Video copyright protection method and system based on blockchain and digital watermark technology

Abstract

The present invention discloses a video copyright protection method and system based on blockchain and digital watermark technology, the method comprising the following steps: storing the robust watermark uploaded by the user into the system database, and obtaining the watermark hash value through a hash algorithm; uploading the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user to the blockchain and generating a block; generating a fragile watermark according to the uploaded robust watermark, and embedding the robust watermark and the fragile watermark into the video file to be registered respectively. The present invention forms a new template for copyright protection by combining blockchain with digital watermark technology, and makes the watermark more closely related to the owner of the work and more traceable. The dual watermark technology of robust watermark and fragile watermark is also used to realize the confirmation of copyright and the location of tampering, and the watermark effect is better.

Description

Technical Field

[0001] The present invention belongs to the field of copyright protection, and in particular relates to a video copyright protection method and system based on blockchain and digital watermark technology. Background Art

[0002] As a part of intellectual property, video copyright has been developing rapidly on the one hand, but infringement is hard to stop on the other. The frequent infringement of video copyright is also related to the characteristics of the video itself, such as easy modification and difficulty in distinguishing the authenticity. Users can easily obtain video files, and can also perform illegal and infringing operations such as interception, copying, tampering and redistribution. Pirated video resources can be found everywhere on the Internet, and the authenticity of video information is difficult to distinguish, and there are many copyright disputes. Research on video copyright protection is also achieving innovation in technology and models.

[0003] At present, blockchain-based digital copyright protection has achieved some results in academia. However, the following problems are common:

[0004] First, the robustness of the video watermark algorithm is poor. The carrier of the video watermark algorithm is video, and the attack methods of video include but are not limited to the attack methods of image. In addition to the attack on the image, the attack on a video also includes frame averaging, frame deletion, frame reorganization and other attacks. Therefore, for video watermark technology, it is required to be able to resist the above attacks. In addition, the robustness against geometric attacks, compression and other attack methods needs to be improved.

[0005] Second, the video watermarking scheme is poorly balanced. The video single watermarking algorithm constrains each other in terms of robustness and fragility. Robust watermarking ensures that when the video is attacked, the watermark information confirming the copyright can still be extracted. Fragile watermarking can locate and quantify tampering when it is detected. Single watermarking systems often cannot take into account both robustness and fragility.

[0006] The third is to use a centralized database. The traditional digital copyright protection system uses a centralized central database, so the security of the data is easily threatened. Nowadays, with the rapid development of the Internet, copyright protection is becoming more and more important, and there are more and more copyright protection issues. The central database cannot meet the growing copyright protection needs of the Internet.

[0007] Fourth, the watermark has a poor correlation with its owner. In order to protect their intellectual property rights, individuals or groups often embed relevant information with certainty and confidentiality, namely watermarks, into the resources they want to protect. However, when the watermark itself is not notarized by a trusted third-party organization, the relationship between the watermark and the individual or group cannot be guaranteed, which leads to a poor correlation between the watermark and the owner. Summary of the invention

[0008] In view of the above problems, the technical solution adopted by the present invention is: a video copyright protection method based on blockchain and digital watermark technology, comprising the following steps:

[0009] The robust watermark uploaded by the user is stored in the system database, and the watermark hash value is obtained through the hash algorithm;

[0010] The watermark hash value and the user ID of the logged-in user are encrypted and uploaded to the blockchain to generate a block;

[0011] A fragile watermark is generated according to the uploaded robust watermark, and the robust watermark and the fragile watermark are respectively embedded into the video file to be registered.

[0012] Optionally, before the step of uploading the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user to the blockchain and generating a block, identity authentication is also included, specifically:

[0013] Use SM3 to perform hash operation on the robust watermark and user ID to obtain the corresponding summary, and use the user ID summary value as the private key, and use SM2 to generate the corresponding public key; encrypt the robust watermark summary value and user ID with the private key; use the public key to decrypt before uploading to the chain. If the decrypted user ID matches the user identity, the authentication is passed, and the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user is uploaded to the blockchain and a block is generated.

[0014] Optionally, before executing the step of storing the robust watermark uploaded by the user into the system database, the following steps are also included:

[0015] The uploaded video file to be registered and the robust watermark are respectively checked for similarity with the video file and the user watermark in the system database;

[0016] If the review is passed, the robust watermark uploaded by the user will be stored in the system database; if the video file fails the review, the operation will be stopped; if the robust watermark fails the review, the user will be required to re-upload a new robust watermark until the review is passed.

