Identity authentication method and related device

Abstract

The embodiment of the invention provides an identity authentication method. The method comprises the following steps: acquiring real-name information indicated by a target object in a real-name authentication process; authenticating an authentication interface of a tool through an event corresponding to the target service; determining a corresponding auditing rule according to an event type corresponding to the event label information, and inputting the multi-modal large model based on the auditing rule for information auditing to obtain an auditing result; if the auditing result indicates that the auditing is passed, comparing the real name information with the identity information indicated by the label content to obtain a pre-confidence identifier; and configuring an identity authentication tag corresponding to the target object based on the pre-confidence identifier to support execution of the target service. The identity authentication process of the confidence blank period is achieved, due to the fact that rechecking of real name authentication and event authentication is adopted, and a multi-mode large model is adopted in the event authentication process to conduct a targeted auditing process of a specific rule, the effectiveness of a conclusion obtained through event authentication is improved, and the accuracy of identity authentication is improved.

Description

Technical Field

[0001] This specification relates to the field of computer application technology, specifically to identity authentication technology within the field of computer application technology, and more specifically to identity authentication methods and related devices. Background Technology

[0002] In logistics scenarios, student users are an important user group, and their business needs are cyclical, so corresponding promotional activities can be configured accordingly. However, in some scenarios, non-student users often use related services illegally by exploiting identity authentication vulnerabilities, especially during the period when student registration information is blank for new students, and the identity information available for reference is limited. Therefore, how to authenticate the identity of student users has become a problem. Summary of the Invention

[0003] This specification provides an identity authentication method and related apparatus to improve the accuracy of identity authentication.

[0004] To achieve the above technical objectives, the embodiments of this specification provide the following technical solutions: Firstly, one embodiment of this specification provides an authentication method, including: Obtain the real-name information indicated by the target object in the real-name authentication process, wherein the real-name authentication process is associated with the target business; The event tag information input by the target object is obtained through the authentication interface of the event authentication tool corresponding to the target business. The event tag information includes tag content of different modalities, and the tag content of different modalities is related to each other. Based on the event type corresponding to the event tag information, the corresponding review rules are determined, and review prompt words are configured based on the review rules. The review prompt words and the event tag information are then input into a multimodal big data model to review the tag content in the event tag information according to the review rules, and the review results are obtained. If the review result indicates that the review is passed, the real-name information is compared with the identity information indicated by the label content to obtain a pre-confidence identifier; The identity authentication tag corresponding to the target object is configured based on the pre-confidence identifier. The identity authentication tag is used to support the execution of the target service. The pre-confidence identifier is a permission identifier for the target object to execute the target service before it is authenticated by the persistent authentication tool.

[0005] Secondly, one embodiment of this specification provides an identity authentication device, comprising: The acquisition unit is used to acquire the real-name information indicated by the target object in the real-name authentication process, wherein the real-name authentication process is associated with the target business. The acquisition unit is further configured to acquire the event tag information input by the target object through the authentication interface of the event authentication tool corresponding to the target business. The event tag information includes tag content of different modalities, and the tag content of different modalities is related to each other. The authentication unit is used to determine the corresponding review rules based on the event type corresponding to the event tag information, and configure review prompt words based on the review rules, so as to input the review prompt words and the event tag information into the multimodal big data model to review the tag content in the event tag information according to the review rules and obtain the review result; The authentication unit is further configured to compare the real-name information with the identity information indicated by the label content if the review result indicates that the review has passed, so as to obtain a pre-confidence identifier; The authentication unit is further configured to configure an identity authentication tag corresponding to the target object based on the pre-confidence identifier. The identity authentication tag is used to support the execution of the target service. The pre-confidence identifier is a permission identifier for the target object to execute the target service before it is authenticated by the persistent authentication tool.

[0006] Optionally, in one possible implementation, the acquisition unit is specifically used to acquire the event-related image input by the target object through the authentication interface of the event authentication tool corresponding to the target business; The acquisition unit is specifically used to rotate the event-associated image according to a preset recognition direction to obtain a rotated image; The acquisition unit is specifically used to perform content region detection on the corrected image to obtain multiple label content regions, so as to determine the event label information input by the target object through the label content regions.

[0007] Optionally, in one possible implementation, the acquisition unit is specifically used to configure area prompt words based on the authentication scenario corresponding to the event authentication tool; The acquisition unit is specifically used to input the region prompt words and the straightened image into the document detection model to mark multiple positioning boxes in the straightened image. The document detection model is based on a lightweight configuration. The acquisition unit is specifically configured to configure the positioning box as the label content area, so as to determine the event label information input by the target object through the label content area.

[0008] Optionally, in one possible implementation, the authentication unit is specifically used to determine event description information based on the event type corresponding to the event tag information; The authentication unit is specifically used to determine the timeliness parameters, content overview, and confidence symbols contained in the event description information; The authentication unit is specifically used to configure the review rules based on the timeliness parameters, the content overview, and the confidence symbols, and to configure review prompt words based on the review rules; The authentication unit is specifically used to input the audit prompt words and the event tag information into the multimodal big data model to audit the tag content in the event tag information according to the audit rules, and obtain the audit result.

[0009] Optionally, in one possible implementation, the authentication unit is specifically used to input the audit prompt words and the event tag information into the multimodal big model, and generate an audit sequence based on the tag content in the event tag information; The authentication unit is specifically used to extract information based on the items included in the audit sequence according to the audit rules, and obtain the information extraction value. The authentication unit is specifically used to configure the audit result through the result information corresponding to the information extraction value.

