Identity verification method and system based on large model

Abstract

The invention discloses an identity verification method and system based on a large model, and the method comprises the steps: collecting the face data of a requester, carrying out the feature extraction, carrying out the digital conversion of the face data after the feature extraction into a feature vector, generating a session identifier, and binding the session identifier with the face data of the requester. Constructing request data, encrypting the request data, sending the encrypted request data to the large model, judging whether the face data is consistent with the user identity stored in the large model, if so, receiving the request, and continuing the session; and if not, stopping the session, and storing the face data of the request. The legal consultation request of the user is processed by using the large model, the identity of the user is verified in combination with a biological feature verification method, and personalized related legal services are provided, so that the security and personalization of the legal services are improved, and the safe use experience of the user is improved.

Description

Technical Field

[0001] The present invention relates to the fields of information technology and legal service technology, and in particular to a method and system for identity authentication based on a large model. Background Art

[0002] With the continuous development of Internet technology and artificial intelligence technology, the legal service industry is gradually transforming towards online and intelligent directions. Traditional legal services mostly rely on offline physical institutions. Users need to go to legal service institutions in person for consultation, signing legal documents, etc. This process is not only time-consuming and labor-intensive, but also has problems such as complex user identity authentication and lack of security in the document signing process. In order to improve the efficiency of legal services, ensure the authenticity of user identities, and provide personalized legal document signing services, it is urgent to develop a new technical solution to achieve remote legal document signing.

[0003] However, traditional legal services have many deficiencies in identity verification and personalized services. Accurate user identity verification is crucial to ensuring the security of legal services. At the same time, providing personalized services that meet user needs can significantly improve user experience. Summary of the invention

[0004] In response to the above-mentioned defects, the technical problem solved by the present invention is to provide a method for identity authentication based on a big model, which can utilize the big model to process users' legal consultation requests, combine biometric verification methods to verify user identities, and provide personalized related legal services, thereby increasing the security and personalization of legal services and improving users' safe usage experience.

[0005] A first aspect of the present invention provides a method for identity authentication based on a large model, which is used for identity authentication of a user requesting access under a large model dialogue, and includes: S101: collecting facial data of the requester and performing feature extraction.

[0006] S102: Convert the facial data after feature extraction into a feature vector, generate a session identifier, and bind or unbind the session identifier to the facial data of the requester.

[0007] S103: construct request data according to the session identifier, encrypt the request data, and send it to the large model. S104: the large model receives the request data, stores the request data, decrypts the request data, and stores the face data in the request data.

[0008] S105: Determine whether the face data is consistent with the user identity stored in the large model. S106: If the face data is consistent, receive the request and respond. S107: If not, return to step S102.

[0009] According to an embodiment of the present invention, the step S102 converts the facial data after feature extraction into a feature vector to generate a session identifier, including writing the feature vector of the facial data into the session identifier.

[0010] According to one embodiment of the present invention, the step S102 binds or unbinds the session identifier with the requester's facial data, including: when a user request for access is issued and a session identifier is generated, the session identifier is bound to the facial data, and the binding result is recorded in the session identifier as a number or code.

[0011] When the large model system determines that the facial data is inconsistent with the stored user identity, it unbinds the session identifier from the requester's facial data, records the unbinding result, a number or code, into the session identifier, and terminates the session.

[0012] According to an embodiment of the present invention, constructing the request data according to the session identifier in step S103 includes determining whether the session identifier is bound to the requester's facial data, and if the session identifier is bound to the requester's facial data, constructing the request data according to the session identifier. If the session identifier is unbound from the requester's facial data, no request data is constructed, and the session is terminated. The request data includes the session identifier and the request instruction.

[0013] According to an embodiment of the present invention, the session identifier also includes a timestamp.

[0014] According to an embodiment of the present invention, the step S106 further includes:

[0015] S106: Set a first time period, and determine whether the difference between the current time and the session timestamp is less than the first time period. If the difference between the current time and the session timestamp is less than the first time period, the session continues. If the difference between the current time and the session timestamp is greater than or equal to the first time period, the session is deemed expired, the session identifier is destroyed, and the process returns to step S101.

