Artwork copyright protection and management method based on NFT

Through the NFT-based art copyright protection and management method, blockchain and Rabin encryption algorithms are used to solve the problems of cumbersome copyright registration and easy information tampering in the existing technology, and the reliable copyright protection and transparent transaction process of artworks are realized. Improve management efficiency and security of user identity authentication.

CN118940229AInactive Publication Date: 2024-11-12QINGHAI NORMAL UNIV
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202410914936.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing copyright protection and management methods for artworks have problems such as cumbersome copyright registration process, easy tampering of copyright information, and opaque copyright transfer and authorization process.

Method used

Adopt NFT-based art copyright protection and management methods, use digital images of art resources, perform feature extraction and metadata generation, cast non-fungible token NFT, and store it on the blockchain, using Rabin The encryption algorithm encrypts and authenticates user identity, simplifies copyright management and transaction processes.

Benefits of technology

It provides reliable copyright protection evidence, enhances the visibility and influence of artworks, ensures the security of user transactions, and simplifies copyright management through smart contract automation, improving management efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940229A_ABST
    Figure CN118940229A_ABST
Patent Text Reader

Abstract

The invention discloses an NFT-based artwork copyright protection and management method, and the method comprises the following steps: obtaining artwork resources, and carrying out the digitalization of the artwork resources, and obtaining a digitalized image; performing feature extraction on the digital image to obtain a feature vector set of the digital image; performing artwork metadata generation and non-homogeneous NFT casting on the digital image to obtain a non-homogeneous token NFT of the artwork resource; according to the feature vector set and feature vector sets of other artwork resources, obtaining a similarity result of the artwork resources and the other artwork resources; if the similarity result is higher than a preset similarity threshold value, the process is ended, and otherwise, the step S6 is executed; the non-homogeneous token NFT and the copyright information are stored on the block chain; through a block chain platform, transaction and management are carried out on the non-homogeneous token NFT, and user identity encryption and authentication are carried out by using a Rabin encryption algorithm in the transaction process. According to the invention, the identification accuracy and high speed of the NFT and the security of the transaction can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of digital copyright management, and more particularly to a method for protecting and managing copyright of artworks based on NFT. Background Art

[0002] With the rise of digital art and the development of blockchain technology, non-fungible tokens have become an important means of art transactions and copyright protection. However, the current copyright protection and management methods for artworks still face many challenges, such as cumbersome copyright registration procedures, easy tampering of copyright information, and opaque copyright transfer and authorization processes. Summary of the invention

[0003] In response to the above-mentioned deficiencies in the prior art, the present invention provides an NFT-based artwork copyright protection and management method, which solves the security problems existing in non-fungible token transactions in the prior art.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] A method for copyright protection and management of artworks based on NFT, comprising the following steps:

[0006] S1. Acquire artwork resources, and digitize the artwork resources to obtain a digital image, where the digital image is an RGB image;

[0007] S2, extracting features from the digitized image to obtain a feature vector set of the digitized image;

[0008] S3. Generate artwork metadata and mint non-homogeneous NFTs for digital images to obtain non-homogeneous tokens of artwork resources;

[0009] S4. Obtain similarity results of the artwork resource and other artwork resources based on the feature vector set and the feature vector set of other artwork resources;

[0010] S5. If the similarity result is higher than the preset similarity threshold, the process ends, otherwise, step S6 is executed;

[0011] S6. Store non-fungible tokens NFT and copyright information on the blockchain;

[0012] S7. Non-fungible tokens (NFTs) are traded and managed through the blockchain platform, and the Rabin encryption algorithm is used to encrypt and authenticate user identities during the transaction process.

[0013] The beneficial effects of the above scheme are:

[0014] (1) The present invention utilizes the immutability of blockchain technology and converts artworks into NFTs. If a copyright dispute occurs, it can provide reliable copyright protection evidence for artwork resources.

[0015] (2) The present invention enhances the visibility and influence of art resources through the transaction and display of NFTs, thereby better promoting the dissemination of art resources.

[0016] (3) The present invention performs identity authentication on users during NFT transactions, providing a safer operating environment for user transactions;

[0017] (4) The present invention simplifies copyright management and transaction processes and improves management efficiency through automated smart contracts.

