Database watermark management system and method, electronic equipment and storage medium

Abstract

The invention discloses a database watermark management system and method, electronic equipment and a storage medium, and relates to the technical field of information security, and the disclosed database watermark management system comprises an intelligent decision layer which is used for analyzing data features of input data and determining a watermark embedding mode of the input data according to an analysis result; the bimodal embedding layer is used for carrying out watermark embedding operation on first data of which the watermark embedding mode is a numeric embedding mode in the input data by adopting a preset improved least significant bit algorithm, and carrying out watermark embedding operation on second data of which the watermark embedding mode is a non-numeric embedding mode in the input data by adopting a preset improved least significant bit algorithm; carrying out watermark embedding on the second data by adopting a multi-system invisible character replacement technology; and the cross-modal traceability layer is used for analyzing the watermark information from the input data embedded with the watermark through a preset double-watermark collaborative verification mechanism. According to the method, the adaptation degree of database watermark embedding can be improved.

Description

Technical Field

[0001] This application relates to the field of information security technology, and in particular to database watermarking management systems, methods, electronic devices and storage media. Background Technology

[0002] With the rapid development of information technology, databases are being used more and more widely in various fields, which puts forward higher requirements for database data security.

[0003] Database watermarking technology refers to embedding watermarks in a database to prevent data from being copied and illegally disseminated, and to enable rapid tracing of data leaks. However, due to the diverse data types, large data volumes, and frequent data updates of databases, current database watermarking technologies are difficult to adapt to the watermark embedding needs of different types of data, resulting in insufficient protection and compromised data availability after watermarking.

[0004] In summary, improving the adaptability of database watermark embedding has become a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0005] The main objective of this application is to provide a database watermark management system, method, electronic device, and storage medium, which aims to improve the adaptability of database watermark embedding.

[0006] To achieve the above objectives, this application proposes a database watermarking management system, which includes:

[0007] The intelligent decision-making layer is used to analyze the data characteristics of the input data and determine the watermark embedding mode of the input data based on the analysis results.

[0008] A dual-modal embedding layer is used to perform watermark embedding operations on the first data in the input data where the watermark embedding mode is numerical, using a preset improved least significant bit algorithm; and to perform watermark embedding on the second data in the input data where the watermark embedding mode is non-numerical, using a multi-base invisible character replacement technique.

[0009] The cross-modal tracing layer is used to parse watermark information from the watermarked input data through a pre-defined dual-watermark collaborative verification mechanism.

[0010] In one embodiment, the intelligent decision-making layer includes:

[0011] A feature extraction unit is used to extract data features from the input data, wherein the data features include structural features, statistical features, and semantic features;

[0012] The decision unit is used to analyze the input data based on the data features using a preset hybrid decision tree to determine the watermark embedding pattern of the input data.

[0013] In one embodiment, the dual-modal embedding layer includes a numerical embedding unit for:

[0014] For the first data in the input data whose watermark embedding mode is numerical embedding mode, a preset improved least significant bit algorithm is used to perform watermark embedding operation on the first data. The improved least significant bit algorithm dynamically adjusts the embedding parameters according to the data characteristics of the first data, and embeds watermark information in the least significant bit of the first data through a dynamic perturbation function. At the same time, a distribution preservation correction mechanism is used to ensure that the distribution characteristics of the first data after watermark embedding remain unchanged.

[0015] In one embodiment, the dual-modal embedding layer includes non-numerical embedding units for:

[0016] For the second data in the input data whose watermark embedding mode is a non-numerical embedding mode, the watermark embedding position of the second data is determined based on the two-factor authentication mechanism and hash key, and the watermark information is converted into invisible characters and then embedded at the watermark embedding position.

[0017] In one embodiment, the cross-modal tracing layer includes a numerical tracing unit for:

[0018] The first data after embedding the watermark is denoised by wavelet transform to obtain the first denoised data.

[0019] The embedded first watermark information is extracted from the first denoised data, and the correlation between the first watermark information and the expected watermark is calculated to verify the effectiveness of the first watermark information.

[0020] In one embodiment, the cross-modal tracing layer includes a non-numerical tracing unit for:

[0021] The second data after embedding the watermark is used to detect control characters based on the Unicode code point traversal algorithm.

[0022] Based on the mapping relationship between control characters and number base values, the detected control characters are converted into a number base value sequence;

[0023] The watermark semantic information in the second data after embedding the watermark is recovered based on the given numerical sequence.

[0024] In one embodiment, the cross-modal tracing layer further includes:

[0025] A numerical compensation unit is used to reconstruct the watermark based on spatial correlation using statistical or interpolation algorithms when the watermark is missing in the first data after the watermark is embedded.

