Security protection method, device and equipment for digital human, storage medium and product

Abstract

The invention discloses a security protection method and device for a digital person, equipment, a storage medium and a product, and relates to the technical field of digital product security, and the method comprises the steps: determining a digital identity sequence of an affiliation party of the digital person; generating local identity information based on the digital identity sequence; and respectively writing the local identity information into different parts of the digital person to obtain an authenticated digital person. It can be understood that the local identity information distributed in different parts of the authenticated digital person can be used for auditing and tracking the authenticated digital person, so that the real owner of the authenticated digital person is determined, and the security of the digital person is ensured.

Description

Technical Field

[0001] The present application relates to the field of digital product security technology, and in particular to a digital human security protection method, device, equipment, storage medium and product. Background Art

[0002] In recent years, with the rise of concepts / products such as "metaverse", "blockchain", and "GPT (Generated Pre-trained Transformer) big model", digital human (virtual human) technology has received increasing attention. Voice broadcast and interactive dialogue digital humans have been applied in multiple industries and scenarios, such as virtual tour guide digital humans in scenic spots, China Mobile's smart assistant digital humans, and smart customer digital humans on shopping platforms. However, as "digital humans" become increasingly popular, the security risks they bring (for example, digital humans are plagiarized and used by others, and it is impossible to trace the source and protect rights, etc.) are also becoming increasingly prominent.

[0003] At present, although there are many existing security prevention and control management strategies on various digital human platforms, they are all mainly based on early warning and cannot effectively protect digital humans.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a digital human security protection method, device, equipment, storage medium and product, aiming to solve the technical problem that most of the existing digital human security prevention and control management strategies are policy-level warnings and cannot effectively protect digital humans.

[0006] To achieve the above purpose, the present application proposes a digital human security protection method, which includes:

[0007] Determine the digital identity sequence of the party to which the digital person belongs;

[0008] generating each local identity information based on the digital identity sequence;

[0009] The local identity information is hidden in different parts of the digital person to obtain a certified digital person.

[0010] Optionally, the step of stego-writing the local identity information into different parts of the digital person to obtain a certified digital person includes:

[0011] For any one of the partial identity information, the partial identity information and the part of the digital human corresponding to the partial identity information are input into a preset fusion model to obtain an authentication part;

[0012] After traversing the local identity information, the authentication digital person is generated based on the obtained authentication parts.

[0013] Optionally, the step of generating each local identity information based on the digital identity sequence includes:

[0014] Splitting the digital identity sequence into sequences of digital identity fragments;

[0015] For any digital identity fragment sequence, encrypt the digital identity fragment sequence to obtain an encrypted fragment sequence;

[0016] Generate an encrypted QR code based on the encrypted fragment sequence, and use the encrypted QR code as local identity information corresponding to the digital identity fragment sequence;

[0017] After traversing the digital identity fragment sequences, the local identity information is obtained.

[0018] Optionally, the step of encrypting the digital identity fragment sequence to obtain an encrypted fragment sequence includes:

[0019] Segmenting the digital identity fragment sequence to obtain translation units;

[0020] For any one of the translation units, the characters in the translation unit are translated into binary values ​​to obtain a binary sequence unit corresponding to the translation unit, wherein the binary sequence unit is configured with a mark bit, the mark bit includes a data type mark bit and a translation direction mark bit, and the mark bit is used for reverse translation of the binary sequence unit;

[0021] Performing encryption processing on the binary sequence unit to obtain an encrypted binary sequence unit, wherein the encrypted binary sequence unit is configured with an encryption action bit, the encryption action bit includes an encryption action and action details, the encryption action includes at least one of assignment, data flipping, sequence permutation, and function mapping, and the encryption action bit is used for decryption of the encrypted binary sequence unit;

[0022] After traversing the translation units, the encrypted fragment sequence is generated based on the obtained encrypted binary sequence units.

[0023] Optionally, the preset fusion model is trained by a preset training sample set, and the digital human safety protection method further includes:

[0024] Based on the training samples in the preset training sample set and the preset extraction model, the preset fusion model is iteratively trained by dual-branch mutual learning until both the preset fusion model and the extraction model converge, wherein the initial model parameters of the preset extraction model are obtained by transposing the model parameters of the preset fusion model that meets the preset first preliminary training conditions.

[0025] Optionally, the step of determining the digital identity sequence of the digital person's owner includes:

[0026] Obtaining the identity card information and the phone number of the belonging party;

[0027] Constructing a primary identity code based on the identity card information and the phone number, and randomly generating an auxiliary identity code;

[0028] Based on the primary identity code, the auxiliary identity code and the domain code of the digital person, the digital identity sequence of the belonging party is constructed, wherein the digital identity sequence of the belonging party and the identity card information of the belonging party will be associated and stored in a preset digital identity library, and the preset digital identity library is used for identity verification.

[0029] Optionally, the digital human security protection method further includes:

[0030] After receiving the verification request, obtaining the identity information of the sender of the verification request;

[0031] Sending a verification request to the communication device corresponding to the identity information, and completing preliminary identity authentication after receiving a verification code carried in the verification request fed back by the communication device;

[0032] If a target digital identity sequence matching the identity information is obtained from the preset digital identity database, the identity verification of the verification request sender is successful.

[0033] In addition, to achieve the above-mentioned purpose, the present application also proposes a digital human safety protection device, the digital human safety protection device comprising:

[0034] A determination module, used to determine the digital identity sequence of the party to which the digital person belongs;

[0035] A generating module, used for generating each local identity information based on the digital identity sequence;

[0036] The writing module is used to write the local identity information into different parts of the digital person to obtain a certified digital person.

[0037] In addition, to achieve the above-mentioned purpose, the present application also proposes an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the digital human security protection method as described above.

[0038] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the digital human security protection method as described above are implemented.

[0039] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the digital human security protection method as described above.

