A blockchain-based art anti-counterfeiting and traceability system and method

Abstract

The present application relates to the field of blockchain anti-counterfeiting technology, and particularly relates to an art piece anti-counterfeiting and tracing system and method based on a blockchain, comprising: obtaining multi-modal feature data of an art piece and generating a unique digital fingerprint through a hash algorithm; continuously obtaining moving trajectory images of the art piece and extracting space-time features to generate a dynamic trajectory password; jointly encrypting the digital fingerprint and the trajectory password with art piece ownership information and writing into an immutable distributed ledger of the blockchain to form blockchain data and bind with a two-dimensional code corresponding to the art piece; scanning the two-dimensional code attached to the art piece to retrieve the blockchain data and compare with the physical features and the trajectory password to verify the art piece. Thus, the history trajectory of the art piece can be combined to continuously track the location of the art piece, and the location of the art piece and other anti-counterfeiting information of the art piece are associated to better trace the art piece, thereby improving the transparency and trustworthiness of the art piece trading market.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of blockchain anti-counterfeiting technology, in particular to an art work anti-counterfeiting and traceability system and method based on blockchain. BACKGROUND

[0002] Art work anti-counterfeiting and traceability refers to using modern information technology to establish a complete identity recognition and tracking system for art works, to ensure the authenticity and traceability of the source of art works. By combining various technical means such as blockchain, RFID tags, digital watermarking, etc., each transfer of art works can be recorded on an unalterable ledger, from the beginning of creation to each transaction, all information can be effectively saved and real-time queried.

[0003] In the above manner, the transfer information data of art works can be established online, but it is not convenient to associate and correspond with the actual art works, reducing the effectiveness of anti-counterfeiting and traceability. SUMMARY

[0004] The present application aims to provide an art work anti-counterfeiting and traceability system and method based on blockchain, which can combine the historical track of art works to continuously track the location of art works, associate the location of art works with other anti-counterfeiting information of art works to better trace the art works, and improve the transparency and trustworthiness of art work transaction market.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides an art work anti-counterfeiting and traceability method based on blockchain, comprising obtaining multi-modal feature data of art works and generating a unique digital fingerprint through a hash algorithm;

[0006] Continuously obtaining moving track images of art works and extracting spatiotemporal features to generate dynamic track passwords;

[0007] Encrypting the digital fingerprint and the track password together with the ownership information of art works, and writing them into the unalterable distributed ledger of the blockchain to form blockchain data and bind them with the corresponding QR code of art works;

[0008] Scanning the QR code attached to the art work to retrieve the blockchain data and compare it with the physical features and track password to verify the art work.

[0009] Among them, the multi-modal feature data includes visible light images and infrared images.

[0010] The specific steps of obtaining multi-modal feature data of art works and generating a unique digital fingerprint through a hash algorithm include:

[0011] Collecting visible light images of art works, and scanning the surface of art works with an infrared spectrometer to obtain infrared images;

[0012] normalizing the visible light image and the infrared image into a uniform size and converting into a gray value matrix;

[0013] extracting deep features in the gray value matrix through a VGG16 network to obtain a total feature matrix;

[0014] processing the total feature matrix through a hash function to obtain a digital fingerprint.

[0015] The specific steps of continuously acquiring the moving trajectory image of the artwork and extracting the space-time features to generate the dynamic trajectory password include:

[0016] acquiring the moving path data of the artwork through an acceleration sensor and a gyroscope;

[0017] extracting the time stamp and geographic coordinates of the key nodes in the moving path data to obtain node information;

[0018] generating the dynamic trajectory password based on the node information using a chaotic encryption algorithm.

[0019] After generating the dynamic trajectory password based on the node information using the chaotic encryption algorithm, the steps further include:

[0020] generating an auxiliary verification code based on environmental sensor collected environmental data of the artwork, the environmental data including temperature data and humidity data, and matching the dynamic trajectory password and the auxiliary verification code simultaneously during verification.

[0021] The specific steps of acquiring the moving path data of the artwork through the acceleration sensor and the gyroscope include:

[0022] recording geographic data every preset time interval through a positioning module;

[0023] if the positioning module signal is lost, automatically switching to indoor positioning and acquiring corresponding geographic data;

[0024] cutting the geographic data according to a time window to obtain the moving path data.