[0010] Optionally, in one possible implementation, the authentication unit is specifically used to compare the real-name information with the identity information indicated by the label content if the audit result indicates that the audit has passed, so as to obtain a real-name confidence identifier. The authentication unit is specifically used to obtain the facial information corresponding to the target object; The authentication unit is specifically used to compare the facial information with the facial information indicated by the label content to obtain a facial confidence identifier; The authentication unit is specifically used to combine the real-name trust identifier and the face trust identifier to obtain the pre-trust identifier.

[0011] Optionally, in one possible implementation, the authentication unit is specifically used to collect business information of the target object performing the express delivery business during the configuration of the pre-confidence identifier when obtaining the identity authentication tag; The authentication unit is specifically used to locate the abnormality through the business information if the business status corresponding to the target object is abnormal, so as to determine the distribution of abnormal express packages. The authentication unit is specifically used for business compliance management based on the distribution of abnormal express shipments.

[0012] Optionally, in one possible implementation, the authentication unit is specifically used to obtain the authentication validity parameters output by the persistent authentication tool after the target object has been authenticated by the persistent authentication tool; The authentication unit is specifically used to configure a confidence identifier based on the authentication timeliness parameter; The authentication unit is specifically configured to configure the identity authentication tag corresponding to the target object based on the confidence identifier. The identity authentication tag is used to support the execution of the target service. The confidence identifier is a permission identifier for the target object to execute the target service within the time range corresponding to the authentication time limit parameter.

[0013] Thirdly, one embodiment of this specification also provides a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the authentication method described above.

[0014] Fourthly, one embodiment of this specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the authentication method described above.

[0015] Fifthly, embodiments of this specification provide a computer program product or computer program, the computer program product including a computer program that can be stored in a computer-readable storage medium or in the cloud; the processor of the computer device reads the computer program, and when the processor executes the computer program, it implements the steps of the above-described authentication method.

[0016] As can be seen from the above technical solution, the identity authentication method provided in this specification obtains the real-name information indicated by the target object in the real-name authentication process, which is associated with the target business; then, through the authentication interface of the event authentication tool corresponding to the target business, it obtains the event tag information input by the target object, and the event tag information includes tag content of different modalities, which are interconnected; and determines the corresponding review rules according to the event type corresponding to the event tag information, and configures review prompt words based on the review rules, so as to input the review prompt words and event tag information into the multimodal big model to review the tag content in the event tag information according to the review rules, and obtain the review result; if the review result indicates that the review is passed, the real-name information is compared with the identity information indicated by the tag content to obtain a pre-confidence identifier; then, based on the pre-confidence identifier, the identity authentication tag corresponding to the target object is configured, which is used to support the execution of the target business, and the pre-confidence identifier is the permission identifier for the target object to execute the target business before being authenticated by the persistent authentication tool. This enables identity authentication during the confidence gap period. By employing verification through real-name authentication and event authentication, and by using a multimodal large model to conduct targeted review of specific rules during the event authentication process, the effectiveness of the conclusions obtained from event authentication is improved, and the accuracy of identity authentication is enhanced. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this specification. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 A network architecture diagram for the operation of an identity authentication system.

[0019] Figure 2 This is a flowchart illustrating an identity authentication process provided in an embodiment of this application.

[0020] Figure 3 This is a flowchart illustrating an identity authentication method provided for one embodiment of this specification.

[0021] Figure 4 This is a schematic diagram illustrating a scenario of an identity authentication method provided for one embodiment of this specification.

[0022] Figure 5 This is a schematic diagram of the functional modules of an identity authentication device provided in one embodiment of this specification.

[0023] Figure 6 This is a schematic diagram of the structure of a computing device provided for one embodiment of this specification. Detailed Implementation

[0024] Unless otherwise defined, the technical or scientific terms used in the embodiments of this specification shall have the ordinary meaning understood by one of ordinary skill in the art to which this specification pertains. The terms "first," "second," and similar terms used in the embodiments of this specification do not indicate any order, quantity, or importance, but are merely used to avoid confusion of constituent elements.

[0025] Unless the context otherwise requires, throughout this specification, "a plurality of" means "at least two," and "including" is interpreted as open-ended or encompassing, that is, "including, but not limited to." In the description of this specification, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this specification. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example.

[0026] It should be understood that the authentication method provided in this application can be applied to systems or programs in terminal devices that include authentication functions, such as logistics management applications. Specifically, the authentication system can run in environments such as... Figure 1 In the network architecture shown, such as Figure 1 The diagram shown illustrates the network architecture of an identity authentication system. As can be seen, the system can provide authentication services to multiple information sources. This is achieved by having a terminal upload identity information, which then triggers the server to perform the corresponding authentication process. This can be understood as... Figure 1 The document shows various terminal devices, which can be computer devices. In real-world scenarios, more or fewer types of terminal devices may participate in the identity authentication process. The specific number and types depend on the actual scenario and are not limited here. Figure 1 The example shows one server, but in real-world scenarios, multiple servers can be involved, especially in multidisciplinary output scenarios. The specific number of servers depends on the actual scenario.

[0027] In this embodiment, the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, and the terminal and server can be connected to form a blockchain network; this application does not impose any restrictions.