[0017] Optionally, the method further comprises the following steps:

[0018] Perform scale-invariant feature transformation on the video file to be verified;

[0019] Extract the watermark from the video file to be verified to obtain the extracted watermark;

[0020] Determine whether the watermark hash value and user ID corresponding to the user watermark can be obtained from the blockchain according to the user ID of the logged-in user. If so, the watermark owner is authenticated successfully.

[0021] The extracted watermark is compared with the user watermark in the system database corresponding to the user ID by the normalized correlation coefficient. If it passes the threshold, the video owner is authenticated successfully.

[0022] Optionally, the method further comprises the following steps:

[0023] Users upload relevant certification documents and conduct copyright complaint analysis on the video files to be verified;

[0024] If the copyright appeal is successful, the video copyright will be transferred to the complaining user, the transaction information will be uploaded to the blockchain, and the data in the system database will be updated.

[0025] Optionally, in the step of embedding the robust watermark and the fragile watermark into the video file to be registered respectively, specifically:

[0026] Perform Hamming code encoding and scrambling on the robust watermark;

[0027] Extract video key frames of the video file to be registered;

[0028] Embed the fragile watermark in the lowest bit of the B component of the video key frame to obtain the B component containing the fragile watermark;

[0029] The G component of the key frame of the video is subjected to secondary Contourlet transform, the low-frequency subband is subjected to QR decomposition, and the R matrix is ​​selected as the embedding carrier to embed the robust watermark. Then, QR inverse decomposition and Contourlet inverse transform are performed to obtain the G component containing the robust watermark.

[0030] And, a video copyright protection system based on blockchain and digital watermark technology, comprising:

[0031] User management module, used for user registration and login, as well as user information management, user video file management and user watermark management;

[0032] The copyright registration module is used to generate a fragile watermark based on the uploaded robust watermark, and embed the robust watermark and the fragile watermark into the video file to be registered respectively;

[0033] The blockchain management module is used to upload the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user to the blockchain and generate blocks.

[0034] Optionally, the copyright registration module is also used for copyright review, specifically:

[0035] The uploaded video file to be registered and the robust watermark are respectively compared with the video file and the user watermark in the system database using the perceptual hash algorithm for similarity review;

[0036] If the review is passed, the robust watermark uploaded by the user will be stored in the system database; if the video file fails the review, the operation will be stopped; if the robust watermark fails the review, the user will be required to re-upload a new robust watermark until the review is passed.

[0037] Optionally, the system further comprises:

[0038] The copyright authentication module is used to extract the watermark in the video file to be verified after performing scale-invariant feature transformation on the video file to obtain the extracted watermark; and determine whether the watermark hash value and user ID corresponding to the user watermark can be obtained from the blockchain according to the user ID of the logged-in user. If so, the watermark owner is successfully authenticated; it is also used to compare the normalized correlation coefficient between the extracted watermark and the user watermark in the system database corresponding to the user ID. If the threshold is passed, the video owner is successfully authenticated.

[0039] Optionally, the system further comprises:

[0040] The copyright appeal module is used to conduct copyright appeal analysis based on the relevant supporting documents uploaded by the user when the user files a copyright appeal request.

[0041] The copyright appeal module is also used to transfer the video copyright to the complaining user when the copyright appeal is successful, upload the transaction information to the blockchain, and update the data in the system database.

[0042] The present invention adopts the above technical solution, so it has the following beneficial effects: a new template for copyright protection is formed by combining blockchain with digital watermark technology, and the watermark is more closely associated with the owner of the work, and the traceability is better. The dual watermark technology of robust watermark and fragile watermark is also adopted to realize the confirmation of copyright and the location of tampering, and the watermark effect is better.

[0043] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A flowchart of a video copyright protection method according to an embodiment of the present invention is shown;

[0046] Figure 2 A flow chart of embedding watermark according to an embodiment of the present invention is shown;

[0047] Figure 3 A flow chart of extracting watermarks according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The video copyright protection system based on blockchain and digital watermarking technology in the embodiment of the present invention includes a user management module, a copyright registration module, a blockchain management module, a copyright authentication module, and a copyright complaint module. Specifically,

[0050] User management module: used for user registration and login, as well as user information management, user video file management and user watermark management.