[0040] One or more technical solutions proposed in this application have at least the following technical effects:

[0041] In the embodiment of the present application, the management platform will determine the digital identity sequence of the digital person's owner; generate each local identity information based on the digital identity sequence; and write each local identity information into different parts of the digital person to obtain a certified digital person. That is, after the digital person is created and generated, the management platform will determine the digital identity sequence of the digital person's owner, generate each local identity information based on the digital identity sequence, and then write each local identity information into different parts of the digital person to obtain a certified digital person. It can be understood that in actual applications, once the certified digital person in this application is plagiarized and used by others, the local identity information distributed in different parts of the certified digital person can be used for the audit and tracking of the certified digital person to determine the true owner of the certified digital person, thereby ensuring the safety of the digital person. Moreover, since in this application, local identity information will be written into different parts of the digital person, it can prevent the digital person from being partially plagiarized or partially tampered with, thereby improving the comprehensiveness of the protection of the digital person. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0044] Figure 1 This is a flowchart of the first embodiment of the digital human security protection method of the present application;

[0045] Figure 2 A schematic diagram of a hidden scene in the digital human security protection method of the present application;

[0046] Figure 3 A schematic diagram of the composition of a digital identity sequence in the digital human security protection method of the present application;

[0047] Figure 4 This is a timing diagram of information written into the platform by the digital identity in the digital human security protection method of this application;

[0048] Figure 5 This is a flow chart of the second embodiment of the digital human security protection method of the present application;

[0049] Figure 6 This is a flowchart of the third embodiment of the digital human security protection method of the present application;

[0050] Figure 7 A schematic diagram of the translation in the digital human security protection method of the present application;

[0051] Figure 8 A schematic diagram of the encryption action in the digital human security protection method of the present application;

[0052] Fig. 9 A schematic diagram of an encryption scenario in the digital human security protection method of this application;

[0053] Fig.10 This is a flowchart of the fourth embodiment of the digital human security protection method of the present application;

[0054] Fig.11 This is a timing diagram of the digital identity authentication process in the digital human security protection method of this application;

[0055] Fig.12 This is a schematic diagram of the structure of the digital human safety protection device in the digital human safety protection method of the present application;

[0056] Fig.13 Schematic diagram of the device structure of the hardware operating environment involved in the digital human security protection method in the embodiment of the present application.

[0057] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0059] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0060] The main solution of the embodiment of the present application is: determine the digital identity sequence of the party to which the digital person belongs; generate each local identity information based on the digital identity sequence; and stego-write each local identity information into different parts of the digital person to obtain a certified digital person.

[0061] In recent years, with the rise of concepts / products such as "metaverse", "blockchain", and "GPT (Generated Pre-trained Transformer) big model", digital human (virtual human) technology has received increasing attention. Digital humans of voice broadcasting and interactive dialogue have been used in multiple industries and scenarios, such as virtual tour guides in scenic spots, China Mobile's smart assistants, and smart customers on shopping platforms. However, as "digital humans" become increasingly popular, the security risks they bring (for example, digital humans are plagiarized and used by others, and it is impossible to trace the source and protect rights, etc.) are also becoming increasingly prominent. At present, although there are many existing security prevention and control management strategies for various digital human platforms, they are all mainly based on policy-level warnings. For example, strengthen user privacy protection, including encrypting user data, limiting data access rights, and regularly conducting security audits and vulnerability repairs to ensure that user data is not leaked or abused; establish a complaint and reporting mechanism, including establishing a dedicated platform customer service team, strengthen user feedback and communication, promptly handle user issues and disputes, and safeguard user rights and interests and the platform's reputation; strengthen user education and awareness, including providing user guides and safety tips, educating users on how to identify false information and fraudulent behavior, and encouraging users to protect personal privacy and data security. However, these methods are difficult to use for post-audit and tracking, and cannot form an effective protection for digital people.

[0062] This application provides a solution. After a digital person is created, the relevant platform can determine the digital identity sequence of the digital person's owner (the party that owns the copyright of the digital person), split the digital identity sequence to obtain partial identity information, and then write each partial identity information into different parts of the digital person, thereby obtaining a certified digital person. It is understandable that once the certified digital person in this application is plagiarized and used by others, the partial identity information distributed in different parts of the certified digital person can be used for the audit and tracking of the certified digital person to determine the true owner of the certified digital person, thereby ensuring the security of the digital person. Moreover, in this application, partial identity information will be written into different parts of the digital person to prevent the digital person from being partially plagiarized or partially tampered with, thereby improving the comprehensiveness of the protection of the digital person.

[0063] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a cloud platform, a computer, a mobile phone, etc., or an electronic device that can realize the above functions.

[0064] Based on this, the present application embodiment provides a digital human security protection method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the digital human security protection method of the present application.

[0065] In this embodiment, the digital human security protection method includes steps S10 to S30:

[0066] Step S10, determining the digital identity sequence of the digital person's owner;

[0067] It should be noted that, in this embodiment, the implementation subject of the above-mentioned digital human security protection method can be a digital human management platform. The digital human owner can be a person, an organization or a company, etc. In this embodiment, in order to clearly explain this solution, the owner is taken as an example.

[0068] Exemplarily, in this embodiment, in order to ensure the security of the digital person, relevant information about the digital person's owner will be hidden in the digital person to characterize the ownership of the digital person. Therefore, it is necessary to digitize the relevant information of the owner first. Exemplarily. In the case where the owner is the owner, the digital identity sequence of the above-mentioned owner can be generated by the personal information of the owner. For example, the digital identity sequence can be constructed by at least one of the owner's ID card information, telephone number or the owner's personal SIM (Subscriber Identity Module) card information. In addition, a random sequence can also be generated for constructing a digital identity sequence. It is worth noting that when using a random sequence, the mapping relationship between the random sequence and the owner needs to be stored. The constructed digital identity sequence can be a sequence composed of numerical values ​​and characters.

[0069] Step S20, generating each local identity information based on the digital identity sequence;

[0070] Exemplarily, the result obtained by dividing the digital identity sequence can be used as each local identity information. Among them, the division method can be random division, and when the digital identity sequence is based on a combination of different types of information (such as identity card information, telephone number and random sequence), it can also be divided according to different types, without limitation here. In addition, in addition to directly using the result obtained by the division as the local identity information, the result obtained by the division can also be encrypted as the local identity information, and the encrypted result can also be converted into a QR code and then used as the local identity information. It can be understood that encryption and conversion into a QR code can prevent the digital identity sequence from being directly exposed.

[0071] Step S30, the local identity information is hidden in different parts of the digital person to obtain a certified digital person.