[0018] Further, step S2 specifically includes:

[0019] S21, graying the digitized image to obtain a grayed image of the artwork resource;

[0020] S22, creating a multi-scale pyramid based on the grayscale image;

[0021] S23, screening each pixel point in each layer of the image in the multi-scale pyramid to determine the feature points in the multi-scale pyramid;

[0022] S24, determining the direction and BRIEF descriptor of each feature point;

[0023] S25. Integrate the position, direction and BRIEF descriptor of the feature point to obtain a feature vector set.

[0024] The above further scheme has the beneficial effect that features can be extracted in different layers of the network to obtain multi-scale information. In the object detection task, the object may appear in the image at different sizes, so it is necessary to detect and locate it at different scales. By merging feature maps at different levels, objects of different sizes can be detected and recognized.

[0025] Further, step S4 specifically includes:

[0026] S41, obtaining the Hamming distance between the BRIEF descriptor in the feature vector set and the BRIEF descriptor in the feature vector set of other artwork resources;

[0027] S42. Determine the Hamming distance as the similarity result.

[0028] Furthermore, step S5 specifically includes:

[0029] S51, setting a Hamming distance threshold, and determining the Hamming distance threshold as a preset similarity threshold;

[0030] S52, when the Hamming distance is less than the Hamming distance threshold, determine that the similarity result is higher than the preset similarity threshold, and end the process; when the Hamming distance is greater than the Hamming distance threshold, determine that the similarity result is less than the preset similarity threshold, and execute step S6.

[0031] The beneficial effects of the above further scheme are: the Hamming distance is applicable to binary data, is simpler and more intuitive, and is more robust to noise and missing data.

[0032] Furthermore, in step S7, the Rabin encryption algorithm is used to encrypt the user identity during the transaction, specifically including:

[0033] S71. Determine a public key and a private key according to two large prime numbers selected by the user;

[0034] S72, performing hash processing on the user's identity information to obtain an identity code value, where the identity code value is an integer;

[0035] S73, using the public key to encrypt the identity code value to obtain an encrypted ciphertext;

[0036] S74. Associate the encrypted ciphertext with the non-fungible token NFT.

[0037] Furthermore, in step S7, the Rabin encryption algorithm is used to perform user identity authentication during the transaction, specifically including:

[0038] S75, solving the decryption equation to obtain a plaintext set;

[0039] S76. Determine, according to the plaintext set, a hash value of each plaintext data in the plaintext set;

[0040] S76, uploading the hash value of each plaintext data in the plaintext set to the authentication server, and comparing the hash value of each plaintext data in the plaintext set with the hash value of the user's identity information through the authentication server;

[0041] S77. If the hash value of all plaintext data in the plaintext set does not match the hash value of the user's identity information, the authentication fails; if the hash value of any plaintext data in the plaintext set matches the hash value of the user's identity information, the authentication succeeds, and the result of the successful authentication is recorded in the blockchain.

[0042] Furthermore, in step S7, during the transaction of non-fungible tokens NFT, the identity authentication process of both parties to the transaction specifically includes:

[0043] S78. The seller signs the transaction information using the private key to obtain a digital signature;

[0044] S79. The purchaser decrypts the digital signature according to the public key and the encryption index to obtain a decryption result;

[0045] S710. If the decryption result is consistent with the hash value of the message sent by the seller, the authentication is successful and the transaction continues; otherwise, the authentication fails and the transaction is terminated.

[0046] The beneficial effect of the above further scheme is that the use of public keys and private keys can provide a secure, private and reliable communication and transaction environment, protecting data from being accessed and tampered with by unauthorized parties.

[0047] Furthermore, in step S23, each pixel point in each layer of the multi-scale pyramid is screened, specifically including:

[0048] S231, determining each pixel point in each layer of the image in the multi-scale pyramid as a candidate key point;

[0049] S232, setting a brightness threshold;

[0050] S233, when the pixel values ​​of the pixels in the 3*3 neighborhood of the candidate key point meet the preset brightness relationship, it is determined that the candidate key point passes the brightness test, otherwise, the candidate key point is eliminated, and the preset brightness relationship is:

[0051] P+L≤M i ≤PL

[0052] Among them, P is the pixel value of the candidate key point, T is the brightness threshold, and M i is the pixel value of the i-th pixel in the 3*3 neighborhood of the candidate key point, i=1,2…48;

[0053] S234, performing a first screening on all candidate key points that pass the brightness test using non-maximum suppression to obtain a first candidate key point;

[0054] S235 , performing a second screening on the first candidate key points to remove nearly duplicate first candidate key points.