[0026] The non-numerical compensation unit is used to predict the watermark location based on the N-gram model (a statistical language model) by means of linguistic statistical features when the second data after the watermark is embedded is detected to have a missing watermark, and to determine the validity of the watermark by means of a Bayesian network.

[0027] Furthermore, to achieve the above objectives, this application also proposes a database watermark management method, applied to the database watermark management system described above. The database watermark management method includes:

[0028] The intelligent decision-making layer analyzes the data characteristics of the input data and determines the watermark embedding mode of the input data based on the analysis results.

[0029] Through a dual-modal embedding layer, for the first data in the input data where the watermark embedding mode is numerical, a preset improved least significant bit algorithm is used to perform watermark embedding on the first data. For the second data in the input data where the watermark embedding mode is non-numerical, a multi-base invisible character replacement technique is used to perform watermark embedding on the second data.

[0030] Through a cross-modal tracing layer, watermark information is parsed from the watermarked input data using a pre-defined dual-watermark collaborative verification mechanism.

[0031] In addition, to achieve the above objectives, this application also proposes an electronic device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the database watermark management method as described above.

[0032] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the database watermark management method described above.

[0033] This application proposes a database watermark management system, which includes: an intelligent decision layer for analyzing the data characteristics of input data and determining the watermark embedding mode of the input data based on the analysis results; a bimodal embedding layer for embedding watermarks into first data with a numerical watermark embedding mode using a preset improved least significant bit algorithm, and embedding watermarks into second data with a non-numerical watermark embedding mode using a multi-base invisible character replacement technique; and a cross-modal tracing layer for parsing watermark information from the watermarked input data through a preset bimodal watermark collaborative verification mechanism.

[0034] In summary, the database watermark management system of this application analyzes the data characteristics of the input data of the database through an intelligent decision layer and determines the appropriate watermark embedding mode. Combined with a dual-modal embedding layer, it adopts an improved least significant bit algorithm for numerical data and uses multi-base invisible character replacement technology for non-numerical data for watermark embedding. Then, through a cross-modal traceability layer, a dual-watermark collaborative verification mechanism is used to ensure the accuracy and reliability of watermark parsing. Overall, it improves the adaptability of database watermark embedding to different data types and ensures that watermarks can be embedded efficiently, stably and covertly in various database data scenarios. Attached Figure Description

[0035] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the system structure provided in Embodiment 1 of the database watermark management system of this application;

[0038] Figure 2 This is another system structure diagram provided for Embodiment 1 of the database watermark management system of this application;

[0039] Figure 3 This is a schematic diagram of the watermark management logic provided in Embodiment 1 of the database watermark management system of this application;

[0040] Figure 4 This is a flowchart illustrating the database watermark management method according to an embodiment of this application.

[0041] Figure 5This is a schematic diagram of the hardware operating environment involved in the database watermark management system in this application embodiment.

[0042] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0043] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0044] With the rapid development of information technology, databases are being used more and more widely in various fields, which puts forward higher requirements for database data security.

[0045] Database watermarking technology refers to embedding watermarks in a database to prevent data from being copied and illegally disseminated, and to enable rapid tracing of data leaks. However, due to the diverse data types, large data volumes, and frequent data updates of databases, current database watermarking technologies are difficult to adapt to the watermark embedding needs of different types of data, resulting in insufficient protection and compromised data availability after watermarking.

[0046] In summary, improving the adaptability of database watermark embedding has become a pressing technical problem that needs to be solved in this field.

[0047] This application provides a solution for a database watermarking management system, comprising: an intelligent decision layer for analyzing the data characteristics of input data and determining the watermark embedding mode of the input data based on the analysis results; a bimodal embedding layer for embedding watermarks into first data with a numerical watermark embedding mode using a preset improved least significant bit algorithm, and embedding watermarks into second data with a non-numerical watermark embedding mode using a multi-base invisible character replacement technique; and a cross-modal tracing layer for parsing watermark information from the watermarked input data through a preset bimodal watermark collaborative verification mechanism. This application improves the adaptability of database watermark embedding.

[0048] In summary, the database watermark management system of this application analyzes the data characteristics of the input data of the database through an intelligent decision layer and determines the appropriate watermark embedding mode. Combined with a dual-modal embedding layer, it adopts an improved least significant bit algorithm for numerical data and uses multi-base invisible character replacement technology for non-numerical data to embed the watermark. Then, through a cross-modal tracing layer, a dual-watermark collaborative verification mechanism is used to ensure the accuracy and reliability of watermark parsing. Overall, it improves the adaptability of database watermark embedding to different data types and ensures that watermarks can be embedded efficiently, stably and covertly in various database data scenarios.

