Methods, devices and storage media for fingerprint verification of porcelain

Abstract

This invention relates to the field of image processing technology, and more particularly to a method, apparatus, and storage medium for verifying the fingerprint of porcelain. The method includes: acquiring a glaze image and mineral element data; determining the texture coefficient of the glaze image; obtaining the diameter of each bubble in a grayscale image, dividing the grayscale image into partitions, obtaining the number of bubbles in each partition, and determining the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition; determining the spectral characteristics of the porcelain; determining the authenticity of the porcelain based on the texture coefficient of the glaze image, the bubble distribution state of the glaze layer, and the spectral characteristics of the porcelain; determining an adjustment factor based on the fluorescence decay time of the porcelain glaze, and determining the mineral element fingerprint of the porcelain based on the acquired mineral element data; and updating the porcelain authenticity determination process based on the elemental mineral fingerprint and the adjustment factor. This invention effectively improves the accuracy of porcelain fingerprint verification.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method, apparatus and storage medium for verifying fingerprints on porcelain. Background Technology

[0002] Traditional methods of authenticating porcelain rely heavily on empirical visual assessment and single physicochemical tests, which suffer from high subjectivity, destructiveness, and limited data dimensionality. With advancements in high-quality counterfeiting technology, traditional methods struggle to accurately identify the microscopic features of the glaze and differences in material composition in modern forgeries.

[0003] In recent years, identification techniques based on optical image analysis and spectral detection have gradually emerged, but they still face the following challenges: quantitative assessment of glaze texture lacks multi-scale feature coupling analysis, making it difficult to effectively distinguish between natural crazing and artificially imitated patterns; statistical models of bubble distribution struggle to handle pore size variations caused by glaze aging and overlapping bubble segmentation errors; and spectral reflectance data is not dynamically correlated with mineral element fingerprints, resulting in insufficient sensitivity to minute differences in chemical composition. Therefore, a non-destructive identification method integrating multimodal data is urgently needed to improve discrimination accuracy and anti-interference capabilities through multi-dimensional feature coupling modeling. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus and storage medium for verifying fingerprints on porcelain, so as to solve at least one of the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for verifying fingerprints on porcelain, comprising:

[0007] Obtain the diameter of each bubble in the grayscale image, divide the grayscale image into partitions, obtain the number of bubbles in each partition, and determine the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition.

[0008] The spectral reflectance of the porcelain surface at wavelength λ was collected, and the spectral fit was calculated to determine the spectral characteristics of the porcelain.

[0009] The authenticity of porcelain can be determined based on the texture coefficient of the glaze image, the distribution of bubbles in the glaze layer, and the spectral characteristics of the porcelain.

[0010] The adjustment factor is determined based on the fluorescence decay time of the porcelain glaze, and the mineral element fingerprint of the porcelain is determined based on the collected mineral element data. The authenticity determination process of the porcelain is updated based on the element and mineral fingerprint and the adjustment factor.

[0011] Optionally, glaze images and mineral element data can be acquired;

[0012] The glaze image is converted into a grayscale image to determine the texture contrast and texture direction dispersion of the glaze image. The texture coefficient of the glaze image is determined based on the texture contrast and texture direction dispersion of the grayscale image.

[0013] Optionally, the glaze image is converted into a grayscale image, and the texture contrast A1 of the grayscale image is calculated;

[0014] Construct the gray-level co-occurrence matrix of the grayscale image and calculate the energy values ​​θ0, θ1, and θ2 in four preset directions (0°, 45°, 90°, 135°). 45 θ 90 θ 135 ;

[0015] The texture orientation dispersion D is calculated based on a multidimensional spatial energy coupling model. The formula for calculating D is as follows:

[0016] In the formula, Dr = {0, 45, 90, 135} is the set of directions;

[0017] The texture coefficient W of the glaze image is determined based on the texture contrast A1 and texture direction dispersion D of the grayscale image.

