Ceramic microstructure digital identity authentication system and method

Abstract

The present invention discloses a digital identity authentication system and method for ceramic microstructures. The ceramic microstructure images are collected through a high-resolution microscope, the feature information is extracted using advanced image processing technology, a digital signature is generated through an encryption algorithm, and the distributed ledger, non-tamperability and encryption technology characteristics of the blockchain are utilized to store the digital signature and related ceramic information on the chain, thereby realizing multi-node joint maintenance and verification. The authentication system and authentication method can ensure the authenticity, integrity and traceability of ceramic microstructure data, effectively solve the problems of authenticity identification and traceability in the ceramic market, provide strong technical support for quality control, cultural relics identification and market regulation in the ceramic industry, and have significant innovation, practicality and reliability.

Description

[Technical field]

[0001] The present invention relates to the technical field of ceramic identification, and in particular to a ceramic microstructure digital identity authentication system and method. [Background Technology]

[0002] As a material with a long history and wide application, ceramics play an important role in many fields such as art, industry, and cultural heritage. However, the current ceramics market is facing many severe challenges.

[0003] (I) Difficulties in authenticity identification

[0004] Limitations of traditional identification methods. In terms of ceramic authenticity identification, traditional methods mainly rely on the experience and judgment of experts, through appearance observations such as color, texture, shape, and simple physical and chemical tests such as hardness testing and component analysis. However, these methods are highly subjective, and different experts may come to different conclusions. Moreover, with the rapid development of modern ceramic technology, high-imitation ceramic products are extremely similar to the originals in appearance and conventional test indicators, making it difficult for traditional identification methods to effectively distinguish the authenticity.

[0005] The current dilemma of microstructure analysis. Although microstructure analysis provides a deeper perspective for ceramic identification, the existing microstructure analysis technology has obvious deficiencies in data management and verification. For example, the storage of microstructure image data lacks a safe and reliable mechanism and is easily tampered with or lost. The data verification process also lacks a unified and authoritative standard, resulting in uneven credibility of identification results.

[0006] (II) Difficulty in product traceability

[0007] Industrial ceramics quality control needs. In the field of industrial ceramics, product quality and performance are directly related to the safety and efficiency of industrial production. However, due to the lack of an effective traceability system, it is difficult for companies to accurately track key information such as the production process, raw material sources, and process parameters of products. This makes it impossible to quickly locate the root cause of quality problems and take effective improvement measures when quality problems arise, which seriously affects the quality stability and batch consistency of industrial ceramics.

[0008] Challenges in the protection and identification of cultural relic ceramics. For cultural relic ceramics, the proliferation of counterfeits not only damages the economic interests of collectors, but also poses a serious threat to the protection and inheritance of cultural heritage. Existing cultural relic ceramic identification technology often seems powerless when faced with complex counterfeiting methods. At the same time, the source and circulation history of cultural relic ceramics lack reliable records and verification methods, making it difficult to ensure their authenticity and legitimacy.

[0009] (III) Opportunities for the introduction of blockchain technology

[0010] With the advent of the digital age, blockchain technology, with its unique characteristics of distributed ledger, non-tamperability, encryption security, etc., provides new ideas and ways to solve the above problems in the ceramic industry. Blockchain technology can provide a safe, transparent, traceable storage and verification environment for ceramic microstructure data, and is expected to fundamentally change the identification and traceability model of the ceramic industry.

[0011] For example, a ceramic traceability and identification method based on blockchain and trusted data space (publication number: CN114971653A) collects ceramic texture images to build a standard library, stores relevant information in a trusted data space and performs security classification, encrypts some information and stores it in the form of blocks on the blockchain to achieve ceramic traceability and identification. However, the present invention uses a high-resolution microscope to obtain more accurate ceramic microstructure images in image acquisition, and deeply analyzes microstructure features through specific algorithms, uses asymmetric encryption to generate digital signatures, and uses a fast comparison algorithm based on hash functions during authentication, which is unique and innovative.

[0012] Another example is the blockchain-based digital identity authentication method and system (publication number: CN116011025B), which discloses a blockchain-based digital identity authentication method, including the steps of an authentication terminal sending a credential request to a user terminal, and the user terminal generating credentials and credential certification according to a credential template. However, the present invention focuses on digital identity authentication of ceramic microstructures, obtains microstructure images through a microscope, and performs preprocessing, feature extraction and other operations, which is different from the application field and specific implementation method of the comparative document.

