminance value sequence.

[0028] Optional, the initial state value of the chaotic iterative function This can be set to a fixed value associated with image features, such as the normalized value of the mean brightness values ​​of all pixels in the image. It can be understood that the construction of the coordinate permutation table can be based on a pseudo-random number generation algorithm. This algorithm takes the image size and a set of initial keys as input, generates a series of non-repeating coordinate pairs, and establishes a mapping relationship from sequential indices to these coordinate pairs. The spatial perturbation seed, after modulo operation, is used as the sequential index. In some embodiments, the selection of the waveform function can be determined based on the shape of the pixel brightness histogram of the current batch of processing. For example, when the histogram exhibits a bimodal distribution, a sawtooth wave function is extracted from the brightness parameter library; when the histogram exhibits a unimodal distribution, a sine wave function is extracted. Optionally, the specific process of dynamically calculating the phase offset and amplitude coefficient can be as follows: normalize the average value of the current batch of pixel brightness values ​​to the [0, 2π] interval as the phase offset, and normalize the standard deviation of the current batch of pixel brightness values ​​to the [0.5, 2.0] interval as the amplitude coefficient. It is understandable that the backtracking operation during the cyclic migration process ensures that all brightness values ​​eventually fall within the valid representation range, preventing information truncation and loss, and maintaining the continuity of the values ​​in the migrated brightness value sequence.

[0029] In one embodiment of the present invention, a fusion kernel and weaving rules are adaptively selected from a texture parameter library corresponding to the texture dimension components, based on the roughness level of the image partition. The roughness level is obtained by calculating and dividing the variance of the pixel texture vectors within the image partition. The selected fusion kernel is applied to the neighborhood texture vector of the current pixel to generate an intermediate texture feature template. The application method involves convolving the fusion kernel with the neighborhood texture vector matrix centered on the current pixel. According to the weaving rules, key feature lines are extracted from the intermediate texture feature template. The key feature lines are the directions of continuous pixels in the template where the gradient magnitude exceeds a predetermined threshold. The key feature lines are then cross-connected and woven with the texture feature lines of adjacent partitions to form a new texture network. The cross-connection points are determined by the cross-connection pattern defined in the weaving rules. Based on the replacement probability in the texture parameter library, some nodes in the new texture network are replaced with preset texture fragments from the parameter library. The replacement operation is determined by generating a random number and comparing it with the replacement probability, thus completing the cross-weaving and generating a set of woven texture vectors.

[0030] In some embodiments, the processing completion times of the initial position disorder, the migrated brightness value sequence, and the woven texture vector set are recorded respectively, and the maximum time difference between the three times is calculated. The maximum time difference is the maximum of the absolute values ​​of the differences between any two times. The initial position disorder, the migrated brightness value sequence, and the woven texture vector set are sampled, and the state association entropy of the initial position disorder, the migrated brightness value sequence, and the woven texture vector set at the same sampling point is calculated. Sampling involves extracting data points from the same spatial location index of the three data sets to form triplet samples. The formula for calculating the state association entropy is:

[0031] in: This represents the calculated state association entropy. This represents the total number of possible triplet state combinations. Indicates the first The empirical probability of a triplet state occurring in all sampled data is calculated, and the state association entropy is used to quantify the degree of coordination of the changes in the three component data. The maximum time difference and the state association entropy are input into a synchronization factor calculation model, which outputs a scalar value as the global synchronization factor. The synchronization factor calculation model can be a pre-defined weighted summation function. ,in It is a global synchronization factor. It is the maximum time difference. and It is a preset weighting coefficient. The larger the global synchronization factor, the more the processing of the three components needs to be adjusted to achieve synchronization.

[0032] Optionally, the roughness level of an image partition can be determined by calculating the standard deviation of the Euclidean distance between all pixel texture vectors within the partition, and divided into three levels: "high," "medium," and "low." Different levels correspond to different fusion kernel sets in the texture parameter library. It can be understood that the crossover patterns defined in the weaving rules can include "grid-like crossovers" and "spiral crossovers," with different patterns determining the geometric topology of the key feature lines during connection. In some embodiments, the random number used for the replacement operation can be a pseudo-random number uniformly distributed in the interval [0,1). When this random number is less than the replacement probability set for the current node in the texture parameter library, the replacement of that node is triggered. Optionally, the sampling process can be equally spaced sampling, i.e., collecting a triplet sample every fixed number of pixels in the spatial dimension to ensure the uniformity of the sampling in spatial distribution. It can be understood that state association entropy... The smaller the value, the more concentrated the state combinations of the three components at the same sampling point, i.e., the higher the degree of coordination; conversely, the larger the value, the higher the degree of coordination. The larger the value, the lower the degree of collaboration.