[0018] Optionally, the number of bubbles in the m-th partition is denoted as Qm, and the diameter of the n-th bubble in the grayscale image is denoted as Zn. The number of bubbles in the m-th partition, Qm, is coupled with the diameter of the n-th bubble in the grayscale image, Zn, to construct the bubble distribution index QF. The expression for QF is:

[0019]

[0020] In the formula, σ(Zn) is the standard deviation of bubble diameter, μ(Zn) is the average bubble diameter, B is the variation factor, μ(Qm) is the average number of bubbles in each zone, M is the number of zones, and Qb is the preset standard number.

[0021] The bubble distribution factor q0 is compared with QF to determine the bubble distribution state of the glaze layer. When QF is less than or equal to the bubble distribution factor q0, the bubble distribution state of the glaze layer is determined to be normal; otherwise, the bubble distribution state of the glaze layer is determined to be abnormal.

[0022] Optionally, reflectance data of the porcelain surface in the visible light range are collected at wavelength intervals of 10 nm, and the spectral reflectance of the porcelain surface at wavelength λ is denoted as Ra(γ), and the spectral fit S is calculated; the spectral characteristics GT of the porcelain are determined based on the spectral fit, GT=(1-S) / △S.

[0023] Optionally, if the texture coefficient W of the glaze image is greater than the texture discrimination factor w0, or the glaze bubble distribution state is abnormal, or the porcelain spectral feature GT is greater than the spectral discrimination factor g0, then the porcelain is determined to be a fake; otherwise, the texture coefficient W of the glaze image, the bubble distribution index QF, and the porcelain spectral feature GT are coupled and analyzed to determine the porcelain fingerprint index ZW, ZW=u1×(w0-W)+u2×(q0-QF) / q0+u3×(g0-GT), where u1 is the texture weight, u2 is the bubble weight, u3 is the spectral weight, and u1+u2+u3=1;

[0024] The fingerprint index ZW of the porcelain is compared with the fingerprint discrimination factor z0. If ZW is less than or equal to z0, the porcelain is determined to be a fake. If ZW is greater than z0, the porcelain is determined to be genuine.

[0025] Optionally, the fluorescence decay time t0 of the porcelain glaze is compared with the preset duration t1. If t0 is less than or equal to t1, the adjustment factor is determined to be 1. If t0 is greater than t1, the adjustment factor is determined to be {1+exp[3×(t0-t1) / (t0+t1)-3]}.

[0026] Optionally, the iron content F1 in the porcelain glaze is compared with the first mineral content discrimination factors f1 and f2. If F1 is greater than or equal to f1 and less than or equal to f2, the iron content in the porcelain glaze is determined to be normal; otherwise, the iron content in the porcelain glaze is determined to be abnormal.

[0027] The titanium content F2 in the porcelain glaze is compared with the second mineral content discrimination factors f3 and f4. If F2 is greater than or equal to f3 and less than or equal to f4, the titanium content in the porcelain glaze is judged to be normal; otherwise, the titanium content in the porcelain glaze is judged to be abnormal.

[0028] When the iron or titanium content in the porcelain glaze is abnormal, the process for determining the authenticity of the porcelain is updated by updating the fingerprint discrimination factor to z1, where z1 = z0 + adjustment factor × β, and β is the mineral element fingerprint coefficient of the porcelain.

[0029] According to another aspect of this application, a porcelain verification device is provided, comprising:

[0030] The acquisition unit collects glaze images and mineral element data.

[0031] The texture analysis unit converts the glaze image into a grayscale image to determine the texture contrast and texture direction dispersion of the glaze image, and determines the texture coefficient of the glaze image based on the texture contrast and texture direction dispersion of the grayscale image.

[0032] The bubble distribution unit obtains the diameter of each bubble in the grayscale image, divides the grayscale image into partitions, obtains the number of bubbles in each partition, and determines the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition.

[0033] The spectral feature construction unit collects the spectral reflectance of the porcelain surface at wavelength λ and calculates the spectral fit to determine the spectral characteristics of the porcelain.

[0034] The verification unit determines the authenticity of porcelain based on the texture coefficient of the glaze image, the distribution state of bubbles in the glaze layer, and the spectral characteristics of the porcelain.