[0016] According to one embodiment of the present invention, step S106 also includes: setting a second time period, the second time period is smaller than the first time period, determining whether the difference between the current time and the session timestamp is equal to the second time period, if the difference between the current time and the session timestamp is equal to the second time period, returning to step S101; if the difference between the current time and the session timestamp is not equal to the second time period, continuing the session.

[0017] A second aspect of the present invention provides a system for identity authentication based on a large model, comprising:

[0018] Collection module: used to collect the requester's facial data and perform feature extraction.

[0019] Generation module; used to convert the facial data after feature extraction into a feature vector, generate a session identifier, and bind or unbind the session identifier with the requester's facial data.

[0020] The sending module is used to construct the request data according to the session identifier, encrypt the request data, and send it to the large model. The receiving module is used to receive the request data, store the request data, decrypt the request data, and store the face data in the request data.

[0021] Judgment module: judge whether the face data is consistent with the user identity stored in the large model.

[0022] Response module: If the facial data is consistent with the facial data, receive the request and respond; if not, unbind the session identifier from the requester's facial data.

[0023] According to an embodiment of the present invention, the session identifier also includes a timestamp.

[0024] According to one embodiment of the present invention, the system also includes: a first setting module, used to set a first setting first time period, determine whether the difference between the current time and the session timestamp is less than the first time period, if the difference between the current time and the session timestamp is less than the first time period, the session continues; if the difference between the current time and the session timestamp is greater than or equal to the first time period, the session is deemed expired, the session identifier is destroyed, and the collection module is returned to collect facial data and perform feature extraction.

[0025] The second setting module sets a second time period, which is smaller than the first time period, and determines whether the difference between the current time and the session timestamp is equal to the second time period. If the difference between the current time and the session timestamp is equal to the second time period, the module returns to the collection module to collect face data and perform feature extraction. If the difference between the current time and the session timestamp is not equal to the second time period, the session continues.

[0026] The third aspect of the present invention provides an intelligent device, including a transmitter, a receiver, a memory and a processor; the memory is used to store computer instructions; the processor is used to execute the computer instructions stored in the memory to implement the above-mentioned large model-based identity authentication method.

[0027] A fourth aspect of the present invention provides a storage medium, comprising: a readable storage medium and computer instructions, wherein the computer instructions are stored in the readable storage medium; the computer instructions are used to implement the above-mentioned large model-based identity authentication method.

[0028] The beneficial effects provided by the present invention are as follows: through a special biometric verification method, the authenticity of the user's identity is ensured and unauthorized access is prevented, while at the same time being able to provide users with better personalized services, including this identity verification method, which completely reduces manual participation and reduces service costs, while allowing users to have a better service experience and improving service efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0030] Figure 1 A flow chart of a method for identity authentication based on a large model disclosed in an embodiment of the present invention;

[0031] Figure 2 A schematic block diagram of a system for identity authentication based on a large model disclosed in an embodiment of the present invention.

[0032] The above drawings show clear embodiments of the present disclosure, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present disclosure to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0033] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0034] The present invention provides a method for identity authentication based on a big model, which can utilize the big model to process the user's legal consultation request, combine the user's signature and voice recognition technology to verify the user's identity, and provide personalized related legal services, thereby increasing the security and personalization of legal services and improving the user's safe use experience.

[0035] like Figure 1 As shown, S101: collect the face data of the requester and perform feature extraction. Here, the face data can be collected by a camera.

[0036] Some large models do not have the ability to directly process images or visual data. They use computer vision models, such as OpenCV, deep learning models (FaceNet or dlib), to complete data collection and face detection, and then convert these data into text or other digital formats for processing by large models.

[0037] Specifically, facial data is collected by a camera, OpenCV or other computer vision libraries are used to collect facial data in the camera video stream, and a deep learning model (such as FaceNet or dlib) is used to extract facial features.