[0025] The specific steps of extracting the time stamp and geographic coordinates of the key nodes in the moving path data to obtain node information include:

[0026] cleaning the moving path data, including filtering invalid data, deleting GPS drift points, and using a cubic spline interpolation method to complete missing time periods;

[0027] identifying the location of the stay point through a space-time clustering algorithm;

[0028] matching the location of the stay point with a geographic fence and an exhibition area to obtain the location of the key node.

[0029] The specific step of generating the auxiliary verification code based on the environment data collected by the environment sensor includes:

[0030] Collecting environment data of the artwork by the environment sensor;

[0031] Extracting environment features based on the environment data, the environment features including average temperature and humidity change rate;

[0032] Combining the selected environment features and then converting them into a fixed-length string through a hash function;

[0033] Encrypting the string using a symmetric encryption algorithm to obtain the auxiliary verification code.

[0034] In a second aspect, the present application also provides an artwork anti-counterfeiting and traceability system based on a blockchain, which includes an artwork fingerprint generation module, a track password generation module, a blockchain generation module, and a verification module.

[0035] The artwork fingerprint generation module is configured to obtain multi-modal feature data of the artwork and generate a unique digital fingerprint through a hash algorithm.

[0036] The track password generation module is configured to continuously obtain moving track images of the artwork and extract spatiotemporal features to generate a dynamic track password.

[0037] The blockchain generation module is configured to jointly encrypt the digital fingerprint and the track password with artwork ownership information and then write them into a tamper-proof distributed ledger of a blockchain to form blockchain data and bind them with a two-dimensional code corresponding to the artwork.

[0038] The verification module is configured to scan the two-dimensional code attached to the artwork to retrieve the blockchain data and compare them with the physical features and the track password to verify the artwork.

[0039] The blockchain-based art anti-counterfeiting and traceability system and method of the present application obtains various characteristic data of the artwork, including but not limited to image, material, process and other multi-modal information, and uses a hash algorithm to convert these information into a unique digital fingerprint. This digital fingerprint serves as the unique identifier of the artwork, ensuring its non-replicability and uniqueness. To enhance the security of the artwork during circulation, this method also includes continuous tracking of the artwork's movement trajectory. By capturing images of the artwork at different time points and locations, the temporal and spatial features contained therein are extracted to generate a dynamic trajectory code. This code not only records the historical location of the artwork, but also reflects its changing course in the time and space dimensions, providing a more solid basis for the authenticity verification of the artwork. The digital fingerprint generated above, together with the dynamic trajectory code and the ownership information of the artwork, are written into the blockchain system through encryption technology. As a decentralized and tamper-proof distributed ledger technology, blockchain can ensure the security and transparency of these information. Once the information is recorded on the blockchain, any attempt to modify these data will become impossible, thereby greatly improving the authenticity and reliability of the artwork information. To facilitate user inquiry and verification, a unique QR code is generated for each artwork. Users only need to scan the QR code attached to the artwork to quickly retrieve the relevant data stored on the blockchain and compare it with the physical characteristics and dynamic trajectory code to confirm the authenticity of the artwork and its ownership. Thus, the historical trajectory of the artwork can be combined to continuously track the location of the artwork, and the location of the artwork and other anti-counterfeiting information of the artwork can be associated to better trace the artwork, thereby improving the transparency and trustworthiness of the art transaction market. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0041] Fig. 1 is a flowchart of a blockchain-based art anti-counterfeiting and traceability method of the present application.

[0042] Fig. 2 is a flowchart of obtaining multi-modal characteristic data of the artwork and generating a unique digital fingerprint through a hash algorithm of the present application.

[0043] Fig. 3 is a flowchart of continuously obtaining movement trajectory images of the artwork and extracting temporal and spatial features to generate a dynamic trajectory code of the present application.

[0044] Fig. 4 is a flowchart of the present application for acquiring a movement path data of an artwork by an acceleration sensor and a gyroscope.

[0045] Fig. 5 is a flowchart of the present application for extracting a timestamp and a geographic coordinate of a key node in the movement path data to obtain node information.

[0046] Fig. 6 is a flowchart of the present application for generating an auxiliary verification code based on environmental data of an environment where the artwork is located collected by an environmental sensor. DETAILED DESCRIPTION

[0047] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the accompanying drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.

