Digital collection copyright intelligent monitoring method and platform based on block chain

Abstract

The invention discloses a digital collection copyright intelligent monitoring method and platform based on a block chain, and relates to the related field of data processing, and the method comprises the steps: carrying out the blocking processing of a target digital collection; performing infringement feature recognition according to the type; calling a content-aware Hash algorithm library to carry out infringement recognition sensitivity test, and constructing an initial Hash algorithm integration architecture; performing monitoring attention configuration, and generating a preset Hash algorithm integration architecture; performing multimode hash fusion processing on the plurality of collection blocks to obtain a plurality of block hash values, and uploading the block hash values to a block chain; monitoring according to the type to obtain a first detection collection, and performing multimode hash fusion processing to generate a plurality of detection hash values; and performing similarity comparison to generate infringement monitoring reminding information. The technical problem that existing digital collection copyright monitoring is insufficient in monitoring precision is solved, and the technical effect of accurately recognizing and monitoring digital collection infringement behaviors is achieved.

Description

Technical Field

[0001] The present application relates to data processing related fields, and in particular to a blockchain-based intelligent copyright monitoring method and platform for digital collections. Background Art

[0002] With the rapid development of digital technology, the creation, trading and collection of digital collections (such as digital artworks, virtual goods, etc.) have become increasingly popular and have gradually become an important part of the global cultural industry. However, the easy duplication, unbounded dissemination and diversity of digital collections have made copyright protection face unprecedented challenges. Most existing copyright monitoring methods rely on a single hash algorithm to identify infringements. Although this method can provide basic protection in some cases, its limitations are becoming increasingly apparent. In particular, when faced with digital collections that have been slightly modified, traditional methods are often unable to effectively detect them, resulting in frequent omissions or false positives, which seriously affects the accuracy and comprehensiveness of copyright protection.

[0003] Among the current related technologies, copyright monitoring of digital collections has the technical problem of insufficient monitoring accuracy. Summary of the invention

[0004] This application provides a blockchain-based intelligent copyright monitoring method and platform for digital collections, divides the target digital collections into blocks, assigns monitoring weights to each block, identifies infringement characteristics according to the type of digital collections, generates preset infringement characteristics, performs sensitivity tests based on these characteristics, builds and optimizes the initial hash algorithm integration architecture, performs multi-mode hash fusion processing on multiple collection blocks through the preset hash algorithm integration architecture, generates multiple block hash values, and uploads the generated block hash values ​​to the blockchain, etc., to achieve the technical effect of accurately identifying and monitoring infringement of digital collections.

[0005] This application provides a blockchain-based intelligent copyright monitoring method for digital collections, including: The target digital collection is processed into blocks to generate multiple collection blocks and multiple block monitoring weights; infringement characteristics are identified according to the type of the target digital collection to generate preset infringement characteristics; based on the preset infringement characteristics, the content-aware hash algorithm library is called to perform infringement identification sensitivity testing, and an initial hash algorithm integration architecture is constructed according to the sensitivity test results; monitoring attention configuration is performed on the initial hash algorithm integration architecture based on the multiple block monitoring weights to generate a preset hash algorithm integration architecture; multi-mode hash fusion processing is performed on the multiple collection blocks through the preset hash algorithm integration architecture to obtain multiple block hash values ​​and upload them to the blockchain; a connection with a preset network platform is established, the first detection collection is monitored and obtained according to the type of the target digital collection, and the preset hash algorithm integration architecture is called to perform multi-mode hash fusion processing to generate multiple detection hash values; similarity comparison is performed on the multiple block hash values ​​and the multiple detection hash values ​​to generate infringement monitoring reminder information.

[0006] In a possible implementation, the target digital collection is divided into blocks to generate multiple collection blocks and multiple block monitoring weights, and the following processing is performed: The target digital collection is divided into blocks according to a predetermined grid division to generate the multiple collection blocks; regional feature density detection is performed on the multiple collection blocks to generate multiple feature densities; complexity identification of color and texture features is performed on the multiple collection blocks to generate multiple complexity indices; weight configuration is performed in combination with the multiple feature densities and the multiple complexity indices to generate the multiple block monitoring weights, wherein the block monitoring weight is proportional to the feature density and the complexity index.

[0007] In a possible implementation, infringement feature identification is performed according to the type of the target digital collection, a preset infringement feature is generated, and the following processing is performed: According to the type of the target digital collection, infringement records of digital collections of the same type are mined to generate a collection infringement record data set; infringement features are extracted from the collection infringement record data set to generate an infringement feature set; similarity comparison is performed on any two groups of features in the infringement feature set, features with similarity greater than a preset similarity are clustered to generate the preset infringement feature.

[0008] In a possible implementation, based on the preset infringement features, the content-aware hash algorithm library is called to perform an infringement identification sensitivity test, an initial hash algorithm integration architecture is constructed according to the sensitivity test results, and the following processing is performed: Step 1: extract the first infringement feature from the preset infringement features, and collect a historical infringement sample data set with the first infringement feature as a constraint, wherein any group of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 2: use multiple content-aware hash algorithms in the content-aware hash algorithm library to hash and compare the historical genuine collection samples and the historical infringing collection samples, respectively, to generate multiple groups of hash value difference information; Step 3: generate multiple first infringement identification sensitivity indicators based on the multiple groups of hash value difference information; Step 4: select the content-aware hash algorithm corresponding to the maximum indicator based on the multiple first infringement identification sensitivity indicators, and map and associate it with the first infringement feature to generate a first content-aware hash algorithm; and so on, iteratively repeat steps 1 to 4 until the second infringement feature from the preset infringement feature to the Nth infringement feature is traversed, and the second content-aware hash algorithm to the Nth content-aware hash algorithm is obtained; the initial hash algorithm integration architecture is generated using the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm.