[0035] The update unit determines the adjustment factor based on the fluorescence decay time of the porcelain glaze, and determines the porcelain mineral element fingerprint based on the collected mineral element data. The authenticity determination process of the porcelain is updated based on the element mineral fingerprint and the adjustment factor.

[0036] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, wherein the computer program is used to control the electronic device in which the computer-readable storage medium is located to perform the porcelain fingerprint verification method during runtime.

[0037] The beneficial effects of this invention are as follows: By integrating the microscopic texture of the glaze, the distribution of bubble clusters, multispectral reflectance, and mineral element fingerprint characteristics, a multidimensional dynamic identification model is constructed. Synchronous acquisition of high-resolution images and mineral data provides a benchmark for microstructure analysis; quantification of texture contrast and directional dispersion reduces interference from artificially imitated textures; modeling of bubble distribution effectively identifies differences in firing processes; spectral reflectance matching accurately captures anomalies in chemical composition; and the dynamic calibration mechanism of fluorescence decay time and mineral elements further enhances adaptability to environmental and process fluctuations. Feature extraction and weight allocation in each step form a complementary verification chain, avoiding the limitations of a single criterion and enhancing the ability to distinguish highly realistic forgeries through multi-parameter collaborative decision-making, ultimately achieving non-destructive, high-precision, and traceable authentication of porcelain. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0039] Figure 1 This is a flowchart illustrating the fingerprint verification method for porcelain in this embodiment.

[0040] Figure 2 This is a flowchart illustrating the texture feature analysis method in this embodiment.

[0041] Figure 3 This is a flowchart illustrating the update method in this embodiment.

[0042] Figure 4 This is a schematic diagram of the porcelain verification device in this embodiment. Detailed Implementation

[0043] To more clearly illustrate the present invention, the following description, in conjunction with preferred embodiments and accompanying drawings, further explains the invention. Similar components in the drawings are indicated by the same reference numerals. Those skilled in the art should understand that the specific description below is illustrative rather than restrictive and should not be construed as limiting the scope of protection of the present invention.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] Specifically, the fingerprint verification method, device and storage medium for porcelain described in this embodiment are applied to the fingerprint verification of monochrome glazed porcelain such as Ru kiln, Guan kiln and Longquan kiln celadon of the Song Dynasty.

[0046] Please see Figure 1 As shown, it is a flowchart illustrating the porcelain fingerprint verification method of this embodiment, including:

[0047] Step S101: Acquire glaze image and mineral element data. The glaze image is a high-resolution microscopic image, and the mineral elements include iron content and titanium content.

[0048] For example, in this embodiment, high-resolution microscopic images can be acquired by a laser confocal microscope, and mineral element data can be acquired by an X-ray fluorescence spectrometer. In this embodiment, no specific limitation is made on the data acquisition method, and those skilled in the art can set it freely according to their needs.

[0049] Step S102: Convert the glaze image into a grayscale image to determine the texture contrast and texture direction dispersion of the glaze image, and determine the texture coefficient of the glaze image based on the texture contrast and texture direction dispersion of the grayscale image.

[0050] Please see Figure 2 As shown, the texture feature analysis method includes:

[0051] Step S201: Convert the glaze image into a grayscale image and calculate the texture contrast of the grayscale image.

[0052] Specifically, step S201 converts the glaze image into a grayscale image and calculates the texture contrast A1 of the grayscale image. The expression for A1 is:

[0053] P(i,j) represents the probability of occurrence of pixel pair (i,j), where i,j are the gray values ​​of the grayscale image.

[0054] Specifically, nonlinear grayscale conversion and brightness adaptive equalization technology are used to eliminate the interference of color deviation on the analysis of the microscopic topology of the glaze surface, accurately extract the layered gradient characteristics formed by the natural flow of the glaze, and avoid misjudging the mechanical luster of artificial polishing as historical patina.

[0055] Please see Figure 2 As shown, the texture feature analysis method further includes:

[0056] Step S202: Construct the gray-level co-occurrence matrix of the gray-level image and calculate the texture direction dispersion. Determine the texture coefficient of the glaze image based on the texture contrast and texture direction dispersion of the gray-level image.

