Digital content ownership traceability and ethical supervision method and system

Abstract

The invention relates to a digital content ownership traceability and ethical supervision method and system, and belongs to the field of digital content ownership management and ethical governance. According to the method, before digital multimedia content is publicly published, a digital watermark technology, a computer vision technology, an ethical risk identification technology and other technologies are used for completing three operations of digital watermark embedding, digital fingerprint generation and ethical risk rating, and results of the three operations and related ownership information are recorded in a data center; when ownership query, source tracing or risk control supervision is needed, ownership identification and source tracing can be carried out through combination of a watermark extraction mode and a fingerprint comparison mode; and when the embedded digital watermark cannot be extracted after being maliciously attacked and damaged, the ownership information can be quickly positioned through the digital fingerprint. The system comprises a server side, a management side and a user side, the user side comprises an SDK client side and a web user side.

Description

Technical Field

[0001] The present invention belongs to the field of computers and relates to a method and system for tracing the ownership of digital content and ethical supervision. Background Art

[0002] Digital content, especially multimedia content, is easy to copy, tamper with, and spread, which brings huge challenges to ownership identification, leak tracing, and copyright protection. Currently, the commonly used solutions mainly include the following:

[0003] (1) DRM (Digital Rights Management) digital copyright protection technology

[0004] The technology protects digital content through encryption and authorization control, limiting the copying, distribution and use of content.

[0005] Advantages: Can effectively prevent unauthorized copying and use.

[0006] Disadvantages: easy to crack and poor user experience.

[0007] (2) Visible image watermark

[0008] The watermark information is embedded into the digital content in a visible way, such as text or pattern on a picture.

[0009] Advantages: easy to identify and can serve as a warning.

[0010] Disadvantages: easy to be removed and affects the user experience.

[0011] (3) Hidden watermark technology

[0012] The watermark information is embedded into the digital content in an invisible way, such as frequency changes in audio signals or slight adjustments in image pixel values.

[0013] Advantages: Hard to detect and remove.

[0014] Disadvantages: The watermark embedding and extraction process is complicated and has a certain impact on content quality.

[0015] (4) Blockchain technology

[0016] The immutability and traceability of blockchain are used to record the ownership information and transaction records of digital content.

[0017] Advantages: high security and strong traceability.

[0018] Disadvantages: complex technology and high cost.

[0019] Limitations of existing technologies:

[0020] The rapid development of self-media and AIGC has made the generation and dissemination of digital content more convenient, but has also brought new challenges in ownership management.

[0021] There are potential ethical risks in the content generation process of self-media and AIGC, such as false information, discriminatory content, etc., which require an effective regulatory mechanism. Summary of the invention

[0022] In view of this, the purpose of the present invention is to provide a digital content ownership tracing and ethical supervision method and system. The method uses digital watermark technology, computer vision technology and ethical risk identification technology to complete the three operations of digital watermark embedding, digital fingerprint generation and ethical risk rating before the digital multimedia content is publicly released, and the results of these three operations and related ownership information are filed in the data center. When ownership query, tracing or risk control supervision is required, ownership identification and tracing can be carried out by combining watermark extraction and fingerprint comparison. The ethical risks of the content can also be quickly queried or the ethical risk status can be monitored and managed as a whole; when the embedded digital watermark is destroyed by malicious attacks and cannot be extracted, the ownership information can also be quickly located by digital fingerprints. The system consists of three major parts: the server, the system management, and the user. The server is responsible for providing specific algorithms and services for digital watermark generation and extraction, fingerprint generation and comparison, and risk rating, and provides management and external services for information related to ownership and security ethics management; the system management is mainly a subsystem provided to management personnel, providing functions or services related to ownership management, security management, ethics management, system management, etc.; the user includes two forms: SDKSDK client and web user.