[0049] Based on this, embodiments of this application provide a database watermark management system, referring to... Figure 1 , Figure 1 This is a schematic diagram of the system structure of the first embodiment of the database watermark management system of this application.

[0050] In this embodiment, the database watermark management system includes:

[0051] The intelligent decision-making layer is used to analyze the data characteristics of the input data and determine the watermark embedding mode of the input data based on the analysis results.

[0052] It should be noted that the input data is the data in the database to be watermarked. Watermark management specifically includes watermark generation, embedding, and traceability operations.

[0053] In this embodiment, the intelligent decision layer is used to analyze the data characteristics of the input data, including but not limited to data type identification, distribution pattern detection, and business weight evaluation. Based on the analysis results, the intelligent decision layer can determine the watermark embedding mode of the input data, that is, determine whether the input data is suitable for a numerical embedding mode and / or a non-numerical embedding mode.

[0054] It should also be noted that data type identification refers to determining whether the input data is numeric (such as integers, floating-point numbers) or non-numeric (such as text, characters). This can be determined by checking the field types (such as INT, VARCHAR, TEXT, etc.) of the input data column and the format of the data content. Distribution pattern detection refers to analyzing the distribution of data in the data column, including statistical characteristics such as central tendency and dispersion, in order to select an appropriate watermark embedding strategy and strength, avoiding excessive impact on the statistical characteristics of the data. This can be achieved by calculating statistical indicators of the data column, such as mean (μ) and variance (σ). 2 The watermark is determined by factors such as kurtosis (K); business weight assessment refers to assigning weights to data columns based on the importance and sensitivity of input data in the business scenario, in order to adjust the strength and strategy of watermark embedding, so as to protect the integrity and privacy of key data. This can be determined by combining domain knowledge and business needs.

[0055] A dual-modal embedding layer is used to perform watermark embedding operations on the first data in the input data where the watermark embedding mode is numerical, using a preset improved least significant bit algorithm; and to perform watermark embedding on the second data in the input data where the watermark embedding mode is non-numerical, using a multi-base invisible character replacement technique.

[0056] In this embodiment, the bimodal embedding layer performs corresponding watermark embedding operations on the input data according to the watermark embedding pattern determined by the intelligent decision layer. Specifically, for the first data with a numerical watermark embedding pattern, the bimodal embedding layer uses a preset improved Least Significant Bit (LSB) algorithm for watermark embedding. It should be noted that the improved LSB algorithm, through a dynamic perturbation function and a distribution-preserving correction mechanism, can maintain the original characteristics and usability of the data to the greatest extent while embedding the watermark.

[0057] For the second data with a non-numerical embedding mode for watermark embedding, the bimodal embedding layer uses a multi-base invisible character replacement technique for watermark embedding. Specifically, the characters controlling the watermark content can be converted into invisible characters, and the embedding position can be dynamically calculated based on the modulo operation and string length to achieve seamless watermark embedding while ensuring the watermark's invisibility and robustness.

[0058] The cross-modal tracing layer is used to parse watermark information from the watermarked input data through a pre-defined dual-watermark collaborative verification mechanism.

[0059] In this embodiment, after receiving a watermark tracing request for the input data, the cross-modal tracing layer uses a preset dual-watermark collaborative verification mechanism to comprehensively verify the data after embedding the watermark by combining the characteristics of numerical and non-numerical watermarks. This verification mechanism can effectively improve the accuracy and reliability of watermark detection, and can accurately extract and verify watermark information even when the input data is subject to a certain degree of interference or attack.

[0060] It should be noted that the dual-watermark collaborative verification mechanism refers to using two different types of watermarks (numerical watermarks and non-numerical watermarks) simultaneously when verifying the validity of watermarked input data. By combining the characteristics and advantages of the two watermarks, the accuracy and reliability of verification are improved, while the credibility of data traceability is enhanced.

[0061] Specifically, for input data that contains both numerical and non-numerical data, numerical and non-numerical watermarks are embedded in different parts of the input data, and they are independent of each other. During the watermark verification or watermark tracing stage, these two types of watermarks are extracted from the input data and verified. If both watermarks pass the verification, the integrity of the watermark can be determined. If one watermark is damaged or cannot be extracted, the other watermark can still provide valid tracing information. Therefore, even if some data is tampered with or damaged, as long as one watermark can be successfully extracted and verified, watermark tracing can be performed, making it more resistant to attacks.