[0051] Copyright registration module: used to generate a fragile watermark according to the uploaded robust watermark, and embed the robust watermark and the fragile watermark into the video file to be registered respectively; the copyright registration module is also used for copyright review, specifically: the uploaded video file to be registered and the robust watermark are respectively reviewed for similarity with the video file and the user watermark in the system database. If the review is passed, the robust watermark uploaded by the user is stored in the system database. If the video file fails to pass the review, the operation is stopped; if the robust watermark fails to pass the review, the user is required to re-upload a new robust watermark until the review is passed; the copyright registration module is also used for identity authentication, and the identity authentication is specifically: using SM3 to perform hash operations on the robust watermark and user ID to obtain the corresponding summary, and using the user ID summary value as the private key, and using SM2 to generate the corresponding public key; encrypting the robust watermark summary value and the user ID using the private key; using the public key to decrypt before uploading to the chain, if the decrypted user ID is consistent with the user identity, the authentication is passed, and the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user is uploaded to the blockchain and a block is generated.

[0052] Blockchain management module: This module mainly includes functions such as node configuration, channel creation, contract deployment, and copyright information chaining. This system uses Fabric to build a blockchain module. For different node configuration information, after the node configuration is completed, a channel is jointly built. Write the information to be chained (the ciphertext generated by encrypting the watermark summary value and the user ID using the national secret algorithm SM2, where the watermark summary value is the watermark hash value obtained after the robust watermark is processed by the hash algorithm) into the smart contract. When the user provides a robust watermark, the video platform will call the copyright information chaining function in the smart contract, upload the chained information to the blockchain platform, and generate a block. At the same time, through the timestamp, it is ensured that every transaction data on the blockchain has a time stamp, and all data stored on the chain is traceable and authentic. In the event of a legal dispute, users can use smart contracts to automatically handle it, saving time and manpower.

[0053] The copyright authentication module is used to extract the watermark in the video file to be verified and obtain the extracted watermark; it is also used to determine whether the watermark hash value and user ID corresponding to the user watermark can be obtained from the blockchain according to the user ID of the logged-in user. If so, the watermark owner authentication is successful; it is also used to compare the normalized correlation coefficient between the extracted watermark and the user watermark in the system database corresponding to the user ID. If the threshold is passed, the video owner authentication is successful.

[0054] Copyright appeal module: used to analyze copyright appeals based on relevant certification documents uploaded by users when users file copyright appeals. The copyright appeal module is also used to transfer the video copyright to the complaining user when the copyright appeal is successful, upload the transaction information to the blockchain, and update the data in the system database.

[0055] Based on the video copyright protection system based on blockchain and digital watermark technology in the above embodiment of the present invention, a video copyright protection method based on blockchain and digital watermark technology is further introduced. After the user registers and logs in, the copyright protection method includes three major sections: copyright registration, copyright certification, and copyright appeal. Corresponding operations can be performed according to user selection. Among them,

[0056] Copyright registration involves the following steps:

[0057] S11: The uploaded video file to be registered and the robust watermark are respectively checked for similarity with the video file and the user watermark in the system database.

[0058] In this embodiment, the image correlation coefficient method is used to extract key frames from registered video files, and the perceptual hash algorithm is used to verify the similarity between the registered video files and the video files in the system database to ensure the originality of the work. The robust watermark provided by the user is also verified for similarity using the perceptual hash algorithm to avoid copyright disputes caused by excessive watermark similarity.

[0059] S12: If the review is passed, the robust watermark uploaded by the user will be stored in the system database; if the video file review is not passed, it is determined that the video file has been registered and the operation is stopped; if the robust watermark review is not passed, the user is required to re-upload a new robust watermark until the review is passed.

[0060] S13: The robust watermark uploaded by the user is stored in the system database, and the watermark hash value is obtained through the hash algorithm.

[0061] S14: The ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user is uploaded to the blockchain and a block is generated.