[0072] Exemplarily, each local identity information is stego-written into different parts of the digital person, thereby obtaining an authenticated digital person. Among them, the stego-writing method can be the least significant bit (LSB) steganography, space domain method, frequency domain method, etc. And in practical applications, the number of local identity information can be the same as or different from the number of parts of the digital person, such as, the local identity information can be less than the parts of the digital person, in which case the local identity information can not be stego-written in some parts of the digital person; the local identity information can also be more than the parts of the digital person, in which case multiple local identity information can be written in some parts of the digital person. Since the present application is to stego-write the identity information into the digital person, the digital person can carry the identity information representing the ownership without affecting the visual effect of the digital person.

[0073] In this embodiment, the management platform will determine the digital identity sequence of the digital person's owner; generate each local identity information based on the digital identity sequence; and write each local identity information into different parts of the digital person to obtain a certified digital person. That is, after the digital person is created, the management platform will determine the digital identity sequence of the digital person's owner, generate each local identity information based on the digital identity sequence, and then write each local identity information into different parts of the digital person to obtain a certified digital person. It can be understood that in actual applications, once the certified digital person in this application is plagiarized and used by others, the local identity information distributed in different parts of the certified digital person can be used for the audit and tracking of the certified digital person to determine the true owner of the certified digital person, thereby ensuring the safety of the digital person. Moreover, since in this application, the local identity information will be written into different parts of the digital person, it can prevent the digital person from being partially plagiarized or partially tampered with, thereby improving the comprehensiveness of the protection of the digital person.

[0074] In a feasible implementation manner, the step of stego-writing the local identity information into different parts of the digital person to obtain a certified digital person includes steps S31 to S32:

[0075] Step S31, for any one of the partial identity information, input the partial identity information and the part of the digital human corresponding to the partial identity information into a preset fusion model to obtain an authentication part;

[0076] Step S32, after traversing the local identity information, generating the authenticated digital person based on the obtained authentication parts.

[0077] It should be noted that, since the method of steganographically writing each local identity information into a certain part of the digital human is the same, in this embodiment, the process of steganographically writing a local identity information is used for explanation.

[0078] Exemplarily, for any one of the local identity information, the local identity information and the part corresponding to the local identity information are input into the preset fusion model to obtain the authentication part. Among them, the parts of the digital person can be the head (such as the hairstyle part), the torso (such as the upper part) and the lower part (the lower part), etc. of the digital person. In addition, for the digital person with a background, the background can also be used as one of the parts of the digital person. The relationship between the local identity information and the parts of the digital person can be set by the technical personnel according to the actual situation. For example, for some local identity information with more content or more complex content, it can be preferentially written into the head of the digital person, that is, the hairstyle part. The specific correspondence between the local identity information and the part is not limited here. The above-mentioned preset fusion model can be a neural network model, and the preset fusion model can be pre-trained to have a hidden writing function. After traversing each local identity information, the authentication digital person can be generated based on each authentication part. For example, the authentication digital person can be directly composed based on each authentication part, or each authentication part can be used to replace the corresponding unauthenticated part of the digital person, so as to obtain the authentication digital person.

[0079] Reference Figure 2 , is a schematic diagram of the hidden writing scenario in this application. The local identity information A to D in the figure can be, encrypted ID number, encrypted last four digits of mobile phone number, encrypted auxiliary identity code, encrypted domain code. And the local identity information A to D will be written into the parts 1 to 4 of the digital person respectively.

[0080] In a feasible implementation manner, the step of determining the digital identity sequence of the digital person's owner includes steps S11 to S13:

[0081] Step S11, obtaining the identity card information and the telephone number of the belonging party;

[0082] Step S12, constructing a primary identity code based on the identity card information and the phone number, and randomly generating an auxiliary identity code;

[0083] Step S13, based on the primary identity code, the auxiliary identity code and the domain code of the digital person, construct the digital identity sequence of the belonging party, wherein the digital identity sequence of the belonging party and the identity card information of the belonging party will be associated and stored in a preset digital identity library, and the preset digital identity library is used for identity verification.

[0084] Exemplarily, in the process of determining the digital identity sequence of the attributable party, the management platform may obtain the ID card information (ID card number) and telephone number of the attributable party. For example, in most practical applications, the attributable party needs to authenticate the digital person. Accordingly, the attributable party applies to the management platform to create a digital identity. After receiving the creation of the digital identity, the management platform may apply to the attributable party to obtain the ID card information and telephone number of the attributable party. After obtaining the authorization of the attributable party, the management platform may try to find the ID card information and telephone number from the corresponding database, or receive the ID card information and telephone number provided by the attributable party. The ID card information and telephone number are then combined to obtain the main identity code, and sequence data is randomly generated as the auxiliary identity code. The main identity code, the auxiliary identity code, and the domain code of the digital person are then combined to obtain the digital identity sequence of the attributable party. Among them, the domain code of the digital person can be determined according to the domain of application of the digital person. The mapping relationship between the attribution domain and the domain code can be pre-constructed, as shown in Table 1 below, which is a mapping table between the attribution domain and the domain code.

[0085] Field of Attribution Field Code Art Creation Area YZ Technology Zone KJ Game Zone YX Film and TV Area YS Anime Zone DM Customer Service / Shopping Guide Area KF / DG other OT

[0086] Table 1

[0087] For example, if the domain to which the digital person belongs is the animation area and the corresponding domain code is DM, then the domain code of the digital person is DM.

[0088] It can be understood that in this embodiment, by enriching the content of the digital identity sequence, the risk of the digital identity sequence being maliciously reconstructed or tampered with by others can be reduced.

[0089] In addition, after the digital identity sequence of the owner is constructed, the digital identity sequence will be associated with the owner's ID card information and stored in a preset digital identity library. The constructed digital identity library can be used for identity verification in the subsequent tracing stage.

[0090] Reference Figure 3, is a schematic diagram of the composition of a digital identity sequence in this application. As shown in the figure, a digital identity sequence can be composed of a primary identity code (consisting of the ID card number and the last four digits of the mobile phone number), a creative field code (i.e., the field code), and an auxiliary identity code.

