Electronic report anti-counterfeiting method by utilizing digital watermark embedding

Abstract

The invention discloses an electronic report anti-counterfeiting method by utilizing digital watermark embedding, which relates to the technical field of data anti-counterfeiting, and comprises the following steps: acquiring an electronic report to be subjected to anti-counterfeiting processing, extracting report characteristics, and calling historical anti-counterfeiting record data of the electronic report of the same family; performing anti-counterfeiting weight distribution and quality weight distribution according to the historical anti-counterfeiting record data to obtain an anti-counterfeiting weight and a quality weight; according to the anti-counterfeiting weight and the quality weight, the digital watermark embedding parameter is optimized, the optimal digital watermark embedding parameter is obtained, and the anti-counterfeiting fitness is calculated by adopting the anti-counterfeiting weight and the quality weight for optimization; and adopting the optimal digital watermark embedding parameter to carry out digital watermark embedding anti-counterfeiting processing on the electronic report. The technical problem that in the prior art, the anti-counterfeiting effect and the data quality cannot be effectively balanced, so that the anti-counterfeiting effect is poor is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data forgery prevention, in particular to an electronic report forgery prevention method using digital watermark embedding. BACKGROUND

[0002] As an important means of electronic report forgery prevention, digital watermark technology has been widely used in various types of electronic document protection. The existing technology usually uses pre-defined rules or fixed parameter sets to embed digital watermarks, such as setting the same watermark strength and embedding position for all reports of a specific category. Although this method is simple to implement, it gradually exposes its limitations in actual application. There are many types of electronic reports, and the security threats faced by different categories of reports differ significantly. The content characteristics and data format of the report also have different requirements for the embedding effect of the watermark. Using static and unified forgery prevention strategies cannot adapt to this dynamic and complex environment, which may lead to high-risk reports being easily forged due to insufficient forgery prevention strength, while low-risk reports suffer unnecessary quality loss due to excessive watermark embedding. At the same time, the inherent contradiction between forgery prevention capability and report quality protection in the watermark embedding process has not been accurately balanced, often relying on manual experience to adjust parameters, which not only is inefficient, but also makes it difficult to ensure the optimality of the results. The above defects lead to the need to improve the adaptability, intelligence level and overall performance of the existing electronic report forgery prevention method. SUMMARY

[0003] The present application provides an electronic report forgery prevention method using digital watermark embedding, which is used to solve the technical problem that the forgery prevention effect and data quality cannot be effectively balanced in the prior art, resulting in poor forgery prevention effect.

[0004] In view of the above problems, the present application provides an electronic report forgery prevention method using digital watermark embedding, which comprises: Obtaining an electronic report to be subjected to forgery prevention processing, extracting report features, and calling historical forgery prevention record data of the same family of electronic reports; According to the historical forgery prevention record data, performing forgery prevention weight distribution and quality weight distribution to obtain forgery prevention weight and quality weight, wherein the weight distribution includes forgery frequency analysis distribution and report change analysis distribution; According to the forgery prevention weight and the quality weight, optimizing the digital watermark embedding parameters to obtain optimal digital watermark embedding parameters, wherein the forgery prevention fitness is calculated using the forgery prevention weight and the quality weight for optimization; Using the optimal digital watermark embedding parameters, performing digital watermark embedding forgery prevention processing on the electronic report.

[0005] Another embodiment of the present application also provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, when the computer program is executed by the processor, the steps of the anti-counterfeiting method for electronic report embedded with digital watermark provided by the present application are implemented.

[0006] Another embodiment of the present application also provides a computer readable storage medium item, comprising a stored computer program, when the computer program is running, the device where the computer readable storage medium is located is controlled to execute the steps of the anti-counterfeiting method for electronic report embedded with digital watermark provided by the present application.