[0028] It is understandable that the aforementioned identity authentication system can run on personal mobile terminals, such as applications for logistics management, or on servers, or on third-party devices to provide identity authentication and obtain the identity authentication processing results of information sources. Specifically, the identity authentication system can run as a program on the aforementioned devices, or as a system component within the aforementioned devices, or as a cloud service program. The specific operating mode depends on the actual scenario and is not limited here.

[0029] In logistics scenarios, student users are an important user group, and their business needs are cyclical, so corresponding promotional activities can be configured accordingly. However, in some scenarios, non-student users often use related services illegally by exploiting identity authentication vulnerabilities, especially during the period when student registration information is blank for new students, and the identity information available for reference is limited. Therefore, how to authenticate the identity of student users has become a problem.

[0030] To address the aforementioned problems, this application proposes an identity authentication method, which is applied to... Figure 2 In the identity authentication process framework shown, as follows: Figure 2 The diagram shown is a flowchart of an identity authentication process provided in this application embodiment. The terminal sends the corresponding identity information to the server through the authentication request, so that the multimodal big model configured in the server performs the event authentication dimension review. After the review is passed, it is verified with the real name information and outputs a pre-confidence identifier to support the user to authenticate the business execution before the persistent authentication tool, thereby improving the accuracy of identity authentication.

[0031] It is understood that the authentication method provided in this application can be a program written as a processing logic in a hardware system, or it can be an authentication device that implements the above processing logic in an integrated or external manner. As one implementation method, the identity authentication device obtains the real-name information indicated by the target object in the real-name authentication process, which is associated with the target business. Then, through the authentication interface of the event authentication tool corresponding to the target business, it obtains the event tag information input by the target object. The event tag information includes tag content of different modalities, and the tag content of different modalities is interconnected. Based on the event type corresponding to the event tag information, it determines the corresponding review rules and configures review prompt words based on the review rules. The review prompt words and event tag information are input into a multimodal big model to review the tag content in the event tag information according to the review rules and obtain the review result. If the review result indicates that the review is passed, the real-name information is compared with the identity information indicated by the tag content to obtain a pre-confidence identifier. Then, based on the pre-confidence identifier, the device configures the identity authentication tag corresponding to the target object. The identity authentication tag is used to support the execution of the target business. The pre-confidence identifier is a permission identifier for the target object to execute the target business before being authenticated by the persistent authentication tool. This enables identity authentication during the confidence gap period. By employing verification through real-name authentication and event authentication, and by using a multimodal large model to conduct targeted review of specific rules during the event authentication process, the effectiveness of the conclusions obtained from event authentication is improved, and the accuracy of identity authentication is enhanced.

[0032] Based on the above process architecture, the authentication method in this application will be described below. Please refer to [link / reference]. Figure 3 , Figure 3 A flowchart of an identity authentication method provided in this application embodiment, which includes at least the following steps: 301. Obtain the real-name information indicated by the target object in the real-name authentication process, which is related to the target business.

[0033] In this embodiment, the target object is the user who uses the target service. Specifically, it can be a user who needs to authenticate their identity to enjoy the corresponding functions of the target service, such as student users or other specific groups of users. The target service is a service for a specific group, which can be a logistics service, such as a courier service for student members or a warehousing service for market cargo owners. The specific form depends on the actual scenario.

[0034] Specifically, the real-name authentication process for the target object can be carried out through documents or through network information retrieval. This real-name authentication process can be recorded as one of the login information of the target business, thereby improving the transparency of business execution.

[0035] 302. Obtain the event tag information input by the target object through the authentication interface of the event authentication tool corresponding to the target business. The event tag information includes tag content of different modalities, and the tag content of different modalities are related to each other.

[0036] In this embodiment, the event authentication tool corresponding to the target business is the authentication method configured for events that must occur before the target object executes the target business. For example, if the target business is a student membership business, the event authentication tool could be an authentication tool configured based on the exam event, and the corresponding event tag information could be information triggered by the specific exam event, such as an admission notice or examination admission ticket. Correspondingly, the authentication interface is the information upload method corresponding to different event authentication tool forms, such as the interface for uploading an admission notice.

[0037] Furthermore, since event tag information is generally in the form of images, but the images usually contain text descriptions with specific identity meanings, and the text descriptions are related to the content displayed in the images, that is, the multimodal tag content is all configured for the event authentication tool.

[0038] In one possible scenario, taking an admission notice as an example, considering that different schools have different image display formats for their admission notices, and different formats may lead to misunderstandings, image preprocessing can be performed to improve the accuracy of subsequent recognition. Specifically, for the process of obtaining the event tag information input by the target object through the authentication interface of the event authentication tool corresponding to the target business, the event-related image input by the target object can be obtained first through the authentication interface of the event authentication tool corresponding to the target business; then, the event-related image is rotated according to a preset recognition direction to obtain a corrected image; and content region detection is performed on the corrected image to obtain multiple tag content regions, so as to determine the event tag information input by the target object through the tag content regions.