[0091] It should be noted that, in a feasible implementation, the management platform can generate an application based on the "digital identity" of the belonging party to create a digital identity sequence for the belonging party. Figure 4 , is a timing diagram of the information written into the platform by the digital identity in this application. As shown in the figure, the writing process is as follows:

[0092] 1. The user (owner) submits the "digital identity" application on the mobile terminal APP, and the "digital identity" application task will be automatically transferred to the digital identity management platform of the management platform;

[0093] 2. The command system of the management platform has a command template inherent in "identity application". The command system sends the information collection command to the mobile terminal APP via the digital identity management platform;

[0094] 3. The mobile terminal APP collects the user's ID card information and SIM card information according to the instructions of the command system;

[0095] 4. The mobile terminal APP transmits the collected ID card information and the last four digits of the mobile phone number to the digital identity management platform. The digital identity management platform writes the ID card information and the last four digits of the mobile phone number into the digital identity database and configures the user ID (for recording relevant information of the user);

[0096] 5. The account management platform of the management platform concatenates the last four digits of the user's ID number and mobile phone number to generate the main identity code. The instruction system randomly generates four random numbers as the auxiliary identity code. The main identity code and the auxiliary identity code are combined to form the user's digital identity unique identifier. The digital identity unique identifier can be directly used as the user's digital identity sequence, or the field code can be added to the main identity code and the auxiliary identity code to form a digital identity sequence.

[0097] In a feasible implementation manner, the preset fusion model is trained by a preset training sample set, and the digital human safety protection method further includes step A1:

[0098] Step A1, based on the training samples in the preset training sample set and the preset extraction model, perform dual-branch mutual learning iterative training on the preset fusion model until the preset fusion model and the extraction model converge, wherein the initial model parameters of the preset extraction model are obtained by transposing the model parameters of the preset fusion model that meets the preset first preliminary training conditions.

[0099] It should be noted that the training samples in the preset training sample set include noise, original image data, and fused image data with noise added to the original image data. The preset fusion model and the preset extraction model have opposite functions, that is, the preset fusion model is used to fuse the original image with the noise, while the extraction model is used to separate the noise from the original image. Accordingly, during the training process, the output of the preset fusion model can be used as the input of the preset extraction model, and the input of the preset fusion model can be used as the label of the output of the preset extraction model, that is, the preset fusion model and the preset extraction model can be trained in a dual-branch mutual learning manner.

[0100] Exemplarily, the preset fusion model can be first trained using the training samples in the preset training sample set to satisfy the preset first preliminary training condition (it can be that the preset fusion model reaches the preset number of training times, or traverses the training samples in the above preset training sample set, or the preset fusion model reaches convergence), and then the model parameters of the preset fusion model that satisfies the preset first preliminary training condition are transposed to obtain the preliminary model parameters of the preset extraction model. After the preset extraction model is configured with the preliminary model parameters, the preset extraction model can be iteratively trained with the preset fusion model in a dual-branch mutual learning training method until both the preset extraction model and the preset fusion model reach convergence, which can indicate that the training of the preset fusion model is completed.

[0101] Reference Figure 5 , is a flow chart of a second embodiment proposed based on the first embodiment of the present application. In this embodiment, the same or similar contents as the above embodiments can refer to the above description and will not be repeated hereafter. The step of performing dual-branch mutual learning iterative training on the preset fusion model based on the training samples in the preset training sample set and the preset extraction model until both the preset fusion model and the extraction model converge includes steps A10 to A30:

[0102] Step A10: Generate sample subsets in sequence based on the preset training sample set.

[0103] Step A20, for any first sample subset among the sample subsets, input the input units of each training sample in the first sample subset into the preset fusion model respectively to obtain each first fusion result, wherein any one of the input units includes the noise and original image data of the corresponding training sample, and the first sample subset is the sample subset corresponding to the model parameter update stage of the preset fusion model.

[0104] Step A30: updating the model parameters of the preset fusion model according to the errors between the first fusion results and the fused image data of the training samples in the first sample subset.

[0105] It should be noted that, in this embodiment, the preset fusion model is trained. The preset fusion model will be trained by the training samples in the pre-training sample set, wherein any training sample will include noise, original image data and fused image data, and the fused image data is obtained by adding the noise in the training sample to the original image data in the training sample, wherein open source steganography tools (such as Stegh i de, OpenStego and Stegosu ite, etc.) can be used to add noise to the original image data to generate fused image data, thereby completing the construction of a training sample.

[0106] Exemplarily, during the training phase of the preset fusion model, random Gaussian noise can be used to initialize the model parameters in the preset fusion model (e.g., when the preset fusion model is a neural network model, the model parameters can be the link weights between neurons) to minimize the time required for model training. The training samples in the preset training sample set can be sorted (or not sorted), and then each sample subset is generated in sequence based on the preset training sample set, such as dividing the preset training sample set into sample subsets of the same size. It should be noted that the first in the above-mentioned first sample subset is only used to characterize that the sample subset is a sample subset in the model parameter update phase of the preset fusion model.

[0107] Since the process of using each sample subset to train the preset fusion model is basically similar, in this embodiment, the process of training the preset fusion model with a first sample subset is taken as an example for explanation. For example, for any first sample subset in each sample subset, the input units of each training sample in the first sample subset are respectively input into the preset fusion model to obtain each first fusion result, that is, inputting an input unit into the preset fusion model can obtain a first fusion result. In any input unit, the noise and original image data of the corresponding training sample corresponding to the input unit, that is, the input unit of a training sample is the noise and original image data in the training sample.

[0108] After obtaining each first fusion result, the model parameters of the preset fusion model are updated based on the error between each first fusion result and the fused image data of each training sample in the above-mentioned first sample subset. After the preset fusion model meets the preset training conditions, it can be represented that the training of the preset fusion model is completed, or it can be represented that the next training stage can be entered. Among them, the preset training condition can be that the preset fusion model converges, or the preset fusion model reaches a preset number of training times (i.e., the number of updates), or all sample subsets are traversed. Accordingly, the preset fusion model that meets the preset training conditions can be directly used for the hidden writing of local identity information, and can also be used for model training in the next stage.

[0109] Exemplarily, in a feasible implementation manner, the error formula for calculating the error between each first fusion result and the fused image data of each training sample in a first sample subset is as follows:

[0110]

[0111] In the formula, F(x j ) represents a first fusion result obtained by inputting the input unit of the j-th training sample in a first sample subset (i.e., the noise and original image data in the j-th training sample) into the preset fusion model, y j is the fused image data of the jth training sample, s is the number of training samples in the first sample subset, E p To calculate the error of the preset fusion model.