[0062] S15: Generate a fragile watermark based on the uploaded robust watermark, and embed the robust watermark and the fragile watermark into the video file to be registered respectively through digital watermark technology. Figure 2 The flowchart of embedding a watermark into a video file is shown as follows:

[0063] S151: Performing Hamming code encoding and scrambling processing on the robust watermark;

[0064] In this embodiment, the robust watermark is pre-processed by adding some check bits through Hamming code encoding to enhance the fault tolerance and error correction capabilities of the watermark, and the Arnold scrambling process of the maximum partition method is used to encrypt the robust watermark;

[0065] S152: extracting a video key frame of the video file to be registered, and separating three color channels R, G, and B of the video key frame from each other;

[0066] S153: embedding a fragile watermark in the lowest bit of component B of the key frame of the video to obtain a B component containing the fragile watermark;

[0067] S154: Perform secondary Contourlet transform on component G of the key frame of the video, perform QR decomposition on the low-frequency subband, select R matrix as the embedding carrier to embed the robust watermark to obtain R', and then perform QR inverse decomposition and Contourlet inverse transform to obtain component G containing the robust watermark. The robustness of the watermark can be improved while ensuring its invisibility.

[0068] This completes the copyright registration of the original work.

[0069] Copyright certification includes the following steps:

[0070] S21: Perform scale-invariant feature transform (SIFT algorithm) processing on the video file to be verified.

[0071] S22: Extract the watermark in the video file to be verified to obtain the extracted watermark. Figure 3 The watermark extraction flow chart shown in the embodiment is as follows:

[0072] S221: extracting the video key frame of the video file to be verified, and separating the three color channels R, G, B of the video key frame containing the double watermark from each other;

[0073] S222: Extracting fragile watermarks, specifically:

[0074] Extract the lowest bit of component B and set it to zero to form a fragile watermark matrix W' 2;

[0075] The three components R, G, and B are divided into 2×2 blocks and the mean is calculated. The highest 4 bits of the mean are extracted and formed into a 2×2 small matrix, which is recorded as T k ;

[0076] For fragile watermark matrix W' 2 is divided into 2×2 blocks, recorded as W' 2k ;like W' 2k and T k If they are equal, the first k If the small matrix is ​​all 0 matrices, it means no tampering; otherwise, it is all 1 matrices, which means tampering.

[0077] S223: Extracting a robust watermark, specifically:

[0078] Perform secondary Contourlet transform on the G component to extract its low-frequency subband;

[0079] The low-frequency subband is divided into 2×2 blocks and subjected to QR decomposition to obtain Q Matrix and R matrix;

[0080] calculate R The coefficient value of the first row and second column of the matrix C' 12 The quantitative value of q'=round(C' 12 / L) , where L is the quantization step size. Perform robust watermarking W 1T Extraction of robust watermark W 1T A pixel value (unencrypted and including checksum) w 1T for:

[0081]

[0082] First, the robust watermark W 1T Decryption is performed using the Arnold descrambling method based on the maximum partitioning method to obtain a robust watermark matrix containing only the check code. W' 1T ;

[0083] For the robust watermark matrix W' 1T Decoding is performed by: converting the robust watermark matrix containing the check code W' 1T becomes a matrix with 7 columns A′ , and calculate the matrix A′ The correction factor for each row in the matrix A′ The element values ​​of each row are corrected and the matrix is ​​finally extracted A′ The first 4 columns of elements in the matrix are converted into a square matrix denoted as W' 1, which is the robust watermark finally extracted and restored.

[0084] S23: Determine whether the watermark hash value and user ID corresponding to the user watermark can be obtained from the blockchain according to the user ID of the logged-in user. If so, the watermark owner is authenticated successfully. Specifically, the system calls the corresponding public key through the user ID to decrypt the ciphertext on the blockchain, and obtains the summary value and user ID of the corresponding watermark, which indicates that the watermark belongs to the logged-in user.

[0085] S24: performing a normalized correlation coefficient comparison between the extracted watermark and the user watermark in the system database corresponding to the user ID. If the correlation coefficient passes the threshold, it indicates that the video to be verified belongs to the logged-in user and the video owner is successfully authenticated.

[0086] Based on the above copyright certification, if the user has doubts about it, he or she can choose to file a copyright complaint. The copyright complaint includes the following steps:

[0087] S31: The user uploads relevant certification documents.

[0088] S32: Perform copyright complaint analysis on the video file to be verified; specifically, the administrator reviews and analyzes the uploaded relevant certification documents to determine the ownership of the video to be verified.