[0038] Convert the extracted facial feature vector into a string format so that it can be passed to the large model later.

[0039] S102: Convert the facial data after feature extraction into a feature vector, generate a session identifier, and bind or unbind the session identifier to the facial data of the requester. The reason for binding with facial data is to provide more personalized services in the future. The services here are not limited to legal services, but can cover or include all services that can be personalized.

[0040] For example, the big model stores all consultations or past services related to facial data identity. The big model is prepared to provide services that are closer to the user's identity, which undoubtedly improves the response speed.

[0041] The action of collecting face data is triggered by a session request. When the user triggers the session request, the collection end starts collecting face data.

[0042] Session identifiers can be generated by combining user ID, timestamp, and random number to ensure uniqueness and security.

[0043] In the step S102, the facial data after feature extraction is digitally converted into a feature vector to generate a session identifier, including writing the feature vector of the facial data into the session identifier.

[0044] Use the user's unique identifier, such as user ID, timestamp, face data, and some random numbers to form a string. Use an encryption algorithm. Here, a hash algorithm (such as SHA-256) can be used to encrypt it to generate a unique hash value as a session identifier. Use the UUID (Universally Unique Identifier) ​​generation function to generate a session identifier. UUID is a standard identifier format that is generated by an algorithm to ensure global uniqueness.

[0045] The generated session identifier will be stored in the server-side session storage (such as a database or cache) together with other user information (such as user ID, facial data, IP address, browser fingerprint, etc.).

[0046] The session identifier is sent to the user's browser or client via a cookie or HTTP header and stored locally. After that, each request from the user will be accompanied by this session identifier for verification and tracking.

[0047] The generation of session identifiers is usually performed by the server. After the user initiates a request or logs in, the server generates a unique session identifier to ensure that each user session is independent and secure.

[0048] After the session identifier is generated, it can be written to an appropriate location for subsequent access and verification. Common writing locations and methods include: storing the session identifier together with the user data on the server side, such as a database, memory cache (such as Redis, Memcached), etc. In Web applications, the session identifier is usually stored in the client browser's cookie, or using local storage.

[0049] When making an API call, a session identifier can usually be sent to the server as part of the request header to verify and identify the user.

[0050] Session identifiers can be set and written in a variety of ways, including server-side storage, client-side storage, request header transmission, etc. Its generation and management need to pay attention to security, ensuring the uniqueness, unpredictability, and timeliness of session identifiers to prevent potential security threats.

[0051] In step S102, the session identifier is bound or unbound to the facial data of the requester, including: when a user requests access and a session identifier is generated, the session identifier is bound to the facial data, and the binding result is recorded in the session identifier as a number or code. When the large model system determines that the facial data is inconsistent with the stored user identity, the session identifier is unbound from the facial data of the requester, and the unbinding result, a number or code, is recorded in the session identifier to terminate the session. For example, the binding result can be recorded as "number 1" as the binding of the session identifier to the facial data, and the unbinding result can be recorded as "number 0" as the unbinding of the session identifier from the facial data.

[0052] S103: Construct request data according to the session identifier, encrypt the request data, and send it to the large model.

[0053] Here, request data is generated based on the binding result of the facial data and the session identifier. If it is recognized that the session identifier and the facial data have been unbound, the session identifier is destroyed and the session is terminated. If it is recognized that the session identifier and the facial data are still in a bound state, the request data is constructed based on the session identifier and the request instruction.

[0054] The action of collecting face data is triggered by a session request. When the user triggers the session request, the collection end starts collecting face data. Therefore, the request data includes the session identifier and this request. The session identifier here includes the user's ID, face data, IP address, browser fingerprint, binding relationship between representation and face data, etc., as well as a timestamp.

[0055] When sending to large models, encrypted transmission must be used to ensure that the request is sent via HTTPS to protect the security of the data during transmission.