[0013] For another example, a blockchain-based identity authentication method and system (application publication number: CN113946875A) mainly involves data access identity authentication between different nodes. However, the present invention is aimed at digital identity authentication of ceramic microstructures, and achieves ceramic identity authentication by acquiring and processing ceramic microstructure images through specific technical means, which is different from the technical solution and application object of the comparative document. [Summary of the invention]

[0014] The present invention overcomes the shortcomings of the prior art, provides a ceramic microstructure digital identity authentication system and method, utilizes the characteristics of blockchain to achieve accurate authentication of the digital identity of ceramic microstructures, and provides innovative solutions for quality control, cultural relic identification, and market regulation in the ceramic industry.

[0015] To achieve the above object, the present invention adopts the following technical solutions:

[0016] A ceramic microstructure digital identity authentication system based on blockchain, characterized by:

[0017] An image acquisition module, used to obtain microstructure images of ceramic samples;

[0018] A feature extraction module, connected to the image acquisition module, is used to extract feature vectors of the microstructure image;

[0019] The digital signature generation module is connected to the feature extraction module, and encrypts the feature vector through an asymmetric encryption algorithm to generate a digital signature of the ceramic microstructure;

[0020] The blockchain network is connected to the image acquisition module, the feature extraction module, the digital signature generation module and the authentication module respectively. The blockchain network includes multiple nodes, and each node jointly maintains a blockchain account book for storing the digital signature and related information of the ceramic sample;

[0021] The authentication module is connected to the blockchain network and is used to compare the digital signature of the ceramic to be authenticated with the data stored on the blockchain to determine the authenticity of the ceramic and related attribute information.

[0022] A ceramic microstructure digital identity authentication system based on blockchain as described above is characterized in that the image acquisition module is a scanning electron microscope or an atomic force microscope.

[0023] The ceramic microstructure digital identity authentication system based on blockchain as described above is characterized by:

[0024] The feature extraction module includes

[0025] An image preprocessing unit, used for denoising and contrast enhancement of microstructure images;

[0026] Feature analysis unit, which uses deep learning-based image analysis algorithms to identify and extract features in ceramic microstructures;

[0027] The feature vector generating unit converts the extracted features into feature vectors in the form of mathematical vectors.

[0028] The ceramic microstructure digital identity authentication system based on blockchain as described above is characterized in that the image preprocessing unit uses a median filtering algorithm for denoising and a histogram equalization method to enhance contrast.

[0029] As described above, a ceramic microstructure digital identity authentication system based on blockchain is characterized in that the deep learning image analysis algorithm in the feature analysis unit adopts a convolutional neural network and is trained with a large amount of ceramic microstructure image data.

[0030] A ceramic microstructure digital identity authentication system based on blockchain as described above is characterized in that the asymmetric encryption algorithm includes an RSA algorithm and an ECC algorithm.

[0031] A ceramic microstructure digital identity authentication system based on blockchain as described above is characterized in that the relevant information stored in the blockchain network node includes the origin, production process, production batch and collection time of the ceramic.

[0032] The ceramic microstructure digital identity authentication system based on blockchain as described above is characterized in that: the blockchain network adopts a consortium chain architecture, and the nodes include ceramic production enterprises, testing agencies and regulatory departments; the data storage in the blockchain network adopts a distributed file system.

[0033] The ceramic microstructure digital identity authentication system based on blockchain as described above is characterized in that: the authentication module adopts a fast comparison algorithm based on hash function to calculate the hash value of the digital signature of the ceramic to be authenticated and the digital signature stored in the blockchain, and judges the authenticity of the ceramic based on the matching degree of the hash value.

[0034] A digital identity authentication method for ceramic microstructures based on blockchain using the above system, characterized in that:

[0035] S1, image acquisition step, using an image acquisition module to obtain a microstructure image of a ceramic sample;

[0036] S2, feature extraction step, performing preprocessing, feature analysis and feature vector generation operations on the collected image through a feature extraction module;

[0037] S3, a digital signature generation step, using a digital signature generation module to encrypt the feature vector to generate a digital signature;

[0038] S4, the chain storage step, uploading the digital signature and related ceramic information to the blockchain network for storage, and multiple nodes jointly verify and record the transaction;

[0039] S5, authentication step, using the authentication module to compare the digital signature of the ceramic to be authenticated with the data stored on the blockchain to obtain the authentication result.

[0040] The beneficial effects of the present invention are:

[0041] The present invention integrates advanced image acquisition and analysis technology with blockchain technology to achieve accurate extraction of ceramic microstructural features, secure generation and reliable authentication of digital identities, and ensure the authenticity, integrity and traceability of data in all aspects of ceramic products such as production, circulation and collection, thereby effectively solving the problems of authenticity identification and product traceability in the ceramic market and promoting the healthy and sustainable development of the ceramic industry. [Drawings]

[0042] Figure 1It is a schematic diagram of the model structure of the present invention. [Specific implementation method]

[0043] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings.