[0009] In a possible implementation, the preset hash algorithm integrated architecture is used to perform multi-mode hash fusion processing on the multiple collection blocks, multiple block hash values ​​are obtained and uploaded to the blockchain, and the following processing is performed: The first content-aware hash algorithm, the second content-aware hash algorithm, and up to the Nth content-aware hash algorithm in the preset hash algorithm integrated architecture are used to perform feature processing in the pre-hashing stage on the multiple collection blocks, respectively, to generate multiple first pre-feature data sets, multiple second pre-feature data sets, and up to multiple Nth pre-feature data sets; hash calculations are performed on the multiple first pre-feature data sets, multiple second pre-feature data sets, and up to multiple Nth pre-feature data sets according to the monitoring attention configuration to generate the multiple block hash values.

[0010] In a possible implementation, the following processing is performed: Read the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index respectively corresponding to the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm; perform weight allocation on the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm based on the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index to obtain a weight allocation result; establish a fusion optimization network based on the weight allocation result, connect it to the preset hash algorithm integrated architecture, fuse the multiple block hash values, and generate a preset fusion hash value; and perform infringement monitoring reminder on the first detected collection with the preset fusion hash value.

[0011] In a possible implementation, the plurality of block hash values ​​are compared with the plurality of detection hash values ​​for similarity, infringement monitoring reminder information is generated, and the following processing is performed: Based on the blockchain, copyright transaction information is obtained; the copyright transaction information is parsed to generate multiple permission features of multiple authorized parties; when the similarity between the multiple block hash values ​​and the multiple detection hash values ​​meets the preset infringement similarity, the first collection circulation information of the first detection collection is read; based on the multiple authorized parties and the multiple permission features, the first collection circulation information is verified, and if the verification fails, the infringement monitoring reminder information is generated.

[0012] This application also provides a blockchain-based intelligent monitoring platform for digital collection copyrights, including: The block processing module is used to perform block processing on the target digital collection and generate multiple collection blocks and multiple block monitoring weights; the infringement feature identification module is used to identify the infringement features according to the type of the target digital collection and generate preset infringement features; the infringement identification sensitivity test module is used to call the content-aware hash algorithm library to perform infringement identification sensitivity test based on the preset infringement features, and build an initial hash algorithm integration architecture according to the sensitivity test results; the monitoring attention configuration module is used to perform monitoring attention configuration on the initial hash algorithm integration architecture based on the multiple block monitoring weights, and generate a preset hash algorithm set The present invention relates to a digital collection storage system comprising: a multi-mode hash fusion processing module, which is used to perform multi-mode hash fusion processing on the multiple collection blocks through the preset hash algorithm integrated architecture, obtain multiple block hash values ​​and upload them to the blockchain; multiple detection hash value generation modules, which are used to establish a connection with a preset network platform, monitor and obtain a first detection collection according to the type of the target digital collection, and call the preset hash algorithm integrated architecture to perform multi-mode hash fusion processing to generate multiple detection hash values; an infringement monitoring reminder information generation module, which is used to perform a similarity comparison between the multiple block hash values ​​and the multiple detection hash values ​​to generate infringement monitoring reminder information.

[0013] The blockchain-based intelligent copyright monitoring method and platform for digital collections proposed in this application first divides the target digital collection into blocks to generate multiple collection blocks and multiple block monitoring weights, then identifies infringement features based on the type of the target digital collection to generate preset infringement features, then calls the content-aware hash algorithm library to perform infringement identification sensitivity testing based on the preset infringement features, builds an initial hash algorithm integration architecture based on the sensitivity test results, and then performs monitoring attention configuration on the initial hash algorithm integration architecture based on multiple block monitoring weights to generate a preset hash algorithm integration architecture, then performs multi-mode hash fusion processing on multiple collection blocks through the preset hash algorithm integration architecture to obtain multiple block hash values ​​and upload them to the blockchain, then establishes a connection with a preset network platform, monitors and obtains the first detection collection according to the type of the target digital collection, calls the preset hash algorithm integration architecture to perform multi-mode hash fusion processing to generate multiple detection hash values, and finally performs similarity comparison between multiple block hash values ​​and multiple detection hash values ​​to generate infringement monitoring reminder information, thereby achieving the technical effect of accurately identifying and monitoring infringement of digital collections. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention are briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the platform according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A flowchart of a blockchain-based intelligent copyright monitoring method for digital collections provided in an embodiment of the present application.

[0016] Figure 2 A schematic diagram of the structure of the blockchain-based intelligent copyright monitoring platform for digital collections provided in an embodiment of the present application.

[0017] Explanation of the accompanying drawings: block processing module 10, infringement feature recognition module 20, infringement recognition sensitivity test module 30, monitoring attention configuration module 40, multi-mode hash fusion processing module 50, multiple detection hash value generation module 60, infringement monitoring reminder information generation module 70. DETAILED DESCRIPTION

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0020] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0021] The present application embodiment provides a blockchain-based digital collection copyright intelligent monitoring method, such as Figure 1 As shown, the method includes: Step S100, divide the target digital collection into blocks to generate multiple collection blocks and multiple block monitoring weights. Specifically, the original data of the target digital collection, such as images, videos or photos, is obtained through the file reading interface. According to the preset block size (such as fixed pixel size or time segment), the target digital collection is divided into multiple small blocks (collection blocks). For example, for images, it is divided by pixel grid; for videos, it is divided by frames and intra-frame areas. Based on the complexity, uniqueness and importance of each block (such as the central area in the image is more important than the edge area), a monitoring weight is assigned to each block through an algorithm.