[0056] At the same time, during the storage and processing of facial features, ensure compliance with privacy regulations (such as GDPR) and obtain user consent.

[0057] The user submits a legal consultation request through the legal consultation system interface. After receiving the request, the system immediately generates a unique session identifier with the current timestamp. The system returns the session identifier and timestamp to the user client through Cookie or HTTP header, and records the session information on the server side.

[0058] S104: The large model receives the request data of the current session, stores the request data, decrypts the request data and stores the face data in the request data.

[0059] S105: Determine whether the facial data is consistent with the user identity stored in the large model. If it is consistent, accept the request and respond to the dialogue; if it is inconsistent, return to step S102.

[0060] The facial data that is inconsistent with the user identity stored in the large model is stored for the purpose of subsequent identity filing, such as targeted investigation of user-related events.

[0061] Timestamp is a digital mark that records a specific time point. It is used to ensure data consistency and security, and is used as incidental information in the session identifier generation process to prevent replay attacks. The continuity and security of the session interaction process can be determined by setting the first time period and calculating the difference between it and the timestamp.

[0062] When a user initiates a legal consultation request, the system generates a timestamp using the server's current practice to ensure the accuracy and consistency of the time. It can be accurate to milliseconds or nanoseconds, or a higher precision timestamp to ensure consistency.

[0063] When generating a session identifier, you can use a combination of a timestamp, a random number, or the user's ID information to enhance uniqueness and security.

[0064] Each time a user initiates a request, the system can check the timestamp to ensure that it is valid within a predefined time window to prevent expiration or duplicate requests.

[0065] Record timestamps in logs for auditing, analyzing, and monitoring user activities. Specific implementation methods:

[0066] Receive legal consultation requests from users;

[0067] Generate a session identifier and current timestamp information according to the user's legal consultation request, and return the session identifier and current timestamp information to the user end, while recording the current session information on the server end,

[0068] When the user continues to submit requests in the same session, they will be accompanied by a session identifier and timestamp.

[0069] Verify the legitimacy of the session identifier and timestamp. If they are valid, process the user request. Also set the session validity period.

[0070] Within a certain period of time, the system automatically logs out the session for operation, improving the security of the session. The session identifier is generated by combining the timestamp and random number to prevent attackers from forging requests. The specific implementation steps are as follows:

[0071] S107: Set a first time period, and determine whether the difference between the current time and the session timestamp is less than the first time period. If the difference between the current time and the session timestamp is less than the first time period, the session continues; if the difference between the current time and the session timestamp is greater than or equal to the first time period, the session is deemed expired, the session identifier is destroyed, and the process returns to step S101.

[0072] Of course, in order to maximize the safety of users, the following methods can be used to strictly prevent and monitor, and strengthen the safety of users in a short period of time and in a quantitative manner.

[0073] Step S106 also includes: setting a second time period, the second time period is smaller than the first time period, determining whether the difference between the current time and the session timestamp is equal to the second time period, and if so, returning to step S101.

[0074] Destroying a session identifier is an important step to ensure that after a user logs out, the system no longer allows access to their session or information. Destroying a session identifier usually involves several aspects, including clearing the session data in storage, invalidating the identifier, and ensuring that it is not misused after the session ends. Here are some common ways to destroy a session identifier:

[0075] Delete the session data from the server-side storage. On the server side, destroying the session identifier usually involves deleting or invalidating the session data. Different storage methods (such as database, cache, memory) have different ways of handling this.

[0076] The first method is to delete the session data in the database. Delete the session data from the memory cache. If you use the memory cache, you can delete the session data through the corresponding cache operation.

[0077] The second method is to delete the session identifier from the client and clear the session identifier stored locally on the client.

[0078] The third method is to use session identifier invalidation, which can be achieved by setting a timeout or expiration flag. The session timeout can be set on the server side so that the session identifier automatically expires after a certain period of time.

[0079] The fourth option is to exit or log out.