[0055] Furthermore, in step S24, the direction of each feature point is determined, which specifically includes:

[0056] S241, for each feature point, using the moment to determine the main direction of the feature point;

[0057] S242, determining the centroid within the preset radius of the feature point according to the pixel points and the main direction within the preset radius of the feature point;

[0058] S243, substituting the coordinate position of the feature point and the coordinate position of the center of mass into a preset direction calculation formula to determine the direction. The preset direction θ calculation formula is:

[0059]

[0060] in, I(x,y) is the grayscale value of the pixel at coordinate position (x,y), p,q={0,1}.

[0061] Further, in step S24, the BRIEF descriptor of each feature point is determined, specifically including:

[0062] S244, determining an initial point pair set according to the pixel points in the neighborhood of the feature point;

[0063] S245, rotating the initial point pair set according to the direction to obtain a point pair set;

[0064] S246, comparing the sizes of the point pair data in the point pair set to determine an initial binary matrix;

[0065] S247, sorting the column vectors of the initial binary matrix to obtain a binary matrix;

[0066] S248. Determine the BRIEF descriptor according to the binary matrix.

[0067] The beneficial effect of the above further scheme is that the use of blockchain for NFT transactions and management can ensure the permanent existence of assets, while providing a safe and transparent trading environment for NFT. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a flowchart of an NFT-based copyright protection and management method for artworks. DETAILED DESCRIPTION

[0069] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0070] like Figure 1 As shown, a method for copyright protection and management of artworks based on NFT includes the following steps:

[0071] S1. Acquire artwork resources, and digitize the artwork resources to obtain a digital image, where the digital image is an RGB image.

[0072] Optionally, in order to ensure the copyright of the NFT and continuously update the iterative database, the NFT needs to be feature extracted before the new NFT is registered and confirmed; to complete the entire NFT registration, confirmation and transaction process, the owner of the NFT needs to be modified, and the timestamp of each time point of the NFT transaction needs to be written. After the transaction is completed, the relevant transaction information is generated and recorded on the blockchain to prevent malicious users from tampering with the information.

[0073] S2. Extract features from the digitized image to obtain a set of feature vectors of the digitized image.

[0074] In this embodiment, step S2 specifically includes:

[0075] S21, graying the digitized image to obtain a grayed image of the artwork resource;

[0076] S22, creating a multi-scale pyramid based on the grayscale image;

[0077] S23, screening each pixel point in each layer of the image in the multi-scale pyramid to determine the feature points in the multi-scale pyramid;

[0078] S24, determining the direction and BRIEF descriptor of each feature point;

[0079] S25. Integrate the position, direction and BRIEF descriptor of the feature point to obtain a feature vector set.

[0080] In this embodiment, in step S23, each pixel point in each layer of the multi-scale pyramid is screened, specifically including:

[0081] S231, determining each pixel point in each layer of the image in the multi-scale pyramid as a candidate key point;

[0082] S232, setting a brightness threshold;

[0083] S233, when the pixel values ​​of the pixels in the 3*3 neighborhood of the candidate key point meet the preset brightness relationship, it is determined that the candidate key point passes the brightness test, otherwise, the candidate key point is eliminated, and the preset brightness relationship is:

[0084] P+L≤M i ≤PL

[0085] Among them, P is the pixel value of the candidate key point, T is the brightness threshold, and M i is the pixel value of the i-th pixel in the 3*3 neighborhood of the candidate key point, i=1,2…48;

[0086] S234, performing a first screening on all candidate key points that pass the brightness test using non-maximum suppression to obtain a first candidate key point;

[0087] S235 , performing a second screening on the first candidate key points to remove nearly duplicate first candidate key points.

[0088] In this embodiment, in step S24, determining the direction of each feature point specifically includes:

[0089] S241, for each feature point, using the moment to determine the main direction of the feature point;

[0090] S242, determining the centroid within the preset radius of the feature point according to the pixel points and the main direction within the preset radius of the feature point;

[0091] S243, substituting the coordinate position of the feature point and the coordinate position of the center of mass into a preset direction calculation formula to determine the direction. The preset direction θ calculation formula is:

[0092]

[0093] in, I(x,y) is the grayscale value of the pixel at coordinate position (x,y), p,q={0,1}.