[0023] In order to achieve the above object, the present invention provides the following technical solutions:

[0024] A digital content ownership tracing and ethical supervision method, the method comprising the following steps:

[0025] S1: Ownership registration;

[0026] S2: Ownership extraction;

[0027] The S1 specifically includes the following steps:

[0028] S11: Setting ownership information and digital content;

[0029] S12: Generate digital works and initiate operation requests;

[0030] S13: The system accepts the request and starts executing S131 to S133 in parallel;

[0031] S131: Digital fingerprint generation;

[0032] S132: Ethical level determination;

[0033] S133: Embedding implicit digital watermark;

[0034] S131 is as follows:

[0035] S1311: extracting a salient area from the digital content, extracting stable feature points from the salient area, and generating a feature-based digital fingerprint, i.e., a first digital fingerprint, according to the feature points;

[0036] S1312: Input the digital content into a fingerprint generation pre-training model, and the model generates a digital fingerprint based on a neural network, that is, a second digital fingerprint;

[0037] S132 specifically includes: inputting the digital content into a pre-trained deep learning model, and the model predicts the ethical level based on the content;

[0038] S133 specifically includes: generating an ownership identifier, encrypting and encoding the ownership identifier into a digital watermark, implicitly embedding the digital watermark into the digital content, and obtaining the digital content after the watermark is embedded; generating a digital fingerprint for the digital content after the watermark is embedded, i.e., a third digital fingerprint; returning the digital content after the watermark is embedded to the user, and the watermark embedding phase ends;

[0039] S14: associating ownership information, digital content, the first digital fingerprint, the second digital fingerprint, the third digital fingerprint, the ethics level and the ownership identifier, and storing and filing them in the background;

[0040] The S2 specifically includes the following steps:

[0041] S21: Initiate an ownership authentication or source tracing operation request for the digital content embedded with the watermark;

[0042] S22: extracting the implicit digital watermark and determining whether the extraction is successful;

[0043] S221: If the extraction is successful, the digital watermark is decoded and decrypted into an ownership identifier, and it is determined whether the decoding and decryption is successful. If the decoding and decryption is successful, the ownership information corresponding to the ownership identifier is extracted, and an ownership certification document is generated. The identification result or traceability result is returned to the user or management personnel according to the management rules; if the decoding and decryption is unsuccessful, enter S222;

[0044] S222: If the extraction is unsuccessful, a digital fingerprint is generated according to the content;

[0045] S2221: extracting stable feature points from the salient area, and generating a feature-based digital fingerprint based on the stable feature points, namely, a first digital fingerprint';

[0046] S2222: The model generates a digital fingerprint based on a neural network, namely, a second digital fingerprint';

[0047] S2223: Generate a digital fingerprint according to the embedded fingerprint generation model, that is, a third digital fingerprint';

[0048] S223: Retrieve similar digital contents from the background data through the first digital fingerprint '~3' and sort them by similarity; the manager receives a list of similar digital contents, performs manual ownership identification or tracing, gives a manual identification conclusion, and returns the identification result or tracing result to the user or manager according to the management rules; at the same time, compress the similar contents and add invisible watermarks and visible watermarks as query results to obtain a subset of the similar digital content set, and feed back the processed most similar digital content to the user according to the management rules.

[0049] Optionally, in S11, the ownership information includes the author, the owner, the creation time and the content description; the digital content includes text, images, audio and video and artificial intelligence generated content AIGC.

[0050] A digital content ownership tracing and ethics supervision system based on the method, the system includes a user end, a service end and a system management end;

[0051] The user end includes the software development kit SDK client and the web client; the SDK client provides SDK interface, watermark embedding component, digital fingerprint generation component and ethical risk assessment component for server authentication; the SDK client is embedded in the AIGC system, and the calculation process of digital fingerprint generation, ethical level judgment and embedded implicit digital watermark is transferred to AIGC to provide results; the web client provides the user end with ownership registration, ownership query and digital content traceability services;

[0052] The server is used to provide cloud services of ownership recordation, ethical risk management, ownership registration, and ownership identification and detection to the SDK client for calling;

[0053] The system management terminal provides a management platform for supervisors to realize data statistics display, system parameter setting and client management.

[0054] The beneficial effects of the present invention are:

[0055] (1) Enhanced reliability of ownership identification:

[0056] The combination of digital watermark and digital fingerprint provides double protection for ownership identification. Even if the digital watermark is destroyed, the digital fingerprint can quickly locate the source of the digital content, improving the accuracy and reliability of ownership identification.

[0057] This method is applicable to various types of content, including text, images, audio and video, and AI-generated content, and has strong versatility.

[0058] (2) Improved efficiency of leak tracing

[0059] Digital fingerprint technology can quickly locate the possible source of digital content, shorten the time for leak tracing and improve efficiency.