[0044] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly. In addition, the descriptions of "preferred", "sub-preferred", etc. in the present invention are only used for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "preferred" or "sub-preferred" may explicitly or implicitly include at least one such feature.

[0045] like Figure 1 As shown, a ceramic microstructure digital identity authentication system based on blockchain includes an image acquisition module, a feature extraction module, a digital signature generation module, a blockchain network and an authentication module.

[0046] Wherein, the image acquisition module is used to obtain the microstructure image of the ceramic sample;

[0047] Specifically, a high-resolution microscope is selected as the core acquisition device, such as a scanning electron microscope (SEM) or an atomic force microscope (AFM), with a resolution of not less than 0.1 microns. This high resolution can clearly present the crystal structure details in the ceramic microstructure, including the shape, size, and arrangement of the crystals, as well as key features such as the morphology of the grain boundaries, the size and distribution of the pores. For example, a scanning electron microscope can provide high-definition microscopic morphological images by emitting an electron beam and detecting the electron signal reflected from the sample surface, which has unique advantages for observing the internal structure of ceramics; an atomic force microscope uses the interaction force between atoms to detect the microscopic morphology of the sample surface, and performs well in analyzing the nano-level structural features of the ceramic surface.

[0048] Specifically, the image acquisition module is closely connected with the feature extraction module, and the acquired ceramic microstructure image is transmitted to the feature extraction module in real time for subsequent processing.

[0049] The feature extraction module is connected to the image acquisition module and is used to extract the feature vector of the microstructure image; the feature extraction module includes an image preprocessing unit, a feature analysis unit and a feature vector generation unit, wherein:

[0050] An image preprocessing unit, used for denoising and contrast enhancement of microstructure images;

[0051] Specifically, in the denoising process, the median filter algorithm is used to perform denoising operations on the collected microstructure images. By selecting the median value in the pixel neighborhood to replace the central pixel value, the median filter can effectively remove common noise types such as verification noise and Gaussian noise in the image, while better retaining the edge and detail information of the image. For example, for ceramic microstructure images with noise interference, the median filter can make the crystal structure and grain boundaries in the image clearer and more discernible.

[0052] Specifically, in the contrast enhancement process, the histogram equalization method is used to enhance the image contrast. This method redistributes the grayscale of the image and stretches the grayscale range of the image, so that different features in the ceramic microstructure can be more clearly presented in the image, which is convenient for subsequent feature analysis. For example, the pores and crystals that were originally difficult to distinguish in the darker or brighter areas of the image have their contrast significantly improved and the features more prominent after histogram equalization.

[0053] Feature analysis unit, which uses deep learning-based image analysis algorithms to identify and extract features in ceramic microstructures;

[0054] Specifically, the algorithm chosen is to use a deep learning image analysis algorithm based on a convolutional neural network (CNN). The CNN algorithm has a powerful automatic feature learning capability and can automatically discover and extract complex feature patterns from a large amount of ceramic microstructure image data. For example, by training ceramic microstructure images from different origins and different processes, CNN can learn the characteristic representations of various crystal morphologies, as well as the inherent laws of grain boundary distribution and pore characteristics.

[0055] Specifically, the training process is to use a large amount of professionally annotated ceramic microstructure image data to train the CNN model. During the training process, the model parameters are continuously adjusted so that the model can accurately identify and extract key features such as crystal morphology, grain boundary distribution, pore size and distribution in the ceramic microstructure. For example, for the recognition of crystal morphology, the model can learn the characteristic differences of different crystal shapes (such as columnar, lamellar, granular, etc.) and accurately locate and classify these crystals in the image.

[0056] The feature vector generating unit converts the extracted features into feature vectors in the form of mathematical vectors.

[0057] Specifically, feature quantification is to convert various features extracted by the feature analysis unit into mathematical vector form. For example, the crystal morphology characteristics are quantified into shape parameter vectors, and the crystal shape is described by calculating the aspect ratio, area, perimeter and other geometric parameters of the crystal; the grain boundary distribution characteristics are represented as boundary point coordinate vectors, and the position information of the grain boundary in the image is recorded; the pore size and distribution characteristics are converted into aperture and position vectors, which specifically describe the size of the pores and the distribution position in the ceramic microstructure. This vector form can concisely and accurately represent the characteristic information of the ceramic microstructure, which is convenient for efficient processing and comparison in the system.

[0058] Specifically, the output of the feature extraction module (i.e., the feature vector) is transmitted to the digital signature generation module, and at the same time, there is data interaction with the blockchain network to record the relevant data in the feature extraction process (such as feature extraction algorithm parameters, training data sources, etc.) on the blockchain to ensure the traceability of the feature extraction process.