[0080] The above method ensures that the session identifier is properly destroyed after the user logs out or the session expires to prevent session abuse or data leakage.

[0081] Here, the first time period is set to prevent session abuse or data leakage due to timeout, and the second time period is set to strengthen user identity authentication, enhance identity access density, ensure user security, avoid loss of visitor information, and ensure user security.

[0082] Big model processing: After completing identity verification, the big model begins to process the user's legal consultation request. Based on a large amount of legal data and case library, the big model analyzes the user's specific problems and provides personalized legal advice or solutions. The big model can also further adjust the content and style of the answer based on the user's historical consultation records and specific preferences.

[0083] The system generates personalized legal service content based on the legal advice generated by the big model and combined with the user's personal information and preferences. These service contents can be detailed legal advice, document templates, legal risk assessment reports, etc.

[0084] All user data and service results are transmitted through encrypted channels and securely stored in the system's database. The system regularly backs up and conducts security reviews on stored data to ensure the privacy and security of user data.

[0085] The second aspect of the present invention provides a system 800 for identity authentication based on a large model, such as Figure 2 As shown, including:

[0086] Collection module 801: used to collect the facial data of the requester and perform feature extraction.

[0087] The generation module 802 is used to convert the facial data after feature extraction into a feature vector, generate a session identifier, and bind or unbind the session identifier with the facial data of the requester.

[0088] The sending module 803 is used to construct the request data according to the session identifier, encrypt the request data, and send it to the large model. The receiving module is used to receive the request data, store the request data, decrypt the request data, and store the face data in the request data.

[0089] Determination module 804: Determine whether the face data is consistent with the user identity stored in the large model.

[0090] Response module 805: If the facial data is consistent with the facial data, receive the request and respond; if not, unbind the session identifier and the requester's facial data.

[0091] According to an embodiment of the present invention, the session identifier also includes a timestamp.

[0092] According to one embodiment of the present invention, the system also includes: a first setting module, used to set a first setting first time period, determine whether the difference between the current time and the session timestamp is less than the first time period, if the difference between the current time and the session timestamp is less than the first time period, the session continues; if the difference between the current time and the session timestamp is greater than or equal to the first time period, the session is deemed expired, the session identifier is destroyed, and the collection module is returned to collect facial data and perform feature extraction.

[0093] The second setting module sets a second time period, which is smaller than the first time period, and determines whether the difference between the current time and the session timestamp is equal to the second time period. If the difference between the current time and the session timestamp is equal to the second time period, the module returns to the collection module to collect face data and perform feature extraction. If the difference between the current time and the session timestamp is not equal to the second time period, the session continues.

[0094] The third aspect of the present invention provides an intelligent device, including a transmitter, a receiver, a memory and a processor; the memory is used to store computer instructions; the processor is used to execute the computer instructions stored in the memory to implement the above-mentioned large model-based identity authentication method.

[0095] A fourth aspect of the present invention provides a storage medium, comprising: a readable storage medium and computer instructions, wherein the computer instructions are stored in the readable storage medium; the computer instructions are used to implement the above-mentioned large model-based identity authentication method.

[0096] The beneficial effects provided by the present invention are as follows: through a special biometric verification method, the authenticity of the user's identity is ensured to prevent unauthorized access; determining a stable user identity is a prerequisite for providing personalized services, and verifying the user's identity is crucial. At the same time, reliable verification of the user's identity provides a very reliable guarantee for the subsequent data model and learning of the large model, and provides a guarantee for more efficient learning of the subsequent model and personalized services. At the same time, manual participation is completely reduced, service costs are reduced, and users have a better service experience, while improving service efficiency.

[0097] Obviously, the above specific implementation cases are only examples for illustrating the application of the present method, rather than limiting the implementation methods. For those skilled in the art, other different forms of changes and modifications can be made on the basis of the above description to study other related issues. Therefore, the protection scope of the present invention should be the protection scope of the claims.