[0048] First Embodiment

[0049] Referring to Figs. 1-6 , the present application provides a blockchain-based artwork anti-counterfeiting traceability method, comprising:

[0050] S101 acquiring multi-modal feature data of the artwork and generating a unique digital fingerprint by a hash algorithm;

[0051] The multi-modal feature data includes visible light images and infrared images.

[0052] The specific steps include:

[0053] S201 collecting visible light images of the artwork, and scanning the surface of the artwork by an infrared spectrometer to obtain infrared images;

[0054] Use a high-resolution camera (such as a single-lens reflex camera or an industrial-grade camera) to take pictures of the artwork, ensuring that the light is evenly distributed to avoid shadows or overexposure affecting the image quality. The shooting angle should be perpendicular to the surface of the artwork to reduce perspective distortion. Use an infrared spectrometer (such as a near-infrared or short-wave infrared imaging device) to scan the surface of the artwork. Infrared images can capture hidden information under the surface of the artwork, such as pigment layer thickness, cracks, repair marks, etc., thus providing more details than visible light.

[0055] S202 normalizing the visible light images and the infrared images to a uniform size and converting them into a gray value matrix;

[0056] The collected visible light image and infrared image are cropped to the same size and normalized. The purpose of normalization is to eliminate the brightness difference caused by different devices, so that the image range is limited between [0, 255] or [0, 1].

[0057] The color image (if the visible light image is color) and the infrared image are converted to grayscale images. The grayscale formula is usually:

[0058] Gray = 0.299R + 0.587G + 0.114B Gray = 0.299R + 0.587G + 0.114B

[0059] Where R, G, B represent the pixel values of red, green, and blue channels respectively. The grayscale image is converted to a two-dimensional matrix form, and each element represents the grayscale value of the corresponding pixel.

[0060] For example, for a visible light image and an infrared image with a resolution of 1024x768, after normalization, they are adjusted to a uniform size of 256x256, and two 256x256 grayscale value matrices are generated respectively.

[0061] S203 extracts the depth features in the grayscale value matrix through the VGG16 network to obtain a total feature matrix;

[0062] Load the pre-trained VGG16 model using a deep learning framework (such as TensorFlow or PyTorch). VGG16 is a classic convolutional neural network with strong feature extraction capability. The grayscale value matrix is input into the VGG16 network. Since VGG16 is originally designed for RGB images, it is necessary to expand the grayscale image to a three-channel input (i.e. copy the grayscale matrix three times). Extract the feature vector at a certain layer of VGG16 (usually a layer before the fully connected layer). Assuming that a feature vector with a length of 4096 is extracted from each image, then the feature vectors of two images (visible light and infrared) are spliced to form a total feature matrix with a length of 8192.

[0063] The visible light image and the infrared image are input into VGG16 respectively to obtain two feature vectors:

[0064] Visible light image feature vector: [0.1, 0.5,..., 0.9] (length 4096)

[0065] Infrared image feature vector: [0.3, 0.2,..., 0.7] (length 4096)

[0066] After splicing, the total feature matrix is obtained:

[0067] [0.1, 0.5,..., 0.9, 0.3, 0.2,..., 0.7] [0.1, 0.5,..., 0.9, 0.3, 0.2,..., 0.7]

[0068] S204 processes the total feature matrix through a hash function to obtain a digital fingerprint.

[0069] The total feature matrix is hashed using a secure hash algorithm such as SHA-256 or MD5. The characteristic of the hash algorithm is that the output is completely different when the input changes slightly, and it is irreversible.

[0070] The total feature matrix is serialized into a string format and then input into the hash function to obtain a fixed-length digital fingerprint. For example, SHA-256 will generate a 256-bit binary string or corresponding hexadecimal representation.

[0071] Suppose the total feature matrix is:

[0072] [0.1, 0.5, 0.9, 0.3, 0.2, 0.7] [0.1, 0.5, 0.9, 0.3, 0.2, 0.7]

[0073] It is serialized into the string "0.1, 0.5, 0.9, 0.3, 0.2, 0.7", and then the digital fingerprint is generated through the SHA-256 hash algorithm.