[0059] The digital signature generation module is connected to the feature extraction module, and encrypts the feature vector through an asymmetric encryption algorithm to generate a digital signature of the ceramic microstructure;

[0060] Specifically, the choice of asymmetric encryption algorithm is to use an asymmetric encryption algorithm, such as the RSA algorithm or the ECC algorithm, to encrypt the feature vector generated by the feature vector generation unit. The RSA algorithm is based on the problem of large integer decomposition and has high security and a wide range of applications; the ECC algorithm is based on the elliptic curve discrete logarithm problem and has a shorter key length and higher computational efficiency at the same security strength. For example, when using the RSA algorithm, by generating a public key and a private key pair, the feature vector is signed using the private key. The generated digital signature contains the characteristic information of the ceramic microstructure and the encrypted authentication information. Only the node with the corresponding public key can verify the authenticity of the digital signature.

[0061] Specifically, the digital signature generation module transmits the generated digital signature to the blockchain network for storage, and at the same time associates the digital signature with relevant information of the ceramic to be authenticated (such as the origin, production process, production batch, collection time, etc. of the ceramic) to ensure the integrity of the information.

[0062] The blockchain network is connected to the image acquisition module, feature extraction module, digital signature generation module and authentication module respectively. The blockchain network includes multiple nodes, and each node jointly maintains a blockchain account book for storing the digital signature and related information of the ceramic sample. The relevant information stored in the blockchain network node includes the origin, production process, production batch and collection time of the ceramic.

[0063] Specifically, the blockchain network adopts a consortium chain architecture, and its nodes include ceramic manufacturers, testing agencies, regulatory authorities and other stakeholders. While ensuring a certain degree of decentralization, the consortium chain can provide better manageability and privacy protection. For example, ceramic manufacturers can upload ceramic microstructure data and related information during the production process to the blockchain, and the testing agency is responsible for verifying the authenticity and accuracy of the data, while the regulatory authorities can supervise and manage the entire process to ensure the compliance and reliability of the data.

[0064] Specifically, the data storage in the blockchain network uses a distributed file system, such as IPFS (InterPlanetary File System) for data storage. IPFS stores data in a content-addressed manner, which is efficient, reliable, and tamper-proof. In the blockchain network, the digital signature and related information of the ceramic microstructure are stored in IPFS in the form of files, and the blockchain ledger records the hash value and other index information of the file for fast data search and verification. For example, when it is necessary to query the microstructure data of a ceramic product, the corresponding file can be quickly located and obtained in IPFS through the index information in the blockchain ledger.

[0065] Specifically, each node jointly maintains the blockchain ledger and participates in data verification, storage, and update operations. Nodes reach consensus through consensus algorithms (such as the PBFT algorithm) to ensure the consistency and immutability of blockchain data. For example, when a new ceramic microstructure digital signature and related information are to be uploaded to the chain, multiple nodes need to verify it, and only information that has been verified by a certain proportion of nodes can be recorded on the blockchain.

[0066] Specifically, the blockchain network interacts with the image acquisition module, feature extraction module, digital signature generation module, and authentication module. The image acquisition module obtains the acquisition-related information of the ceramic microstructure image (such as acquisition time, acquisition device number, etc.), receives the feature extraction process data of the feature extraction module, stores the digital signature and related information generated by the digital signature generation module, and provides data query and comparison services for the authentication module.

[0067] Among them, the authentication module is connected to the blockchain network and is used to compare the digital signature of the ceramic to be authenticated with the data stored on the blockchain to determine the authenticity of the ceramic and related attribute information.

[0068] Specifically, the comparison algorithm uses a fast comparison algorithm based on a hash function to calculate the hash value of the digital signature to be authenticated and the digital signature stored in the blockchain, and judges the authenticity of the ceramic according to the matching degree of the hash value. The hash function maps the digital signature to a hash value of fixed length, which is one-way and anti-collision. For example, using the SHA-256 hash function, the digital signature to be authenticated and the digital signature stored in the blockchain are converted into 256-bit hash values ​​respectively, and then the two hash values ​​are compared. If the hash values ​​match completely, it is considered that the ceramic to be authenticated has the same microstructural characteristics as the corresponding ceramic recorded in the blockchain, and the authentication is true; otherwise, the authentication is false.

[0069] Specifically, the authentication module obtains data from the blockchain network for comparison, and interacts with the user interface or other external systems (such as ceramic sales platforms, museum cultural relics management systems, etc.), receives the digital signature or related information of the ceramics to be authenticated, and feeds back the authentication results to the corresponding system.

[0070] A digital identity authentication method for ceramic microstructures based on blockchain using the above system, comprising:

[0071] S1, image acquisition step, using an image acquisition module to obtain a microstructure image of a ceramic sample;

[0072] Specifically, the image acquisition module uses a high-resolution microscope to obtain the microstructure image of the ceramic sample and transmits the image to the feature extraction module.