[0007] The one or more technical solutions provided in the present application have at least the following technical effects or advantages: The present application provides an anti-counterfeiting method for electronic report embedded with digital watermark, by obtaining the electronic report to be processed and extracting the report features, and simultaneously calling the historical anti-counterfeiting record data of the electronic report of the same family, and then dynamically distributing the anti-counterfeiting weight and the quality weight according to the counterfeit record and the report change information in the historical data, and optimizing the digital watermark embedding parameters based on these weights, and finally completing the watermark embedding with the optimal parameters, the anti-counterfeiting ability and the overall quality stability of the electronic report are significantly improved. Compared with the traditional method, the technical solution provided by the present application significantly improves the adaptability and intelligent level of the anti-counterfeiting process, achieves the technical effect of effectively improving the anti-counterfeiting reliability of the electronic report while maintaining the original quality of the report in complex environment, and ensures the sustainability and scalability of the watermark embedding through the adaptive mechanism driven by historical data, which is suitable for various electronic report scenarios, and provides a more efficient and reliable solution for anti-counterfeiting processing. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced, and obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0009] Figure 1 A flowchart of an anti-counterfeiting method for electronic report embedded with digital watermark provided by an embodiment of the present application.

[0010] Figure 2 A flowchart of anti-counterfeiting weight distribution and quality weight distribution in an anti-counterfeiting method for electronic report embedded with digital watermark provided by an embodiment of the present application. DETAILED DESCRIPTION

[0011] The application provides an anti-counterfeiting method for an electronic report embedded with a digital watermark.

[0012] The technical solutions in the embodiments of the application will be clearly and completely described in connection with the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0013] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units need not be limited to only those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to the process, method, product or device.

[0014] As shown in the embodiments, Figure 1 The application provides an anti-counterfeiting method for an electronic report embedded with a digital watermark, and the method comprises the following steps: S10: obtaining an electronic report to be subjected to anti-counterfeiting processing, extracting report features, and calling historical anti-counterfeiting record data of electronic reports of the same family.

[0015] In the digital watermark anti-counterfeiting processing of the electronic report, the formulation of the anti-counterfeiting strategy lacks sufficient information support. The existing method usually adopts a "one-size-fits-all" strategy, that is, the same or similar anti-counterfeiting strength and watermark embedding mode are applied to all electronic reports, while the actual counterfeiting risk faced by electronic reports of different categories and different purposes and the possible significant differences in the content characteristics of the electronic reports are ignored. This indiscriminate processing mode leads to low efficiency of the configuration of anti-counterfeiting resources, and cannot achieve the key protection of high-risk reports, and may also cause unnecessary quality loss to low-risk reports.

[0016] The step S10 in the method provided by the embodiments of the application comprises the following steps: Obtaining an electronic report to be subjected to anti-counterfeiting processing; Extracting report features of the electronic report, wherein the report features comprise a report category; According to the report features, extracting all electronic reports of the same family with the same report features within a historical time, and historical digital watermark embedding parameters and counterfeiting record parameters of each electronic report of the same family, and integrating to obtain historical anti-counterfeiting record data, wherein the counterfeiting record parameters comprise whether there is counterfeiting.

[0017] In the embodiments of the present application, an electronic report to be subjected to anti-counterfeiting processing is acquired, wherein the electronic report to be subjected to anti-counterfeiting processing can have multiple formats, such as doc, pdf, and the like.

[0018] A report feature of the electronic report is extracted, wherein the report feature includes a report category. Illustratively, key feature extraction can be performed on the title and content of the report, for example, the report title contains the keyword “communication equipment operation report”, and the content of the electronic report is related parameters of communication equipment operation, such as communication protocol type, communication signal strength, and communication channel parameters, and the like, and the key feature of the report is extracted as “communication equipment operation report”.

[0019] According to the report feature, all the same family electronic reports having the same report feature in the historical time are extracted, and the historical digital watermark embedding parameters and the forgery record parameters of each same family electronic report are extracted, and the historical anti-counterfeiting record data is obtained by integration, wherein the forgery record parameters include whether there is forgery. Specifically, the electronic reports in the historical time are queried, all the same family electronic reports having the same report feature are screened, for example, all the electronic reports of the “communication equipment operation report” category are screened and extracted, and the communication equipment operation reports using the same format template are screened as the same family electronic reports of the electronic report to be processed. The corresponding historical digital watermark embedding parameters, such as watermark length, are acquired. The corresponding forgery record parameters are acquired, wherein the forgery record parameters include whether there is forgery, such as “there is forgery” and “there is no forgery”.