[0074] S102 continuously acquires the moving trajectory image of the artwork and extracts the spatiotemporal features to generate a dynamic trajectory password;

[0075] The specific steps include:

[0076] S301 obtains the moving path data of the artwork through an acceleration sensor and a gyroscope;

[0077] An intelligent sensor module is embedded in the artwork packaging, including a three-axis acceleration sensor and a three-axis gyroscope. These sensors can monitor the motion state of the artwork in real time, such as acceleration changes, direction angles, and rotation angles. The specific steps include:

[0078] S401 records geographic data every preset time interval through a positioning module;

[0079] Use the GPS module to record the geographic position information (longitude, latitude, height) of the artwork at regular intervals. The preset time interval is usually between 1 second and 10 seconds, depending on the application scenario.

[0080] Suppose the artwork records position data every 5 seconds during transportation:

[0081] Time point t1: longitude = 116.397, latitude = 39.916

[0082] Time point t2: longitude = 116.398, latitude = 39.917.

[0083] S402 If the positioning module signal is lost, automatically switch to indoor positioning and obtain the corresponding geographic data;

[0084] When the GPS module cannot receive satellite signals (such as entering a building or a tunnel), the system will automatically switch to a backup positioning scheme.

[0085] Wi-Fi positioning estimates location using the MAC addresses and signal strengths of surrounding Wi-Fi hotspots. Bluetooth beacon triangulation determines location by deploying Bluetooth beacons. Inertial navigation combines data from acceleration sensors and gyroscopes to calculate relative displacement.

[0086] Indoor positioning data is seamlessly connected with previous GPS data to form complete mobile path data.

[0087] When the artwork enters a museum, the GPS signal is lost, and the system switches to Wi-Fi positioning, recording the following data:

[0088] Time point t3: location = museum A exhibition hall entrance.

[0089] S403 Divide the geographic data by time window to obtain the mobile path data.

[0090] According to the preset time window (such as 30 seconds or 1 minute), the geographic data is divided into multiple segments. Each segment contains all geographic coordinates within that time period.

[0091] Interpolate or fit the geographic data within each time window to generate a smooth mobile path. Output the path data within each time window for subsequent spatiotemporal feature extraction.

[0092] Suppose the artwork records the following location data within 1 minute:

[0093] Time point t1: longitude = 116.397, latitude = 39.916

[0094] Time point t2: longitude = 116.398, latitude = 39.917

[0095] Time point t3: longitude = 116.399, latitude = 39.918

[0096] After time window segmentation and path fitting, a smooth path from the starting point to the ending point is obtained.

[0097] S302 extracts the time stamp and geographic coordinates of the key nodes in the mobile path data to obtain node information;

[0098] The specific steps include:

[0099] S501 cleans the mobile path data, including filtering invalid data, deleting GPS drift points, and using a cubic spline interpolation method to complete the missing period;

[0100] Check the validity of each record, such as whether there are outliers (such as longitude exceeding the range [-180, 180] [-180, 180] or latitude exceeding the range [-90, 90] [-90, 90]). Delete data with repeated or missing time stamps. Use the speed threshold method to detect drift points. Assuming that the average moving speed of the artwork cannot exceed a certain upper limit (such as 100 km / h), if the speed between the two adjacent positions is much higher than the upper limit, it is considered to be a drift point. After deleting the drift points, the valid geographic coordinate data is retained.

[0101] For missing data caused by signal loss, use a cubic spline interpolation method to generate a smooth trajectory curve. Cubic spline interpolation can reasonably estimate the position of the missing period while ensuring continuity and smoothness.

[0102] S502 identifies the location of the stay point by a spatio-temporal clustering algorithm;

[0103] Use a density-based clustering algorithm (such as DBSCAN) to divide the mobile path data into clusters. DBSCAN can divide dense points into clusters according to time and space dimensions, thereby identifying stay points.

[0104] In the parameter setting, the spatial distance threshold is set to 50 meters within the same position. The time window is set to more than 5 minutes as an effective stay point. Calculate the geographic coordinates and time range of the center point of each cluster as the location and duration of the stay point.

[0105] Suppose a segment of mobile path data contains the following points:

[0106] P1: longitude = 116.397, latitude = 39.916, time = t1;

[0107] P2: longitude = 116.3972, latitude = 39.9161, time = t2;

[0108] P3: longitude = 116.3973, latitude = 39.9162, time = t3;

[0109] After clustering by DBSCAN, a stay point is identified:

[0110] Stay point location longitude = 116.3971, latitude = 39.9161

[0111] The residence time range is [t1, t3].