[0112] The error is calculated through the above process and the back propagation algorithm is used to update the weights (ie, model parameters) in the preset fusion model.

[0113] Exemplarily, in a feasible implementation manner, the preset fusion model parameter (weight) update formula is as follows:

[0114]

[0115] In the formula, E p is the error of the preset fusion model, P w (k) represents the weight of the preset fusion model at the kth step of training, P w (k+1) represents the weight of the preset fusion model in the k+1th step of training, and α is the learning rate.

[0116] In a feasible implementation manner, the preset fusion model is configured with an extraction model, and after the step of updating the model parameters of the preset fusion model by using the errors between the first fusion results and the fused image data of each training sample in the first sample subset, the method further includes steps A40 to A90:

[0117] Step A40, configuring initial model parameters of a preset extraction model based on the transposition results of the model parameters in the preset fusion model when the preset fusion model satisfies the preset first preliminary training condition;

[0118] It should be noted that, in this embodiment, the preset fusion model is also configured with an extraction model, and the extraction model can be used to assist the training of the preset fusion model. In practical applications, the preset fusion model and the extraction model can be respectively the positive neural network and the negative neural network of the dual-branch mutual learning neural network to simulate the bidirectional cognitive process of the biological nervous system. In the training process, "cause" and "result" are used as objects of mutual learning. The above-mentioned preset first preliminary training condition can be that the preset fusion model reaches the preset number of training times, or the training samples in the above-mentioned preset training sample set are traversed, or the preset fusion model reaches convergence.

[0119] Exemplarily, in this embodiment, when the preset fusion model satisfies the first preliminary training condition, the initial model parameters of the extraction model can be generated based on the model parameters in the preset fusion model, that is, the model parameters in the preset fusion model in this case are transposed, and the transposed results are used as the initial model parameters of the extraction model.

[0120] In a feasible manner, the configuration formula of the initial model parameters of the above extraction model is as follows:

[0121] N w (initial) = P w (k+1) T

[0122] Where P w (k+1) represents the model parameters of the preset fusion model at the k+1th step of training (or, the model parameters of the preset fusion model after completing the preliminary training), T represents the transposition, N w (Initial) Characterizes the initial model parameters of the extraction model.

[0123] Step A50, for any second sample subset in the sample subsets, inputting the input units of each training sample in the second sample subset into the preset fusion model respectively to obtain each second fusion result;

[0124] It should be noted that the above-mentioned second sample subset is used to characterize the training stage of the sample subset (such as for the training of the extraction model), thereby forming a difference with the above-mentioned first sample subset. Similarly, the process of using each sample subset to train the extraction model is basically similar. In this embodiment, the process of training a preset fusion model with a second sample subset is taken as an example for explanation. For any second sample subset in each sample subset, the input unit of each training sample in the second sample subset is respectively input into the preset fusion model to obtain each second fusion result, that is, the input unit of a training sample in the second sample subset is input into the preset fusion model to obtain a second fusion result. Similarly, the input unit of the training sample in the second sample subset includes the noise and original image data of the corresponding training sample.

[0125] Step A60, inputting each of the second fusion results into a preset extraction model to obtain each extraction result;

[0126] It should be noted that, in this embodiment, the preset fusion model is used to stego-write local identity information into different parts of the digital person. Accordingly, in the training stage, the second fusion result obtained by the preset fusion model is actually the result of fusing the noise in the sample into the original image data in the sample. It can be understood that since the initial model parameters of the extraction model are the result of the transposition of the model parameters of the trained preset fusion model. Therefore, in this embodiment, the function of the extraction model will be opposite to the preset fusion model, and the extraction model is used to extract noise or original image data from the fusion result (i.e., the image data after adding noise). Accordingly, each second fusion result is input into the extraction model respectively to obtain each extraction result, and inputting a second fusion result only into the extraction model can obtain an extraction result. Among them, the extraction result can be noise, original image data, or noise and original image data.

[0127] Step A70, updating the model parameters of the extraction model based on the errors between the extraction results and the original data of the training samples in the second sample subset, wherein when the extraction results are noise extraction results, the original data are noise, when the extraction results are image extraction results, the original data are original image data, and when the extraction results are noise extraction results and image extraction results, the original data are noise and original image data;

[0128] Exemplarily, based on the error between each extraction result and the original data of each training sample in the second sample subset, the model parameters of the extraction model are updated. Accordingly, when each extraction result is a noise extraction result obtained by extracting noise, the original data of each training sample in the second sample subset is noise; when each extraction result is an image extraction result obtained by extracting original image data, the original data of each training sample in the second sample subset is original image data; when each extraction result is a noise extraction result and an image extraction result obtained by extracting noise and original image data, the original data of each training sample in the second sample subset is noise and original image data.

[0129] Step A80, when the extraction model reaches the preset second preliminary training condition and the preset fusion model and the extraction model are not converged, disrupting the order of the training samples in the preset training sample set to obtain a new preset training sample set;

[0130] Exemplarily, the second preliminary training condition corresponding to the extraction model may be reaching a preset number of training times, or traversing each second sample subset. When the extraction model reaches the preset second preliminary training condition, and any one of the preset fusion model and the extraction model has not converged, the training samples in the preset training sample set are shuffled to obtain a new preset training sample set.

[0131] Step A90, based on the new preset training sample set, returning to execute the step of sequentially generating each sample subset based on the preset training sample set until both the preset fusion model and the extraction model converge.

[0132] Exemplarily, based on the new preset training sample set, the process returns to the step of sequentially generating each sample subset from the preset training sample set, until both the preset fusion model and the extraction model have completed convergence.

[0133] In a feasible implementation manner, the calculation formula for the error between the extraction result and the original data of the training sample in the second sample subset is as follows:

[0134]

[0135] In the formula, G(y ij ) represents the extraction result obtained by inputting the second fusion result corresponding to the jth training sample in the second sample subset into the extraction model, x j is the noise, original image data, or noise and original image data of the jth training sample, s is the number of training samples in the second sample subset, E n To calculate the error of the extracted model.