[0057] Specifically, step S202 constructs the gray-level co-occurrence matrix of the gray-level image and calculates the energy values ​​θ0, θ1, θ2, and θ3 in four preset directions (0°, 45°, 90°, 135°). 45 θ 90 θ 135 The energy value is calculated using the following formula:

[0058]

[0059] In the formula, N is the number of gray levels and N = 256, P d (i,j) represents the joint probability distribution of gray values ​​along direction d, where d = 0, 45, 90, 135;

[0060] The texture orientation dispersion D is calculated based on a multidimensional spatial energy coupling model. The formula for calculating D is as follows:

[0061] In the formula, k is a numerical index, Dr = {0, 45, 90, 135}, and Dr is the set of directions;

[0062] The texture coefficient of the glaze image is determined based on the texture contrast A1 and texture direction dispersion D of the grayscale image. The expression for the texture coefficient is W=x1×|A1-A| / A+x2×D / Db;

[0063] In the formula, x1 is the contrast weight, x2 is the dispersion weight, x1+x2=1, A is the contrast factor, and Db is the dispersion factor.

[0064] Specifically, by using a gray-level co-occurrence matrix to couple multi-directional energy distribution, the flow directionality of the glaze layer is decoupled from the difference in texture generation caused by mechanical processing of imitations. The focus is on analyzing the degree of disorder in crack distribution. Natural cracks are random in direction, like the branching of tree branches. Imitations often have cracks with overly neat directions due to mechanical engraving. This can be identified by calculating the degree of directional difference.

[0065] For example, in this embodiment, the contrast weight can be set to 0.6, the dispersion weight can be set to 0.4, the contrast factor can be set to 120, and the dispersion factor can be set to 0.45. This embodiment does not specifically limit the setting of the above data, and those skilled in the art can set them freely according to their needs.

[0066] For example, in this embodiment, the cv2.cvtColor function of Python-OpenCV 4.5 can be used to map an RGB image to a grayscale channel. The energy value θd can be calculated by setting the distance to 1 and the direction Dr = [0,45,90,135] using the graycomatrix function of Scikit-image. This embodiment does not impose specific limitations on the above settings, and those skilled in the art can set them freely according to their needs.

[0067] Please continue reading. Figure 1 As shown, the porcelain fingerprint verification method further includes:

[0068] Step S103: Obtain the diameter of each bubble in the grayscale image, divide the grayscale image into partitions, obtain the number of bubbles in each partition, and determine the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition.

[0069] For example, in this embodiment, when obtaining the diameter of each bubble and the number of bubbles in each partition of the grayscale image, the Otsu algorithm is used to automatically determine the threshold, and morphological closing operation (3×3 circular kernel) is used to fill the internal voids of the bubbles. Canny edge detection combined with the watershed algorithm is used to separate overlapping bubbles (minimum aperture 5μm). The equivalent circle diameter is calculated based on the contour. This can be achieved using tools such as OpenCV 4.5 (Python interface) and Scikit-image. In this embodiment, no specific limitation is made on the method of obtaining the diameter of each bubble and the number of bubbles in each partition of the grayscale image. Those skilled in the art can set it freely according to their needs.

[0070] Specifically, in step S103, the number of bubbles in the m-th partition is denoted as Qm, the diameter of the n-th bubble in the grayscale image is denoted as Zn, and the number of bubbles in the m-th partition Qm is coupled with the diameter of the n-th bubble in the grayscale image Zn for analysis to construct the bubble distribution index QF. The expression for QF is:

[0071]

[0072] In the formula, σ(Zn) is the standard deviation of bubble diameter, μ(Zn) is the average bubble diameter, B is the variation factor, μ(Qm) is the average number of bubbles in each zone, M is the number of zones, and Qb is the preset standard number.

[0073] The bubble distribution factor q0 is compared with QF to determine the bubble distribution state of the glaze layer. When QF is less than or equal to the bubble distribution factor q0, the bubble distribution state of the glaze layer is determined to be normal; otherwise, the bubble distribution state of the glaze layer is determined to be abnormal.