[0098] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0099] The electronic device and other embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, rather than to limit them. Although the embodiments of the present invention have been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

[0102] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be regarded as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0103] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for identity verification based on a large model, characterized in that: The method is used for user identity authentication of a request for access in a large model dialogue, and includes: S101: Collecting the requester's facial data and performing feature extraction; S102: Convert the facial data after feature extraction into a feature vector, generate a session identifier, and bind or unbind the session identifier to the facial data of the requester; S103: construct request data according to the session identifier, encrypt the request data, and send it to the big model; S104: the large model receives the request data, stores the request data, decrypts the request data and stores the face data in the request data; S105: Determine whether the face data is consistent with the user identity stored in the large model; S106: If the facial data is consistent with the facial data, receive the request and respond; if not, return to step S102.

2. The method according to claim 1, characterized in that In the step S102, the facial data after feature extraction is converted into a feature vector to generate a session identifier, including writing the feature vector of the facial data into the session identifier.

3. The method according to claim 1, characterized in that The step S102 of binding or unbinding the session identifier with the facial data of the requester includes: When a user requests access and a session identifier is generated, the session identifier is bound to the facial data, and the binding result is recorded in the session identifier as a number or code; When the large model system determines that the facial data is inconsistent with the stored user identity, it unbinds the session identifier from the requester's facial data, records the unbinding result, a number or code, into the session identifier, and terminates the session.

4. The method according to claim 3, characterized in that The step S103 constructs the request data according to the session identifier, including determining whether the session identifier is bound to the requester's facial data, and if the session identifier is bound to the requester's facial data, constructing the request data according to the session identifier; If the session identifier is unbound from the requester's facial data, the request data is not constructed and the session is terminated; The request data includes a session identifier and a request instruction.

5. The method according to claim 4, characterized in that The session identifier also includes a timestamp.

6. The method according to claim 5, characterized in that The step S106 further includes: S106: Set a first time period, and determine whether the difference between the current time and the session timestamp is less than the first time period. If the difference between the current time and the session timestamp is less than the first time period, the session continues. If the difference between the current time and the session timestamp is greater than or equal to the first time period, the session is deemed expired, the session identifier is destroyed, and the process returns to step S101.

7. The method according to claim 6, characterized in that The step S106 further includes: A second time period is set, where the second time period is smaller than the first time period, and whether the difference between the current time and the session timestamp is equal to the second time period is determined. If the difference between the current time and the session timestamp is equal to the second time period, the process returns to step S101. If the difference between the current time and the session timestamp is not equal to the second time period, the process continues the session.

8. A system for identity verification based on a large model, characterized in that: The system comprises: Collection module: used to collect the requester's facial data and perform feature extraction; A generation module; used to convert the facial data after feature extraction into a feature vector, generate a session identifier, and bind or unbind the session identifier with the facial data of the requester; Sending module: used for constructing request data according to the session identifier, encrypting the request data, and sending it to the big model; Receiving module: used for receiving the request data, storing the request data, decrypting the request data and storing the face data in the request data; Judgment module: judge whether the face data is consistent with the user identity stored in the large model; Response module: If the facial data is consistent with the facial data, receive the request and respond; if not, unbind the session identifier from the requester's facial data.

9. The system according to claim 8, characterized in that The session identifier also includes a timestamp.

10. The system according to claim 9, characterized in that The system further comprises: A first setting module is used to set a first time period, determine whether the difference between the current time and the session timestamp is less than the first time period, if the difference between the current time and the session timestamp is less than the first time period, the session continues; if the difference between the current time and the session timestamp is greater than or equal to the first time period, the session is deemed expired, the session identifier is destroyed, and the collection module is returned to collect face data and perform feature extraction; The second setting module sets a second time period, which is smaller than the first time period, and determines whether the difference between the current time and the session timestamp is equal to the second time period. If the difference between the current time and the session timestamp is equal to the second time period, the module returns to the collection module to collect face data and perform feature extraction. If the difference between the current time and the session timestamp is not equal to the second time period, the session continues.

Images