[0020] By acquiring the historical anti-counterfeiting record data, a key change from isolated decision to data-driven intelligent decision of anti-counterfeiting processing is realized. Specifically, by extracting the features of the current electronic report and calling the historical anti-counterfeiting record data of the same family report accordingly, a rich and relevant information base is constructed for subsequent anti-counterfeiting decision. This not only enables the anti-counterfeiting method to recognize the category to which the current report belongs, but also enables the anti-counterfeiting method to obtain the security status of the report in history, for example, whether it belongs to a high-forgery type, thereby providing a solid data support for subsequent differentiated and refined anti-counterfeiting resource allocation.

[0021] S20: According to the historical anti-counterfeiting record data, anti-counterfeiting weight distribution and quality weight distribution are performed to obtain anti-counterfeiting weights and quality weights, wherein the weight distribution includes counterfeit times analysis distribution and report change analysis distribution. Specifically, it needs to be clear which of the two possible conflicting goals, maximizing anti-counterfeiting effect and minimizing the impact on report quality, should be inclined. The traditional method often relies on manually setting fixed weights or making simple binary judgments, and cannot dynamically and finely reflect the actual risk level faced by different report groups and their sensitivity to quality changes. For example, for the report category that has been frequently counterfeited historically, it should be given a higher anti-counterfeiting priority; and for those reports whose file size itself fluctuates more, their tolerance to the slight changes caused by watermark embedding may be higher.

[0022] After obtaining the historical anti-counterfeiting record data of the same family electronic report, how to extract key indicators from these data that can effectively guide the current anti-counterfeiting decision and quantify them into operational parameters is a technical problem to be solved. The historical data itself is complex, and if it cannot be effectively analyzed and quantified, it cannot be converted into actual decision-making ability.

[0023] The step S20 in the method provided by the embodiments of the present application, as shown in Figure 2 , includes: extracting the counterfeit record parameters in the historical anti-counterfeiting record data, statistically obtaining the counterfeit times, performing counterfeit times analysis distribution, obtaining first anti-counterfeiting weights and first quality weights; Among them, extracting the counterfeit record parameters in the historical anti-counterfeiting record data, statistically obtaining the counterfeit times, performing counterfeit times analysis distribution, obtaining first anti-counterfeiting weights and first quality weights, includes: According to the counterfeit times, the counterfeit rate is calculated and obtained; According to the counterfeit rate, the first anti-counterfeiting weight is configured, and the first quality weight is calculated and obtained; extracting the report size information of all same family electronic reports in the historical anti-counterfeiting record data, performing report change analysis distribution, obtaining second anti-counterfeiting weights and second quality weights; Among them, extracting the report size information of all same family electronic reports in the historical anti-counterfeiting record data, performing report change analysis distribution, obtaining second anti-counterfeiting weights and second quality weights, includes: extracting the report size information of all same family electronic reports in the historical anti-counterfeiting record data to obtain a same family report size information set; According to the same family report size information set, the report size fluctuation amplitude is calculated and obtained; According to the report size fluctuation amplitude, the second quality weight is configured, and the second anti-counterfeiting weight is calculated and obtained; According to the first anti-counterfeiting weight, the first quality weight, the second anti-counterfeiting weight and the second quality weight, the anti-counterfeiting weight and the quality weight are calculated and obtained.

[0024] In the embodiment of the application, the counterfeiting record parameters in the historical anti-counterfeiting record data are extracted, the number of counterfeiting times is counted and obtained, the number of counterfeiting times is analyzed and distributed, and the first anti-counterfeiting weight and the first quality weight are obtained.

[0025] Specifically, according to the number of counterfeiting times, the counterfeiting rate is calculated and obtained. The counterfeiting rate = the number of existing counterfeiting reports ÷ the total number of electronic reports of the same family. For example, the number of existing counterfeiting reports is 5, and the total number of electronic reports of the same family is 100, then the counterfeiting rate = 5 ÷ 100 = 0.05. The more the number of electronic reports of the same family is counterfeited, the greater the counterfeiting rate is, and the more encryption is needed.