[0094] In this embodiment, in step S24, determining the BRIEF descriptor of each feature point specifically includes:

[0095] S244, determining an initial point pair set according to the pixel points in the neighborhood of the feature point;

[0096] S245, rotating the initial point pair set according to the direction to obtain a point pair set;

[0097] S246, comparing the sizes of the point pair data in the point pair set to determine an initial binary matrix;

[0098] S247, sorting the column vectors of the initial binary matrix to obtain a binary matrix;

[0099] S248. Determine the BRIEF descriptor according to the binary matrix.

[0100] Exemplarily, to extract features from a digitized image, the digitized image of the artwork resource may be read first. For example, a color digitized image is converted into a grayscale image by a linear weighting method, because feature detection is usually performed in a grayscale space to reduce the amount of calculation, including the following steps: reading the values ​​of the red, green, and blue channels of each pixel in the digitized image, R, G, and B respectively; a set of weighting coefficients obtained according to the brightness perceived by the human eye are W respectively. R , W G , W B , where W R =0.299,W G =0.587,W B =0.114; the gray value Gray of each pixel after conversion is obtained by the linear weighted formula, and the linear weighted formula is:

[0101] Gray=W R ×R+W G×G+W B ×B

[0102] Finally, the calculated grayscale value is assigned to the corresponding pixel in the grayscale image.

[0103] Exemplarily, a digitized image is continuously scaled according to a certain ratio to create a multi-scale pyramid of the digitized image (ie, NFT image) so as to detect key points at different scales and enhance scale invariance.

[0104] For example, for each pixel p in each layer of the NFT image in the multi-scale pyramid, it is taken as a candidate key point. A circular neighborhood centered on p is determined, which consists of N = 48 pixels and corresponds to a radius of r = 3 pixels; a brightness threshold T is determined, and for each candidate key point p, the pixels in the surrounding neighborhood are checked. If there are at least 3 or continuous pixels whose brightness values ​​are greater than the brightness value of p plus the threshold T, or less than the brightness value of p minus the threshold T, then p can pass the brightness test. Among the candidate key points that pass the brightness test, non-maximum suppression is performed, and only when the brightness value of the pixel is a local maximum, the secondary candidate key point is retained as a first candidate key point. The first candidate key point is detected twice to achieve more accurate positioning to improve the stability of feature matching, while removing nearly duplicate key points to ensure that each key point is unique.

[0105] For example, for each key point (i.e., feature point), the main direction is calculated based on the surrounding sampling points, and the direction with the largest change in grayscale intensity is usually selected. The moment m can be used pq To determine the direction of the feature point; I(x,y) is the gray value of the pixel at the coordinate position (x,y), p,q = {0,1}; calculate the centroid C of the feature point with r as the radius; the calculation formula of C is: The coordinates of the feature point to the centroid form a vector as the direction θ of the feature point:

[0106] For example, randomly select a set of pixel pairs around the key point (i.e., feature point) to randomly select sampling points, such as (x1, y1), (x2, y2); for each pair of sampling points, calculate their gray intensity ratio to generate a binary string, if (x1, y1) < (x2, y2), the binary feature is 0, otherwise it is 1; combine all binary strings to form a descriptor that can describe the local features around the key point of the NFT. For example:

[0107] In order to enhance the rotation invariance of feature description, the calculated key points include the direction angle of the feature points; the original BRIEF algorithm selects n pairs of point pairs D in the neighborhood of the key points. Rotate D by the rotation angle θ to get a new point pair set D θ , D θ =R θ D, compare the size of the point pairs at the new point pair set position to form a binary string descriptor. In order to enhance the distinguishability, point pairs are selected in each S = 31 × S = 31 neighborhood of 300k feature points in M ​​ways; the size of the point pairs is compared to form a 300k*M binary matrix Q; each column of the binary matrix Q represents the binary number obtained by a certain method of selecting 300k points; the average value of each column of the binary matrix Q is calculated, and the column vectors of the Q matrix are reordered according to the distance from the average value to 0.5 to form a matrix T; the first column vector of the matrix T is placed in the matrix R; the next column vector of T is taken to calculate the correlation with all the column vectors in the matrix R. If the correlation coefficient is less than the set threshold, the column vector in the matrix T is moved to the matrix R; then the next column vector is calculated to calculate the correlation with all the column vectors in the matrix R. If the correlation coefficient is less than the set threshold, the column vector in the matrix T is moved to the matrix R. The operation is repeated until the number of vectors in the matrix R is 256; thus, the BRIEF descriptor is obtained.