[0073] S2, feature extraction step, performing preprocessing, feature analysis and feature vector generation operations on the collected image through a feature extraction module;

[0074] Specifically, the feature extraction module preprocesses the image (denoising and contrast enhancement), and then uses a deep learning algorithm to perform feature analysis to extract features such as crystal morphology, grain boundary distribution, pore size and distribution in the ceramic microstructure, and converts these features into feature vectors.

[0075] S3, a digital signature generation step, using a digital signature generation module to encrypt the feature vector to generate a digital signature;

[0076] Specifically, the digital signature generation module encrypts the feature vector to generate a digital signature, and associates the digital signature with relevant information of the ceramic (origin, production process, production batch, collection time, etc.).

[0077] S4, the chain storage step, uploading the digital signature and related ceramic information to the blockchain network for storage, and multiple nodes jointly verify and record the transaction;

[0078] Specifically, the digital signature and related information are uploaded to the blockchain network. The nodes in the blockchain network verify the data. After the verification is passed, the data is stored using a distributed file system (such as IPFS), and the index information of the data (such as file hash value, etc.) is recorded in the blockchain ledger.

[0079] S5, authentication step, using the authentication module to compare the digital signature of the ceramic to be authenticated with the data stored on the blockchain to obtain the authentication result.

[0080] Specifically, when ceramics need to be authenticated, the authentication module obtains the digital signature of the ceramic to be authenticated (which can be generated by re-collecting the image and going through the above processing flow, or obtained from existing records), and then queries the corresponding digital signature data from the blockchain network. Through a comparison algorithm based on a hash function, the hash value of the digital signature to be authenticated and the digital signature stored in the blockchain is calculated, the matching degree of the hash value is compared, the authentication result (true or false) is obtained, and the result is fed back to the relevant parties.

[0081] The specific implementation of this case is as follows:

[0082] 1. Image Acquisition

[0083] 1.1. Microscope selection and parameter adjustment

[0084] According to the specific type of ceramic sample and the identification requirements, carefully select a suitable high-resolution microscope. For artistic ceramics, especially ceramic products that focus on surface microstructure and decorative details, atomic force microscopy (AFM) is preferred. AFM can provide surface morphology images with nanometer resolution, which can clearly show the microscopic arrangement of crystals on the ceramic surface, the fine structure of grain boundaries, and the microscopic characteristics of the decorative layer. When using AFM, it is necessary to accurately adjust the scanning parameters, such as scanning range, scanning rate, and the force between the needle tip and the sample surface. The scanning range should be reasonably set according to the size of the ceramic sample and the size of the area to be observed, usually between a few microns and tens of microns; the scanning rate should be moderate, too fast may lead to a decrease in image resolution, too slow will increase the acquisition time, generally controlled at a few to dozens of scanning lines per second; the force between the needle tip and the sample surface should be adjusted to a level that can obtain a clear image without causing damage to the sample surface, usually in the nanonewton to micronewton range.

[0085] For industrial ceramics or ceramic products that need to observe the internal microstructure, scanning electron microscopy (SEM) is a more appropriate choice. SEM has a large depth of field and high resolution, and can deeply observe the crystal structure, pore distribution and grain boundary characteristics of ceramics. When using SEM, pretreatment should be performed according to the conductivity and surface state of the sample. If the sample is not conductive, conductive treatment such as gold spraying or carbon spraying is required to avoid charge accumulation affecting the image quality. The thickness of gold spraying or carbon spraying should be uniform and moderate, generally controlled between a few nanometers and tens of nanometers. At the same time, the acceleration voltage, working distance and beam current of the electron beam should be adjusted. The acceleration voltage is generally between 1kV and 30kV. The appropriate value is selected according to the composition and structural characteristics of the sample. The lower acceleration voltage is suitable for observing the surface morphology, and the higher acceleration voltage can be used to analyze the internal structure of the sample; the working distance is usually between a few millimeters and tens of millimeters. The appropriate working distance can ensure the clarity and depth of field of the image; the beam current size should be adjusted according to the conductivity of the sample and the required image brightness, generally between a few nanoamperes and tens of nanoamperes.

[0086] Sample preparation

[0087] For different types of ceramic samples, corresponding sample preparation methods need to be adopted. For bulk ceramic samples, they must be cut and polished before observation under a microscope. High-precision cutting equipment should be used during cutting to ensure that the cut surface is flat and smooth. The cutting thickness should be controlled according to the depth of field of the microscope and the observation requirements, generally between a few hundred microns and a few millimeters. The polishing process should gradually use polishing sandpaper of different grits, from coarse to fine, and finally use polishing liquid for mirror polishing to remove scratches and surface damage caused during the cutting process, so that the sample surface reaches a microscopic level of flatness.