[0039] The process of rotating the image according to a preset recognition direction—that is, straightening the image to align the text—is crucial. Many uploaded images have text orientations of 90°, 180°, or 270°, affecting the recognition performance of large multimodal models. Therefore, image orientation straightening is performed first. Specifically, a high-efficiency, lightweight neural network model designed for document orientation classification tasks (such as the PaddlePaddle open-source PP-LCNet_x1_0_doc_ori) can be used. This type of model is built on a PP-LCNet backbone network optimized for CPU performance. Its core function is to automatically determine the orientation of scanned or captured document images (e.g., 0°, 90°, 180°, 270°), thus providing crucial preprocessing capabilities for document scanning apps and OCR systems. This model maintains high accuracy while being extremely lightweight, and features high speed and low computational resource consumption, making it ideal for integration into mobile or edge computing devices to achieve efficient automatic document orientation correction.

[0040] Furthermore, considering that some users' photos of admission notices may be small, or have their long side parallel to the short side of the photo or perpendicular to the field edge, in actual production applications, to ensure service stability, images are usually scaled to limit the maximum side size. This can lead to smaller, harder-to-read text. Therefore, document region detection is needed to improve text clarity within the image size limit. Thus, for the process of performing content region detection on a corrected image to obtain multiple labeled content regions, and then determining the event label information input by the target object through these labeled content regions, we can first configure region prompts based on the authentication scenario corresponding to the event authentication tool; then, the region prompts and the corrected image are input into a document detection model to mark multiple bounding boxes in the corrected image. The document detection model is based on a lightweight configuration; and these bounding boxes are configured as labeled content regions to determine the event label information input by the target object.

[0041] For the document region detection mentioned above, a lightweight document detection model (such as LLMDet's tiny model) is used. The text region prompt is set to "offer letter document" and the model can generate a detection box for the admission letter document region. The detection box is used to crop the image and obtain the cropped image for subsequent information extraction.

[0042] Specifically, the lightweight document detection model's core advantage lies in its ability to accurately locate and select key text regions (such as student name, admitted major, university name, and registration date) from complex admission notice images with extremely high speed and precision. Through innovative multi-scale feature fusion and hierarchical prediction head design, this model significantly reduces computational load and model parameters while still handling scenarios characterized by diverse admission notice layouts, dense text, and complex backgrounds.

[0043] As can be seen, this embodiment, after image preprocessing, shows a significant improvement in recognition accuracy compared to directly applying a multimodal large model.

[0044] 303. Determine the corresponding review rules based on the event type corresponding to the event tag information, and configure review prompt words based on the review rules. Input the review prompt words and event tag information into the multimodal big data model to review the tag content in the event tag information according to the review rules and obtain the review results.

[0045] In this embodiment, the audit rules are the information authentication rules configured for the event types corresponding to different event tag information, that is, to audit the accuracy and validity of the information in the event tag information used to indicate the identity of the target object.

[0046] Specifically, since event tag information includes tag content in different modalities, the review process adopts a multimodal large model. Therefore, for the review process, the event description information can be determined first based on the event type corresponding to the event tag information; then, the timeliness parameters, content overview, and confidence symbols contained in the event description information can be determined; and review rules can be configured based on the timeliness parameters, content overview, and confidence symbols, and review prompt words can be configured based on the review rules; then, the review prompt words and event tag information are input into the multimodal large model to review the tag content in the event tag information according to the review rules, and the review result is obtained.

[0047] For example, in a scenario where the event tag information is "admission notice," the admission notice review rules could include: 1. Certificates issued in the current year (validity parameter), for example, 2025. Except for prospective university students, other current students have student registration information on the student registration website, and this information should be used for authentication first. Therefore, the certificate is limited to those issued in the current year.

[0048] 2. Includes name (content summary). Do not censor or obscure.

[0049] 3. Includes an official seal (credibility symbol). The inclusion of an official seal can exclude many cases of non-admission notices or blank admission notice templates. Furthermore, the official seal has legal validity, cannot be forged by individuals, and possesses a certain legal deterrent effect.

[0050] 4. The content conforms to the general format of an admission notice (content summary), for example, "Student XX, you have been admitted to our school's XX major."

[0051] Specifically, for the review process based on a multimodal large model, it can be carried out by rule-based information extraction. That is, firstly, the review prompt words and event tag information are input into the multimodal large model, and a review sequence is generated based on the tag content in the event tag information; then, information is extracted according to the review rules based on the items contained in the review sequence to obtain the information extraction value; and the review result is configured through the result information corresponding to the information extraction value.

[0052] For example, in a scenario where the event tag information is an admission notice, the information extraction process can utilize a multimodal large-scale model (e.g., Qwen2.5-VL-7B) to determine the document's date, student's name, whether it bears an official seal, and whether it is indeed an admission notice. The method of use is to input an image and a text prompt containing the above verification items; the model then provides the extracted information. For example: { "Is it a 2025 document?": true, Student Name: Zhang San "Student's complete name": true, "Stamped": true "It is a university admission notice": true, "School Level": "University" } For information review, the information extraction results provided by the model are used to perform post-processing of the rules in conjunction with the review rules, and a final review conclusion of True / False is given.

[0053] In another possible scenario, taking the event tag information as an example of an examination admission ticket, the verification rules for the examination admission ticket may include: 1. Certificates issued in the current year (validity parameter), for example, 2025. Except for prospective university students, other current students have student registration information on the student registration website, and this information should be used for authentication first. Therefore, the certificate is limited to those issued in the current year.

[0054] 2. Includes name (content summary). Do not censor or obscure.

[0055] 3. Includes a passport photo (content overview). Verifying that the admission ticket matches the examinee is one of the authentication methods for entering the college entrance examination, therefore, admission tickets generally include a passport photo.