[0033] In one embodiment of the present invention, the global synchronization factor is compared with a preset synchronization threshold. The preset synchronization threshold can be a fixed value set based on experience, such as 5.0. If the global synchronization factor is greater than the synchronization threshold, a cooperative reordering operation is triggered. In the cooperative reordering operation, the variation range benchmark of elements in the spatial parameter library, the brightness parameter library, and the texture parameter library is first determined based on the magnitude of the global synchronization factor. The variation range benchmark can be expressed by the formula:

[0034] in: Indicates the benchmark for the range of change. Represents the global synchronization factor. This indicates the preset synchronization threshold. A preset proportional coefficient is used to map the difference to a specific adjustment range. Guided by the aforementioned change range benchmark, the cross-dimensional correlation mapping table is traversed. When adjusting an element in the spatial parameter library, the values ​​of affected elements in the brightness and texture parameter libraries are simultaneously calculated and pre-adjusted based on the chain reaction weights recorded in the cross-dimensional correlation mapping table. The adjustment amount is the product of the change range benchmark and the corresponding chain reaction weight. After completing one round of traversal adjustment based on the correlation mapping, updated spatial parameter libraries, updated brightness parameter libraries, and updated texture parameter libraries are generated. Referring to Table 1, a simplified fragment of the cross-dimensional correlation mapping table is shown to illustrate the chain reaction of element adjustments.

[0035] Table 1: Cross-Dimensional Association Mapping Table

[0036] In some embodiments, the synchronization threshold can be dynamically set according to the security level requirements of different images. Higher security levels correspond to lower synchronization thresholds, making cooperative reordering operations easier to trigger. It can be understood that the calculation of the change amplitude benchmark ensures that the adjustment intensity of parameter library elements is proportional to the degree to which the global synchronization factor exceeds the threshold; the larger the global synchronization factor, the more significant the parameter changes. During the traversal adjustment process, the spatial parameter library, brightness parameter library, and texture parameter library can be traversed according to a fixed element identifier order. After adjusting an element in one library, the associated element values ​​in the other two libraries are immediately updated according to the cross-dimensional association mapping table, forming a chain adjustment effect. After completing one round of traversal adjustment based on the association mapping, the element values ​​in all parameter libraries have been updated according to the global synchronization factor and association rules, generating updated spatial parameter libraries, updated brightness parameter libraries, and updated texture parameter libraries.

[0037] The initial position disorder is used as a new spatial dimension component input. The generation of the spatial perturbation seed and the position disorder induction process are repeated using the updated spatial parameter library to generate a position disorder. This process is logically consistent with the generation of the initial position disorder, but the parameters used are from the updated spatial parameter library. The migrated luminance value sequence is used as a new luminance dimension component input. The generation of the luminance modulation waveform and the cyclic migration process are repeated using the updated luminance parameter library to generate a luminance sequence. This process is logically consistent with the generation of the migrated luminance value sequence, but the parameters used are from the updated luminance parameter library. The woven texture vector set is used as a new texture dimension component input. The generation of the texture fusion template and the cross-weaving process are repeated using the updated texture parameter library to generate a texture network. This process is logically consistent with the generation of the woven texture vector set, but the parameters used are from the updated texture parameter library. The position disorder, the luminance sequence, and the texture network are recombine according to the original pixel correspondence, which refers to the correspondence between the coordinates, luminance values, and texture vector indices of each pixel during the parsing stage, to generate a final pixel matrix containing 3D encryption information. The final anti-counterfeiting encoding sequence is obtained by performing binary encoding conversion on the final pixel matrix. The binary encoding conversion is to convert the coordinates, brightness values ​​and texture vector components of each pixel in the final pixel matrix into a fixed-length binary bit stream and then concatenate them in order.