[0022] In a possible implementation, the target digital collection is divided into blocks to generate multiple collection blocks and multiple block monitoring weights. Step S100 further includes step S110, dividing the target digital collection into blocks according to a predetermined division grid to generate the multiple collection blocks. Specifically, a two-dimensional (for images) or three-dimensional (for videos) division grid is determined according to the size of the target digital collection (such as the pixel size of the image, the time length and resolution of the video) and the preset block size. According to the division grid, the target digital collection is divided into multiple collection blocks of equal size, that is, the collection block is a single part obtained after the target digital collection is divided, and each collection block contains a part of the collection data. Step S120, regional feature density detection is performed on the multiple collection blocks respectively to generate multiple feature densities. Specifically, the SIFT feature extraction algorithm is applied to each collection block to extract key points or feature descriptors in the collection block, and the number of feature points or the density of feature descriptors in each collection block is counted as the feature density of the collection block. The feature density reflects the complexity and uniqueness of the content of the collection block. Step S130, respectively identify the complexity of color and texture features of the multiple collection blocks, and generate multiple complexity indices. Specifically, perform color and texture analysis on each collection block, use histogram analysis to quantify the complexity of color and texture, and calculate the complexity index of each collection block based on the color and texture analysis results. The complexity index is a numerical value that quantifies the complexity of color and texture in the collection block, reflecting the richness and variability of the content of the collection block. Step S140, combine the multiple feature densities and the multiple complexity indices, perform weight configuration, and generate the multiple block monitoring weights, wherein the block monitoring weight is proportional to the feature density and complexity index. Specifically, set a weight calculation rule, which takes the feature density and complexity index as input, generates the block monitoring weight by weighted summation, and assigns a block monitoring weight to each collection block according to the weight calculation rule, and the block monitoring weight is proportional to the feature density and complexity index, that is, the higher the feature density and the greater the complexity of the block, the higher the block monitoring weight. This implementation method allocates block monitoring weights by comprehensively considering the feature density and complexity of the collection blocks, which more accurately reflects the importance of each collection block in copyright monitoring.

[0023] Step S200, identify infringement features according to the type of the target digital collection and generate preset infringement features. Specifically, the type of the target digital collection is determined by file type detection and content analysis (such as texture and color distribution in images, and motion patterns in videos). According to the type, features related to the type are selected from the preset infringement feature library, such as common tampering methods, common copy and paste traces, etc.

[0024] In a possible implementation, infringement feature identification is performed according to the type of the target digital collection to generate a preset infringement feature. Step S200 further includes step S210, mining the infringement records of digital collections of the same type according to the type of the target digital collection to generate a collection infringement record data set. Specifically, the type of the target digital collection to be analyzed is determined, such as artwork, photographic work, music, video, etc., and a database storing digital collection infringement records is accessed. The database stores digital collection infringement records from various channels such as copyright protection agencies, legal litigation records, and network monitoring platforms. In the database, all relevant infringement records are screened out according to the type of the target digital collection, including detailed information such as description of the infringed work, information of the infringer, infringement method, and infringement time, and the screened infringement records are organized into a collection infringement record data set. Step S220, extracting infringement features from the collection infringement record data set to generate an infringement feature set. Specifically, a series of quantifiable features are defined according to common patterns of infringement behavior, such as the degree of modification of the infringing work, the form of infringement (such as direct copying, partial copying, adaptation, etc.), the time span of infringement, the scope of infringement dissemination, etc. By using technical means such as text analysis, image processing, and data analysis, these defined feature values ​​are extracted from the collection infringement record data set, and the extracted feature values ​​are organized into a feature set, where each feature corresponds to one or more specific values ​​or descriptions. Step S230, perform a similarity comparison on any two groups of features in the infringement feature set, cluster the features whose similarity is greater than a preset similarity, and generate the preset infringement features. Specifically, the cosine similarity calculation method is used to calculate the similarity of any two groups of features in the infringement feature set. A preset similarity threshold is set to determine whether two groups of features are similar enough to be considered as the same type of infringement features. Features with a similarity greater than a preset threshold are grouped together to form different infringement feature clusters. Based on the clustering results, representative preset infringement features are summarized and summarized, and these features are used for infringement identification and early warning. This implementation method ensures the pertinence and accuracy of the analysis results by mining infringement records based on the type of digital collections. It comprehensively reflects the diversity and complexity of infringement behaviors by defining and extracting a series of infringement features. It automatically summarizes representative preset infringement features through similarity comparison and cluster analysis, thereby improving the accuracy of the generation of preset infringement features, and thus improving the efficiency and accuracy of infringement identification.

[0025] Step S300: Based on the preset infringement features, call the content-aware hash algorithm library to perform infringement identification sensitivity test, and build an initial hash algorithm integration architecture according to the sensitivity test results. Specifically, according to the preset infringement features, select an algorithm (such as aHash, dHash, pHash, etc.) from the content-aware hash algorithm library, wherein the content-aware hash algorithm library is a library that brings together multiple hash algorithms, each hash algorithm is designed to identify specific features or changes in digital content, and has different applicable scenarios. For example, aHash is suitable for capturing global changes, dHash is suitable for detecting changes in edge features, and pHash is suitable for capturing the overall structure and features of an image. Test each algorithm to evaluate its recognition effect (such as sensitivity, false alarm rate, and false negative rate) in different infringement scenarios, select the algorithm combination with the best performance based on the test results, and build an initial hash algorithm integration architecture, that is, the initial hash algorithm integration architecture is a framework built based on the content-aware hash algorithm library, which selects and integrates multiple suitable hash algorithms according to the type of target digital collections, expected infringement features, and sensitivity test results, and this architecture is used to achieve efficient hash processing and fusion of digital collection blocks.