[0062] In one feasible embodiment, such as Figure 2 As shown, the intelligent decision-making layer includes:

[0063] The feature extraction unit is used to extract data features from the input data, including structural features, statistical features, and semantic features.

[0064] In this embodiment, the feature extraction unit is used to extract data features from the input data. The extracted data features include structural features, statistical features, and semantic features. Among them, structural features mainly reflect the organization and basic attributes of the input data, such as data type (e.g., numeric, text), the proportion of unique values, and the null value rate; statistical features reflect the numerical distribution and central tendency of the input data, such as the variance and kurtosis of the numerical column and the word frequency entropy of the text column; semantic features reflect the meaning and importance of the input data in a specific business domain, such as the sensitivity score of the data field based on the domain dictionary.

[0065] The decision unit is used to analyze the input data based on data features through a pre-defined hybrid decision tree in order to determine the watermark embedding pattern of the input data.

[0066] In this embodiment, the decision-making unit performs in-depth analysis of the input data based on the data features extracted by the feature extraction unit through a preset hybrid decision tree. The hybrid decision tree integrates multiple factors such as data feature entropy and business semantic weights, breaking through the limitation of traditional decision tree algorithms (such as C4.5) that only rely on information gain rate.

[0067] Based on the analysis results of the hybrid decision tree, the decision unit comprehensively considers various factors of data characteristics and finally determines the watermark embedding mode of the input data, that is, whether the data is more suitable for numerical watermark embedding mode or non-numerical watermark embedding mode, thus providing clear guidance for the watermark embedding operation of the dual-modal embedding layer.

[0068] For example, a database table contains two fields: age and description. The data type of age is INT (integer type), and the data type of description is VARCHAR (variable-length string type). The feature extraction unit extracts the data features of each field, including structural features, statistical features, and semantic features. In a feasible implementation scenario, the decision-making rule of the decision-making unit is: if the data type is INT and the proportion of unique values ​​is <0.5, then the numeric watermark embedding mode is selected; if the data type is INT and the proportion of unique values ​​is ≥0.5, but the variance is <100, then... If the data type is INT, and the unique value ratio is ≥0.5 with a variance ≥100, then a non-numerical watermark embedding mode is chosen because the data distribution is too scattered, and numerical watermarks may not be stable enough. If the data type is VARCHAR, and the unique value ratio is ≥0.8 with an average length L>50, then a non-numerical watermark embedding mode is chosen. If the data type is VARCHAR, and the unique value ratio is <0.8 or the average length L≤50, then a numerical watermark embedding mode is chosen because the text content is relatively simple, and numerical watermarks can be used to improve robustness. Then, based on the hybrid decision tree analysis of this decision rule and the data characteristics of each field, it is determined that the field 'age' uses a numerical watermark embedding mode, and the field 'description' uses a non-numerical watermark embedding mode.

[0069] Therefore, by comprehensively considering various factors of data characteristics, such as data type, proportion of unique values, variance, average length, word frequency entropy, and sensitivity score, the intelligent decision layer can select the most suitable watermark embedding mode for each field in the database, thereby improving the flexibility and adaptability of watermark embedding, ensuring data integrity and availability, and enhancing the robustness and invisibility of the watermark.

[0070] In one feasible embodiment, such as Figure 2 As shown, the bimodal embedding layer includes numerical embedding units for:

[0071] For the first data in the input data where the watermark embedding mode is numerical, a preset improved least significant bit algorithm is used to perform watermark embedding on the first data. The improved least significant bit algorithm dynamically adjusts the embedding parameters according to the data characteristics of the first data and embeds watermark information in the least significant bit of the first data through a dynamic perturbation function. At the same time, a distribution preservation correction mechanism is used to ensure that the distribution characteristics of the first data after watermark embedding remain unchanged.

[0072] In this embodiment, the numerical embedding unit performs watermark embedding operation on the first data in the input data whose watermark embedding mode is numerical embedding mode. Specifically, it adopts a preset improved least significant bit algorithm. This algorithm can dynamically adjust the embedding parameters according to the data characteristics of the first data to adapt to different data characteristics, improve the flexibility and robustness of watermark embedding, and embed watermark information in the least significant bit of the first data through a dynamic perturbation function to achieve effective watermark embedding. At the same time, a distribution-preserving correction mechanism is used to ensure that the distribution characteristics of the first data after watermark embedding remain unchanged, thereby minimizing the impact on the original characteristics and usability of the data while embedding the watermark.