[0088] For powdered ceramic samples, they need to be dispersed in a suitable medium, such as alcohol or water, and then dropped on a glass slide and dried naturally or at low temperature to form a thin film sample. During the dispersion process, ultrasonic oscillation and other methods can be used to ensure that the powder is evenly dispersed to avoid agglomeration that affects the observation effect. At the same time, attention should be paid to controlling the thickness of the film, which is generally between a few microns and tens of microns. Too thick may cause blurred images, and too thin may not fully present the microscopic structural characteristics of the sample.

[0089] 2. Feature extraction

[0090] Image preprocessing

[0091] Implementation of median filtering algorithm: When performing median filtering denoising on the collected microstructure image, first determine the size of the filter window. The selection of the filter window size should be determined according to the type and degree of noise in the image and the scale of the ceramic microstructure features. For images with severe noise and relatively large microstructure features, a larger filter window, such as 5×5 or 7×7 pixels, can be selected; for images with light noise and fine microstructure features, a filter window of 3×3 pixels is generally selected. During the filtering process, for each pixel in the image, the pixel values ​​in its neighborhood (within the filter window) are sorted by size, and the middle value is taken as the new value of the pixel. In this way, noise points such as inspection noise and Gaussian noise in the image can be effectively removed, while the edge and detail information of the ceramic microstructure are well preserved.

[0092] Histogram equalization operation: When using the histogram equalization method to enhance the contrast of an image, first calculate the grayscale histogram of the image and count the frequency of each grayscale level. Then calculate the cumulative distribution function (CDF) based on the grayscale histogram, and map the original grayscale value to a new grayscale range through the CDF. For example, for an 8-bit grayscale image (grayscale value range is 0-255), the grayscale value distribution can be more evenly extended to the entire grayscale range through histogram equalization, so that the grayscale difference between features such as crystals, grain boundaries, and pores in the originally low-contrast ceramic microstructure image is increased, so that it is displayed more clearly in the image. In actual operation, the functions provided by programming languages ​​(such as the OpenCV library in Python) can be used to implement histogram equalization operations to improve processing efficiency.

[0093] 2.2. Feature analysis

[0094] CNN model training and optimization: When using a deep learning image analysis algorithm based on a convolutional neural network (CNN), it is first necessary to collect a large amount of ceramic microstructure image data of different types, origins, and production processes to construct a training set. The scale of the training set should be large enough to ensure that the CNN model can learn rich feature patterns. It is generally recommended to contain tens of thousands to hundreds of thousands of images. During the training process, the structure of the CNN model should be reasonably designed. For example, select appropriate parameters such as the number of convolutional layers, convolution kernel size, and step size. The number of convolutional layers is generally between 3 and 10 layers, and the convolution kernel size can be selected from common sizes such as 3×3 and 5×5, and the step size is usually 1 or 2. At the same time, it is necessary to configure a suitable pooling layer, such as a maximum pooling layer, whose pooling window size is generally 2×2 or 3×3, to reduce the data dimension and reduce the amount of calculation. In addition, it is necessary to select a suitable activation function (such as the ReLU function) and an optimization algorithm (such as the stochastic gradient descent algorithm and its variants, such as Adagrad, Adadelta, Adam, etc.) to accelerate the convergence speed of the model and improve the performance of the model. During the training process, cross-validation and other techniques should be used to evaluate and optimize the model, and the model parameters should be adjusted according to the evaluation results until the model achieves better performance indicators, such as accuracy and recall rate.

[0095] Feature extraction process: When the trained CNN model performs feature analysis on the preprocessed image, the image is first input into the convolution layer, which performs convolution operation with the image through the convolution kernel to extract local features in the image, such as edges, textures, etc. Then, the pooling layer downsamples the output of the convolution layer to further extract the main information of the features and reduce the amount of data. Next, the fully connected layer integrates the output of the pooling layer to obtain a vector representation of the features such as crystal morphology, grain boundary distribution, pore size and distribution in the ceramic microstructure. For example, for crystal morphology features, the CNN model can output parameter vectors representing the crystal shape, such as aspect ratio, circularity, symmetry, etc., by learning the characteristic patterns of different crystal shapes in the image; for grain boundary distribution features, the model can output parameters such as grain boundary length, curvature, and distribution density; for pore size and distribution features, the model can provide information such as the average diameter, maximum diameter, number of pores, and position distribution of pores in the image.