[0112] S503 matches the residence point position with the geofence and exhibition area to obtain the key node position.

[0113] A number of geofence areas are predefined, such as museum entrances, warehouse boundaries, exhibition halls, etc. Each geofence is represented by a polygon composed of a set of geographic coordinates.

[0114] The geographic coordinates of each residence point are matched with the geofence. If the residence point is located within a certain geofence, it is marked as the key node corresponding to the fence.

[0115] If the residence point is not within any predefined geofence, it is manually annotated according to its surrounding environment.

[0116] The timestamp, geographic coordinates, and belonging area of each key node are output.

[0117] Suppose the geofence of a certain museum is defined as follows:

[0118] Exhibition Hall A: [(116.397, 39.916), (116.398, 39.916), (116.398, 39.917), (116.397, 39.917)]

[0119] Warehouse B: [(116.400, 39.915), (116.401, 39.915), (116.401, 39.916), (116.400, 39.916)]

[0120] The position of a certain residence point is:

[0121] Longitude = 116.3971, Latitude = 39.9161

[0122] After matching, it is confirmed that the residence point belongs to Exhibition Hall A, and the final output is:

[0123] Key node information: Time = t2, Position = (116.3971, 39.9161), Area = Exhibition Hall A.

[0124] S303 generates a dynamic trajectory password using a chaotic encryption algorithm based on the node information.

[0125] The timestamp and geographic coordinates of the key nodes obtained from S302 are taken as input data. These information represent some important location points in the movement process of the artwork. Chaos encryption algorithm relies on the characteristics of chaotic system, which is extremely sensitive to initial conditions and can produce seemingly random but actually completely repeatable results. Commonly used chaotic systems include Logistic mapping, Chua circuit, etc. Some features in the node information (such as the specific part of the timestamp or geographic coordinates) are used as the initial value of the chaotic system. This step ensures that each generated dynamic trajectory password is unique.

[0126] According to the selected chaotic system iteration calculation, the node information is processed in turn through the system to obtain a series of chaotic sequences. Then, these chaotic sequences are converted into binary string form to finally form the dynamic trajectory password.

[0127] S304 generates an auxiliary verification code based on the environmental data collected by the environmental sensor, including temperature data and humidity data, and the dynamic trajectory password and the auxiliary verification code need to be matched at the same time during verification.

[0128] The specific steps include:

[0129] S601 collects environmental data of the artwork based on environmental sensors;

[0130] High-precision temperature and humidity sensors are installed in the environment where the artwork is stored or transported. These sensors should be able to monitor and record the temperature and humidity information of the location in real time. According to the actual needs, set the data collection frequency, for example, collect data every 5 minutes, to ensure that the changing trend of environmental conditions can be captured. The collected data is transmitted to the central server for processing through wireless network or other ways. To ensure the authenticity of the data, encryption transmission technology can be used to prevent data tampering.

[0131] Suppose an artwork is in the process of transportation, and the environmental sensor records temperature and humidity data every 5 minutes:

[0132] Time point t1: temperature = 22℃, humidity = 45%

[0133] Time point t2: temperature = 21.8℃, humidity = 46%.

[0134] S602 extracts environmental features based on environmental data, including average temperature and humidity change rate;

[0135] Calculate the average value of all temperature readings in the selected time period to get the average temperature in that period. Calculate the change in humidity between adjacent time points, then divide by the time interval to get the humidity change rate. This can help identify periods of high humidity fluctuations, which may indicate environmental instability.

[0136] For the two sets of data mentioned above, let's assume we are interested in the time period from t1 to t2:

[0137] Average temperature = (22 + 21.8) / 2 = 21.9 (22 + 21.8) / 2 = 21.9°C

[0138] Humidity change rate = (46 - 45) / 5 = 0.2% / minute

[0139] S603 combines the selected environmental features and then converts them into a fixed-length string through a hash function;

[0140] The calculated average temperature and humidity change rate are combined into a feature vector. For example, if the average temperature is 21.9°C and the humidity change rate is 0.2% / minute, the feature vector can be represented as [21.9, 0.2].