[0026] According to the counterfeiting rate, the first anti-counterfeiting weight is configured, and the first quality weight is calculated and obtained. Specifically, the first anti-counterfeiting weight = the counterfeiting rate × the amplification coefficient. The amplification coefficient is used to amplify the influence of the counterfeiting rate, so as to prevent the first anti-counterfeiting weight from being too small due to the too small counterfeiting rate, which affects the subsequent steps. Illustratively, the amplification coefficient can be set to 3. Further, when the counterfeiting rate is large, the counterfeiting rate can not be amplified, for example, when the counterfeiting rate is greater than or equal to 0.2, the counterfeiting rate is not amplified, and the counterfeiting rate is directly taken as the first anti-counterfeiting weight. Illustratively, when the counterfeiting rate is 0.05, the first anti-counterfeiting weight = 0.05 × 3 = 0.15, and when the counterfeiting rate is 0.2, the first anti-counterfeiting weight = 0.2. According to the first anti-counterfeiting weight, the first quality weight is obtained, and the first quality weight = 1 - the first anti-counterfeiting weight. For example, the first anti-counterfeiting weight is 0.15, then the first quality weight = 1 - 0.15 = 0.85.

[0027] The report size information of all electronic reports of the same family in the historical anti-counterfeiting record data is extracted, the report change analysis distribution is performed, and the second anti-counterfeiting weight and the second quality weight are obtained.

[0028] Specifically, the report size information of all electronic reports of the same family in the historical anti-counterfeiting record data is extracted, and the same family report size information set is obtained. The report size information of the electronic report can be obtained by the file size of the electronic report, for example, five report size information with sizes [150KB, 155KB, 148KB, 160KB, 152KB] can be collected.

[0029] The report size fluctuation range is calculated based on the family of report sizes. For example, first, the mean report size in the family of report sizes is calculated, then the standard deviation of the report sizes in the family of report sizes is calculated, and finally, the standard deviation is divided by the mean to obtain the report size fluctuation range. For instance, if the report sizes are [150KB, 155KB, 148KB, 160KB, 152KB], the calculated mean is 153KB, and the standard deviation is 4.20KB. Therefore, the report size fluctuation range = 4.19 ÷ 153 = 0.03.

[0030] Based on the fluctuation range of the report size, a second quality weight is configured, and a second anti-counterfeiting weight is calculated. The second quality weight equals the fluctuation range of the report size. The greater the fluctuation range, the greater the imperceptibility of anti-counterfeiting measures, and the more important the report data quality; thus, the larger the second quality weight. Based on the second quality weight, the second anti-counterfeiting weight is calculated as: Second Anti-counterfeiting Weight = 1 - Second Quality Weight. For example, if the second quality weight is 0.03, then the second anti-counterfeiting weight = 1 - 0.03 = 0.97.

[0031] The anti-counterfeiting weight and quality weight are calculated based on the first anti-counterfeiting weight, the first quality weight, the second anti-counterfeiting weight, and the second quality weight. For example, the anti-counterfeiting weight = (first anti-counterfeiting weight + second anti-counterfeiting weight) ÷ 2, and the quality weight = (first quality weight + second quality weight) ÷ 2. For instance, if the first anti-counterfeiting weight is 0.15, the first quality weight is 0.85, the second anti-counterfeiting weight is 0.97, and the second quality weight is 0.03, then the anti-counterfeiting weight = (0.15 + 0.97) ÷ 2 = 0.56, and the quality weight = (0.85 + 0.03) ÷ 2 = 0.44. The anti-counterfeiting weight and quality weight obtained through weighted processing can more accurately reflect the anti-counterfeiting and quality requirements of electronic archives, providing a more precise decision-making basis for subsequent digital watermark embedding parameters of electronic archives.