[0096] 2.3. Feature vector generation

[0097] Feature quantification method: When converting the various features extracted by the feature analysis unit into mathematical vector form, it is necessary to select a suitable quantification method according to the characteristics of different features. For crystal morphology features, in addition to the geometric parameters such as aspect ratio, circularity, symmetry, etc. mentioned above, parameters such as crystal orientation angle and interplanar spacing can also be calculated to further describe the spatial orientation and structural characteristics of the crystal. The complexity of grain boundary distribution characteristics can be quantified by the fractal dimension of the grain boundary. The higher the fractal dimension, the more complex the grain boundary. The pore size and distribution characteristics can be represented by parameters such as the volume fraction of pores and the pore spacing distribution function. These quantified features are combined into feature vectors in a certain order. For example, they can be arranged in the order of crystal morphology features, grain boundary distribution features, pore size and distribution features. The parameters inside each feature are also sorted according to certain rules to form a feature vector in a unified format for subsequent encryption processing and data comparison.

[0098] 3. Digital signature generation and blockchain storage

[0099] 3.1. Digital signature generation

[0100] Application of asymmetric encryption algorithm: When using an asymmetric encryption algorithm (such as RSA algorithm or ECC algorithm) to encrypt a feature vector to generate a digital signature, a key pair must be generated first. For the RSA algorithm, the generation of a key pair involves selecting two large prime numbers (usually 1024 bits or 2048 bits), calculating their product as the modulus, and then selecting the appropriate public key exponent and private key exponent based on the modulus. When generating a digital signature, the private key is used to encrypt the feature vector. The encryption process can use the standard encryption algorithm in the public key cryptography system (such as the PKCS#1 standard). For the ECC algorithm, the appropriate elliptic curve equation and base point must be selected first, and then the private key and public key are generated based on the base point. When using the private key to sign the feature vector, a standard algorithm such as the Elliptic Curve Digital Signature Algorithm (ECDSA) is used for calculation. The generated digital signature must contain sufficient information, including not only the encrypted feature vector, but also the signature algorithm identifier, key information, etc., to ensure the integrity and verifiability of the digital signature.

[0101] 3.2. Blockchain storage

[0102] Alliance chain node settings and operations: In the blockchain network based on the alliance chain architecture, the nodes of ceramic manufacturers, testing agencies, regulatory authorities, etc. need to be set up and operated accordingly. The ceramic manufacturer node is responsible for uploading the ceramic microstructure image data collected during the production process, the digital signature generated after processing, and related ceramic information (such as origin, production process, production batch, collection time, etc.) to the blockchain network. Before uploading, the data must be formatted and packaged to meet the requirements of the blockchain network. After receiving the data, the testing agency node must verify the authenticity and accuracy of the data. The verification content includes whether the image data is complete, whether the feature extraction process is correct, and whether the digital signature is valid. The regulatory department node is responsible for supervising and managing the entire process, formulating and implementing relevant policies and standards, and ensuring the compliance operation of the blockchain network. The nodes communicate and collaborate through consensus algorithms (such as the PBFT algorithm). When a node makes a request for data to be uploaded to the chain, other nodes must verify and vote on the request. Only when more than a certain proportion (such as 2 / 3) of the nodes agree, the data can be recorded in the blockchain ledger.

[0103] Distributed File System (IPFS) storage: When using IPFS for data storage, the digital signature and related ceramic information must first be converted into a file format suitable for IPFS storage, such as JSON format or binary file format. Then, upload the file to the IPFS network, and IPFS will generate a unique hash value (Content Identifier, CID) for the file. This hash value will be used as the index of the file in IPFS for subsequent search and access. In the blockchain ledger, the hash value of the file and the metadata related to the file (such as data source, upload time, identification of the ceramic product to which it belongs, etc.) are recorded. When it is necessary to query or obtain ceramic microstructure data, the hash value in the blockchain ledger can be used to quickly locate and download the corresponding file in IPFS, and then parse and process it.

[0104] 4. Certification process

[0105] 4.1. Ceramic treatment to be certified

[0106] Data collection and signature generation: When authenticating a ceramic product, first, the microstructure image of the ceramic must be obtained according to the above-mentioned image acquisition and feature extraction methods, and the corresponding digital signature must be generated through the digital signature generation module. During the image acquisition process, it is necessary to ensure that the acquisition conditions (such as microscope parameter settings, sample preparation methods, etc.) are as consistent as possible with the conditions when the data was previously stored in the system to ensure data comparability. The feature extraction process should also use the same algorithm and parameters to ensure that the extracted features are consistent. After the digital signature is generated, it is prepared as the data to be authenticated and compared with the data stored on the blockchain.