[0136] Similarly, the weights (i.e., model parameters) in the preset fusion model can be updated through the back propagation algorithm. The formula for extracting the model parameter (weight) update is as follows:

[0137]

[0138] In the formula, E n is the error of the preset fusion model, N w (k) represents the weight of the model extracted at the kth step of training, N w (k+1) represents the weight of the model extracted at the k+1th step of training, and α is the learning rate.

[0139] Reference Figure 6 , is a flowchart of a third embodiment proposed based on the first and second embodiments of the present application. In this embodiment, the same or similar contents as those in the above embodiments can be referred to the above introduction and will not be repeated hereafter. The step of generating each local identity information based on the digital identity sequence includes steps S21 to S24:

[0140] Step S21, splitting the digital identity sequence into digital identity fragment sequences;

[0141] Step S22, for any digital identity fragment sequence, encrypt the digital identity fragment sequence to obtain an encrypted fragment sequence;

[0142] Step S23, generating an encrypted two-dimensional code based on the encrypted fragment sequence, and using the encrypted two-dimensional code as local identity information corresponding to the digital identity fragment sequence;

[0143] Step S24, after traversing the digital identity fragment sequences, obtaining the local identity information.

[0144] It should be noted that, in the embodiment, in order to prevent the information of the owner from being exposed, encryption is performed during the process of generating the local identity information in the embodiment.

[0145] Exemplarily, the digital identity sequence is split into digital identity fragment sequences, which can be split randomly or according to type (such as ID card information, telephone number, random sequence, and field code). For any digital identity fragment sequence obtained by splitting, the digital identity fragment sequence is encrypted to obtain an encrypted fragment sequence. Among them, the specific confidentiality method can be set by technical personnel according to actual needs, which will not be repeated here. After obtaining the encrypted fragment, an encrypted QR code can be generated based on the confidential fragment (an open source QR code generation tool can be used), and the generated QR code is the local identity information corresponding to the digital identity fragment sequence. After traversing each digital identity fragment sequence, each local identity information can be obtained.

[0146] In a feasible implementation manner, the step of encrypting the digital identity fragment sequence to obtain an encrypted fragment sequence includes steps S221 to S224:

[0147] Step S221, segmenting the digital identity fragment sequence to obtain translation units;

[0148] Step S222: for any one of the translation units, the characters in the translation unit are translated into binary values ​​to obtain a binary sequence unit corresponding to the translation unit, wherein the binary sequence unit is configured with a mark bit, the mark bit includes a data type mark bit and a translation direction mark bit, and the mark bit is used for reverse translation of the binary sequence unit;

[0149] Step S223, encrypting the binary sequence unit to obtain an encrypted binary sequence unit, wherein the encrypted binary sequence unit is configured with an encryption action bit, the encryption action bit includes an encryption action and action details, the encryption action includes at least one of assignment, data flipping, sequence permutation, and function mapping, and the encryption action bit is used for decrypting the encrypted binary sequence unit;

[0150] Step S224, after traversing the translation units, generating the encrypted fragment sequence based on the obtained encrypted binary sequence units.

[0151] Exemplarily, the digital identity fragment sequence is further segmented to obtain each translation unit. The segmentation method can be set by a technician, such as every two characters in the digital identity fragment sequence are a translation unit. For any one of the translation units, the characters in the translation unit are translated into binary values. Similarly, the translation method can also be set by a technician, such as, if the characters in the digital identity fragment sequence are decimal numbers, the translation rule is decimal to binary, if the characters in the digital identity fragment sequence are letters, the letters in the digital identity fragment sequence can be mapped to binary values ​​using a pre-set mapping rule. The binary value obtained by translating the characters in the translation unit is (or constitutes) the binary sequence unit corresponding to the translation unit. It should be noted that the mark bit of each binary sequence will also be generated during the translation process. The mark bit may include a data type mark bit and a translation direction mark bit, wherein the data type mark bit indicates whether the translation unit used when generating the binary sequence belongs to a numerical type or a character type, and the translation direction mark bit indicates whether the direction of reading the data in the translation unit when generating the binary sequence is from left to right or from right to left. The marker bit can be used for reverse translation of the binary sequence unit, that is, to restore the binary unit to the translation unit, thereby forming a sequence of digital identity fragments.

[0152] like Figure 7As shown, it is a translation schematic diagram in the present application. As shown in the figure, it is now necessary to translate the digital identity fragment sequence 31103, and each translation unit is set to include two characters. Data is obtained from left to right, and the translation units are 31, 10, and 30, respectively. The translation unit 31 is now translated to obtain binary values ​​of 00000011 and 00000001, respectively, and combined to obtain a binary sequence unit, and then the mark bit of the binary sequence unit is set. As shown in the figure, the type of the mark bit includes not only a translation direction mark bit and a value type mark bit, but also a sequence mark bit (usually a digital identity fragment sequence can only set one sequence mark bit), and the sequence mark bit can characterize the information type of the digital identity fragment sequence to which the binary sequence unit belongs, such as sequence 1 corresponds to the digital identity fragment sequence of identity card information, sequence 2 corresponds to the digital identity fragment sequence of telephone number information (such as the last four digits of the number), sequence 3 corresponds to the digital identity fragment sequence of auxiliary identity code, sequence 4 corresponds to the digital identity fragment sequence of the field code, etc. Exemplarily, the sequence mark can be represented by a two-bit binary digit value, such as "10", where the first "1" indicates that the two characters obtained contain the sequence bit, and "0" does not, and the second "0" indicates that the "first" bit of the two characters is the sequence bit, and "1" indicates that the "second" bit of the two characters is the sequence bit. The value type mark is used to identify the value type of the two-digit number obtained, and is represented by two binary bits. "00" indicates that both characters are numerical, "01" indicates that the first character is numerical and the second character is character, "10" indicates that the first character is character and the second character is numerical, and "11" indicates that both characters are numerical. The translation direction mark is used to represent the order in which the two characters in the original digital identity fragment sequence are obtained. It is represented by one binary bit, "1" indicates that the order in which the two characters in the original digital identity fragment sequence are obtained is from left to right, and "0" indicates that the order in which the two characters are obtained is from right to left. It should be noted that the binary values ​​of each translation unit can be stored together, and each 16 binary codes in the corresponding combined and stored data correspond to one translation unit; similarly, the corresponding mark bits of each translation unit can be stored together, and each 5 binary codes in the combined and stored data correspond to one translation unit.