[0032] By introducing counterfeit frequency analysis and allocation, high-risk report categories can be automatically identified, and these categories can be assigned higher anti-counterfeiting weights, thereby significantly improving the targeting and effectiveness of anti-counterfeiting measures. Simultaneously, by analyzing report changes, the sensitivity of reports to quality changes can be assessed, allowing for more flexible adjustment of watermark parameters. The allocation process of quality weights and anti-counterfeiting weights is essentially a process of quantifying and strategizing historical experience, providing a clear objective for subsequent optimization algorithms—that is, simultaneously pursuing strong anti-counterfeiting effects with high anti-counterfeiting weight requirements and maintaining quality under high quality weight requirements. This provides a precise and quantitative decision-making basis for ultimately achieving the optimal balance between anti-counterfeiting and quality.

[0033] S30: Optimize the digital watermark embedding parameters according to the anti-counterfeiting weight and quality weight to obtain the optimal digital watermark embedding parameters, wherein the anti-counterfeiting fitness is calculated using the anti-counterfeiting weight and quality weight for optimization.

[0034] The embedding parameters of a digital watermark directly determine the strength of its anti-counterfeiting capabilities and its impact on the quality of the original report; there is usually a trade-off between these two aspects. Traditional parameter setting methods often rely on empirical formulas or exhaustive trial and error, which is not only inefficient but also difficult to guarantee finding the globally optimal solution. More importantly, they lack a unified optimization objective that can simultaneously respond to both anti-counterfeiting and quality requirements. Without a scientific optimization objective, it is impossible to effectively combine the quality weights and anti-counterfeiting weights obtained from the aforementioned steps to derive a specific and executable digital watermark embedding scheme.

[0035] Step S30 in the method provided in this application embodiment includes: The first watermark embedding parameters are randomly set, wherein the first watermark embedding parameters include the length of the digital watermark; The first watermark embedding parameter is input into the anti-counterfeiting predictor, the first anti-counterfeiting rate is obtained by outputting the first report quality parameter, and the first anti-counterfeiting fitness is calculated. The training steps for the anti-counterfeiting predictor include: Based on the historical anti-counterfeiting data of electronic reports, a set of sample watermark embedding parameters was collected, and the proportion of anti-counterfeiting completed under different sample watermark embedding parameters was collected and labeled as the sample anti-counterfeiting rate set. Using the sample watermark embedding parameter set and sample anti-counterfeiting rate set as supervised training data, an anti-counterfeiting predictor is trained based on machine learning. The process of obtaining the first report quality parameters and calculating the first anti-counterfeiting fitness includes: Get the historical maximum watermark embedding parameters; The ratio of the historical maximum watermark embedding parameter to the first watermark embedding parameter is calculated and used as the first report quality parameter; Continue optimizing until convergence, and obtain the optimal digital watermark embedding parameters with the greatest anti-counterfeiting adaptability.

[0036] In this embodiment, a first watermark embedding parameter is randomly set, wherein the first watermark embedding parameter includes a digital watermark length. For example, a random number generator is used to randomly generate a digital watermark length within the range of digital watermark lengths, which is then used as the first watermark length. For instance, the range of digital watermark lengths can be set from 128 bits to 1024 bits.

[0037] The first watermark embedding parameter is input into the anti-counterfeiting predictor, the first anti-counterfeiting rate is output, the first report quality parameter is processed, and the first anti-counterfeiting fitness is calculated.

[0038] Specifically, first, obtain the anti-counterfeiting predictor.

[0039] The training steps for the anti-counterfeiting predictor include: Based on historical anti-counterfeiting data of electronic reports, a set of sample watermark embedding parameters is collected, and the proportion of anti-counterfeiting completed under different sample watermark embedding parameters is collected and labeled as the sample anti-counterfeiting rate set. The anti-counterfeiting completion rate is calculated as follows: the number of electronic reports with embedded watermarks and an anti-counterfeiting record parameter of "no counterfeiting" ÷ the total number of electronic reports in the same family. For example, if the number of electronic reports with embedded watermarks and an anti-counterfeiting record parameter of "no" is 85, and the total number of electronic reports in the same family is 100, then the anti-counterfeiting completion rate is 85 ÷ 100 = 0.85, i.e., the anti-counterfeiting rate is 0.85.

[0040] Based on machine learning, an anti-counterfeiting predictor is constructed. For example, an anti-counterfeiting predictor is constructed based on a decision tree, with a maximum depth of 5 layers, using mean squared error as the loss function, and a minimum number of samples per leaf node set to 2.