[0074] Specifically, the size and density of bubbles in the glaze are statistically analyzed, and the typical characteristics of antique porcelain and modern imitations are compared. For example, the glaze of old porcelain has a longer melting time, and the bubbles are usually more dispersed; modern, fast-fired imitations have more bubbles and a more chaotic distribution, and traces of "fake bubbles" blown by machines can also be seen.

[0075] For example, in this embodiment, the number of partitions can be set to 12, the variation factor can be set to 0.15, the number of preset standards can be set to 50, and the bubble distribution factor can be set to 1.2. This embodiment does not specifically limit the above settings, and those skilled in the art can set them freely according to their needs.

[0076] Please continue reading. Figure 1 As shown, the porcelain fingerprint verification method further includes:

[0077] Step S104: Collect the spectral reflectance of the porcelain surface at wavelength λ and calculate the spectral fit to determine the spectral characteristics of the porcelain.

[0078] Specifically, in step S104, reflectance data of the porcelain surface in the visible light range are collected at wavelength intervals of 10 nm, and the spectral reflectance of the porcelain surface at wavelength λ is denoted as Ra(γ). The expression for the spectral fit is set as follows:

[0079]

[0080] The spectral characteristics of porcelain, GT, are determined based on the spectral fit, where GT = (1-S) / ΔS.

[0081] Where Rs(γ) is the preset spectral reflectance of the porcelain surface at wavelength λ, and ΔS is the fit factor.

[0082] Specifically, different colors of light are used to illuminate the glaze surface, and the changes in reflected light are analyzed to accurately determine whether the glaze formula matches the corresponding era. For example, the cobalt material in Ming dynasty blue and white porcelain has a unique blue color. Even if modern imitations use similar materials, the reflection under light will have subtle differences due to different firing processes.

[0083] For example, in this embodiment, the reflectance data of the porcelain surface in the visible light range (390-780nm) can be collected using a spectrophotometer. For the preset spectral reflectance of the porcelain surface at wavelength λ, 50 archaeologically certified porcelain pieces of the same type can be used as samples. The samples must have a clear year (e.g., carbon-14 dating of 1100–1150), be well-preserved, and show no signs of repair. Ocean Insight can be used to collect the data. An FX spectrophotometer (wavelength 380–780 nm) was used. Five measurement points were selected for each sample (avoiding cracked and stained areas). Reflectance was collected at 10 nm intervals, and abnormal fluctuation data (such as abnormal points with a sudden increase of 20% in reflectance) were removed. The reflectance of each sample was standardized according to the white board calibration value (98%). The average reflectance of all samples was calculated for each wavelength point, and the average values ​​were connected to form the reflectance curve of the genuine product and stored in the database as the Rs(λ) benchmark. In this embodiment, the above data acquisition method and the setting method of the preset spectral reflectance of the porcelain surface at wavelength λ are not specifically limited. Those skilled in the art can set them freely according to their needs.

[0084] For example, in this embodiment, the fit factor can be set to 0.1. This embodiment does not specifically limit the setting of the fit factor, and those skilled in the art can set it freely according to their needs.

[0085] Please continue reading. Figure 1 As shown, the porcelain verification method further includes:

[0086] Step S105: Determine the authenticity of the porcelain based on the texture coefficient of the glaze image, the distribution state of bubbles in the glaze layer, and the spectral characteristics of the porcelain.

[0087] Specifically, if the texture coefficient W of the glaze image is greater than the texture discrimination factor w0, or the glaze bubble distribution state is abnormal, or the porcelain spectral feature GT is greater than the spectral discrimination factor g0, then the porcelain is determined to be a fake; otherwise, the texture coefficient W of the glaze image, the bubble distribution index QF, and the porcelain spectral feature GT are coupled and analyzed to determine the porcelain fingerprint index ZW, ZW=u1×(w0-W)+u2×(q0-QF) / q0+u3×(g0-GT), where u1 is the texture weight, u2 is the bubble weight, u3 is the spectral weight, and u1+u2+u3=1;

[0088] The fingerprint index ZW of the porcelain is compared with the fingerprint discrimination factor z0. If ZW is less than or equal to z0, the porcelain is determined to be a fake. If ZW is greater than z0, the porcelain is determined to be genuine.