[0038] Optionally, during the recombination process, the position scrambling provides new pixel coordinates, the brightness sequence provides new brightness values ​​at those coordinates, and the texture network provides a new texture vector description at those coordinates. These three are combined according to the same original index. It can be understood that the final anti-counterfeiting encoding sequence is a long binary sequence that integrates the spatial, brightness, and texture information after two rounds of 3D encryption processing. In some embodiments, the second processing step of generating the position scrambling, brightness sequence, and texture network can be viewed as a further obfuscation and enhancement of the first processing result, utilizing new parameters dynamically adjusted according to the synchronization state. Optionally, during binary encoding conversion, for numerical data such as coordinates and brightness values, a standard integer-to-binary string conversion can be used; for multi-component data such as texture vectors, each floating-point component can be quantized into an integer before binary conversion.

[0039] See Figure 4This is a trend chart of the global synchronization factor, showing the changes in the global synchronization factor throughout the seven stages of the encryption process in the multi-parameter variable-order 3D encrypted binary anti-counterfeiting printing method, and comparing it with the preset synchronization threshold (5.0). Initially, the components of each dimension are just decomposed, the data state is stable, and the synchronization requirement is low. Parameter libraries map each dimension to establish independent parameter libraries, and no complex interactions have yet occurred. During the first encryption, each dimension executes encryption operations independently, and an asynchronous trend begins to emerge. Synchronous calculations calculate the time difference and state correlation entropy of each dimension's processing completion, reaching their peak, triggering cooperative variable ordering. The second encryption uses the updated parameter library, and the processing of each dimension tends to synchronize again. Finally, the encoding encryption is completed, generating anti-counterfeiting codes, and the system state returns to stability. After cooperative variable ordering, the global synchronization factor significantly decreases to 4.1 in the "second encryption" stage, proving that parameter adjustment effectively improves the synchronization of each dimension's components.

[0040] In one embodiment of the present invention, the final anti-counterfeiting encoding sequence is subjected to block hash calculation to obtain a set of hash values. The block size can be set to 512 consecutive bits as a data block. The hash algorithm used is SHA-256, and the hash values ​​are used as the original feature fingerprint. Key parameter trajectories generated during the encryption process are extracted. These key parameter trajectories include the evolution path of the spatial perturbation seed, the morphological sequence of the brightness modulation waveform, and the change record of the texture fusion template. The evolution path of the spatial perturbation seed is the sequence of intermediate state values ​​generated by the chaotic iterative function during the generation of the spatial perturbation seed for each pixel. The morphological sequence of the brightness modulation waveform is a record of the function type and parameter combination of the brightness modulation waveform in each round of processing. The change record of the texture fusion template is the identifier sequence for selecting the fusion kernel and weaving rule from the texture parameter library each time. The original feature fingerprint and the key parameter trajectories are mixed and fed into a compression encoder. The mixing method involves interleaving the data blocks of the key parameter trajectories with the data blocks of the original feature fingerprint.

[0041] The compression encoder uses a combination of run-length encoding and differential encoding to compress the mixed data, outputting a compact bitstream, which is the identifier code. In a specific compression process, the mixed data is first differentially encoded, and the difference value... From the formula:

[0042] in: Indicates the first The difference between the data points. Represents the first in the mixed data sequence The original values ​​of each data point This represents the original value of the preceding data point, for the first data point of the sequence. Its difference value is defined as Next, run-length encoding is performed on the differentially encoded sequence. Run-length encoding encodes and stores the consecutively repeated difference values ​​and their number of consecutive occurrences as a (value, length) pair.