[0073] It should be noted that the improved least significant bit algorithm proposed in this embodiment is used to embed watermark information in numerical data in a database. This watermark embedding process includes the following steps:

[0074] Dynamically adjust embedding parameters: The watermark embedding parameters are dynamically adjusted based on data characteristics (such as data distribution, variance, mean, etc.). These parameters may include watermark strength, embedding bit length, etc. For example, if the data variance is large, a smaller embedding strength can be selected to avoid excessively affecting the statistical characteristics of the data; if the data variance is small, the embedding strength can be appropriately increased to improve the robustness of the watermark.

[0075] Least significant bit embedding: Watermark information is embedded in the least significant bit of numerical data. The least significant bit refers to the least important bit of a value. For example, in the binary representation of an integer, the least significant bit is the least significant bit. By embedding the watermark in the least significant bit, the watermark information can be hidden while minimizing the impact on the original data.

[0076] Dynamic perturbation function: To further improve the robustness and security of watermarks, a dynamic perturbation function is used to adjust the embedding process. The dynamic perturbation function can dynamically adjust the watermark embedding method and intensity according to the specific characteristics of the data. For example, the perturbation intensity can be reduced for values ​​near the edge of the data distribution to avoid pushing these values ​​into unreasonable ranges.

[0077] Distribution Preservation Correction Mechanism: After embedding the watermark, the distribution preservation correction mechanism ensures that the distribution characteristics of the data remain unchanged. That is, it ensures that the data after embedding the watermark is statistically consistent with the original data (such as mean, variance, distribution shape, etc.). For example, if the mean of the original data is 30 and the variance is 100, after embedding the watermark, the mean and variance of the data should be kept as close as possible to the range of 30 and 100.

[0078] In one feasible embodiment, such as Figure 2 As shown, the bimodal embedding layer includes non-numerical embedding units for:

[0079] For the second data in the input data where the watermark embedding mode is a non-numerical embedding mode, the watermark embedding position of the second data is determined based on the two-factor authentication mechanism and hash key, and the watermark information is converted into invisible characters and then embedded at the watermark embedding position.

[0080] In this embodiment, the non-numerical embedding unit performs watermark embedding operation on the second data in the input data whose watermark embedding mode is non-numerical. Specifically, the non-numerical embedding unit determines the watermark embedding position of the second data based on a two-factor authentication mechanism and a hash key. The two-factor authentication mechanism enhances the security of watermark embedding, while the hash key ensures the accuracy and uniqueness of the embedding position. Then, the watermark information is converted into invisible characters and embedded at the determined watermark embedding position. By utilizing the characteristics of invisible characters, the watermark does not affect the normal display and use of the data after embedding, while improving the concealment and invisibility of the watermark.

[0081] It should be noted that in this embodiment, the two-factor authentication mechanism refers to verifying the legitimacy of an identity or operation by combining two independent factors. This ensures that only users who possess both specific keys can correctly embed or extract the watermark. These two factors can be a hash key K1 (used to generate the hash value for the watermark content and embedding position) and a hash key K2 (used to determine the specific embedding position of the watermark). The embedding position of the watermark can be calculated using the hash key and the hash value generated from the text data, thus achieving randomization and unpredictability of the watermark position. Furthermore, to ensure the watermark's concealment, the watermark information is converted into invisible characters. Invisible characters refer to characters that are not displayed or are difficult to detect in the text, such as control words in Unicode. These characters are not noticeable to the user when the data is displayed, but can be extracted using specific detection algorithms.

[0082] In one feasible embodiment, such as Figure 2 As shown, the cross-modal tracing layer includes numerical tracing units, used for:

[0083] The first data after embedding the watermark is denoised by wavelet transform to obtain the first denoised data.

[0084] The embedded first watermark information is extracted from the first denoised data, and the correlation between the first watermark information and the expected watermark is calculated to verify the effectiveness of the first watermark information.

[0085] In this embodiment, the numerical tracing unit is used to verify the tracing of the first data after embedding the watermark. First, the first data is denoised using wavelet transform to remove potential noise interference, resulting in denoised first data. Wavelet transform is a signal processing technique that can effectively separate noise components in the first data, improving the accuracy of watermark extraction. Then, the embedded first watermark information is extracted from the denoised first data, and the correlation between the extracted watermark information and the expected watermark (i.e., the watermark generated during watermark embedding) is calculated. Through correlation analysis, the similarity between the extracted watermark information and the expected watermark can be quantitatively evaluated, thereby verifying the validity of the watermark information in the first data. If the correlation reaches a preset threshold, the watermark verification is considered successful, and the tracing information of the data is reliable.