[0108] The position (x, y), direction and BRIEF descriptor of the feature point are integrated into a complete feature vector, and a feature vector list (i.e., a feature vector set) is created for all key points in the NFT image, each of which contains all relevant information of the key point.

[0109] S3. Generate artwork metadata and mint non-homogeneous NFTs for digital images to obtain non-homogeneous token NFTs of artwork resources.

[0110] For example, an artist may create metadata for a digitized artwork that includes copyright information, creator information, distribution information, and the like.

[0111] For example, blockchain technology is used to combine metadata with digitized artworks to create a unique non-fungible token NFT.

[0112] S4. Obtain similarity results between the artwork resource and other artwork resources based on the feature vector set and the feature vector set of other artwork resources.

[0113] In this embodiment, step S4 specifically includes:

[0114] S41, obtaining the Hamming distance between the BRIEF descriptor in the feature vector set and the BRIEF descriptor in the feature vector set of other artwork resources;

[0115] S42. Determine the Hamming distance as the similarity result.

[0116] S5. If the similarity result is higher than the preset similarity threshold, the process ends; otherwise, step S6 is executed.

[0117] In this embodiment, step S5 specifically includes:

[0118] S51, setting a Hamming distance threshold, and determining the Hamming distance threshold as a preset similarity threshold;

[0119] S52, when the Hamming distance is less than the Hamming distance threshold, determine that the similarity result is higher than the preset similarity threshold, and end the process; when the Hamming distance is greater than the Hamming distance threshold, determine that the similarity result is less than the preset similarity threshold, and execute step S6.

[0120] S6. Store non-fungible tokens NFT and copyright information on the blockchain.

[0121] For example, a Hamming distance threshold T is set S , only when the Hamming distance between the descriptors is less than T S They are considered matched only when . The Hamming distance is used to evaluate the similarity between feature vectors. The smaller the Hamming distance, the higher the possibility of feature matching and the higher the similarity. If the similarity passes, the detailed information and copyright information of the NFT will be registered and stored on the blockchain to ensure its immutability and transparency.

[0122] S7. Non-fungible tokens (NFTs) are traded and managed through the blockchain platform, and the Rabin encryption algorithm is used to encrypt and authenticate user identities during the transaction process.

[0123] In this embodiment, in step S7, the Rabin encryption algorithm is used to encrypt the user identity during the transaction process, specifically including:

[0124] S71. Determine a public key and a private key according to two large prime numbers selected by the user;

[0125] S72, performing hash processing on the user's identity information to obtain an identity code value, where the identity code value is an integer;

[0126] S73, using the public key to encrypt the identity code value to obtain an encrypted ciphertext;

[0127] S74. Associate the encrypted ciphertext with the non-fungible token NFT.

[0128] In this embodiment, in step S7, the Rabin encryption algorithm is used to perform user identity authentication during the transaction process, specifically including:

[0129] S75, solving the decryption equation to obtain a plaintext set;

[0130] S76. Determine, according to the plaintext set, a hash value of each plaintext data in the plaintext set;

[0131] S76, uploading the hash value of each plaintext data in the plaintext set to the authentication server, and comparing the hash value of each plaintext data in the plaintext set with the hash value of the user's identity information through the authentication server;

[0132] S77. If the hash value of all plaintext data in the plaintext set does not match the hash value of the user's identity information, the authentication fails; if the hash value of any plaintext data in the plaintext set matches the hash value of the user's identity information, the authentication succeeds, and the result of the successful authentication is recorded in the blockchain.

[0133] In this embodiment, in step S7, during the transaction of non-fungible tokens NFT, the identity authentication process of both parties to the transaction specifically includes:

[0134] S78. The seller signs the transaction information using the private key to obtain a digital signature;

[0135] S79. The purchaser decrypts the digital signature according to the public key and the encryption index to obtain a decryption result;

[0136] S710. If the decryption result is consistent with the hash value of the message sent by the seller, the authentication is successful and the transaction continues; otherwise, the authentication fails and the transaction is terminated.