[0153] Furthermore, after obtaining the binary sequence unit, the binary sequence unit is encrypted to obtain an encrypted binary sequence unit. In the process of encryption, the encryption action used for the binary value (0 or 1) in the binary sequence unit may include at least one of assignment, data flipping, sequence permutation, and function mapping. Correspondingly, the encrypted binary sequence unit obtained by encryption processing can also be configured with encryption action bits, and the encryption action bits include the encryption actions used in the encryption process and the action details. Similarly, the encryption actions include at least one of assignment, data flipping, sequence permutation, and function mapping. As shown in Table 2 below, the binary encoding corresponding to each encryption action in this application.

[0154]

[0155]

[0156] Table 2

[0157] Among them, the above assignments correspond to setting to 0 and setting to 1.

[0158] The action details are the specific operations during encryption. For example, the sequence replacement may include the starting position of the replacement sequence and the replaced sequence, as well as the sequence length, etc. After traversing each translation unit, each encrypted binary sequence unit can be obtained, and each encrypted binary sequence unit can be combined into an encrypted fragment sequence, or the result of translating each encrypted binary sequence unit into decimal can be used as the encrypted fragment sequence.

[0159] For example, refer to Figure 8 , is a schematic diagram of the encryption action bits in the present application. As shown in the figure, the set-0 mark is 0, which indicates that the set-0 action (a type of assignment action) is not performed, the set-1 mark is 0, which indicates that the set-1 action (a type of assignment action) is not performed, the flip mark is 1, which indicates that the data flip action is performed, and the replacement mark is 1, which indicates that the sequential replacement action is performed. Then 010 is the replacement start bit 1, which can represent the starting bit of the replacement sequence, and then 100 is the replacement start bit 2, which can represent the starting bit of the replaced sequence. 111 is the total number of replacement bits (that is, the length of the replacement sequence), and the function mapping mark is 1, which indicates that the function mapping action is performed.

[0160] Reference Fig. 9 , is a schematic diagram of the encryption scenario in this application. As shown in the figure, the numerical bit in the figure is the above-mentioned binary sequence unit, or is composed of the above-mentioned binary sequence unit, and correspondingly, the encrypted numerical bit in the figure is the above-mentioned encrypted binary sequence unit, or is composed of the above-mentioned encrypted binary sequence unit. The translation in the figure is to translate binary into decimal.

[0161] In addition, it should be noted that the above contents are all processes for generating a certified digital person. Once a certified digital person is stolen, evidence can be obtained from the stolen certified digital person. For example, the certified digital person (or a part of the certified digital person) can be input into an extraction model corresponding to the above preset fusion model. The extraction model can be used to extract each local identity information from the certified digital person, and the encrypted action bits and mark bits in each local identity information can be used to restore it to a digital identity sequence. The digital identity sequence can be used for the attribution of the party.

[0162] Reference Fig.10 , is a flowchart of the fourth embodiment of the present application proposed based on the first embodiment, the second embodiment, and the third embodiment of the present application. In this embodiment, the same or similar contents as the above embodiments can be referred to the above introduction, and will not be repeated later. The digital human security protection method also includes steps S01 to S03:

[0163] Step S01, after receiving a verification request, obtaining the identity information of the sender of the verification request;

[0164] Step S02, sending a verification request to the communication device corresponding to the identity information, and completing preliminary identity verification after receiving the verification code carried in the verification request fed back by the communication device;

[0165] Step S03: If a target digital identity sequence matching the identity information is obtained from the preset digital identity database, the identity verification of the verification request sender is successful.

[0166] It should be noted that this embodiment is mainly a verification stage, that is, the user who needs to be verified can send a verification request to the management platform through a communication device. After receiving the verification request, the management platform can obtain the identity information of the sender of the verification request, such as the identity information can be carried in the verification request, or it can be obtained from the communication device. The management platform can determine the phone number of the user corresponding to the identity information based on the identity information. Through the phone number, the management platform can send a verification request to the communication device used by the user. Correspondingly, a verification code is carried in the verification request, and the user needs to feedback the verification code through the communication device. After the platform receives the verification code, it can indicate that the sender of the verification request has completed the preliminary identity verification. Further, the target digital identity sequence of the identity information of the sender of the verification request is queried from the preset digital identity library. If the target digital identity sequence can be queried, it indicates that the identity verification is successful, otherwise, the verification fails.

[0167] In another feasible implementation, the management platform can verify the user's identity based on the user's "digital identity" authentication application. Fig.11, is a timing diagram of the digital identity authentication process in this application. As shown in the figure, the authentication verification process is as follows:

[0168] 1. The user applies for "digital identity" authentication on the mobile terminal APP, and the authentication task work order will be automatically transferred to the digital identity management platform of the management platform;

[0169] 2. The command system of the management platform splices the instructions for the "identity authentication" application task, and sends the information collection instructions to the mobile terminal APP via the digital identity management platform;

[0170] 3. After the user's mobile terminal APP receives the verification request from the management platform, the command system of the management platform will first send a random 8-digit string security code to the terminal SIM card SMS. The string is randomly composed of numbers, uppercase and lowercase letters, and special symbols, such as "A3@900S"; then the user needs to enter the security code received by the mobile phone in the interactive page popped up by the terminal APP to complete the preliminary identity verification;

[0171] 4. The platform applies to the mobile terminal to collect ID card information and SIM card information. After obtaining the user's permission, the mobile terminal APP feeds back the user's ID card information and SIM card information to the digital identity management platform;

[0172] 5. The digital identity management platform traverses the digital identity database and searches for the user ID and digital identity that matches the ID number and mobile phone number of the user to be verified. If a match is found, the identified user ID is fed back to the mobile terminal APP to complete the user's identity authentication; otherwise, the feedback is "unable to identify".

[0173] This application provides a digital human safety protection device, referring to Fig.12 , the safety protection device of the digital human comprises:

[0174] A determination module 10, used to determine the digital identity sequence of the party to which the digital person belongs;

[0175] A generating module 20, configured to generate each local identity information based on the digital identity sequence;

[0176] The writing module 30 is used to write the local identity information into different parts of the digital person to obtain a certified digital person.