[0056] 4. Does the examination subject fully conform to the National College Entrance Examination (NCEE) subjects (content overview), such as Chinese, Mathematics, English, Physics, Chemistry, Biology, Politics, History, and Geography? This method can eliminate many non-NCEE admission tickets, such as those for the CET-4 or CET-6 exams.

[0057] 5. Whether it is a nationally unified higher education entrance examination admission ticket (confidence symbol). This information is usually included in the title of the admission ticket.

[0058] Correspondingly, the information extraction process for the examination admission ticket utilizes a multimodal large-scale model to determine the document's date, student's name, whether it includes a photo, examination subjects, and whether it is indeed a college entrance examination admission ticket. The method of use is to input an image along with a text prompt containing the above-mentioned verification items; the model then provides the extracted information. For example: { "Year of Certificate": "2025" Student Name: Zhang San "Contains student name": true, "All are subjects tested in the National College Entrance Examination": true, "This is the college entrance examination admission ticket": true, "Includes passport-style photos": false } Correspondingly, information review involves using the information extraction results provided by the model, combining them with review rules for post-processing, and giving a final review conclusion of True / False.

[0059] By utilizing multimodal large models, different types / forms of event tag information can be identified, making the identity authentication process generalizable, such as admission notices from different universities and examination admission tickets from different regions.

[0060] 304. If the review result indicates that the review is approved, the real-name information will be compared with the identity information indicated by the label content to obtain the pre-confidence identifier.

[0061] In this embodiment, the audit result indicates that all audits are true. At this point, it is necessary to verify whether the real-name information provided by the user matches the event information to prevent users from using downloaded images for authentication. Specifically, the user's real-name authentication name is compared with the name extracted from the multimodal large model; if they match, it is considered to be the user's identification document.

[0062] Specifically, the identity verification process can involve dual verification using both real name and facial recognition. If the verification result indicates approval, the real name information is compared with the identity information indicated by the label content to obtain a real name confidence identifier. Then, the facial information corresponding to the target object is obtained and compared with the facial information indicated by the label content to obtain a facial confidence identifier. Finally, the real name confidence identifier and facial confidence identifier are combined to obtain a pre-confidence identifier. That is, for the name verification process, it determines whether the name extracted by the model matches the name verified in the real name authentication. For the facial verification process, it compares whether the ID card photo and the admission ticket photo are of the same person. A facial verification method (such as CompareFace) is used to detect and compare the faces appearing in the two photos, and the model provides a result indicating whether they match.

[0063] By employing dual authentication and multi-dimensional verification, the accuracy of identity verification for target objects when performing target business can be improved. For example, student identity verification can be completed before students enroll.

[0064] 305. Configure the identity authentication label corresponding to the target object based on the pre-confidence identifier. The identity authentication label is used to support the execution of the target business. The pre-confidence identifier is the permission identifier for the target object to execute the target business before it is authenticated by the persistent authentication tool.

[0065] In this embodiment, the pre-confidence identifier is a permission identifier for the target object to execute the target business before being authenticated by the persistent authentication tool. It is a phased confidence identifier, which only supports the time period between the verification authentication by the event authentication tool and real-name authentication and the authentication by the persistent authentication tool. This is because the verification process of the event authentication tool and real-name authentication can only verify the accuracy of the identity at the current moment. If the identity of the target object changes after being authenticated, it cannot proceed. Therefore, the auxiliary authentication by the persistent authentication tool is required. The persistent authentication tool dynamically records the identity information required by the target object during the execution of the target business, thereby avoiding business execution disorder caused by identity changes.

[0066] Specifically, for scenarios targeting students, persistent authentication tools could be student registration verification websites, such as the China Higher Education Student Information System (CHESICC). These websites typically only begin authenticating students some time after registration, usually with a delay of about two months compared to the time it takes to receive admission notices / exam admission tickets. Before enrollment, students have a potentially large demand for item delivery. If student membership authentication can be implemented during the gap in student registration information, it can help attract student members, cultivate brand recognition and usage habits, and increase the activity of student membership services.

[0067] Furthermore, to enhance the security of parcels during the pre-established trust identifier configuration period, when the target business is express delivery, the parcels involved can be recorded during this period to facilitate tracing after anomalies are discovered. Specifically, during the pre-established trust identifier configuration period, the identity authentication tag collects business information about the target object performing the express delivery business; if the business status corresponding to the target object is abnormal, the anomaly is located using the business information to determine the distribution of abnormal parcels; and then business compliance management is performed based on the distribution of abnormal parcels.

[0068] Among these, business compliance management can involve identifying the sender or recipient of abnormal packages, determining whether they are involved in any violations, and intervening in the relevant parties in a timely manner to prevent the expansion of business loopholes and improve the stability of the target business execution.

[0069] In one possible scenario, such as Figure 4 As shown, Figure 4 This diagram illustrates a scenario of an identity authentication method provided in one embodiment of this specification. The diagram shows the execution scenario of this embodiment, specifically the support scenario of an event authentication tool and a persistent authentication tool for a target service. After dual authentication via real-name authentication and event authentication, a pre-configured confidence identifier is used for phased management to maintain the operation of the target service. After the target object is authenticated by the persistent authentication tool, the authentication validity period parameters output by the persistent authentication tool are obtained. Then, a confidence identifier is configured based on the authentication validity period parameters, that is, the pre-configured confidence identifier is replaced with the confidence identifier. Furthermore, an identity authentication tag corresponding to the target object is configured based on the confidence identifier. The identity authentication tag is used to support the execution of the target service, and the confidence identifier is a permission identifier for the target object to execute the target service within the time range corresponding to the authentication validity period parameters. This achieves a full-cycle management process for providing target service functions when the target object belongs to a target group.