[0043] In some embodiments, during the printing plate-making stage, the identification code is embedded into a designated area or the entire image of the printed product in the form of a micro-dot matrix or a spectral watermark. After the identification code is embedded, the identification code on the printed product is collected. After initialization, the status is confirmed. The image information of the identification code is acquired and the collected image is displayed. Subsequently, the identification code is parsed to extract the feature fingerprint and the key parameter trajectory. The authenticity of the printed product is verified by comparing the feature fingerprint and the key parameter trajectory. The parsing result is displayed to complete the verification of the anti-counterfeiting code sequence. When embedded in the form of a micro-dot matrix, the binary code stream of the identification code is converted into a series of tiny dots arranged in a specific pattern in space. The diameter of these dots is smaller than the minimum resolution of the human eye, and they are printed with specific ink in the non-main image area of ​​the printed product. When embedded in the form of a spectral watermark, the binary code stream of the identification code is modulated into the mid-to-high frequency coefficients of the printed image after frequency domain transformation. The value of the selected coefficient is adjusted to carry the encoding information, and then an inverse transformation is performed to generate a watermarked printed image.

[0044] Optionally, when performing block hash calculation on the final anti-counterfeiting encoding sequence, if the length of the last data block is less than 512 bits, it is padded to 512 bits using a predefined padding rule before calculation. It can be understood that the extraction of key parameter trajectories records the dynamic parameter changes during the encryption process; these trajectories, together with the final anti-counterfeiting encoding sequence, constitute complete anti-counterfeiting information. The combination of run-length encoding and differential encoding in the compression encoder is particularly suitable for processing sequences with continuous repeating values ​​and small numerical variations, effectively reducing the data volume of the identifier encoding. In some embodiments, the arrangement pattern of the micro-dot matrix can be a simple rectangular grid or a pseudo-random pattern dynamically generated based on the identifier encoding information. Reading the dot matrix requires dedicated image acquisition and recognition equipment. Optionally, the embedding strength of the spectral watermark needs to be pre-tested and adjusted to ensure that the watermark information can resist noise interference introduced by printing, scanning, and other processes while remaining invisible to the human eye. It can be understood that embedding the identifier encoding into the printed product completes the transfer of anti-counterfeiting information from digital encrypted information to the physical printed product, providing data evidence on a physical carrier for verifying the authenticity of the printed product.

[0045] See Figure 5This is a trend chart showing the changes in the eigenvalues ​​of the spatial, luminance, and texture components in a multi-parameter variable-order 3D encrypted binary anti-counterfeiting printing method. It illustrates the changes in these eigenvalues ​​throughout the encryption process, from the original data to the final encoding. In the three core encryption stages—"positional disordering," "luminance migration," and "texture weaving"—the changes in the eigenvalues ​​of each component are independent. In the "synchronous update" stage, the eigenvalues ​​of the spatial and luminance components increase synchronously, intuitively demonstrating the regulatory effect of the parameter library's collaborative variable ordering on each component under the drive of the global synchronization factor. The eigenvalue of the luminance component is consistently much higher than the other two dimensions, reflecting that optical intensity information is the core carrier of anti-counterfeiting codes in anti-counterfeiting printing. This chart clearly quantifies the dynamic changes of each component during the 3D encryption process, not only demonstrating the impact of each stage of processing on different dimensions but also highlighting the regulatory role of synchronous updates throughout the entire process.

[0046] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method, characterized in that, The method includes: The pixel information of the original printed image is analyzed and decomposed into independent spatial dimension components, luminance dimension components, and texture dimension components. The spatial dimension component, the brightness dimension component, and the texture dimension component are respectively mapped using an encrypted parameter library, which consists of a set of interconnected and dynamically adjustable encrypted elements. For the spatial dimension component, a spatial perturbation seed is generated based on the elements in its encryption parameter library, and the spatial perturbation seed is used to induce the pixel positions to generate an initial position disorder. For the brightness dimension component, a brightness modulation waveform is generated based on the elements in its encrypted parameter library, and the brightness modulation waveform is used to drive the brightness value to cyclically migrate. For the texture dimension component, a texture fusion template is generated based on the elements in its encrypted parameter library, and the texture fusion template is used to guide the texture features to be cross-woven. The components that have completed the initial position disordering, cyclic migration and cross weaving processes are synchronized, and the global synchronization factor is calculated based on the synchronization state of the components. Based on the global synchronization factor, a cooperative reordering operation is performed on all encrypted elements in the encrypted parameter library to update the encrypted parameter library of each component. Based on the updated encryption parameter library, a new round of perturbation, modulation, and weaving is performed on the spatial dimension component, the brightness dimension component, and the texture dimension component to generate the final anti-counterfeiting encoding sequence.

2. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 1, characterized in that, The step of parsing the pixel information of the original printed image, decomposing the pixel information into independent spatial dimension components, luminance dimension components, and texture dimension components, includes: Obtain the pixel matrix of the original printed image, wherein each pixel in the pixel matrix contains coordinates, brightness value and neighborhood texture descriptor; Extract the coordinate set of all pixels from the pixel matrix to form a spatial dimension component, which is used to characterize the geometric distribution information of the pixels; The set of brightness values ​​of all pixels is extracted from the pixel matrix to form a brightness dimension component, which is used to characterize the optical intensity information of the image. Based on the neighborhood texture descriptor of each pixel, the texture vector of each pixel is generated by calculating the consistency of gradient directions within the neighborhood. The set of texture vectors of all pixels constitutes the texture dimension component, which is used to characterize the structural information of the image surface.

3. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 2, characterized in that, The encrypted parameter library mapping is performed on the spatial dimension component, the brightness dimension component, and the texture dimension component, respectively, including: A spatial parameter library is established for the spatial dimension components. The spatial parameter library contains a series of variable displacement vectors, rotation angles, and scaling factors. There is a non-linear coupling relationship between the displacement vectors, rotation angles, and scaling factors. A luminance parameter library is established for the luminance dimension components. The luminance parameter library contains a series of variable waveform functions, phase offsets, and amplitude coefficients. There are modulation constraint relationships among the waveform functions, phase offsets, and amplitude coefficients. A texture parameter library is established for the texture dimension components. The texture parameter library contains a series of variable fusion kernels, weaving rules and replacement probabilities. There are weaving dependencies among the fusion kernels, weaving rules and replacement probabilities. After initializing the encrypted parameter library, a cross-dimensional association mapping table is established between the spatial parameter library, the brightness parameter library, and the texture parameter library. The cross-dimensional association mapping table is used to record the chain effect weights of changes in elements in one parameter library on elements in the other two parameter libraries.

4. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 3, characterized in that, For the aforementioned spatial dimension components, a spatial perturbation seed is generated based on elements in its encryption parameter library, and the spatial perturbation seed is used to induce a preliminary position disorder of pixel positions, including: From the spatial parameter library corresponding to the spatial dimension components, a set of displacement vectors and rotation angles are dynamically selected as basic perturbation elements based on the coordinate values ​​of the current pixel. The basic perturbation element is input into a chaotic iteration function, which uses the state value of the previous pixel after processing as the initial input to iteratively generate the spatial perturbation seed specific to the current pixel. The coordinates of the current pixel are nonlinearly transformed using the spatial perturbation seed. The transformation rule includes finding the corresponding target coordinates in a preset coordinate permutation table using the spatial perturbation seed as an index. Traverse all pixels and perform coordinate transformations sequentially according to the spatial perturbation seed. The set of all transformed coordinates constitutes the initial position disorder.

5. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 4, characterized in that, For the aforementioned luminance dimension component, a luminance modulation waveform is generated based on elements in its encrypted parameter library, and the luminance value is driven to cyclically migrate using the luminance modulation waveform, including: Extract a waveform function from the brightness parameter library corresponding to the brightness dimension component, and dynamically calculate the phase offset and amplitude coefficient of the waveform function based on the statistical brightness characteristics of the pixels in the current processing batch. By combining the waveform function, the phase offset, and the amplitude coefficient, a complete brightness modulation waveform is synthesized. The original luminance value sequence to be processed is convolved with the luminance modulation waveform, and the convolution result generates a set of luminance shift values. The brightness migration amounts are sequentially superimposed onto the corresponding original brightness values. If the superposition result exceeds the brightness representation range, the value is folded back from the range boundary to complete the cyclic migration of brightness values ​​and generate a sequence of migrated brightness values.

6. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 5, characterized in that, For the texture dimension components, a texture fusion template is generated based on elements in its encryption parameter library, and the texture fusion template is used to guide the cross-weaving of texture features, including: From the texture parameter library corresponding to the texture dimension components, the fusion kernel and weaving rules are adaptively selected according to the roughness level of the image partition; The selected fusion kernel is applied to the neighborhood texture vector of the current pixel to generate an intermediate texture feature template; According to the weaving rules, key feature lines are extracted from the intermediate texture feature template, and the key feature lines are cross-connected and woven with the texture feature lines of adjacent partitions to form a new texture network; Based on the replacement probabilities in the texture parameter library, some nodes in the new texture network are replaced with preset texture fragments from the parameter library to complete the cross weaving and generate a set of woven texture vectors.

7. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 6, characterized in that, The process of synchronizing the components that have completed initial position scrambling, cyclic migration, and cross-weaving, and calculating the global synchronization factor based on the synchronization state of the components, includes: Record the processing completion times of the initial position disorder, the migrated brightness value sequence, and the woven texture vector set, respectively, and calculate the maximum time difference between the three times; The initial position disorder, the migrated brightness value sequence, and the woven texture vector set are sampled, and the state association entropy of the initial position disorder, the migrated brightness value sequence, and the woven texture vector set at the same sampling point is calculated. The state association entropy is used to quantify the degree of coordination of the changes in the three component data. The maximum time difference and the state association entropy are input into a synchronization factor calculation model. The synchronization factor calculation model outputs a scalar value as the global synchronization factor. The larger the global synchronization factor, the more the processing of the three components needs to be adjusted to achieve synchronization.

8. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 7, characterized in that, Based on the global synchronization factor, a cooperative reordering operation is performed on all encrypted elements in the encrypted parameter library to update the encrypted parameter library of each component, including: The global synchronization factor is compared with a preset synchronization threshold. If the global synchronization factor is greater than the synchronization threshold, a cooperative reordering operation is triggered. In the cooperative reordering operation, the variation range benchmark of elements in the spatial parameter library, brightness parameter library, and texture parameter library is first determined according to the magnitude of the global synchronization factor. Guided by the aforementioned change range benchmark, when adjusting an element in the spatial parameter library by traversing the cross-dimensional correlation mapping table, the values ​​of the affected elements in the brightness parameter library and texture parameter library are simultaneously calculated and pre-adjusted based on the chain influence weight recorded in the cross-dimensional correlation mapping table. After completing one round of traversal adjustment based on association mapping, an updated spatial parameter library, an updated brightness parameter library, and an updated texture parameter library are generated.

9. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 8, characterized in that, Based on the updated encryption parameter library, a new round of perturbation, modulation, and weaving is performed on the spatial dimension component, the brightness dimension component, and the texture dimension component to generate the final anti-counterfeiting encoding sequence, including: The initial position disorder is used as a new spatial dimension component input. Using the updated spatial parameter library, the generation of spatial perturbation seeds and the position disorder induction process are repeated to generate position disorder. The migrated luminance value sequence is used as the input of the new luminance dimension component. Using the updated luminance parameter library, the generation and cyclic migration process of the luminance modulation waveform is repeated to generate a luminance sequence. The woven texture vector set is used as the input of new texture dimension components. The updated texture parameter library is used to repeat the generation and cross-weaving process of the texture fusion template to generate a texture network. The position disorder, the brightness sequence, and the texture network are recombined according to the original pixel correspondence to generate a final pixel matrix containing three-dimensional encryption information. The final pixel matrix is ​​converted into binary code to obtain the final anti-counterfeiting code sequence.

10. The multi-parameter variable-order three-dimensional encrypted binary anti-counterfeiting printing method as described in claim 9, characterized in that, It also includes: extracting feature fingerprints from the final anti-counterfeiting code sequence, compressing the feature fingerprints into an identification code, and embedding the identification code into the printed product, specifically including: The final anti-counterfeiting encoding sequence is subjected to block hash calculation to obtain a set of hash values, which are then used as the original feature fingerprint. Extract the key parameter trajectory generated during the encryption process. The key parameter trajectory includes the evolution path of the spatial perturbation seed, the morphological sequence of the brightness modulation waveform, and the change record of the texture fusion template. The original feature fingerprint is mixed with the key parameter trajectory and then fed into a compression encoder; The compression encoder uses a combination of run-length encoding and differential encoding to compress the mixed data and output a compact bit stream, which is the identifier encoding. During the printing plate-making stage, the identification code is embedded into a designated area or the full-size image of the printed product in the form of a micro-dot matrix or spectral watermark.

Images