[0026] In a possible implementation, based on the preset infringement feature, the content-aware hash algorithm library is called to perform an infringement identification sensitivity test, and an initial hash algorithm integration architecture is constructed according to the sensitivity test result. Step S300 further includes step S310, step 1: extracting the first infringement feature from the preset infringement feature, and collecting a historical infringement sample data set with the first infringement feature as a constraint, wherein any set of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with a corresponding relationship. Specifically, any one feature is selected from the preset infringement feature set as the first infringement feature (such as watermark removal, cropping, rotation, etc. of the image). According to the selected first infringement feature, a data set containing historical genuine collection samples and historical infringing collection samples is collected from the digital collection infringement record database, and these samples have a clear corresponding relationship, that is, each pair of samples represents an authentic collection and its corresponding infringing copy. Step S320, step 2: using multiple content-aware hash algorithms in the content-aware hash algorithm library to perform hash processing and comparison on the historical genuine collection samples and the historical infringing collection samples, respectively, to generate multiple sets of hash value difference information. Specifically, multiple algorithms in the content-aware hash algorithm library are used to hash the samples of historical genuine collections and historical infringing collections respectively to generate their respective hash values. The hash values ​​of the same pair of samples (genuine and infringing) are compared, the differences between them are calculated, and multiple sets of hash value difference information are generated. Step S330, step three: multiple first infringement identification sensitivity indicators are generated according to the multiple sets of hash value difference information. Specifically, according to the multiple sets of hash value difference information, the sensitivity indicators of each hash algorithm when identifying the first infringement feature are calculated, including the average value, standard deviation, minimum / maximum difference, etc. of the difference rate, wherein the larger the hash value difference information, the higher the sensitivity of the hash algorithm. Step S340, step four: based on the multiple first infringement identification sensitivity indicators, the content-aware hash algorithm corresponding to the maximum indicator is selected, and mapped and associated with the first infringement feature to generate a first content-aware hash algorithm. Specifically, the hash algorithm with the highest infringement identification sensitivity indicator is selected as the best algorithm for the first infringement feature, and the algorithm is mapped and associated with the first infringement feature to generate a specific content-aware hash algorithm (first content-aware hash algorithm). Step S350, and so on, iteratively repeat steps 1 to 4 until the second infringement feature in the preset infringement feature is traversed to the Nth infringement feature, and the second content-aware hash algorithm is obtained to the Nth content-aware hash algorithm. Specifically, steps S310 to S340 are repeatedly performed for each feature in the preset infringement feature set until all preset infringement features are traversed, and each iteration generates a content-aware hash algorithm for a specific infringement feature.Step S360, generating the initial hash algorithm integrated architecture with the first content-aware hash algorithm, the second content-aware hash algorithm, and finally the Nth content-aware hash algorithm. Specifically, all content-aware hash algorithms for different infringement features are combined to form an initial hash algorithm integrated architecture, which includes multiple hash algorithms, each of which has the greatest sensitivity to a specific infringement feature. This implementation method improves the accuracy of infringement detection by performing a sensitivity test on each preset infringement feature and selecting the hash algorithm that best suits the feature.

[0027] Step S400, based on the multiple block monitoring weights, the initial hash algorithm integrated architecture is monitored and configured to generate a preset hash algorithm integrated architecture. Specifically, the block monitoring weights are mapped to the initial hash algorithm integrated architecture, different hash processing weights are assigned to each collection block, and the initial hash algorithm integrated architecture is fine-tuned according to the weight distribution to ensure that high-weight collection blocks receive more detailed hash processing.

[0028] Step S500, multi-mode hash fusion processing is performed on the multiple collection blocks through the preset hash algorithm integrated architecture to obtain multiple block hash values ​​and upload them to the blockchain. Specifically, the preset hash algorithm integrated architecture is applied to each collection block for multi-mode hash fusion processing to generate a unique block hash value, and the block hash value and related information (such as block monitoring weight, digital collection ID) are uploaded to the blockchain network using a blockchain client (such as an Ethereum wallet, a Hyperledger Fabric node), and smart contracts are used to achieve data immutability and traceability.

[0029] In a possible implementation, the plurality of collection blocks are subjected to multi-mode hash fusion processing through the preset hash algorithm integrated architecture, and a plurality of block hash values ​​are obtained and uploaded to the blockchain. Step S500 further includes step S510, wherein the plurality of collection blocks are subjected to feature processing in the pre-hashing stage through the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm in the preset hash algorithm integrated architecture, respectively, to generate a plurality of first pre-feature data sets, a plurality of second pre-feature data sets, and a plurality of Nth pre-feature data sets. Specifically, for each collection block, the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm are respectively applied to perform feature processing in the pre-hashing stage, and feature information useful for hash calculation, such as edges, corners, textures, etc., is extracted from the block, and these features constitute the pre-feature data set. Step S520, hash calculation is performed on the plurality of first pre-feature data sets, the plurality of second pre-feature data sets, and the plurality of Nth pre-feature data sets according to the monitoring attention configuration, to generate the plurality of block hash values. Specifically, the pre-feature data set generated in step S510 is weighted according to the monitoring weights of multiple blocks, so that the features of the high-weight area contribute more to the final hash value, thereby improving the sensitivity of the hash value to important content. Based on the weighted pre-feature data set, the corresponding hash algorithm is applied to perform hash calculations to generate block hash values, and the generated multiple block hash values ​​are uploaded to the blockchain for subsequent verification, retrieval, and copyright monitoring operations. This implementation method combines the advantages of multiple hash algorithms through multi-mode hash fusion processing, improves the accuracy of hash values, and makes hash values ​​more sensitive to changes in important content through monitoring attention configuration and weighted processing, which is conducive to quickly locating key information in copyright monitoring scenarios.