[0070] In summary, this embodiment obtains the real-name information indicated by the target object in the real-name authentication process, which is associated with the target business. Then, through the authentication interface of the event authentication tool corresponding to the target business, it obtains the event tag information input by the target object. The event tag information includes tag content of different modalities, and the tag content of different modalities is interconnected. Based on the event type corresponding to the event tag information, it determines the corresponding review rules and configures review prompt words based on the review rules. The review prompt words and event tag information are input into a multimodal big model to review the tag content in the event tag information according to the review rules and obtain the review result. If the review result indicates that the review is passed, the real-name information is compared with the identity information indicated by the tag content to obtain a pre-confidence identifier. Then, based on the pre-confidence identifier, it configures the identity authentication tag corresponding to the target object. This identity authentication tag is used to support the execution of the target business. The pre-confidence identifier is the permission identifier for the target object to execute the target business before being authenticated by the persistent authentication tool. This enables identity authentication during the confidence gap period. By employing verification through real-name authentication and event authentication, and by using a multimodal large model to conduct targeted review of specific rules during the event authentication process, the effectiveness of the conclusions obtained from event authentication is improved, and the accuracy of identity authentication is enhanced.

[0071] It should be noted that the various embodiments described in this specification emphasize the parts that differ from other embodiments, and the embodiments can be explained by comparison with each other. Any combination of the various embodiments described in this specification based on general technical knowledge is covered within the scope of this specification.

[0072] In one exemplary embodiment of this specification, an identity authentication device 500 is also provided, such as... Figure 5 As shown, Figure 6 This is a functional module diagram of an identity authentication device provided in one embodiment of the present specification. The authentication device 500 includes: The acquisition unit 501 is used to acquire the real-name information indicated by the target object in the real-name authentication process, wherein the real-name authentication process is associated with the target business. The acquisition unit 501 is further configured to acquire the event tag information input by the target object through the authentication interface of the event authentication tool corresponding to the target business. The event tag information includes tag content of different modalities, and the tag content of different modalities is related to each other. The authentication unit 502 is used to determine the corresponding review rules according to the event type corresponding to the event tag information, and configure review prompt words based on the review rules, so as to input the review prompt words and the event tag information into the multimodal big data model to review the tag content in the event tag information according to the review rules and obtain the review result; The authentication unit 502 is further configured to compare the real-name information with the identity information indicated by the label content if the audit result indicates that the audit has passed, so as to obtain a pre-confidence identifier; The authentication unit 502 is further configured to configure an identity authentication tag corresponding to the target object based on the pre-confidence identifier. The identity authentication tag is used to support the execution of the target service. The pre-confidence identifier is a permission identifier for the target object to execute the target service before it is authenticated by the persistent authentication tool.

[0073] Optionally, in one possible implementation, the acquisition unit 501 is specifically used to acquire the event-related image input by the target object through the authentication interface of the event authentication tool corresponding to the target business; The acquisition unit 501 is specifically used to rotate the event-associated image according to a preset recognition direction to obtain a rotated image; The acquisition unit 501 is specifically used to perform content region detection on the straightened image to obtain multiple label content regions, so as to determine the event label information input by the target object through the label content regions.

[0074] Optionally, in one possible implementation, the acquisition unit 501 is specifically used to configure area prompt words based on the authentication scenario corresponding to the event authentication tool; The acquisition unit 501 is specifically used to input the region prompt words and the straightened image into the document detection model to mark multiple positioning boxes in the straightened image. The document detection model is based on a lightweight configuration. The acquisition unit 501 is specifically used to configure the positioning box as the label content area, so as to determine the event label information input by the target object through the label content area.

[0075] Optionally, in one possible implementation, the authentication unit 502 is specifically used to determine event description information based on the event type corresponding to the event tag information; The authentication unit 502 is specifically used to determine the timeliness parameters, content overview and confidence symbols contained in the event description information; The authentication unit 502 is specifically used to configure the audit rules based on the timeliness parameters, the content overview, and the confidence symbols, and to configure audit prompt words based on the audit rules; The authentication unit 502 is specifically used to input the audit prompt words and the event tag information into the multimodal big data model to audit the tag content in the event tag information according to the audit rules, and obtain the audit result.

[0076] Optionally, in one possible implementation, the authentication unit 502 is specifically used to input the audit prompt words and the event tag information into the multimodal big model, and generate an audit sequence based on the tag content in the event tag information; The authentication unit 502 is specifically used to extract information based on the items included in the audit sequence according to the audit rules, and obtain the information extraction value. The authentication unit 502 is specifically used to configure the audit result through the result information corresponding to the information extraction value.

[0077] Optionally, in one possible implementation, the authentication unit 502 is specifically used to compare the real-name information with the identity information indicated by the label content if the audit result indicates that the audit has passed, so as to obtain a real-name confidence identifier. The authentication unit 502 is specifically used to obtain the facial information corresponding to the target object; The authentication unit 502 is specifically used to compare the face information with the face information indicated by the label content to obtain a face confidence identifier. The authentication unit 502 is specifically used to combine the real-name trust identifier and the face trust identifier to obtain the pre-trust identifier.