[0141] The feature vector is hashed using a secure hash algorithm (such as SHA-256) to convert it into a fixed-length string. This is done to ensure that even slight changes in input will result in significantly different outputs, thereby increasing the security of the verification code.

[0142] S604 encrypts the string using a symmetric encryption algorithm and obtains the secondary verification code.

[0143] A symmetric encryption algorithm (such as AES) is selected, and a key is determined. Symmetric encryption means that the same key is used for encryption and decryption, so it is necessary to ensure the secure management of the key. The hash string obtained in the previous step is encrypted using the selected key. The encrypted result is the secondary verification code. In the verification phase, both the dynamic trajectory password and the secondary verification code need to be provided. The system first regenerates the secondary verification code based on the current environmental data and compares it with the provided verification code. Only if both match is considered a successful verification.

[0144] S103 encrypts the digital fingerprint and the trajectory password together with the artwork ownership information and writes it into the tamper-proof distributed ledger of the blockchain, forming blockchain data and binding it with the artwork's corresponding QR code;

[0145] A unique identifier generated from the multi-modal feature data of the artwork (such as visible light images and infrared images). A dynamic password generated based on key node information in the artwork's movement path. Includes the owner's identity information (such as name or institution name), transaction records, current status, etc.

[0146] The prepared data is encrypted using a symmetric encryption algorithm (such as AES) or an asymmetric encryption algorithm (such as RSA). To ensure security, it is recommended to use an asymmetric encryption algorithm, where the public key is used for encryption and the private key is used for decryption.

[0147] If a symmetric encryption algorithm is used, a random key is generated and properly secured. If an asymmetric encryption algorithm is used, a pair of public and private keys is generated, with the public key being publicly distributed and the private key being secured by a trusted party (such as the artwork owner or regulatory authority). The data is encrypted using the selected encryption algorithm and key, resulting in an encrypted string.

[0148] S104 scans the artwork's attached QR code to retrieve blockchain data and compares it with the physical features and trajectory password to verify the artwork.

[0149] Each artwork will be attached with a unique QR code label, which is encrypted and contains the artwork's unique identifier on the blockchain (such as a digital fingerprint or transaction hash). The QR code usually uses anti-counterfeiting technology (such as microprinting, anti-copy coating, etc.) to prevent being counterfeited or replaced.

[0150] Use a special device that supports blockchain query function (such as a smartphone or professional scanner) to scan the artwork's attached QR code. During the scanning process, the device will parse the information in the QR code and send it to the blockchain network for data query. According to the identification information in the QR code, all relevant data of the artwork stored in the blockchain ledger are extracted. These data include but are not limited to:

[0151] Digital fingerprint: a unique identifier generated from the artwork's multi-modal features.

[0152] Dynamic trajectory password: a dynamic password generated based on the artwork's movement path.

[0153] Ownership information: records the history of ownership changes and the current status of the artwork.

[0154] Environmental data: environmental information such as temperature and humidity during the transportation or exhibition of the artwork.

[0155] Then use high-precision equipment to re-collect the multi-modal feature data of the artwork, such as visible light images and infrared images. These data reflect the current physical state of the artwork. The collected multi-modal feature data is processed according to the same algorithm process as when stored in the blockchain, generating a real-time digital fingerprint. This step ensures that the generated digital fingerprint is comparable to the original digital fingerprint stored in the blockchain.

[0156] The real-time generated digital fingerprint is compared with the digital fingerprint stored on the blockchain. If they are completely consistent, it means that the physical state of the artwork has not changed; if there is a difference, it may indicate that the artwork has undergone restoration, damage or forgery. If the digital fingerprint is found to be inconsistent, the system will trigger an alarm and prompt the user to further check the state of the artwork. At the same time, the system will record the results of this verification for subsequent tracking and analysis.

[0157] Through the GPS module or other positioning devices, the current geographical location information of the artwork is collected, and a new dynamic trajectory password is generated combined with its historical moving path. The newly generated dynamic trajectory password is compared with the historical trajectory password stored on the blockchain. If they are consistent, it means that the moving path of the artwork is as expected; if there is a deviation, it may indicate that the artwork has undergone unauthorized movement or tampering. If the trajectory password does not match, the system will further analyze the reason for the difference. For example, whether it is due to data loss caused by signal loss, or whether there is illegal transportation behavior. According to the analysis results, the system will generate a detailed report for the relevant personnel to refer to.