[0030] Step S600, establish a connection with a preset network platform, monitor and obtain a first detection collection according to the type of the target digital collection, and call the preset hash algorithm integrated architecture to perform multi-mode hash fusion processing to generate multiple detection hash values. Specifically, establish a connection with a preset network platform through technical means such as API or web crawler to obtain digital collection data on the platform. The preset network platform is an online platform that can provide digital collection-related services and allow data transmission, processing and analysis, such as an online market, a digital collection trading platform, a copyright monitoring center, etc. According to the type of the target digital collection, the same or similar type of collection is screened out from the preset network platform as the first detection collection. The first detection collection is the digital collection that is currently being monitored to check whether there is any infringement. The preset hash algorithm integrated architecture is applied to the screened first detection collection for hash processing to generate multiple detection hash values.

[0031] Step S700, perform a similarity comparison between the multiple block hash values ​​and the multiple detection hash values, and generate infringement monitoring reminder information. Specifically, compare the multiple block hash values ​​of the target digital collection stored on the blockchain with the multiple detection hash values ​​of the first detection collection obtained from the preset network platform one by one. Use a hash value similarity calculation library (such as hashlib, scipy.spatial.distance, etc. in Python) to calculate the similarity between each pair of hash values, set a similarity threshold to determine whether the two hash values ​​are similar enough to indicate possible infringement. If the similarity of a pair of hash values ​​exceeds this threshold, it is considered that there is a potential infringement relationship between the two collection blocks. For all hash value pairs whose similarity exceeds the threshold, collect relevant information (such as the collection block location, the source of the detection collection, the similarity score, etc.), and generate detailed infringement monitoring reminder information based on this information, including possible infringement locations, infringement degree assessments, and recommended follow-up measures. The embodiment of the present application adopts technical means such as dividing the target digital collection into blocks and assigning a monitoring weight to each block, identifying infringement characteristics according to the type of digital collection, generating preset infringement characteristics, performing sensitivity tests based on these characteristics, constructing and optimizing the initial hash algorithm integration architecture, performing multi-mode hash fusion processing on multiple collection blocks through the preset hash algorithm integration architecture, generating multiple block hash values, and uploading the generated block hash values ​​to the blockchain, thereby achieving the technical effect of accurately identifying and monitoring infringement of digital collections.

[0032] In a possible implementation, the multiple block hash values ​​are compared with the multiple detection hash values ​​for similarity, and infringement monitoring reminder information is generated. Step S700 further includes step S710, obtaining copyright transaction information based on the blockchain. Specifically, the blockchain network is connected, and the API or SDK of the blockchain is used to query the copyright transaction information related to the target digital collection. The copyright transaction information is a detailed record of the copyright transfer or authorization transaction of the target digital collection, which is an important basis for confirming the ownership of the copyright, including the transaction time, transaction parties, transaction amount, transaction content, etc. Step S720, the copyright transaction information is parsed to generate multiple permission features of multiple authorized parties. Specifically, the copyright transaction information is parsed to identify all authorized parties (i.e., the original owner of the copyright and subsequent authorized persons), and for each authorized party, corresponding permission features are generated according to its transaction information, including authorization scope (such as region, time limit), usage method (such as copying, distribution, display, etc.). Step S730, when the similarity between the multiple block hash values ​​and the multiple detection hash values ​​meets the preset infringement similarity, the first collection circulation information of the first detection collection is read. Specifically, if there is at least one set of hash values ​​whose similarity exceeds the preset infringement similarity threshold (such as 90%), it is considered that there may be infringement. After confirming the infringement, the circulation information related to the first detection collection (such as the current holder, circulation history, transaction records, etc.) is read. Step S740, based on the multiple authorized parties and the multiple permission characteristics, the circulation information of the first collection is verified. If the verification fails, the infringement monitoring reminder information is generated. Specifically, according to the circulation information of the first detection collection, check whether the current holder has legal authorization, traverse all authorized parties and their permission characteristics, and confirm whether the current holder's behavior exceeds the authorization scope. If any violation is found (such as unauthorized use, copying, distribution, etc.), the verification fails. In the case of verification failure, the infringement monitoring reminder information is generated and sent to the copyright owner or relevant institutions. This implementation method ensures the accuracy and effectiveness of infringement monitoring by parsing copyright transaction information and verifying the permission characteristics of the authorized party.