[0086] In one feasible embodiment, such as Figure 2 As shown, the cross-modal tracing layer includes non-numerical tracing units, used for:

[0087] The second data after embedding the watermark is used to detect control characters based on the Unicode code point traversal algorithm.

[0088] Based on the mapping relationship between control characters and number base values, the detected control characters are converted into a number base value sequence;

[0089] The semantic information of the watermark in the second data after embedding the watermark is recovered based on the hexadecimal numerical sequence.

[0090] In this embodiment, the non-numerical tracing unit is used to perform tracing verification on the second data after embedding the watermark. First, control characters are detected in the second data based on the Unicode code point traversal algorithm. The Unicode code point traversal algorithm can efficiently identify invisible control characters embedded in text data. These control characters are the carriers of embedded watermark information. Then, based on the preset mapping relationship between control characters and base values, the detected control characters are converted into corresponding base value sequences. The mapping relationship between control characters and base values ​​is predefined in the watermark embedding process and is used to encode the watermark information into invisible characters and perform reverse decoding during tracing. Finally, by decoding the base value sequence, the embedded watermark information is restored, thereby realizing the recovery and verification of tracing information such as data source and copyright.

[0091] In one feasible embodiment, such as Figure 2 As shown, the cross-modal tracing layer also includes:

[0092] A numerical compensation unit is used to reconstruct the watermark based on spatial correlation using statistical or interpolation algorithms when the watermark is missing in the first data after the watermark is embedded.

[0093] In this embodiment, when the numerical compensation unit detects that the first data after embedding the watermark has a missing watermark, it analyzes the relationship between adjacent data points based on the spatial correlation of the data, and uses statistical or interpolation algorithms to calculate and supplement the missing watermark information, thereby reconstructing the complete watermark information and improving the integrity of the watermark and the accuracy of tracing its origin.

[0094] After reconstructing the watermark, the watermark information is extracted again and compared with the expected watermark information to verify the reconstruction effect. If the correlation between the reconstructed watermark and the expected watermark reaches a certain threshold, the reconstruction is considered successful.

[0095] The non-numerical compensation unit is used to predict the watermark location based on the N-gram model using linguistic statistical features when a watermark missing is detected in the second data after the watermark is embedded, and to determine the validity of the watermark using a Bayesian network.

[0096] In this embodiment, when the non-numerical compensation unit detects a missing watermark in the second data after embedding the watermark, it predicts the watermark location based on the N-gram model using linguistic statistical features, and compensates for and verifies the missing watermark information. The N-gram model is a linguistic statistical method that can predict the possible location of the missing watermark information based on the sequence features of characters or words in the text data. Then, the validity of the watermark is determined through a Bayesian network. The Bayesian network is a probabilistic inference model that can comprehensively consider multiple factors (such as the prior probability of watermark embedding, statistical features of the data, etc.) to evaluate the validity of the predicted watermark information. Through Bayesian inference, the posterior probability of the watermark information under given data conditions is calculated, thereby judging the validity and reliability of the watermark.

[0097] For example, to facilitate understanding of the database watermark management process in the above embodiments, please refer to... Figure 3 , Figure 3This diagram illustrates the logic of database watermark management. When data input to the database is detected, an intelligent decision layer analyzes the data to determine the watermark embedding mode. Then, a dual-modal embedding layer uses a numerical watermark embedding mode for numerical data (the first data), specifically employing an improved least significant bit algorithm. For non-numerical data (the second data), a non-numerical watermark embedding mode is used, specifically employing invisible character replacement. This results in watermarked data. When tracing the source of this data is required, a cross-modal tracing layer performs watermark detection. Wavelet transform is applied to detected numerical data, and Unicode code point traversal is performed on detected non-numerical data. A confidence fusion algorithm is then used for verification. If verification passes, the watermark tracing result is directly output. If a watermark attack is detected, an anti-attack compensation strategy is implemented. For numerical data, watermark reconstruction is performed using statistical or super-value methods, and for textual data, N-gram prediction is used. Finally, the tracing result is obtained.

[0098] Thus, the database watermark management system of this application analyzes the data characteristics of the input data of the database through the intelligent decision layer and determines the appropriate watermark embedding mode. Combined with the dual-modal embedding layer, it adopts an improved least significant bit algorithm for numerical data and uses multi-base invisible character replacement technology for non-numerical data to embed the watermark. Then, through the cross-modal traceability layer, the accuracy and reliability of watermark parsing are ensured by the dual-watermark collaborative verification mechanism. Overall, it improves the adaptability of database watermark embedding to different data types and ensures that the watermark can be embedded efficiently, stably and covertly in various database data scenarios.