[0078] Optionally, in one possible implementation, the authentication unit 502 is specifically used to collect business information of the target object performing the express delivery business during the configuration of the pre-confidence identifier when obtaining the identity authentication tag; The authentication unit 502 is specifically used to locate the abnormality through the business information if the business status corresponding to the target object is abnormal, so as to determine the distribution of abnormal express packages. The authentication unit 502 is specifically used for business compliance management based on the distribution of abnormal express shipments.

[0079] Optionally, in one possible implementation, the authentication unit 502 is specifically used to obtain the authentication validity parameters output by the persistent authentication tool after the target object has been authenticated by the persistent authentication tool; The authentication unit 502 is specifically used to configure a confidence identifier based on the authentication timeliness parameter; The authentication unit 502 is specifically used to configure the identity authentication tag corresponding to the target object based on the confidence identifier. The identity authentication tag is used to support the execution of the target service. The confidence identifier is a permission identifier for the target object to execute the target service within the time range corresponding to the authentication time limit parameter.

[0080] Specifically, the acquisition unit and authentication unit in this embodiment can correspond to physical components. For example, the authentication unit can be a processing module such as a CPU, GPU, or FPGA. The specific physical component can be any component or combination of components with the above functions. The specific method depends on the actual scenario and is not limited here.

[0081] The aforementioned authentication device obtains the real-name information indicated by the target object during the real-name authentication process, which is associated with the target business. Then, through the authentication interface of the event authentication tool corresponding to the target business, it obtains the event tag information input by the target object. This event tag information includes tag content in different modalities, and the tag content in different modalities is interconnected. Based on the event type corresponding to the event tag information, it determines the corresponding review rules and configures review prompts based on the review rules. The review prompts and event tag information are then input into a multimodal big data model to review the tag content in the event tag information according to the review rules, obtaining a review result. If the review result indicates that the review is passed, the real-name information is compared with the identity information indicated by the tag content to obtain a pre-confidence identifier. Furthermore, based on the pre-confidence identifier, it configures the identity authentication tag corresponding to the target object. This identity authentication tag is used to support the execution of the target business, and the pre-confidence identifier is a permission identifier for the target object to execute the target business before being authenticated by the persistent authentication tool. This enables identity authentication during the confidence gap period. By employing verification through real-name authentication and event authentication, and by using a multimodal large model to conduct targeted review of specific rules during the event authentication process, the effectiveness of the conclusions obtained from event authentication is improved, and the accuracy of identity authentication is enhanced.

[0082] For specific limitations regarding the authentication device, please refer to the limitations regarding the authentication method above, which will not be repeated here. Each unit module in the aforementioned authentication device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0083] Another embodiment of this application also proposes a computing device, see [link to relevant documentation] Figure 6 As shown, an exemplary embodiment of this specification also provides a computing device, including: a memory and a processor, the memory storing a computer program, the processor executing the computer program to perform the steps of the authentication method according to various embodiments of this specification described above.

[0084] The internal structure of the computing device can be as follows: Figure 6As shown, the computing device includes a processor, memory, network interface, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it follows the steps of the authentication methods according to various embodiments of this specification as described in the above embodiments.

[0085] The processor may include the main processor, as well as baseband chips, modems, etc.

[0086] The memory stores a program that executes the technical solution of this invention, and may also store an operating system and other critical business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0087] The processor can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0088] Input devices may include devices that receive data and information input by the user, such as keyboards, mice, cameras, scanners, light pens, voice input devices, touch screens, pedometers, or gravity sensors.

[0089] Output devices may include devices that allow information to be output to the user, such as displays, printers, speakers, etc.

[0090] The communication interface may include any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0091] The processor executes programs stored in memory and calls other devices, which can be used to implement various steps of any of the authentication methods provided in the above embodiments of this application.

[0092] The computing device may also include a display component and a voice component. The display component may be a liquid crystal display screen or an e-ink display screen. The input device of the computing device may be a touch layer covering the display component, or a button, trackball or touchpad set on the casing of the computing device, or an external keyboard, touchpad or mouse, etc.

[0093] Those skilled in the art will understand that Figure 6 The structures shown are merely block diagrams of some structures related to the solutions in this specification and do not constitute a limitation on the computing devices on which the solutions in this specification are applied. Specific computing devices may include more or fewer components than those shown in the figures, or combine certain components, or have different component arrangements.

[0094] In addition to the methods and devices described above, the authentication methods provided in the embodiments of this specification can also be computer program products, which include computer programs that, when run by a processor, cause the processor to perform the steps in the authentication methods according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0095] The computer program product described herein can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments described herein. These programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0096] Furthermore, embodiments of this specification also provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor of the steps in the authentication methods according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0098] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] The embodiments described above are merely illustrative of several implementation methods outlined in this specification. While the descriptions are specific and detailed, they should not be construed as limiting the scope of the solutions provided in this specification. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this specification, and these all fall within the scope of protection of this specification. Therefore, the scope of protection for this patent should be determined by the appended claims.