[0089] S33: If the copyright complaint is successful, the video copyright is transferred to the complaining user, and the transaction information is uploaded to the blockchain, and the data in the system database is updated. This allows copyright protection. If the administrator finds that the evidence is insufficient, the complaint will be returned as failed.

[0090] In this embodiment, compared with the prior art, the details are as follows:

[0091] To address the problem of poor algorithm robustness, this solution combines Contourlet transform, QR decomposition, SIFT algorithm and Arnold algorithm using maximum segmentation method to reduce the impact of video resolution on algorithm feasibility and improve the robustness of video watermarking against geometric attacks.

[0092] To solve the problem of poor balance of watermarking schemes, this scheme performs secondary Contourlet transform and QR decomposition on the G component of the video frame in the RGB color space. Then, a suitable position is found to embed the robust watermark, and a fragile watermark is embedded in the B component to obtain a video frame with double watermarks. A blind watermarking algorithm is used to extract the fragile watermark and the robust watermark to achieve functions such as invisibility, tampering location, and anti-attack.

[0093] To address the problem of centralized databases, a decentralized, open, transparent, convenient and secure copyright management system based on blockchain is adopted to achieve integrated registration, traceability, theft complaints and certification.

[0094] To solve the problem of poor correlation between watermark and its owner, the hash value of the watermark on the chain and the owner's user ID are encrypted by the national secret algorithm to associate the watermark with its owner. Through the timestamp, every transaction data on the blockchain is guaranteed to have a time stamp, and all data stored on the chain is traceable and authentic. In the event of a legal dispute, users can use smart contracts to automatically handle it, saving time and manpower.

[0095] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A video copyright protection method based on blockchain and digital watermark technology, characterized in that: The following steps are involved: The robust watermark uploaded by the user is stored in the system database, and the watermark hash value is obtained through the hash algorithm; The watermark hash value and the user ID of the logged-in user are encrypted and uploaded to the blockchain to generate a block; Generate a fragile watermark according to the uploaded robust watermark, and embed the robust watermark and the fragile watermark into the video file to be registered respectively; The method further comprises the following steps: Perform scale-invariant feature transformation on the video file to be verified; Extract the watermark from the video file to be verified to obtain the extracted watermark; Determine whether the watermark hash value and user ID corresponding to the user watermark can be obtained from the blockchain according to the user ID of the logged-in user. If so, the watermark owner is authenticated successfully. The extracted watermark is compared with the user watermark in the system database corresponding to the user ID by normalized correlation coefficient. If it passes the threshold, the video owner is authenticated successfully. The step of extracting the watermark in the video file to be verified to obtain the extracted watermark comprises: Extract the video key frame of the video file to be verified, and separate the three color channels R, G, B of the video key frame containing the double watermark from each other; Extract the fragile watermark, specifically: Extract the lowest bit of component B and set it to zero to form a fragile watermark matrix W' 2; The three components R, G, and B are divided into 2×2 blocks and the mean is calculated. The highest 4 bits of the mean are extracted and formed into a 2×2 small matrix, which is recorded as T k ; The fragile watermark matrix W'2 is divided into 2×2 blocks, denoted as W' 2k ; If W' 2k With T k If they are equal, the kth small matrix of the tampering location matrix diagram is a matrix of all 0s, indicating no tampering; otherwise, it is a matrix of all 1s, indicating tampering; Extract the robust watermark, specifically: Perform secondary Contourlet transform on the G component to extract its low-frequency subband; Divide the low frequency subband into 2×2 blocks and perform QR decomposition to obtain Q matrix and R matrix; Calculate the coefficient value C' of the 1st row and 2nd column of the R matrix 12 The quantized value q'=round(C' 12 / L), where L is the quantization step size; Perform robust watermarking W 1T Extraction of; Extracted robust watermark W 1T A pixel value w 1T for: First, the robust watermark W 1T Decryption is performed using the Arnold descrambling method based on the maximum partitioning method to obtain a robust watermark matrix W' containing only the check code. 1T ; For the robust watermark matrix W' 1T Decoding is performed by: converting the robust watermark matrix W' containing the check code 1T It becomes a matrix A′ with 7 columns, and calculates the correction factor of each row in the matrix A′, corrects the element value of each row of the matrix A′, and finally extracts the first 4 columns of the matrix A′ and turns it into a square matrix recorded as W'1, which is the robust watermark finally extracted and restored.