[0033] In a possible implementation, step S700 further includes step S750, reading the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index corresponding to the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm, respectively. Specifically, the infringement identification sensitivity index of each content-aware hash algorithm (the first to the Nth) is obtained from the configuration file. Step S760, weighting the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm is weighted based on the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index to obtain a weight allocation result. Specifically, according to the infringement identification sensitivity index of each algorithm, a weighted average strategy is adopted to allocate weights to them, and the weight reflects the importance of the algorithm in the fusion process. The result of the weight allocation is a weight vector, in which each element corresponds to the weight of a content-aware hash algorithm. Step S770, based on the weight allocation result, a fusion optimization network is established, connected to the preset hash algorithm integrated architecture, and the multiple block hash values ​​are merged to generate a preset fusion hash value. Specifically, based on the weight distribution result, a fusion optimization network is constructed. This network is a weighted summer for fusing the hash values ​​of multiple content-aware hash algorithms into a preset fusion hash value. The fusion optimization network is connected to the preset hash algorithm integrated architecture, and the fusion optimization network is used to fuse multiple block hash values ​​to generate a preset fusion hash value. Step S780, using the preset fusion hash value to perform infringement monitoring reminders on the first detected collection. Specifically, the preset fusion hash value is compared with the detection hash value of the first detected collection (generated through the same fusion process). If the similarity between the two exceeds a preset threshold, it is considered that there is potential infringement, and infringement monitoring reminder information is generated and sent to the copyright owner or relevant stakeholders. This implementation method fuses multiple block hash values ​​into a more representative preset fusion hash value through a fusion optimization network, thereby achieving comprehensive monitoring of the copyright status of digital collections and improving the accuracy of infringement monitoring.

[0034] In the above, refer to Figure 1 The following describes in detail the blockchain-based intelligent monitoring method for copyright of digital collections according to an embodiment of the present invention. Figure 2 The present invention describes a blockchain-based intelligent copyright monitoring platform for digital collections according to an embodiment of the present invention.

[0035] The blockchain-based digital collection copyright intelligent monitoring platform according to the embodiment of the present invention is used to solve the technical problem of insufficient monitoring accuracy in existing digital collection copyright monitoring, and achieve the technical effect of accurately identifying and monitoring digital collection infringement. The blockchain-based digital collection copyright intelligent monitoring platform includes: a block processing module 10, an infringement feature recognition module 20, an infringement recognition sensitivity test module 30, a monitoring attention configuration module 40, a multi-mode hash fusion processing module 50, a plurality of detection hash value generation modules 60, and an infringement monitoring reminder information generation module 70.

[0036] The block processing module 10 is used to perform block processing on the target digital collection, generate multiple collection blocks and multiple block monitoring weights; the infringement feature identification module 20 is used to perform infringement feature identification according to the type of the target digital collection, and generate a preset infringement feature; the infringement identification sensitivity test module 30 is used to call the content-aware hash algorithm library to perform infringement identification sensitivity test based on the preset infringement feature, and build an initial hash algorithm integrated architecture according to the sensitivity test result; the monitoring attention configuration module 40 is used to perform monitoring attention configuration on the initial hash algorithm integrated architecture based on the multiple block monitoring weights, and generate a preset hash algorithm integrated architecture; the multi-mode hash fusion processing module 50 is used to perform multi-mode hash fusion processing on the multiple collection blocks through the preset hash algorithm integrated architecture, obtain multiple block hash values ​​and upload them to the blockchain; the multiple detection hash value generation module 60 is used to establish a connection with a preset network platform, monitor and obtain a first detection collection according to the type of the target digital collection, and call the preset hash algorithm integrated architecture to perform multi-mode hash fusion processing to generate multiple detection hash values; the infringement monitoring reminder information generation module 70 is used to perform similarity comparison between the multiple block hash values ​​and the multiple detection hash values, and generate infringement monitoring reminder information.

[0037] The specific configuration of the block processing module 10 will be described in detail below. As described above, the target digital collection is processed in blocks to generate multiple collection blocks and multiple block monitoring weights. The block processing module 10 may further include: a block processing unit for processing the target digital collection in blocks according to a predetermined grid division to generate the multiple collection blocks; a regional feature density detection unit for performing regional feature density detection on the multiple collection blocks to generate multiple feature densities; a complexity identification unit for performing complexity identification of color and texture features on the multiple collection blocks to generate multiple complexity indicators; a weight configuration unit for combining the multiple feature densities and the multiple complexity indicators to perform weight configuration to generate the multiple block monitoring weights, wherein the block monitoring weight is proportional to the feature density and the complexity indicator.

[0038] The specific configuration of the infringement feature identification module 20 will be described in detail below. As described above, the infringement feature identification is performed according to the type of the target digital collection to generate a preset infringement feature. The infringement feature identification module 20 may further include: an infringement record mining unit for mining the infringement records of digital collections of the same type according to the type of the target digital collection to generate a collection infringement record data set; an infringement feature extraction unit for extracting infringement features from the collection infringement record data set to generate an infringement feature set; a feature clustering unit for performing a similarity comparison on any two sets of features in the infringement feature set, clustering features with a similarity greater than a preset similarity, and generating the preset infringement feature.