Claims

1. An identity authentication method, characterized in that, include: Obtain the real-name information indicated by the target object in the real-name authentication process, wherein the real-name authentication process is associated with the target business; The event tag information input by the target object is obtained through the authentication interface of the event authentication tool corresponding to the target business. The event tag information includes tag content of different modalities, and the tag content of different modalities is related to each other. Based on the event type corresponding to the event tag information, the corresponding review rules are determined, and review prompt words are configured based on the review rules. The review prompt words and the event tag information are then input into a multimodal big data model to review the tag content in the event tag information according to the review rules, and the review results are obtained. If the review result indicates that the review is passed, the real-name information is compared with the identity information indicated by the label content to obtain a pre-confidence identifier; The identity authentication tag corresponding to the target object is configured based on the pre-confidence identifier. The identity authentication tag is used to support the execution of the target service. The pre-confidence identifier is a permission identifier for the target object to execute the target service before it is authenticated by the persistent authentication tool.

2. The method according to claim 1, characterized in that, The step of obtaining the event tag information input by the target object through the authentication interface of the event authentication tool corresponding to the target business includes: The event-related image input by the target object is obtained through the authentication interface of the event authentication tool corresponding to the target business. The event-associated image is rotated according to a preset recognition direction to obtain a properly aligned image; Multiple labeled content regions are obtained by performing content region detection on the straightened image, and the event label information input by the target object is determined through the labeled content regions.

3. The method according to claim 2, characterized in that, The step of performing content region detection on the corrected image to obtain multiple labeled content regions, and determining the event label information input by the target object through the labeled content regions, includes: Based on the authentication scenario configuration area prompts corresponding to the event authentication tool; The region prompt and the straightened image are input into the document detection model to mark multiple bounding boxes in the straightened image. The document detection model is based on a lightweight configuration. The positioning box is configured as the label content area to determine the event label information input by the target object through the label content area.

4. The method according to claim 1, characterized in that, The process involves determining the corresponding review rules based on the event type corresponding to the event tag information, configuring review prompts based on the review rules, and inputting the review prompts and the event tag information into a multimodal big data model to review the tag content in the event tag information according to the review rules, thereby obtaining the review result, including: Determine the event description information based on the event type corresponding to the event tag information; Determine the timeliness parameters, content summary, and confidence symbols included in the event description information; Configure the review rules based on the timeliness parameters, the content overview, and the confidence symbols, and configure review prompt words based on the review rules; The review prompts and event tag information are input into a multimodal big data model to review the tag content in the event tag information according to the review rules, and the review result is obtained.

5. The method according to claim 4, characterized in that, The step of inputting the review prompts and the event tag information into a multimodal big data model to review the tag content in the event tag information according to the review rules, and obtaining the review result, includes: Input the review prompts and event tag information into the multimodal big data model, and generate a review sequence based on the tag content in the event tag information; Based on the items included in the audit sequence, information is extracted according to the audit rules to obtain information extraction values; The audit results are configured based on the result information corresponding to the extracted information values.

6. The method according to claim 1, characterized in that, If the review result indicates that the review has passed, the real-name information is compared with the identity information indicated by the label content to obtain a pre-confidence identifier, including: If the review result indicates that the review is passed, the real name information is compared with the identity information indicated by the label content to obtain a real name confidence identifier; Obtain the facial information corresponding to the target object; The facial information is compared with the facial information indicated by the label content to obtain a facial confidence identifier; The pre-confidence identifier is obtained by combining the real-name confidence identifier and the face confidence identifier.

7. The method according to any one of claims 1-6, characterized in that, The target business is express delivery service, and the method further includes: During the configuration of the pre-established trust identifier, the identity authentication tag is obtained by collecting business information of the target object performing the express delivery service. If the business status corresponding to the target object is abnormal, the abnormality is located through the business information to determine the distribution of abnormal packages; Business compliance management is based on the aforementioned abnormal package distribution.

8. The method according to any one of claims 1-6, characterized in that, The method further includes: After the target object is authenticated by the persistent authentication tool, the authentication validity period parameter output by the persistent authentication tool is obtained; Configure a confidence identifier based on the authentication timeliness parameters; The identity authentication tag corresponding to the target object is configured based on the confidence identifier. The identity authentication tag is used to support the execution of the target service. The confidence identifier is the permission identifier for the target object to execute the target service within the time range corresponding to the authentication time limit parameter.

9. An identity authentication device, characterized in that, include: The acquisition unit is used to acquire the real-name information indicated by the target object in the real-name authentication process, wherein the real-name authentication process is associated with the target business. The acquisition unit is further configured to acquire the event tag information input by the target object through the authentication interface of the event authentication tool corresponding to the target business. The event tag information includes tag content of different modalities, and the tag content of different modalities is related to each other. The authentication unit is used to determine the corresponding review rules based on the event type corresponding to the event tag information, and configure review prompt words based on the review rules, so as to input the review prompt words and the event tag information into the multimodal big data model to review the tag content in the event tag information according to the review rules and obtain the review result; The authentication unit is further configured to compare the real-name information with the identity information indicated by the label content if the review result indicates that the review has passed, so as to obtain a pre-confidence identifier; The authentication unit is further configured to configure an identity authentication tag corresponding to the target object based on the pre-confidence identifier. The identity authentication tag is used to support the execution of the target service. The pre-confidence identifier is a permission identifier for the target object to execute the target service before it is authenticated by the persistent authentication tool.

10. A computing device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the authentication method according to any one of claims 1 to 8.

Images