[0137] For example, the user can randomly select two large prime numbers: p and q, where p = q = 3 (mod) 4, and calculate the public key n = p × q, where the private key consists of the prime numbers p and q; the public key is published, and the private key is kept confidential by the user. The user's identity information is hashed using the SHA-256 hash algorithm to obtain an identity encoding value; this algorithm takes the received string as input and generates a fixed-size hash value h, which is mapped to an integer m; the user can use the public key n to encrypt his or her identity information encoding value m to obtain an encrypted ciphertext; encryption is performed by calculating c = m 2 modn, where c is the encrypted ciphertext. The encrypted ciphertext c of the encrypted identity information is stored on the blockchain and associated with the user's NFT assets.

[0138] For example, when a transaction requires user identity verification, decryption is performed to find all possible plaintext sets x i, solve the decryption equation x 2 ≡c(modn); The authentication server needs to determine which plaintext is the valid user identity information. The authentication server will calculate the hash value of each plaintext in the received plaintext set and compare it with the hash value of the user's claimed identity information; after determining the correct plaintext, the authentication server confirms the user's identity and authorizes the user to conduct transactions or other operations; the authentication results are recorded on the blockchain to ensure the transparency and immutability of the identity authentication process.

[0139] Exemplarily, the user (i.e., the seller) can use his own private key to sign the information copyright statement or transaction information to obtain a digital signature; and the encryption exponent e=2 can be selected because this can simplify the calculation process; the process of using the private key (p and q) to generate a signature is: calculate the digital signature group (r and s);

[0140] r=m e modn

[0141] s=(m -1 modn)×(rh×r)modn

[0142] Among them, m -1 is the multiplicative inverse of m modulo n.

[0143] Digital signatures (r and s) can be attached to messages by users or stored on the blockchain in association with transactions.

[0144] When the buyer needs to verify the seller's identity, he can use the public key (n and e) to calculate v, v = (s × r) e modn. If v is equal to the hash value h of the message sent by the seller, the signature is valid, which means that the message is indeed from the claimed user and has not been tampered with during transmission; the logic of identity encryption authentication is integrated into the smart contract to automatically perform identity authentication and copyright protection; when a transaction occurs, the smart contract automatically triggers the identity authentication process to ensure the legitimacy of the identities of both parties to the transaction.

[0145] The image feature extraction algorithm in the present invention adopts the ORB feature extraction algorithm, which has high accuracy and high speed for the recognition of NFT images; in addition, the Rabin encryption algorithm is used to authenticate the user identity, which has extremely high authentication efficiency; finally, the use of blockchain for NFT transactions and management can ensure the permanent existence of assets, while providing a safe and transparent trading environment for NFT.

[0146] Those skilled in the art will appreciate that the embodiments herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the invention.

Claims

1. A method for copyright protection and management of artworks based on NFT, It is characterized in that The method comprises: S1. Acquire artwork resources, and digitize the artwork resources to obtain a digitized image, where the digitized image is an RGB image; S2, extracting features from the digitized image to obtain a feature vector set of the digitized image; S3. Generate artwork metadata and mint non-homogeneous NFT for the digitized image to obtain a non-homogeneous token NFT for the artwork resource; S4. Obtaining similarity results between the artwork resource and the other artwork resources based on the feature vector set and the feature vector set of the other artwork resources; S5. If the similarity result is higher than the preset similarity threshold, the process ends, otherwise, step S6 is executed; S6. Storing the non-fungible token NFT and copyright information on the blockchain; S7. The non-fungible token NFT is traded and managed through the blockchain platform, and the Rabin encryption algorithm is used to encrypt and authenticate the user identity during the transaction.

2. The method according to claim 1, It is characterized in that The step S2 specifically includes: S21, graying the digitized image to obtain a grayed image of the artwork resource; S22, creating a multi-scale pyramid according to the grayscale image; S23, screening each pixel point in each layer of the image in the multi-scale pyramid to determine the feature points in the multi-scale pyramid; S24, determining the direction and BRIEF descriptor of each feature point; S25, integrating the position of the feature point, the direction and the BRIEF descriptor to obtain the feature vector set.

3. The method according to claim 2, It is characterized in that The step S4 specifically includes: S41, obtaining the Hamming distance between the BRIEF descriptor in the feature vector set and the BRIEF descriptor in the feature vector set of the other artwork resource; S42: Determine the Hamming distance as the similarity result.