[0099] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. On this basis, this embodiment proposes a database watermark management method, referring to... Figure 4 , Figure 4 This is a flowchart illustrating the database watermark management method of this application.

[0100] This database watermark management method is applied to a database watermark management system as described in Embodiment 1 above. The database watermark management method includes steps S10 to S30:

[0101] Step S10: The intelligent decision-making layer analyzes the data characteristics of the input data and determines the watermark embedding mode of the input data based on the analysis results.

[0102] The intelligent decision-making layer analyzes the data characteristics of the input data, including but not limited to data type identification, distribution pattern detection, and business weight evaluation. Based on the analysis results, the intelligent decision-making layer can determine the watermark embedding mode of the input data, that is, determine whether the input data is suitable for a numerical embedding mode and / or a non-numerical embedding mode.

[0103] Step S20: Through the dual-modal embedding layer, for the first data in the input data where the watermark embedding mode is numerical, a preset improved least significant bit algorithm is used to perform watermark embedding operation on the first data; for the second data in the input data where the watermark embedding mode is non-numerical, a multi-base invisible character replacement technique is used to perform watermark embedding on the second data.

[0104] The bimodal embedding layer performs watermark embedding operations on the input data according to the watermark embedding pattern determined by the intelligent decision layer. Specifically, for the first data with a numerical watermark embedding pattern, the bimodal embedding layer uses a preset improved Least Significant Bit (LSB) algorithm for watermark embedding. It should be noted that the improved LSB algorithm, through a dynamic perturbation function and a distribution-preserving correction mechanism, can maintain the original characteristics and usability of the data to the greatest extent while embedding the watermark.

[0105] For the second data with a non-numerical embedding mode for watermark embedding, the bimodal embedding layer uses a multi-base invisible character replacement technique for watermark embedding. Specifically, the characters controlling the watermark content can be converted into invisible characters, and the embedding position can be dynamically calculated based on the modulo operation and string length to achieve seamless watermark embedding while ensuring the watermark's invisibility and robustness.

[0106] Step S30: Through the cross-modal tracing layer, the watermark information is parsed from the input data after the watermark is embedded through the preset dual watermark collaborative verification mechanism.

[0107] Upon receiving a watermark tracing request for input data, the cross-modal tracing layer uses a pre-defined dual-watermark collaborative verification mechanism. This mechanism combines the characteristics of numerical and non-numerical watermarks to comprehensively verify the watermarked data. This verification mechanism effectively improves the accuracy and reliability of watermark detection, and can accurately extract and verify watermark information even when the input data is subject to a certain degree of interference or attack.

[0108] Thus, the database watermark management system of this application analyzes the data characteristics of the input data of the database through the intelligent decision layer and determines the appropriate watermark embedding mode. Combined with the dual-modal embedding layer, it adopts an improved least significant bit algorithm for numerical data and uses multi-base invisible character replacement technology for non-numerical data to embed the watermark. Then, through the cross-modal traceability layer, the accuracy and reliability of watermark parsing are ensured by the dual-watermark collaborative verification mechanism. Overall, it improves the adaptability of database watermark embedding to different data types and ensures that the watermark can be embedded efficiently, stably and covertly in various database data scenarios.

[0109] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the database watermark management system of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0110] This application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to execute the database watermark management system in the first embodiment described above.

[0111] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic devices in these embodiments may include, but are not limited to, mobile terminals such as laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0112] like Figure 5As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. The communication device 1009 allows the electronic device to exchange data with other devices wirelessly or via wired communication. Although the diagrams show electronic devices with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented alternatively.

[0113] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0114] The electronic device provided in this application, employing the database watermark management system described in the above embodiments, can improve the adaptability of database watermark embedding. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the database watermark management system provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the database watermark management system of the previous embodiment, and will not be repeated here.

[0115] It should be understood that the various parts disclosed in the embodiments of this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0116] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0117] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the database watermark management system described in the above embodiments.

[0118] The computer-readable storage medium provided in this application embodiment may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0119] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0120] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by an electronic device, the electronic device causes the following: Through an intelligent decision-making layer, it analyzes the data characteristics of the input data and determines the watermark embedding mode of the input data based on the analysis results; Through a dual-modal embedding layer, for the first data in the input data where the watermark embedding mode is numerical, it performs watermark embedding operations on the first data using a preset improved least significant bit algorithm; for the second data in the input data where the watermark embedding mode is non-numerical, it performs watermark embedding on the second data using a multi-base invisible character replacement technique; Through a cross-modal tracing layer, it parses the watermark information from the watermarked input data using a preset dual-watermark collaborative verification mechanism.