2. The video copyright protection method based on blockchain and digital watermarking technology as claimed in claim 1 is characterized in that: Before executing the step of uploading the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user to the blockchain and generating a block, identity authentication is also included, specifically: Use SM3 to perform hash operation on the robust watermark and user ID to obtain the corresponding summary, and use the user ID summary value as the private key, and use SM2 to generate the corresponding public key; encrypt the robust watermark summary value and user ID with the private key; use the public key to decrypt before uploading to the chain. If the decrypted user ID matches the user identity, the authentication is passed, and the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user is uploaded to the blockchain and a block is generated.

3. The video copyright protection method based on blockchain and digital watermarking technology as claimed in claim 1 is characterized in that: Before executing the step of storing the robust watermark uploaded by the user into the system database, the following steps are also included: The uploaded video file to be registered and the robust watermark are respectively checked for similarity with the video file and the user watermark in the system database; If the review is passed, the robust watermark uploaded by the user will be stored in the system database; if the video file fails the review, the operation will be stopped; if the robust watermark fails the review, the user will be required to re-upload a new robust watermark until the review is passed.

4. The video copyright protection method based on blockchain and digital watermarking technology as claimed in claim 1 is characterized in that: The method further comprises the following steps: Users upload relevant certification documents and conduct copyright complaint analysis on the video files to be verified; If the copyright appeal is successful, the video copyright will be transferred to the complaining user, the transaction information will be uploaded to the blockchain, and the data in the system database will be updated.

5. The video copyright protection method based on blockchain and digital watermarking technology according to any one of claims 1 to 4, characterized in that: In the steps of embedding the robust watermark and the fragile watermark into the video file to be registered, the specific steps are as follows: Perform Hamming code encoding and scrambling on the robust watermark; Extract video key frames of the video file to be registered; Embed the fragile watermark in the lowest bit of the B component of the video key frame to obtain the B component containing the fragile watermark; The G component of the key frame of the video is subjected to secondary Contourlet transform, the low-frequency subband is subjected to QR decomposition, and the R matrix is ​​selected as the embedding carrier to embed the robust watermark. Then, QR inverse decomposition and Contourlet inverse transform are performed to obtain the G component containing the robust watermark.

6. A video copyright protection system based on blockchain and digital watermark technology, used to execute the method described in any one of claims 1 to 5, characterized in that: include: User management module, used for user registration and login, as well as user information management, user video file management and user watermark management; The copyright registration module is used to generate a fragile watermark based on the uploaded robust watermark, and embed the robust watermark and the fragile watermark into the video file to be registered respectively; The blockchain management module is used to upload the ciphertext generated by encrypting the watermark hash value and the user ID of the logged-in user to the blockchain and generate blocks; Wherein, the system further comprises: The copyright authentication module is used to extract the watermark in the video file to be verified after performing scale-invariant feature transformation on the video file to obtain the extracted watermark; and determine whether the watermark hash value and user ID corresponding to the user watermark can be obtained from the blockchain according to the user ID of the logged-in user. If so, the watermark owner is successfully authenticated; it is also used to compare the normalized correlation coefficient between the extracted watermark and the user watermark in the system database corresponding to the user ID. If the threshold is passed, the video owner is successfully authenticated.

7. The video copyright protection system based on blockchain and digital watermarking technology as claimed in claim 6, characterized in that: The copyright registration module is also used for copyright review, specifically: The uploaded video file to be registered and the robust watermark are respectively compared with the video file and the user watermark in the system database using the perceptual hash algorithm for similarity review; If the review is passed, the robust watermark uploaded by the user will be stored in the system database; if the video file fails the review, the operation will be stopped; if the robust watermark fails the review, the user will be required to re-upload a new robust watermark until the review is passed.

8. The video copyright protection system based on blockchain and digital watermarking technology as claimed in claim 7, characterized in that: The system further comprises: The copyright appeal module is used to analyze the copyright appeal based on the relevant supporting documents uploaded by the user when the user files a copyright appeal request; The copyright appeal module is also used to transfer the video copyright to the complaining user when the copyright appeal is successful, upload the transaction information to the blockchain, and update the data in the system database.

Images