[0041] The anti-counterfeiting predictor is trained in a supervised manner using a set of sample watermark embedding parameters as input and a set of sample anti-counterfeiting rates as supervision, until convergence. Specifically, the set of sample watermark embedding parameters and the set of sample anti-counterfeiting rates are divided, with 80% used as the training dataset and 20% as the validation dataset. The anti-counterfeiting predictor is trained in a supervised manner using data from the training dataset, and validated using the validation dataset after a preset number of training iterations. For example, if a sample watermark embedding parameter not used in the training is input, and the error between the output anti-counterfeiting rate and the sample anti-counterfeiting rate is within ±3%, the training is considered complete, and the anti-counterfeiting predictor is obtained.

[0042] The first watermark embedding parameter is input into the anti-counterfeiting predictor, and the first anti-counterfeiting rate is output. The first report quality parameter is obtained through processing, and the first anti-counterfeiting fitness is calculated.

[0043] Specifically, based on historical anti-counterfeiting record data, the watermark embedding parameters are iterated through to obtain the historical maximum watermark embedding parameter.

[0044] The ratio of the historical maximum watermark embedding parameter to the first watermark embedding parameter is calculated and used as the first report quality parameter. For example, if the maximum watermark embedding parameter is 640 bits and the first watermark embedding parameter is 512 bits, then the first report quality parameter = 640 ÷ 512 = 1.25.

[0045] Based on the first report quality parameter, the anti-counterfeiting adaptability is obtained. For example, the anti-counterfeiting adaptability can be obtained by weighting the anti-counterfeiting rate and the first report quality parameter. The weight can be set according to the requirements of the anti-counterfeiting rate and report quality. For example, for the "Communication Equipment Operation Report," where security is more important, a larger anti-counterfeiting weight is set, such as 0.7. Then, the quality weight = 1 - 0.7 = 0.3. The anti-counterfeiting adaptability = anti-counterfeiting weight × anti-counterfeiting rate + quality weight × first report quality parameter. When the anti-counterfeiting rate is 0.85 and the first report quality parameter is 1.25, the anti-counterfeiting adaptability = 0.7 × 0.85 + 0.3 × 1.25 = 0.97. A higher anti-counterfeiting adaptability indicates a better anti-counterfeiting effect, meaning that while ensuring the anti-counterfeiting effect, the loss in report quality is also smaller.

[0046] Continue to obtain the first watermark embedding parameters and perform iterative optimization. For example, obtain the second watermark embedding parameters based on the neighborhood of the first watermark embedding parameters, obtain and compare the anti-counterfeiting fitness of the two, and select the watermark embedding parameter with the larger anti-counterfeiting fitness for the next iterative optimization until convergence. For example, if no larger anti-counterfeiting fitness is obtained after 20 consecutive iterations, the current digital watermark embedding parameters are taken as the optimal digital watermark embedding parameters to obtain the optimal digital watermark embedding parameters with the largest anti-counterfeiting fitness.

[0047] By utilizing preset report transmission characteristics for digital watermark impairment prediction and report quality prediction, this step enables a proactive assessment of the potential consequences of various parameter combinations before actual watermark embedding, significantly improving the efficiency and accuracy of optimization. Furthermore, by combining anti-counterfeiting weights and quality weights to calculate a unified anti-counterfeiting fitness as the optimization objective, the optimization process no longer solely pursues maximizing anti-counterfeiting effectiveness or minimizing quality impact, but rather dynamically seeks the optimal balance between the two. When the anti-counterfeiting weight is high, the optimization algorithm tends to seek parameters that better enhance anti-counterfeiting capabilities; when the quality weight is high, it focuses more on protecting report quality. This optimization mechanism ensures that the final optimal digital watermark embedding parameters are not only theoretically optimal, but also the most practical and customized solution closely aligned with the specific risk conditions and quality requirements of current electronic reports.

[0048] S40: Using the optimal digital watermark embedding parameters, perform digital watermark embedding anti-counterfeiting processing on the electronic report.