[0039] The specific configuration of the infringement identification sensitivity test module 30 will be described in detail below. As described above, based on the preset infringement features, the content-aware hash algorithm library is called to perform infringement identification sensitivity testing, and the initial hash algorithm integration architecture is constructed according to the sensitivity test results. The infringement identification sensitivity test module 30 may further include: Step 1: the execution unit is used to extract the first infringement feature in the preset infringement features, and collect historical infringement sample data sets based on the first infringement feature as a constraint, wherein any group of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 2: the execution unit is used to use a plurality of content-aware hash algorithms in the content-aware hash algorithm library to perform hash processing and comparison on the historical genuine collection samples and the historical infringing collection samples, respectively, to generate multiple groups of hash value difference information; Step 3: the execution unit is used to extract the first infringement feature in the preset infringement features, and collect historical infringement sample data sets based on the first infringement feature as a constraint, wherein any group of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 4: the execution unit is used to perform hash processing and comparison on the historical genuine collection samples and the historical infringing collection samples, respectively, to generate multiple groups of hash value difference information; Step 5: the execution unit is used to extract the first infringement feature in the preset infringement features, and collect historical infringement sample data sets based on the first infringement feature as a constraint, wherein any group of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 6: the execution unit is used to extract the first infringement feature in the preset infringement features, and collect historical infringement sample data sets based on the first infringement feature as a constraint, wherein any group of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 7: the execution unit is used to extract the first infringement feature in the preset infringement features, and collect historical infringement sample data sets based on the first infringement feature as a constraint, wherein any group of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 8: the execution unit is used to extract the first The execution unit of step three is used to generate multiple first infringement identification sensitivity indicators according to the multiple groups of hash value difference information; the execution unit of step four is used to screen the content-aware hash algorithm corresponding to the maximum indicator based on the multiple first infringement identification sensitivity indicators, and map and associate it with the first infringement feature to generate a first content-aware hash algorithm; the iteration unit is used to iteratively repeat steps one to four by analogy, until the second infringement feature to the Nth infringement feature in the preset infringement feature is traversed, and the second content-aware hash algorithm to the Nth content-aware hash algorithm is obtained; the initial hash algorithm integrated architecture generation unit is used to generate the initial hash algorithm integrated architecture with the first content-aware hash algorithm, the second content-aware hash algorithm to the Nth content-aware hash algorithm.

[0040] The specific configuration of the multi-mode hash fusion processing module 50 will be described in detail below. As described above, the multi-mode hash fusion processing is performed on the multiple collection blocks through the preset hash algorithm integrated architecture, and multiple block hash values ​​are obtained and uploaded to the blockchain. The multi-mode hash fusion processing module 50 may further include: a pre-stage feature processing unit is used to perform feature processing of the pre-stage of hash processing on the multiple collection blocks respectively through the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm in the preset hash algorithm integrated architecture, and generate multiple first pre-feature data sets, multiple second pre-feature data sets, and multiple Nth pre-feature data sets; a hash calculation unit is used to perform hash calculation on the multiple first pre-feature data sets, multiple second pre-feature data sets, and multiple Nth pre-feature data sets according to the monitoring attention configuration to generate the multiple block hash values.

[0041] The specific configuration of the infringement monitoring reminder information generation module 70 will be described in detail below. As described above, the multiple block hash values ​​are compared with the multiple detection hash values ​​for similarity, and the infringement monitoring reminder information is generated. The infringement monitoring reminder information generation module 70 may further include: an infringement identification sensitivity index reading unit is used to read the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index corresponding to the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm respectively; a weight allocation unit is used to weight the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm based on the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index to obtain a weight allocation result; a fusion unit is used to establish a fusion optimization network based on the weight allocation result, connect with the preset hash algorithm integrated architecture, fuse the multiple block hash values, and generate a preset fusion hash value; an infringement monitoring reminder unit is used to perform infringement monitoring reminders on the first detection collection with the preset fusion hash value.

[0042] Among them, the infringement monitoring reminder information generation module 70 may further include: a copyright transaction information acquisition unit for acquiring copyright transaction information based on the blockchain; a copyright transaction information parsing unit for parsing the copyright transaction information and generating multiple permission characteristics of multiple authorized parties; a first collection circulation information reading unit for reading the first collection circulation information of the first detected collection when the similarity between the multiple block hash values ​​and the multiple detection hash values ​​meets the preset infringement similarity; a verification unit for verifying the first collection circulation information based on the multiple authorized parties and the multiple permission characteristics, and if the verification fails, generating the infringement monitoring reminder information.

[0043] The blockchain-based intelligent copyright monitoring platform for digital collections provided in the embodiments of the present invention can execute the blockchain-based intelligent copyright monitoring method for digital collections provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0044] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0045] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A blockchain-based intelligent copyright monitoring method for digital collections, characterized in that: include: Divide the target digital collection into blocks to generate multiple collection blocks and multiple block monitoring weights; Identify infringement features according to the type of the target digital collection and generate preset infringement features; Based on the preset infringement features, the content-aware hash algorithm library is called to perform an infringement identification sensitivity test, and an initial hash algorithm integration architecture is constructed according to the sensitivity test results; Performing monitoring attention configuration on the initial hash algorithm integrated architecture based on the multiple block monitoring weights to generate a preset hash algorithm integrated architecture; Perform multi-mode hash fusion processing on the multiple collection blocks through the preset hash algorithm integrated architecture to obtain multiple block hash values ​​and upload them to the blockchain; Establishing a connection with a preset network platform, monitoring and acquiring a first detection collection according to the type of the target digital collection, and calling the preset hash algorithm integration architecture to perform multi-mode hash fusion processing to generate multiple detection hash values; A similarity comparison is performed between the multiple block hash values ​​and the multiple detection hash values ​​to generate infringement monitoring reminder information.

2. The blockchain-based intelligent copyright monitoring method for digital collections according to claim 1, characterized in that: The target digital collection is divided into blocks to generate multiple collection blocks and multiple block monitoring weights, including: Divide the target digital collection into blocks according to a predetermined division grid to generate the plurality of collection blocks; Performing regional feature density detection on the multiple collection blocks respectively to generate multiple feature densities; Performing complexity identification of color and texture features on the plurality of collection blocks respectively to generate a plurality of complexity indicators; The plurality of feature densities and the plurality of complexity indices are combined to perform weight configuration to generate the plurality of block monitoring weights, wherein the block monitoring weight is proportional to the feature density and the complexity indices.