[0089] Specifically, the three-dimensional dynamic threshold determination, which integrates texture, bubbles, and spectrum, avoids misjudgment based on a single feature and improves the confidence level of distinguishing between genuine and fake products in complex aging scenarios.

[0090] For example, in this embodiment, u1 can be set to 0.4, u2 can be set to 0.2, u3 can be set to 0.4, w0 can be set to 0.85, g0 can be set to 0.9, and z0 can be set to 0.18. This embodiment does not specifically limit the setting of the above data, and those skilled in the art can set it freely according to their needs.

[0091] Please continue reading. Figure 1 As shown, the porcelain verification method further includes:

[0092] Step S106: Determine the adjustment factor based on the fluorescence decay time of the porcelain glaze, and determine the porcelain mineral element fingerprint based on the collected mineral element data. Update the porcelain authenticity determination process based on the element mineral fingerprint and the adjustment factor.

[0093] Please see Figure 3 As shown, the update method includes:

[0094] Step S301: Determine the adjustment factor based on the fluorescence decay time of the porcelain glaze.

[0095] Specifically, in step S106, the fluorescence decay time t0 of the porcelain glaze is compared with the preset duration t1. If t0 is less than or equal to t1, the adjustment factor is determined to be 1. If t0 is greater than t1, the adjustment factor is determined to be {1+exp[3×(t0-t1) / (t0+t1)-3]}.

[0096] Specifically, by dynamically generating an adjustment factor based on the fluorescence decay time of the glaze, the problem of misjudgment of optical features caused by glaze aging or differences in storage environment in traditional identification methods is effectively solved. By coupling the fluorescence decay time characteristics with the mineral element fingerprint, this adjustment mechanism can adapt to the natural aging law of the glaze of genuine products, avoid spectral or texture data drift caused by changes in the relaxation time of the glass phase on the glaze surface, and significantly improve the anti-interference ability against high-quality counterfeit porcelain.

[0097] For example, in this embodiment, the fluorescence decay time of the porcelain glaze can be collected by a time-resolved fluorescence spectrometer; this embodiment does not specifically limit the collection method, and those skilled in the art can freely set it according to their needs.

[0098] For example, in this embodiment, the preset duration can be set to 120ns; this embodiment does not specifically limit the setting of the preset duration, and those skilled in the art can set it freely according to their needs.

[0099] Please continue reading. Figure 3 As shown, the update method includes:

[0100] Step S302: Determine the mineral element fingerprint of the porcelain based on the collected mineral element data, and update the porcelain authenticity determination process based on the element mineral fingerprint and adjustment factor.

[0101] Specifically, in step S302, the iron content F1 in the porcelain glaze is compared with the first mineral content discrimination factors f1 and f2. If F1 is greater than or equal to f1 and less than or equal to f2, the iron content in the porcelain glaze is determined to be normal; otherwise, the iron content in the porcelain glaze is determined to be abnormal.

[0102] The titanium content F2 in the porcelain glaze is compared with the second mineral content discrimination factors f3 and f4. If F2 is greater than or equal to f3 and less than or equal to f4, the titanium content in the porcelain glaze is judged to be normal; otherwise, the titanium content in the porcelain glaze is judged to be abnormal.

[0103] When the iron or titanium content in the porcelain glaze is abnormal, the process for determining the authenticity of the porcelain is updated by updating the fingerprint discrimination factor to z1, where z1 = z0 + adjustment factor × β, and β is the mineral element fingerprint coefficient of the porcelain.

[0104] Specifically, by introducing dual discrimination thresholds for the content of first and second mineral elements and a dynamic fingerprint correction mechanism, precise interception of modern synthetic glazes is achieved. This method can not only capture abnormal fluctuations in trace elements in the glaze formula (such as the imbalance of iron and titanium ratio caused by modern chemical additives), but also dynamically adjust the global discrimination criteria according to the degree of elemental anomaly. Combined with the real-time calibration of the porcelain mineral element fingerprint by the adjustment factor, it ensures that even when there are conflicts in mineral traceability data, the system can still output more robust identification conclusions based on a multi-dimensional evidence chain, thus maintaining high-precision identification capabilities in the complex scenario of constantly iterating high-counterfeiting technology.