[0049] In this embodiment, optimal digital watermark embedding parameters are used to perform digital watermark embedding anti-counterfeiting processing on the electronic report. For example, if the optimal digital watermark embedding parameter is 512 bits, the watermark information is added to the document structure as an invisible annotation or metadata tag for embedding. The embedded digital watermark is 512 bits long and is used for anti-counterfeiting. By using optimal digital watermark embedding parameters for anti-counterfeiting processing, it is ensured that the electronic report is endowed with anti-counterfeiting features that are highly matched with its anti-counterfeiting and quality requirements. The final generated electronic report with electronic watermark not only carries strong targeted anti-counterfeiting capabilities based on historical data analysis, but also maintains its original data quality and usability to the maximum extent, achieving a harmonious unity between anti-counterfeiting strength and report quality.

[0050] In summary, the embodiments of this application have at least the following technical effects: This application proposes an electronic report anti-counterfeiting method using digital watermark embedding. It acquires the electronic report to be processed and extracts its report features, while simultaneously accessing historical anti-counterfeiting record data from related electronic reports. Then, it dynamically allocates anti-counterfeiting and quality weights based on forgery records and report change information in the historical data. Based on these weights, it optimizes the digital watermark embedding parameters, ultimately using the optimal parameters to complete the watermark embedding, significantly improving the anti-counterfeiting capability and overall quality stability of the electronic report. Specifically, traditional anti-counterfeiting methods often rely on static watermark parameters, which cannot be dynamically adjusted according to the actual risks and characteristics of the electronic report. This results in inconsistent anti-counterfeiting effects when facing different forgery situations or report types, and may also affect the visual quality or usability of the report due to excessive watermark embedding. This application analyzes the historical anti-counterfeiting records of related electronic reports, including factors such as the number of forgeries and changes in report size, and automatically allocates anti-counterfeiting and quality weights. This balances anti-counterfeiting and quality requirements when optimizing watermark parameters, making the watermark embedding more suitable for practical application scenarios. Specifically, the allocation of anti-counterfeiting weights takes into account the frequency of forgery, thereby strengthening anti-counterfeiting capabilities in high-risk reports. The allocation of quality weights, on the other hand, is based on the fluctuation range of report size, ensuring that watermark embedding does not unnecessarily interfere with the report structure. During parameter optimization, this application utilizes an anti-counterfeiting predictor to predict watermark damage and report quality, combined with anti-counterfeiting fitness calculations, iteratively adjusting parameters such as watermark length until the optimal solution with the best anti-counterfeiting effect and minimal quality impact is found. This method not only improves the accuracy of watermark embedding but also enhances the processing capability for diverse electronic reports, avoiding the subjectivity and errors of manual intervention. Compared to traditional methods, the technical solution provided in this application significantly improves the adaptability and intelligence level of anti-counterfeiting processing, achieving the technical effect of effectively improving the reliability of electronic report anti-counterfeiting while maintaining the original quality of the report in complex environments. Through a historical data-driven adaptive mechanism, the sustainability and scalability of watermark embedding are ensured, making it applicable to various electronic report scenarios and providing a more efficient and reliable solution for anti-counterfeiting processing.

[0051] Another embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the electronic report anti-counterfeiting method using digital watermark embedding according to any embodiment of the present invention.

[0052] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more module units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0053] The electronic device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The electronic device may include, but is not limited to, a processor and a memory.

[0054] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0055] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the electronic report anti-counterfeiting method using digital watermark embedding described in any of the above-described method embodiments of the present invention.

[0056] The modules / units integrated in the aforementioned device / electronic device, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0057] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0058] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0059] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for preventing counterfeiting in electronic reports using embedded digital watermarks, characterized in that, The method includes: Obtain the electronic report to be processed for anti-counterfeiting, extract the report features, and retrieve historical anti-counterfeiting record data of the same electronic report family; Based on the historical anti-counterfeiting record data, anti-counterfeiting weight allocation and quality weight allocation are performed to obtain anti-counterfeiting weight and quality weight. The weight allocation includes allocation based on the number of counterfeit cases and allocation based on the report change analysis. Based on the anti-counterfeiting weight and quality weight, the digital watermark embedding parameters are optimized to obtain the optimal digital watermark embedding parameters. Among them, the anti-counterfeiting adaptability is calculated using the anti-counterfeiting weight and quality weight for optimization. The electronic report is subjected to digital watermark embedding anti-counterfeiting processing using the optimal digital watermark embedding parameters.

2. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 1, characterized in that, Obtain the electronic report to be processed for anti-counterfeiting, extract report features, and retrieve historical anti-counterfeiting record data from similar electronic reports, including: Obtain the electronic report to be processed for anti-counterfeiting; Extract the report features of the electronic report, wherein the report features include the report category; Based on the reported characteristics, all family-type electronic reports with the same reported characteristics within a historical period are extracted, along with the historical digital watermark embedding parameters and forgery record parameters for each family-type electronic report. These are then integrated to obtain historical anti-counterfeiting record data, wherein the forgery record parameters include whether forgery exists.

3. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 1, characterized in that, Based on the historical anti-counterfeiting record data, anti-counterfeiting weight allocation and quality weight allocation are performed to obtain anti-counterfeiting weight and quality weight, including: Extract the counterfeit record parameters from the historical anti-counterfeiting record data, count the number of counterfeits, perform counterfeits count analysis and allocation, and obtain the first anti-counterfeiting weight and the first quality weight. Extract the report size information of all family-type electronic reports from the historical anti-counterfeiting record data, perform report change analysis and allocation, and obtain the second anti-counterfeiting weight and the second quality weight. The anti-counterfeiting weight and quality weight are calculated based on the first anti-counterfeiting weight, the first quality weight, the second anti-counterfeiting weight, and the second quality weight.

4. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 3, characterized in that, Extract the counterfeit record parameters from the historical anti-counterfeiting record data, count the number of counterfeits, perform counterfeit count analysis and allocation, and obtain the first anti-counterfeiting weight and the first quality weight, including: The forgery rate is calculated based on the number of forgeries. Based on the counterfeiting rate, a first anti-counterfeiting weight is configured, and a first quality weight is calculated.

5. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 3, characterized in that, Extract the report size information of all family-specific electronic reports from the historical anti-counterfeiting record data, perform report change analysis and allocation, and obtain the second anti-counterfeiting weight and the second quality weight, including: Extract the report size information of all family-related electronic reports from the historical anti-counterfeiting record data to obtain a family-related report size information set; The fluctuation range of report size is calculated based on the aforementioned set of report size information from the same family. Based on the fluctuation range of the reported size, a second quality weight is configured, and a second anti-counterfeiting weight is calculated.

6. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 1, characterized in that, Based on the anti-counterfeiting weight and quality weight, the digital watermark embedding parameters are optimized to obtain the optimal digital watermark embedding parameters, including: The first watermark embedding parameters are randomly set, wherein the first watermark embedding parameters include the length of the digital watermark; The first watermark embedding parameter is input into the anti-counterfeiting predictor, the first anti-counterfeiting rate is obtained by outputting the first report quality parameter, and the first anti-counterfeiting fitness is calculated. Continue optimizing until convergence, and obtain the optimal digital watermark embedding parameters with the greatest anti-counterfeiting adaptability.

7. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 6, characterized in that, The first report quality parameters are obtained through processing, and the first anti-counterfeiting fitness level is calculated, including: Get the historical maximum watermark embedding parameters; The ratio of the historical maximum watermark embedding parameter to the first watermark embedding parameter is calculated and used as the first report quality parameter.

8. The method for anti-counterfeiting electronic reports using embedded digital watermarks according to claim 6, characterized in that, The training steps for the anti-counterfeiting predictor include: Based on the historical anti-counterfeiting data of electronic reports, a set of sample watermark embedding parameters was collected, and the proportion of anti-counterfeiting completed under different sample watermark embedding parameters was collected and labeled as the sample anti-counterfeiting rate set. Using the sample watermark embedding parameter set and sample anti-counterfeiting rate set as supervised training data, an anti-counterfeiting predictor is trained based on machine learning.

9. An electronic device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the electronic report anti-counterfeiting method using digital watermark embedding as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the electronic report anti-counterfeiting method using digital watermark embedding as described in any one of claims 1-8.

Images