[0158] After completing the comparison of the digital fingerprint and the dynamic trajectory password, the system will comprehensively evaluate all the verification results. Only when both verifications pass, the artwork can be considered authentic and complete. After the verification is completed, the system will provide clear feedback information to the user. For example:

[0159] Verification passed: the identity of the artwork is real, no abnormalities found.

[0160] Partial anomaly: there are slight differences in some features or trajectories, which need to be further verified.

[0161] Verification failed: the artwork may have been forged, tampered with or moved illegally, and immediate action is required.

[0162] Second embodiment

[0163] The present application provides a blockchain-based artwork anti-counterfeiting and traceability system, which comprises an artwork fingerprint generation module, a trajectory password generation module, a blockchain generation module and a verification module; the artwork fingerprint generation module is used to obtain multi-modal feature data of the artwork and generate a unique digital fingerprint through a hash algorithm; the trajectory password generation module is used to continuously obtain moving trajectory images of the artwork and extract spatiotemporal features to generate a dynamic trajectory password; the blockchain generation module is used to encrypt the digital fingerprint and the trajectory password together with the ownership information of the artwork, and then write them into the tamper-proof distributed ledger of the blockchain, forming blockchain data and binding them with the corresponding two-dimensional code of the artwork; the verification module is used to scan the two-dimensional code attached to the artwork to retrieve the blockchain data and compare it with the physical features and trajectory password to verify the artwork.

[0164] In this embodiment, the artwork fingerprint generation module captures visible light images of the artwork using a high-resolution camera and scans the artwork surface using an infrared spectrometer to obtain infrared images. These multi-modal feature data can reveal the details of the artwork under different light conditions. The collected image data is unified in size and converted to grayscale values to form a standardized input format for subsequent processing. Deep learning models such as VGG16 are used to extract deep features from grayscale images, and these features are converted into unique digital fingerprints using a hashing algorithm. This digital fingerprint is a unique identity of the artwork.

[0165] The trajectory password generation module uses acceleration sensors and gyroscopes to record the position changes of the artwork in real-time during transportation or exhibition. In addition, GPS positioning technology is combined to update the geographical location information regularly, ensuring the accuracy and integrity of the trajectory data. The collected geographical coordinates and timestamps are analyzed to identify key nodes (such as stopping points), and dynamic trajectory passwords are generated based on the time and spatial distribution characteristics of these nodes. This password not only reflects the actual movement path of the artwork, but also adds an additional security layer.

[0166] The blockchain generation module combines the digital fingerprint and dynamic trajectory password generated by the above two modules with the ownership information of the artwork (such as the owner's name, transaction history, etc.), and uses advanced encryption technology to protect it before writing it into the distributed ledger of the blockchain. This ensures the data's immutability. Each artwork will be assigned a unique QR code that links to the relevant data on the blockchain. The purpose of this is to make it easy for anyone to access detailed information about the artwork by scanning the QR code.

[0167] The verification module allows users to quickly access relevant information stored on the blockchain by scanning the QR code attached to the artwork using a dedicated device when verification is required. The system will re-collect multi-modal feature data of the artwork in its current state and generate a new digital fingerprint. At the same time, a new trajectory password will also be generated based on the latest geographical location information. Then, the old and new data are compared to confirm the authenticity of the artwork and whether it has undergone unauthorized movement. According to the comparison result, the system will give the corresponding verification conclusion. If everything is normal, it means that the artwork is authentic and reliable; if there is any abnormality, further investigation is required.

[0168] The blockchain-based artwork anti-counterfeiting and traceability system integrates artwork fingerprint generation, trajectory password generation, blockchain writing, and verification, etc. It realizes the effective management of the whole life cycle of the artwork. It not only improves the transparency of the artwork market, but also greatly enhances the security protection capability of the artwork.

[0169] The above disclosed is only a preferred embodiment of the present application, of course, cannot be limited by this to limit the scope of the present application, the person skilled in the art can understand that the implementation of all or part of the above processes, and according to the equivalent changes of the claims of the present application, still belongs to the scope covered by the present application.