[0105] For example, in this embodiment, f1 can be set to 2.5%, f2 can be set to 5.5%, f3 can be set to 0.2%, f4 can be set to 0.9%, and β can be set to 0.06. This embodiment does not specifically limit the setting of the above data, and those skilled in the art can set them freely according to their needs.

[0106] Please see Figure 4 As shown, the porcelain verification device includes:

[0107] The acquisition unit collects glaze images and mineral element data.

[0108] The texture analysis unit converts the glaze image into a grayscale image to determine the texture contrast and texture direction dispersion of the glaze image, and determines the texture coefficient of the glaze image based on the texture contrast and texture direction dispersion of the grayscale image.

[0109] The bubble distribution unit obtains the diameter of each bubble in the grayscale image, divides the grayscale image into partitions, obtains the number of bubbles in each partition, and determines the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition.

[0110] The spectral feature construction unit collects the spectral reflectance of the porcelain surface at wavelength λ and calculates the spectral fit to determine the spectral characteristics of the porcelain.

[0111] The verification unit determines the authenticity of porcelain based on the texture coefficient of the glaze image, the distribution state of bubbles in the glaze layer, and the spectral characteristics of the porcelain.

[0112] The update unit determines the adjustment factor based on the fluorescence decay time of the porcelain glaze, and determines the porcelain mineral element fingerprint based on the collected mineral element data. The authenticity determination process of the porcelain is updated based on the element mineral fingerprint and the adjustment factor.

[0113] This application also provides a computer-readable storage medium, which is a tangible physical storage medium that can store the aforementioned computer program and various types of data used in the program; the physical storage medium includes, but is not limited to, existing physical storage media or combinations thereof, such as random access memory, read-only memory, optical disk, and hard disk.

[0114] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable programs, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable programs, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0115] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A method for fingerprint verification of porcelain, characterized in that, include: Obtain the diameter of each bubble in the grayscale image, divide the grayscale image into partitions, obtain the number of bubbles in each partition, and determine the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition. The spectral reflectance of the porcelain surface at wavelength λ was collected, and the spectral fit was calculated to determine the spectral characteristics of the porcelain. The authenticity of porcelain can be determined based on the texture coefficient of the glaze image, the distribution of bubbles in the glaze layer, and the spectral characteristics of the porcelain. The adjustment factor is determined based on the fluorescence decay time of the porcelain glaze, and the mineral element fingerprint of the porcelain is determined based on the collected mineral element data. The authenticity determination process of the porcelain is updated based on the element and mineral fingerprint and the adjustment factor. If the texture coefficient W of the glaze image is greater than the texture discrimination factor w0, or the distribution state of the glaze bubble is abnormal, or the spectral feature GT of the porcelain is greater than the spectral discrimination factor g0, then the porcelain is determined to be a fake. Conversely, the texture coefficient W and bubble distribution index QF of the glaze image are coupled with the spectral characteristics GT of the porcelain to determine the porcelain fingerprint index ZW, ZW=u1×(w0-W)+u2×(q0-QF) / q0+u3×(g0-GT), where u1 is the texture weight, u2 is the bubble weight, u3 is the spectral weight, u1+u2+u3=1, and q0 is the bubble distribution factor; The fingerprint index ZW of the porcelain is compared with the fingerprint discrimination factor z0. If ZW is less than or equal to z0, the porcelain is determined to be a fake. If ZW is greater than z0, the porcelain is determined to be genuine. The fluorescence decay time t0 of the porcelain glaze is compared with the preset duration t1. If t0 is less than or equal to t1, the adjustment factor is set to 1. If t0 is greater than t1, the adjustment factor is set to {1+exp[3×(t0-t1) / (t0+t1)-3]}. The iron content F1 in the porcelain glaze is compared with the first mineral content discrimination factors f1 and f2. If F1 is greater than or equal to f1 and less than or equal to f2, the iron content in the porcelain glaze is judged to be normal; otherwise, the iron content in the porcelain glaze is judged to be abnormal. The titanium content F2 in the porcelain glaze is compared with the second mineral content discrimination factors f3 and f4. If F2 is greater than or equal to f3 and less than or equal to f4, the titanium content in the porcelain glaze is judged to be normal; otherwise, the titanium content in the porcelain glaze is judged to be abnormal. When the iron or titanium content in the porcelain glaze is abnormal, the process for determining the authenticity of the porcelain is updated by updating the fingerprint discrimination factor to z1, where z1 = z0 + adjustment factor × β, and β is the mineral element fingerprint coefficient of the porcelain.