[0060] Even if digital content has been attacked by screenshots, cropping, scaling, compression, etc., digital fingerprint technology can still be effectively identified, ensuring the accuracy of traceability.

[0061] (3) Effective management and control of ethical risks

[0062] By conducting ethical risk ratings on digital content, potential risks can be discovered and addressed in a timely manner, avoiding the spread of false information, discriminatory content, and other negative information.

[0063] The system management end provides a management platform for regulators, which can realize data statistics display, system parameter setting and client management, build a multi-level supervision mechanism, and effectively ensure the healthy development of digital content.

[0064] (4) Adapting to the development of AIGC

[0065] This method can be integrated into the AIGC platform through the SDK client to achieve automatic watermark embedding and ethical risk assessment of AIGC content, thereby ensuring the legality and security of AIGC content.

[0066] Through the supervision of AIGC content, we can effectively prevent the ethical risks brought by AIGC and promote the healthy development of AIGC technology.

[0067] (5) Easy to use and manage

[0068] The web client provides users with convenient rights registration, rights query and digital content traceability services, and is simple and easy to use.

[0069] The system management end provides regulators with functions such as data statistics display, system parameter settings and client management to facilitate system management and maintenance.

[0070] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0072] Figure 1 It is a flowchart of the ownership registration in the method of the present invention;

[0073] Figure 2 The flowchart of the ownership extraction in the method of the present invention is as follows;

[0074] Figure 3 This is a system structure diagram of the present invention. DETAILED DESCRIPTION

[0075] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0076] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0077] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0078] The present invention achieves a breakthrough in digital content ownership traceability and ethical supervision by combining digital watermarks, digital fingerprints, and ethical risk identification. The following are specific implementation methods:

[0079] (1) Ownership registration stage

[0080] Users submit digital content ownership information through the Web client or SDK client, including the author, owner, creation time, content description, and the digital content itself.

[0081] The system generates an ownership identification based on the ownership information, converts it into a digital watermark using encryption coding technology, and then implicitly embeds the digital watermark into the digital content.

[0082] The system extracts salient areas and stable feature points from digital content to generate feature-based digital fingerprints. At the same time, it uses a pre-trained deep learning model to generate a neural network-based digital fingerprint.

[0083] The system inputs digital content into a pre-trained deep learning model and predicts the ethical level based on the content.

[0084] The system links ownership information, digital content, digital fingerprints, ethical levels and other information, and stores and files them in the background data center.

[0085] (2) Rights extraction stage

[0086] The user initiates an ownership authentication or traceability operation request for the digital content embedded with the watermark.

[0087] The system attempts to extract the digital watermark from the digital content and determines whether the extraction is successful.

[0088] If the extraction is successful, the system will decode the digital watermark into an ownership identification, extract the corresponding ownership information, and generate an ownership certificate document.

[0089] If the extraction is unsuccessful, the system will generate a digital fingerprint based on the digital content and identify similar content by comparing it with the digital fingerprint in the background data.

[0090] Managers conduct manual identification or tracing based on the list of similar content and give conclusions.

[0091] The system returns the ownership identification or traceability results to the user or manager.

[0092] (3) Ethical risk monitoring

[0093] The system monitors the ethical risks of digital content in real time and issues warnings based on preset warning values.

[0094] Managers manually interpret the warning content and handle it according to the rules.

[0095] The system management end provides data statistics display function to help managers understand the overall ethical risk situation.

[0096] (4) System architecture

[0097] User end: includes Web user end and SDK client, providing users with services such as ownership registration, ownership query, and digital content traceability.

[0098] Server: Provides cloud services such as digital watermark generation and extraction, fingerprint generation and comparison, and risk rating.

[0099] System management end: provides a management platform for regulators to realize functions such as data statistics display, system parameter setting and client management.

[0100] (5) Application scenarios

[0101] Digital content creators or copyright holders can use this system to protect copyright and prevent infringement.

[0102] Enterprises or institutions can use this system to trace leaks and track down leakers.

[0103] The AIGC platform can use this system to trace the origin of AIGC content and manage ethical risks, promoting the healthy development of AIGC technology.

[0104] Before the digital multimedia content is publicly released, digital watermark technology, computer vision technology and ethical risk identification technology are used to complete the three operations of digital watermark embedding, digital fingerprint generation and ethical risk rating, and the results of these three operations and related ownership information are filed in the data center. The specific steps are as follows: Figure 1 shown.