[0107] 4.2. Comparison and certification

[0108] Hash function comparison operation: When the authentication module uses a fast comparison algorithm based on a hash function (such as SHA-256) for authentication, the hash value of the digital signature to be authenticated is first calculated. Then, the hash value of the stored digital signature related to the ceramic product is queried from the blockchain network (searched in the blockchain account book through the identification information of the ceramic, such as the number, batch, etc.). The two hash values ​​are compared bit by bit. If the two hash values ​​are exactly the same, it is preliminarily judged that the ceramic to be authenticated has the same microstructural characteristics as the corresponding ceramic recorded in the blockchain, and the authentication is true. However, considering the possibility of extremely low probability events such as hash collisions and possible minor errors in practical applications (such as subtle differences in microscope imaging conditions, minor changes in sample preparation, etc.), further auxiliary verification is required. For example, other relevant information of the ceramic (such as origin, production process, etc.) can be compared for consistency, or some key features in the image can be manually reviewed to ensure the accuracy of the authentication result. If the hash values ​​are different, the authentication is directly false. The authentication results will be fed back to relevant parties in a timely manner, such as ceramic sales platforms, cultural relics appraisal agencies, consumers, etc., so that they can take corresponding measures. At the same time, relevant data in the authentication process (such as comparison time, comparison results, information of nodes involved in the comparison, etc.) can also be recorded on the blockchain for subsequent query and audit.

[0109] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. All equivalent structural changes made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or directly or indirectly applied in other related technical fields, are included in the patent protection scope of the present invention.

Claims

1. A digital identity authentication system for ceramic microstructures based on blockchain, characterized by: Included An image acquisition module, used to obtain microstructure images of ceramic samples; A feature extraction module, connected to the image acquisition module, is used to extract feature vectors of the microstructure image; The digital signature generation module is connected to the feature extraction module, and encrypts the feature vector through an asymmetric encryption algorithm to generate a digital signature of the ceramic microstructure; The blockchain network is connected to the image acquisition module, the feature extraction module, the digital signature generation module and the authentication module respectively. The blockchain network includes multiple nodes, and each node jointly maintains a blockchain account book for storing the digital signature and related information of the ceramic sample; The authentication module is connected to the blockchain network and is used to compare the digital signature of the ceramic to be authenticated with the data stored on the blockchain to determine the authenticity of the ceramic and related attribute information.

2. According to a blockchain-based ceramic microstructure digital identity authentication system according to claim 1, it is characterized in that: The image acquisition module is a scanning electron microscope or an atomic force microscope.

3. A blockchain-based ceramic microstructure digital identity authentication system according to claim 2, characterized in that: The feature extraction module includes An image preprocessing unit, used for denoising and contrast enhancement of microstructure images; Feature analysis unit, which uses deep learning-based image analysis algorithms to identify and extract features in ceramic microstructures; The feature vector generating unit converts the extracted features into feature vectors in the form of mathematical vectors.

4. A blockchain-based ceramic microstructure digital identity authentication system according to claim 3, characterized in that: The image preprocessing unit uses a median filter algorithm to perform denoising and a histogram equalization method to enhance contrast.

5. According to a blockchain-based ceramic microstructure digital identity authentication system according to claim 3, it is characterized in that: The deep learning image analysis algorithm in the feature analysis unit uses a convolutional neural network and is trained with a large amount of ceramic microstructure image data.

6. According to a blockchain-based ceramic microstructure digital identity authentication system according to claim 1, it is characterized by: Asymmetric encryption algorithms include RSA algorithm and ECC algorithm.

7. A blockchain-based ceramic microstructure digital identity authentication system according to claim 1, characterized in that: The relevant information stored in the blockchain network nodes includes the origin, production process, production batch and collection time of the ceramics.

8. A blockchain-based ceramic microstructure digital identity authentication system according to claim 1, characterized in that: The blockchain network adopts a consortium chain architecture, and the nodes include ceramic manufacturers, testing agencies and regulatory departments; the data storage in the blockchain network adopts a distributed file system.

9. A ceramic microstructure digital identity authentication system based on blockchain according to claim 1, characterized in that: The authentication module uses a fast comparison algorithm based on hash functions to calculate the hash value of the digital signature of the ceramic to be authenticated and the digital signature stored in the blockchain, and determines the authenticity of the ceramic based on the degree of match between the hash values.

10. A method for digital identity authentication of ceramic microstructures based on blockchain using the system according to any one of claims 1 to 9, characterized in that: Included S1, image acquisition step, using an image acquisition module to obtain a microstructure image of a ceramic sample; S2, feature extraction step, performing preprocessing, feature analysis and feature vector generation operations on the collected image through a feature extraction module; S3, a digital signature generation step, using a digital signature generation module to encrypt the feature vector to generate a digital signature; S4, the chain storage step, uploading the digital signature and related ceramic information to the blockchain network for storage, and multiple nodes jointly verify and record the transaction; S5, authentication step, using the authentication module to compare the digital signature of the ceramic to be authenticated with the data stored on the blockchain to obtain the authentication result.

Images