4. The method according to claim 3, It is characterized in that The step S5 specifically includes: S51, setting a Hamming distance threshold, and determining the Hamming distance threshold as the preset similarity threshold; S52. When the Hamming distance is less than the Hamming distance threshold, determine that the similarity result is higher than the preset similarity threshold, and end the process; when the Hamming distance is greater than the Hamming distance threshold, determine that the similarity result is less than the preset similarity threshold, and execute step S6.

5. The method according to claim 4, It is characterized in that In step S7, the Rabin encryption algorithm is used to encrypt the user identity during the transaction, specifically including: S71. Determine a public key and a private key according to two large prime numbers selected by the user; S72, performing hash processing on the user's identity information to obtain an identity code value, where the identity code value is an integer; S73, using the public key to encrypt the identity code value to obtain an encrypted ciphertext; S74. Associate the encrypted ciphertext with the non-fungible token NFT.

6. The method according to claim 5, It is characterized in that In step S7, the Rabin encryption algorithm is used to perform user identity authentication during the transaction, specifically including: S75, solving the decryption equation to obtain a plaintext set; S76. Determine, according to the plaintext set, a hash value of each plaintext data in the plaintext set; S76, uploading the hash value of each plaintext data in the plaintext set to the authentication server, and comparing the hash value of each plaintext data in the plaintext set with the hash value of the user's identity information through the authentication server; S77. If the hash value of all plaintext data in the plaintext set and the hash value of the user's identity information do not match, the authentication fails; if the hash value of any plaintext data in the plaintext set matches the hash value of the user's identity information, the authentication succeeds, and the result of the successful authentication is recorded in the blockchain.

7. The method according to claim 6, It is characterized in that In step S7, during the transaction of the non-fungible token NFT, the identity authentication process of the two parties to the transaction specifically includes: S78. The seller signs the transaction information using the private key to obtain a digital signature. S79, the purchaser decrypts the digital signature according to the public key and the encryption index to obtain a decryption result; S710. If the decryption result is consistent with the hash value of the message sent by the seller, the authentication is successful and the transaction continues; otherwise, the authentication fails and the transaction is terminated.

8. The method according to claim 2, It is characterized in that In step S23, screening each pixel point in each layer of the multi-scale pyramid image specifically includes: S231, determining each pixel point in each layer of the image in the multi-scale pyramid as a candidate key point; S232, setting a brightness threshold; S233: When the pixel values ​​of the pixels in the 3*3 neighborhood of the candidate key point satisfy the preset brightness relationship, it is determined that the candidate key point passes the brightness test; otherwise, the candidate key point is eliminated. The preset brightness relationship is: P+L≤M i ≤P-L Among them, P is the pixel value of the candidate key point, T is the brightness threshold, and M i is the pixel value of the i-th pixel in the 3*3 neighborhood of the candidate key point, i=1,2…48; S234, performing a first screening on all candidate key points that pass the brightness test using non-maximum suppression to obtain a first candidate key point; S235: Perform a second screening on the first candidate key points to remove nearly duplicate first candidate key points.

9. The method according to claim 2, It is characterized in that In step S24, determining the direction of each feature point specifically includes: S241, for each of the feature points, using a moment to determine the main direction of the feature point; S242, determining the centroid within the preset radius of the feature point according to the pixel points and the main direction within the preset radius of the feature point; S243, substituting the coordinate position of the feature point and the coordinate position of the centroid into a preset direction calculation formula to determine the direction, the preset direction θ calculation formula is: in, I(x,y) is the grayscale value of the pixel at coordinate position (x,y), p,q={0,1}.

10. The method according to claim 9, It is characterized in that In step S24, determining the BRIEF descriptor of each feature point specifically includes: S244, determining an initial point pair set according to the pixel points in the neighborhood of the feature point; S245, rotating the initial point pair set according to the direction to obtain a point pair set; S246, comparing the sizes of the point pair data in the point pair set to determine an initial binary matrix; S247, sorting the column vectors of the initial binary matrix to obtain a binary matrix; S248. Determine the BRIEF descriptor according to the binary matrix.

Citation Information

Patent Citations

  • Method for uniformly extracting feature points from ORB

    CN111160371A

  • Digital asset right confirmation and transaction method based on NFR

    CN116342284A

  • Optimization method and device of ORB feature matching algorithm and computer equipment

    CN117197712A

  • Block chain digital signature method and system based on asymmetric encryption

    CN117201113A

  • Jianzhu glaze similarity judgment method based on texture features

    CN117237673A