[0105] When there is a need for ownership inquiry, tracing or risk control supervision, ownership identification and tracing can be carried out by combining watermark extraction and fingerprint comparison. The ethical risks of the content can also be quickly queried or the ethical risk status can be monitored and managed as a whole. When the embedded digital watermark is damaged by malicious attacks and cannot be extracted, the ownership information can be quickly located through the digital fingerprint. The advantage of this is that when the digital watermark is damaged to the point where the implicit watermark cannot be extracted, the digital fingerprint can be used to quickly locate the possible source of the digital content, providing multiple technical supports for ownership identification and digital content tracing. The specific steps are as follows: Figure 2 shown.

[0106] When ethical risks reach the warning value, the system can notify the management personnel of the ethical incident in a variety of ways, and the management personnel will make corresponding treatment according to the rules after manual interpretation. In addition, the management personnel can monitor the overall situation of ethical risks through the system management terminal.

[0107] To implement the above method, the present invention also discloses a corresponding management system, the structure of which is as follows: Figure 3 shown.

[0108] The system includes a user end, a service end and a system management end;

[0109] The user end includes the software development kit SDK client and the web client; the SDK client provides SDK interface, watermark embedding component, digital fingerprint generation component and ethical risk assessment component for server authentication; the SDK client is embedded in the AIGC system, and the calculation process of digital fingerprint generation, ethical level judgment and embedded implicit digital watermark is transferred to AIGC to provide results; the web client provides the user end with ownership registration, ownership query and digital content traceability services;

[0110] The server is used to provide cloud services of ownership recordation, ethical risk management, ownership registration, and ownership identification and detection to the SDK client for calling;

[0111] The system management terminal provides a management platform for supervisors to realize data statistics display, system parameter setting and client management.

[0112] Embodiment 1 (Digital Content Ownership Identification and Copyright Protection)

[0113] The method of the present invention can be used for copyright protection of digital content. The general process is as follows: the digital content creator or copyright holder uploads the digital work and copyright information to the system through the Web client. The system automatically completes the three operations of digital watermark embedding, digital fingerprint generation and ethical risk rating and feeds back the digital content embedded with implicit watermark to the copyright holder. The management party can monitor the ethical risks of the content. After receiving the embedded watermark, the copyright holder can publish the digital content work to the outside world. When an infringement dispute occurs, the infringing digital content can be submitted to the system through the Web client. The system will Figure 2 The process can provide the conclusion of ownership identification. When copyright transaction or circulation occurs, it can be realized directly by modifying the ownership registration information in the system, without modifying the ownership identification or re-embedding the digital watermark.

[0114] Embodiment 2 (Digital Content Leakage Tracing)

[0115] The method of the present invention can be applied to the tracing of digital content leaks. First, the digital content is registered in the system before being sent to the viewer, and then the digital content with an implicit watermark is sent to the viewer. When the viewer privately leaks the content (before the leak, the digital content with the implicit watermark may be tampered with by attacking, such as screen photography, cropping, scaling, compression, etc.), when it is necessary to trace the leaker of the digital content, the leaked digital content can be submitted to the platform, and the platform can be used as follows: Figure 2 The method shown is used to trace the author of the leak. The method combining digital watermark and digital fingerprint can greatly improve the accuracy and speed of tracing.

[0116] Example 3 (AIGC traceability and ethical risk management)