Claims

1. A blockchain-based anti-counterfeiting and traceability method for artworks, characterized in that, it comprises: acquiring multi-modal feature data of an artwork and generating a unique digital fingerprint through a hash algorithm; continuously acquiring moving trajectory images of the artwork and extracting spatiotemporal features to generate a dynamic trajectory password; encrypting the digital fingerprint and the trajectory password together with the artwork ownership information, and writing them into the tamper-proof distributed ledger of a blockchain to form blockchain data and bind them with a two-dimensional code corresponding to the artwork; scanning the two-dimensional code attached to the artwork to retrieve the blockchain data and compare them with the physical features and the trajectory password to verify the artwork; the specific steps of continuously acquiring moving trajectory images of the artwork and extracting spatiotemporal features to generate a dynamic trajectory password comprise: acquiring moving path data of the artwork through an acceleration sensor and a gyroscope; extracting the timestamp and geographic coordinates of the key nodes in the moving path data to obtain node information; generating a dynamic trajectory password based on the node information using a chaotic encryption algorithm; after generating the dynamic trajectory password based on the node information using a chaotic encryption algorithm, the steps further comprise: collecting environmental data of the artwork using an environmental sensor to generate an auxiliary verification code, wherein the environmental data includes temperature data and humidity data, and the dynamic trajectory password and the auxiliary verification code need to be matched simultaneously during verification. 2.The blockchain-based anti-counterfeiting and traceability method for artworks according to claim 1, characterized in that, the multi-modal feature data includes visible light images and infrared images. 3.The blockchain-based anti-counterfeiting and traceability method for artworks according to claim 2, characterized in that, the specific steps of acquiring multi-modal feature data of an artwork and generating a unique digital fingerprint through a hash algorithm comprise: collecting visible light images of the artwork and scanning the surface of the artwork using an infrared spectrometer to obtain infrared images; normalizing the visible light images and the infrared images to a uniform size and converting them into a grayscale value matrix; extracting deep features in the grayscale value matrix using a VGG16 network to obtain a total feature matrix; processing the total feature matrix using a hash function to obtain a digital fingerprint. 4.The blockchain-based anti-counterfeiting and traceability method for artworks according to claim 3, characterized in that, the specific steps of acquiring moving path data of the artwork through an acceleration sensor and a gyroscope comprise: recording geographic data every preset time interval through a positioning module; when the positioning module signal is lost, automatically switching to indoor positioning and acquiring corresponding geographic data; dividing the geographic data into time windows to obtain moving path data. 5.The blockchain-based anti-counterfeiting and traceability method for artworks according to claim 4, characterized in that, the specific steps of extracting the timestamp and geographic coordinates of the key nodes in the moving path data to obtain node information comprise: cleaning the moving path data, including filtering invalid data, deleting GPS drift points, and using a cubic spline interpolation method to complete the missing time period; identifying the location of the stop point through a spatiotemporal clustering algorithm; matching the location of the stop point with the geographic fence and the exhibition area to obtain the location of the key node.

6. The blockchain-based anti-counterfeiting and traceability method for artworks according to claim 5, characterized in that, the specific step of generating the auxiliary verification code based on the environmental data collected by the environmental sensor includes: collecting environmental data of the artworks by the environmental sensor; extracting environmental features based on the environmental data, the environmental features including average temperature and humidity change rate; combining the selected environmental features and then converting them into a fixed-length string through a hash function; encrypting the string using a symmetric encryption algorithm to obtain the auxiliary verification code.

7. A blockchain-based anti-counterfeiting and traceability system for artworks, which adopts the blockchain-based anti-counterfeiting and traceability method for artworks according to any one of claims 1-6. The system comprises an artwork fingerprint generation module, a trajectory password generation module, a blockchain generation module, and a verification module. The artwork fingerprint generation module is configured to obtain multi-modal feature data of the artwork and generate a unique digital fingerprint through a hash algorithm. The trajectory password generation module is configured to continuously obtain moving trajectory images of the artwork and extract spatiotemporal features to generate a dynamic trajectory password. The blockchain generation module is configured to encrypt the digital fingerprint and the trajectory password together with the ownership information of the artwork, and then write them into the tamper-proof distributed ledger of the blockchain to form blockchain data and bind them with the corresponding two-dimensional code of the artwork. The verification module is configured to scan the two-dimensional code attached to the artwork to retrieve the blockchain data and compare them with the physical features and the trajectory password to verify the artwork.

Images