[0121] Computer program code for performing the operations of the embodiments of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0122] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0123] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0124] The readable storage medium provided in this application embodiment is a computer-readable storage medium. This computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for executing the aforementioned database watermark management system, which can improve the adaptability of database watermark embedding. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application embodiment are the same as the beneficial effects of the database watermark management system provided in the above embodiments, and will not be repeated here.

[0125] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the database watermark management method described above.

[0126] The computer program product provided in this application can improve the adaptability of database watermark embedding. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the database watermark management system provided in the above embodiments, and will not be repeated here.

[0127] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A database watermark management system, characterized in that, The database watermark management system includes: The intelligent decision-making layer is used to analyze the data characteristics of the input data and determine the watermark embedding mode of the input data based on the analysis results. A dual-modal embedding layer is used to perform watermark embedding operations on the first data in the input data where the watermark embedding mode is numerical, using a preset improved least significant bit algorithm; and to perform watermark embedding on the second data in the input data where the watermark embedding mode is non-numerical, using a multi-base invisible character replacement technique. The cross-modal tracing layer is used to parse watermark information from the watermarked input data through a pre-defined dual-watermark collaborative verification mechanism.

2. The database watermark management system as described in claim 1, characterized in that, The intelligent decision-making layer includes: A feature extraction unit is used to extract data features from the input data, wherein the data features include structural features, statistical features, and semantic features; The decision unit is used to analyze the input data based on the data features using a preset hybrid decision tree to determine the watermark embedding pattern of the input data.

3. The database watermark management system as described in claim 1, characterized in that, The dual-modal embedding layer includes numerical embedding units for: For the first data in the input data whose watermark embedding mode is numerical embedding mode, a preset improved least significant bit algorithm is used to perform watermark embedding operation on the first data. The improved least significant bit algorithm dynamically adjusts the embedding parameters according to the data characteristics of the first data, and embeds watermark information in the least significant bit of the first data through a dynamic perturbation function. At the same time, a distribution preservation correction mechanism is used to ensure that the distribution characteristics of the first data after watermark embedding remain unchanged.

4. The database watermark management system as described in claim 1, characterized in that, The dual-modal embedding layer includes non-numerical embedding units for: For the second data in the input data whose watermark embedding mode is a non-numerical embedding mode, the watermark embedding position of the second data is determined based on the two-factor authentication mechanism and hash key, and the watermark information is converted into invisible characters and then embedded at the watermark embedding position.

5. The database watermark management system as described in claim 1, characterized in that, The cross-modal tracing layer includes numerical tracing units for: The first data after embedding the watermark is denoised by wavelet transform to obtain the first denoised data. The embedded first watermark information is extracted from the first denoised data, and the correlation between the first watermark information and the expected watermark is calculated to verify the effectiveness of the first watermark information.

6. The database watermark management system as described in claim 1, characterized in that, The cross-modal tracing layer includes non-numerical tracing units for: The second data after embedding the watermark is used to detect control characters based on the Unicode code point traversal algorithm. Based on the mapping relationship between control characters and number base values, the detected control characters are converted into a number base value sequence; The watermark semantic information in the second data after embedding the watermark is recovered based on the given numerical sequence.

7. The database watermark management system as described in claim 1, characterized in that, The cross-modal tracing layer also includes: A numerical compensation unit is used to reconstruct the watermark based on spatial correlation using statistical or interpolation algorithms when the watermark is missing in the first data after the watermark is embedded. The non-numerical compensation unit is used to predict the watermark location based on the N-gram model using linguistic statistical features when a watermark missing is detected in the second data after the watermark is embedded, and to determine the validity of the watermark using a Bayesian network.

8. A database watermark management method, characterized in that, The database watermark management method is applied to the database watermark management system as described in any one of claims 1 to 7, and the database watermark management method includes: The intelligent decision-making layer analyzes the data characteristics of the input data and determines the watermark embedding mode of the input data based on the analysis results. Through a dual-modal embedding layer, for the first data in the input data where the watermark embedding mode is numerical, a preset improved least significant bit algorithm is used to perform watermark embedding on the first data. For the second data in the input data where the watermark embedding mode is non-numerical, a multi-base invisible character replacement technique is used to perform watermark embedding on the second data. Through a cross-modal tracing layer, watermark information is parsed from the input data after watermarking is embedded through a pre-set dual-watermark collaborative verification mechanism.

9. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the database watermark management method as described in claim 8.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the database watermark management method as described in claim 8.

Images