[0117] The AIGC platform integrates functions into the AIGC system through the SDK client. After AI generates digital content, it must first call the regulatory SDK client to add a digital watermark before returning it to the user. If it exceeds a certain ethical risk level, it will be sent to the AIGC platform content review department for manual review to decide whether to feedback to the user. When the user obtains the digital content with an embedded digital watermark and distributes or tampered with it, if there is a need to verify whether the digital content is generated by AI or to verify the rights and responsibilities of the content generator (such as active application, public inquiry, dispute, infringement, etc.), the regulatory SDK client must be called to add a digital watermark before returning it to the user. Figure 2 The processing flow can quickly locate the creator and hold the perpetrator accountable according to law. Through the method and system of the present invention, the regulator can manage and monitor the development of AIGC and ethical risks from the overall to the individual level.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A digital content ownership tracing and ethical supervision method, characterized by: The method comprises the following steps: S1: Ownership registration; S2: Ownership extraction; The S1 specifically includes the following steps: S11: Setting ownership information and digital content; S12: Generate digital works and initiate operation requests; S13: The system accepts the request and starts executing S131 to S133 in parallel; S131: Digital fingerprint generation; S132: Ethical level determination; S133: Embedding implicit digital watermark; S131 is as follows: S1311: extracting a salient area Saliency from the digital content, extracting stable feature points from the salient area, and generating a feature-based digital fingerprint, i.e., a first digital fingerprint, according to the feature points; S1312: Input the digital content into the fingerprint generation pre-training model, and the model generates a digital fingerprint based on a neural network, that is, a second digital fingerprint; the fingerprint generation pre-training model performs model training through contrastive learning, and finally obtains a digital fingerprint; S132 specifically includes: inputting the digital content into a pre-trained deep learning model, and the model predicts the ethical level based on the content; S133 specifically includes: generating an ownership identifier, encrypting and encoding the ownership identifier into a digital watermark, implicitly embedding the digital watermark into the digital content, and obtaining the digital content after the watermark is embedded; generating a digital fingerprint for the digital content after the watermark is embedded, i.e., a third digital fingerprint; returning the digital content after the watermark is embedded to the user, and the watermark embedding phase ends; S14: associating ownership information, digital content, the first digital fingerprint, the second digital fingerprint, the third digital fingerprint, the ethics level and the ownership identifier, and storing and filing them in the background; The S2 specifically includes the following steps: S21: Initiate an ownership authentication or source tracing operation request for the digital content embedded with the watermark; S22: extracting the implicit digital watermark and determining whether the extraction is successful; S221: If the extraction is successful, the digital watermark is decoded and decrypted into an ownership identifier, and it is determined whether the decoding and decryption is successful. If the decoding and decryption is successful, the ownership information corresponding to the ownership identifier is extracted, and an ownership certification document is generated. The identification result or traceability result is returned to the user or management personnel according to the management rules; if the decoding and decryption is unsuccessful, enter S222; S222: If the extraction is unsuccessful, a digital fingerprint is generated according to the content; S2221: extracting stable feature points from the salient area, and generating a feature-based digital fingerprint based on the stable feature points, namely, a first digital fingerprint'; S2222: The model generates a digital fingerprint based on a neural network, namely, a second digital fingerprint'; S2223: Generate a digital fingerprint according to the embedded fingerprint generation model, namely the third digital fingerprint '; the embedded fingerprint generation model adopts a digital fingerprint generation method, including MD5, SHA1, and Rabin fingerprint algorithms; S223: Retrieve similar digital content from the background data through the first digital fingerprint '~3' and sort by similarity; The sorting adopts the classification sorting method, and searches for the records corresponding to the third digital fingerprint through the complete and accurate matching algorithm; searches for the records corresponding to the first digital fingerprint and the second digital fingerprint through similarity sorting, and sorts the returned records from high to low in terms of similarity; the similarity calculation method includes L2 norm, cosine similarity or similarity calculation based on edit distance; The manager receives a list of similar digital content, conducts manual identification or tracing of ownership, gives a manual identification conclusion, and returns the identification result or tracing result to the user or manager according to the management rules; at the same time, the similar content is compressed and invisible watermarks and visible watermarks are added as query results to obtain a subset of the similar digital content set, and the processed most similar digital content is fed back to the user according to the management rules.

2. A digital content ownership tracing and ethical supervision method according to claim 1, characterized in that: In S11, the ownership information includes the author, the owner, the creation time and the content description; the digital content includes text, images, audio and video and artificial intelligence generated content AIGC.

3. A digital content ownership tracing and ethics supervision system based on the method of claim 1 or 2, characterized in that: The system includes a user end, a service end and a system management end; The user end includes the software development kit SDK client and the web client; the SDK client provides SDK interface, watermark embedding component, digital fingerprint generation component and ethical risk assessment component for server authentication; the SDK client is embedded in the AIGC system, and the calculation process of digital fingerprint generation, ethical level judgment and embedded implicit digital watermark is transferred to AIGC to provide results; the web client provides the user end with ownership registration, ownership query and digital content traceability services; The server is used to provide cloud services of ownership recordation, ethical risk management, ownership registration, and ownership identification and detection to the SDK client for calling; The system management terminal provides a management platform for supervisors to realize data statistics display, system parameter setting and client management.

Images