2. The method for verifying fingerprints on porcelain according to claim 1, characterized in that, Also includes: Collect glaze images and mineral element data; The glaze image is converted into a grayscale image to determine the texture contrast and texture direction dispersion of the glaze image. The texture coefficient of the glaze image is determined based on the texture contrast and texture direction dispersion of the grayscale image.

3. The method for verifying fingerprints on porcelain according to claim 2, characterized in that, The glaze image is converted into a grayscale image, and the texture contrast A1 of the grayscale image is calculated. Construct the gray-level co-occurrence matrix of the grayscale image and calculate the energy values ​​θ0, θ1, and θ2 in four preset directions (0°, 45°, 90°, 135°). 45 θ 90 θ 135 ; The texture orientation dispersion D is calculated based on a multidimensional spatial energy coupling model. The formula for calculating D is as follows: ; In the formula, Dr={0,45,90,135} is the set of directions; The texture coefficient W of the glaze image is determined based on the texture contrast A1 and texture direction dispersion D of the grayscale image.

4. The method for verifying fingerprints on porcelain according to claim 3, characterized in that, Let Qm be the number of bubbles in the m-th partition, and Zn be the diameter of the n-th bubble in the grayscale image. We then perform a coupled analysis of Qm and Zn to construct the bubble distribution index QF. The expression for QF is: ; In the formula, σ(Zn) is the standard deviation of bubble diameter, μ(Zn) is the average bubble diameter, B is the variation factor, μ(Qm) is the average number of bubbles in each zone, M is the number of zones, and Qb is the preset standard number. The bubble distribution factor q0 is compared with QF to determine the bubble distribution state of the glaze layer. When QF is less than or equal to the bubble distribution factor q0, the bubble distribution state of the glaze layer is determined to be normal; otherwise, the bubble distribution state of the glaze layer is determined to be abnormal.

5. The method for verifying fingerprints on porcelain according to claim 4, characterized in that, The reflectance data of the porcelain surface in the visible light range were collected at wavelength intervals of 10 nm, and the spectral reflectance of the porcelain surface at wavelength λ was denoted as Ra(γ), and the spectral fit S was calculated. The spectral characteristics GT of porcelain are determined based on the spectral fit, where GT = (1-S) / ΔS, and ΔS is the fit factor.

6. A porcelain verification device, applied to the porcelain fingerprint verification method as described in claim 1, characterized in that, include: The acquisition unit collects glaze images and mineral element data. The texture analysis unit converts the glaze image into a grayscale image to determine the texture contrast and texture direction dispersion of the glaze image, and determines the texture coefficient of the glaze image based on the texture contrast and texture direction dispersion of the grayscale image. The bubble distribution unit obtains the diameter of each bubble in the grayscale image, divides the grayscale image into partitions, obtains the number of bubbles in each partition, and determines the bubble distribution state of the glaze layer based on the bubble diameter and the number of bubbles in each partition. The spectral feature construction unit collects the spectral reflectance of the porcelain surface at wavelength λ and calculates the spectral fit to determine the spectral characteristics of the porcelain. The verification unit determines the authenticity of porcelain based on the texture coefficient of the glaze image, the distribution state of bubbles in the glaze layer, and the spectral characteristics of the porcelain. The update unit determines the adjustment factor based on the fluorescence decay time of the porcelain glaze, and determines the porcelain mineral element fingerprint based on the collected mineral element data. The authenticity determination process of the porcelain is updated based on the element mineral fingerprint and the adjustment factor.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is used to control the electronic device on which the computer-readable storage medium is located to perform the porcelain fingerprint verification method according to any one of claims 1-5 during runtime.

Images