3. The blockchain-based intelligent copyright monitoring method for digital collections according to claim 1, characterized in that: Infringement characteristics are identified according to the type of the target digital collection, and preset infringement characteristics are generated, including: Mining infringement records of digital collections of the same type according to the type of the target digital collection to generate a collection infringement record data set; Extracting infringement features from the infringement record data set of the collection to generate an infringement feature set; A similarity comparison is performed on any two groups of features in the infringement feature set, and features with similarities greater than a preset similarity are clustered to generate the preset infringement features.

4. The blockchain-based intelligent copyright monitoring method for digital collections according to claim 1, characterized in that: Based on the preset infringement features, the content-aware hash algorithm library is called to perform infringement identification sensitivity testing, and an initial hash algorithm integration architecture is constructed according to the sensitivity test results, including: Step 1: extract the first infringement feature from the preset infringement features, and collect a historical infringement sample data set with the first infringement feature as a constraint, wherein any set of historical infringement sample data includes historical genuine collection samples and historical infringing collection samples with corresponding relationships; Step 2: using a plurality of content-aware hash algorithms in the content-aware hash algorithm library to perform hash processing and comparison on the samples of the historical genuine collections and the samples of the historical infringing collections, respectively, to generate a plurality of sets of hash value difference information; Step 3: generating a plurality of first infringement identification sensitivity indicators according to the plurality of sets of hash value difference information; Step 4: Based on the multiple first infringement identification sensitivity indicators, a content-aware hash algorithm corresponding to the maximum indicator is selected, and mapped and associated with the first infringement feature to generate a first content-aware hash algorithm; By analogy, iteratively repeat steps 1 to 4 until the second infringement feature to the Nth infringement feature in the preset infringement feature is traversed, and the second content-aware hash algorithm to the Nth content-aware hash algorithm is obtained; The initial hash algorithm integrated architecture is generated using the first content-aware hash algorithm, the second content-aware hash algorithm, and up to the Nth content-aware hash algorithm.

5. The blockchain-based intelligent copyright monitoring method for digital collections as claimed in claim 4 is characterized in that: The plurality of collection blocks are subjected to multi-mode hash fusion processing through the preset hash algorithm integrated architecture to obtain a plurality of block hash values ​​and upload them to the blockchain, including: The first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm in the preset hash algorithm integrated architecture are used to perform feature processing on the plurality of collection blocks in the pre-hashing stage, respectively, to generate a plurality of first pre-feature data sets, a plurality of second pre-feature data sets, and a plurality of Nth pre-feature data sets; According to the monitoring attention configuration, hash calculations are performed on the multiple first pre-feature data sets, the multiple second pre-feature data sets, and up to the multiple Nth pre-feature data sets to generate the multiple block hash values.

6. The blockchain-based intelligent copyright monitoring method for digital collections as claimed in claim 5, characterized in that: Also includes: Read the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index respectively corresponding to the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm; Based on the first infringement identification sensitivity index, the second infringement identification sensitivity index, and the Nth infringement identification sensitivity index, weights are assigned to the first content-aware hash algorithm, the second content-aware hash algorithm, and the Nth content-aware hash algorithm to obtain a weight assignment result; Based on the weight distribution result, a fusion optimization network is established, connected to the preset hash algorithm integrated architecture, and the hash values ​​of the multiple blocks are merged to generate a preset fusion hash value; The preset fusion hash value is used to perform infringement monitoring and reminder on the first detected collection.

7. The blockchain-based intelligent copyright monitoring method for digital collections according to claim 1, characterized in that: Performing a similarity comparison between the multiple block hash values ​​and the multiple detection hash values ​​to generate infringement monitoring reminder information includes: Based on the blockchain, obtaining copyright transaction information; Parsing the copyright transaction information to generate multiple permission features of multiple authorized parties; When the similarity between the plurality of block hash values ​​and the plurality of detection hash values ​​meets a preset infringement similarity, reading the first collection circulation information of the first detection collection; The circulation information of the first collection is verified based on the multiple authorized parties and the multiple authority features, and if the verification fails, the infringement monitoring reminder information is generated.

8. The blockchain-based digital collection copyright intelligent monitoring platform is characterized by: The platform is used to implement the blockchain-based intelligent monitoring method for copyright of digital collections according to any one of claims 1 to 7, and the platform includes: A block processing module, which is used to perform block processing on the target digital collection to generate multiple collection blocks and multiple block monitoring weights; An infringement feature identification module, the infringement feature identification module is used to identify infringement features according to the type of the target digital collection and generate a preset infringement feature; An infringement identification sensitivity test module, wherein the infringement identification sensitivity test module is used to call a content-aware hash algorithm library to perform an infringement identification sensitivity test based on the preset infringement features, and to construct an initial hash algorithm integration architecture according to the sensitivity test results; A monitoring attention configuration module, wherein the monitoring attention configuration module is used to perform monitoring attention configuration on the initial hash algorithm integrated architecture based on the multiple block monitoring weights to generate a preset hash algorithm integrated architecture; A multi-mode hash fusion processing module, which is used to perform multi-mode hash fusion processing on the multiple collection blocks through the preset hash algorithm integrated architecture, obtain multiple block hash values ​​and upload them to the blockchain; A plurality of detection hash value generation modules, wherein the plurality of detection hash value generation modules are used to establish a connection with a preset network platform, monitor and obtain a first detection collection according to the type of the target digital collection, and call the preset hash algorithm integration architecture to perform multi-mode hash fusion processing to generate a plurality of detection hash values; The infringement monitoring reminder information generation module is used to perform similarity comparison between the multiple block hash values ​​and the multiple detection hash values ​​to generate infringement monitoring reminder information.

Images