[0177] The digital human safety protection device provided by the present application adopts the digital human safety protection method in the above embodiment, aiming to solve the technical problem that most of the existing digital human safety prevention and control management strategies are policy-level warnings and cannot effectively protect digital humans. Compared with the prior art, the beneficial effects of the digital human safety protection device provided by the embodiment of the present application are the same as the beneficial effects of the digital human safety protection method provided by the above embodiment 1, and the other technical features of the digital human safety protection device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0178] The present application provides an electronic device, which includes: at least one processor; and a memory that is communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the digital human security protection method in the above-mentioned embodiment one.

[0179] Reference below Fig.13 , which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present application. The electronic devices in the embodiments of the present application may include but are not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Displays), PMPs (Portable Multimedia Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.13 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0180] like Fig.13As shown, the electronic device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read On ly Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the electronic device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an electronic device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0181] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0182] The electronic device provided by the present application adopts the digital human security protection method in the above embodiment, which can solve the technical problem that the existing data display method is difficult to intuitively and clearly complete data visualization. Compared with the prior art, the beneficial effects of the electronic device provided by the present application are the same as the beneficial effects of the digital human security protection method provided by the above embodiment, and the other technical features in the electronic device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0183] It should be understood that the various parts disclosed in this application can be implemented by 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 one or more embodiments or examples in a suitable manner.

[0184] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0185] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the digital human security protection method in the above-mentioned embodiment.

[0186] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read On ly Memory), an erasable programmable read-only memory (EPROM: Erasable Programmab le Read On ly Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read On ly Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0187] The computer-readable storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0188] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by an electronic device, the electronic device is enabled to: provide security protection for digital humans.

[0189] The computer program code for performing the operation of the present application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "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 an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0190] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0191] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0192] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned digital human security protection method, and can solve the technical problem that the existing data display method is difficult to intuitively and clearly complete data visualization. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the digital human security protection method provided in the above-mentioned embodiment, and will not be repeated here.

[0193] The present application also provides a computer program product, including a computer program, which implements the steps of the digital human security protection method as described above when executed by a processor.

[0194] The computer program product provided by this application can solve the technical problem of digital human security protection. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the digital human security protection method provided by the above embodiment, which will not be repeated here.

[0195] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A digital human security protection method, characterized in that: The digital human safety protection method comprises: Determine the digital identity sequence of the party to which the digital person belongs; generating each local identity information based on the digital identity sequence; The local identity information is hidden in different parts of the digital person to obtain a certified digital person.

2. The digital human safety protection method according to claim 1, characterized in that: The step of stego-writing the local identity information into different parts of the digital person to obtain a certified digital person comprises: For any one of the partial identity information, the partial identity information and the part of the digital human corresponding to the partial identity information are input into a preset fusion model to obtain an authentication part; After traversing the local identity information, the authentication digital person is generated based on the obtained authentication parts.

3. The digital human safety protection method according to claim 1, characterized in that: The step of generating each local identity information based on the digital identity sequence comprises: Splitting the digital identity sequence into sequences of digital identity fragments; For any digital identity fragment sequence, encrypt the digital identity fragment sequence to obtain an encrypted fragment sequence; Generate an encrypted QR code based on the encrypted fragment sequence, and use the encrypted QR code as local identity information corresponding to the digital identity fragment sequence; After traversing the digital identity fragment sequences, the local identity information is obtained.

4. The digital human safety protection method according to claim 3, characterized in that: The step of encrypting the digital identity fragment sequence to obtain an encrypted fragment sequence comprises: Segmenting the digital identity fragment sequence to obtain translation units; For any one of the translation units, the characters in the translation unit are translated into binary values ​​to obtain a binary sequence unit corresponding to the translation unit, wherein the binary sequence unit is configured with a mark bit, the mark bit includes a data type mark bit and a translation direction mark bit, and the mark bit is used for reverse translation of the binary sequence unit; Performing encryption processing on the binary sequence unit to obtain an encrypted binary sequence unit, wherein the encrypted binary sequence unit is configured with an encryption action bit, the encryption action bit includes an encryption action and action details, the encryption action includes at least one of assignment, data flipping, sequence permutation, and function mapping, and the encryption action bit is used for decryption of the encrypted binary sequence unit; After traversing the translation units, the encrypted fragment sequence is generated based on the obtained encrypted binary sequence units.

5. The digital human safety protection method according to claim 2, characterized in that: The preset fusion model is trained by a preset training sample set, and the digital human safety protection method further includes: Based on the training samples in the preset training sample set and the preset extraction model, the preset fusion model is iteratively trained by dual-branch mutual learning until both the preset fusion model and the extraction model converge, wherein the initial model parameters of the preset extraction model are obtained by transposing the model parameters of the preset fusion model that meets the preset first preliminary training conditions.

6. The digital human safety protection method according to claim 1, characterized in that: The step of determining the digital identity sequence of the digital person's belonging party comprises: Obtaining the identity card information and the phone number of the belonging party; Constructing a primary identity code based on the identity card information and the phone number, and randomly generating an auxiliary identity code; Based on the primary identity code, the auxiliary identity code and the domain code of the digital person, the digital identity sequence of the belonging party is constructed, wherein the digital identity sequence of the belonging party and the identity card information of the belonging party will be associated and stored in a preset digital identity library, and the preset digital identity library is used for identity verification.

7. The digital human safety protection method according to claim 1, characterized in that: The digital human safety protection method further includes: After receiving the verification request, obtaining the identity information of the sender of the verification request; Sending a verification request to the communication device corresponding to the identity information, and completing preliminary identity authentication after receiving the verification code carried in the verification request fed back by the communication device; If a target digital identity sequence matching the identity information is obtained from the preset digital identity database, the identity verification of the verification request sender is successful.

8. A safety protection device for a digital human, characterized in that: The safety protection device of the digital human comprises: A determination module, used to determine the digital identity sequence of the party to which the digital person belongs; A generating module, used for generating each local identity information based on the digital identity sequence; The writing module is used to write the local identity information into different parts of the digital person to obtain a certified digital person.

9. An electronic device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the digital human security protection method as described in any one of claims 1 to 7.

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, the steps of the digital human security protection method according to any one of claims 1 to 7 are implemented.

11. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the digital human safety protection method as claimed in any one of claims 1 to 7 are implemented.

Images