Big data processing terminal based on user identification authentication

Abstract

The invention discloses a big data processing terminal based on user identification and authentication. Comprising a user information registration and original image acquisition module, a user registration quality grading verification analysis module, a user face key feature selection coordinate positioning module, a user login security authentication coordination module, a user authorization access verification guarantee module and a data processing terminal. The method aims at obtaining registration information and face images of different registered users in advance, conducting multi-dimensional verification on the registration information, conducting initial coordinate system construction on the different face images of the users, quantifying quality scores of the different registered users on the basis of the verification result of the registration information, conducting quality grading on the different registered users, and conducting quality grading on the different registered users. Different facial feature recognition modes are automatically triggered, multi-feature authentication range set selection is performed on facial images of different registered users, coordinate positioning data of a multi-feature authentication range set are extracted, and a multi-feature point differentiation permission migration range is set based on the multi-feature authentication range set to perform facial multi-feature point recognition authentication.

Description

A big data processing terminal based on user identification and authentication Technical Field

[0001] This invention relates to the field of user identification, specifically a big data processing terminal based on user identification and authentication. Background Technology

[0002] With the rapid development of big data and artificial intelligence technologies, facial recognition, as the most natural and non-invasive biometric method, has become a core technology for user authentication. Currently, mainstream facial recognition authentication systems typically employ a static "registration-authentication" separation model. This means that during user registration, facial images are captured and feature templates are extracted and stored in a database. During login authentication, identity verification is performed by calculating the similarity between real-time facial features and the stored templates. Most existing systems adopt a "one-size-fits-all" authentication strategy, applying the same complex facial feature extraction algorithm and uniform similarity threshold to all users. This approach ignores the inherent credit value and risk signals of different users' registration information, preventing the system from establishing differentiated trust baselines at the authentication stage and causing a mismatch of security authentication resources. Furthermore, applying the same security standards to all users lacks... The method dynamically adjusts the authentication strength according to the user's risk level. This application aims to obtain the registration information and facial images of different registered users in advance, perform multi-dimensional verification of the registration information and construct an initial coordinate system for different facial images of users, quantify the quality score of different registered users based on the registration information verification results, classify different registered users according to quality, automatically trigger different facial feature recognition modes, select a multi-feature authentication range set for the facial images of different registered users, extract the coordinate positioning data of the authentication range set under different facial feature recognition modes, set a multi-feature point differentiated permission migration range based on the multi-feature authentication range set to perform facial multi-feature point recognition authentication, directly link registration quality with facial feature extraction strategy, provide differentiated facial verification for users of different levels, and build a multi-user intelligent authentication system that can intelligently identify risks and accurately allocate resources. Summary of the Invention

[0003] The purpose of this invention is to provide a big data processing terminal based on user identification and authentication to solve the problems in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a big data processing terminal based on user identification and authentication, comprising a user information registration and original image acquisition module, a user registration quality grading verification and analysis module, a user facial key feature selection and coordinate positioning module, a user login security authentication coordination module, a user authorized access verification and protection module, and a data processing terminal; the user information registration and original image acquisition module acquires the registration information submitted by the user at the data processing terminal, assigns a unique identifier to the registered user, and simultaneously acquires the registered user's facial image, performing standardized processing on the facial image; the user registration quality grading verification and analysis module obtains the user's registration information form and performs multi-dimensional verification and quantitative analysis of accuracy and completeness, generating a comprehensive registration quality score for each user for grading; the user facial key feature selection... The coordinate positioning module performs facial feature recognition decision-making based on the quality level of each user. Facial feature recognition decisions include basic facial feature recognition modes and complex facial feature recognition modes. It pre-sets reference points for the user's face, establishes a relative coordinate system for different facial features, and selects multiple target feature points on the user's face for coordinate positioning according to different facial recognition modes. The user login security authentication coordination module determines the secure permission migration range of the facial target feature point positioning coordinates selected within the basic feature authentication range set and the complex feature authentication range set, associates them with user registration information, and constructs a user face recognition authentication association database. The user authorized access verification and protection module collects the user's identity identifier and facial target feature points from the face image during login to verify login authorization, and marks recently successfully verified users in a secure cache.

[0005] Further configuration: The user information registration and raw image acquisition module includes a user information registration submission initial identification submodule and a user face image acquisition preprocessing submodule. The user information registration submission initial identification submodule collects user registration information based on a pre-set personal information form in the terminal backend. It extracts the name and ID number from each successfully registered user's information, concatenates the name (in pinyin) with the ID number to form a concatenated string, and uses an encrypted hash function to calculate a fixed-length hash value for each user's concatenated string. Simultaneously, the data processing terminal backend randomly assigns a unique serial number to each registered user. The unique serial number assigned to each registered user is combined with the hash value generated from the user's concatenated string to generate... The system includes a unique identifier for each registered user, which is then aggregated. The user face image acquisition and preprocessing submodule includes a camera invocation unit. This unit is used by the camera on the data processing terminal to acquire face images of different successfully registered users, capturing multiple frames of each registered user's face. Preprocessing of these frames includes, but is not limited to, face region cropping, size normalization, grayscale conversion, illumination compensation, and noise filtering. Defective face images are removed, and the preprocessed face image for each registered user is output. The number of preprocessed face images is less than or equal to a set threshold. Finally, the unique identifier of each registered user is matched and aggregated with the preprocessed face images.

[0006] Further configuration: The user registration quality grading verification and analysis module includes a multi-dimensional verification and analysis sub-module for registration forms and a comprehensive user registration quality grading sub-module. The multi-dimensional verification and analysis sub-module acquires registration information from different users, validates the format of required fields, and connects to an external identity verification database (provided by a third-party access service). The external database is used to verify each piece of identity information under different user registration forms, extracting the number of successfully verified registrations for each user across different forms. The percentage of successfully verified registrations in the total number of registrations is calculated and marked as the accuracy of the user's registration information. Users with accuracy below a set threshold are removed, and registration failure data, along with verification failure data, is sent to their clients. Further configuration: The comprehensive user registration quality grading sub-module collects registrations with accuracy greater than or equal to a set accuracy threshold. The data processing terminal's backend administrator pre-assigns different weights to each piece of registration information within the registration form, setting a preset weight coefficient for each list of registration information within the form. A comprehensive quality weight score is calculated for the registration information of each registered user. The list of registration information within each registration form is defined as follows: If the registration information for a certain list has been filled in and is in a valid format, and has been successfully verified through an external identity verification database, then the current list of registration information is considered complete. If the registration information for a certain list is not filled in, is in an invalid format, or fails to be verified through an external identity verification database, then the current registration information list will be set to 1. The score is set to 0, and the registration information of each registered user is weighted with a comprehensive quality score. According to the formula: The comprehensive quality weight score of each registered user's registration information is calculated according to the above formula. This score is then compared with the weight score threshold preset by the data processing terminal backend administrator. Registered users whose comprehensive quality weight score is greater than or equal to the weight score threshold are marked as trustworthy users, while registered users whose comprehensive quality weight score is less than the weight score threshold are marked as risky users.

[0007] Further configuration: The user facial key feature selection and coordinate positioning module includes a user face initial coordinate system correspondence construction submodule and a facial target feature point selection and coordinate positioning submodule. The user face initial coordinate system correspondence construction submodule acquires the facial images of registered users marked as trustworthy and those marked as risky. The data processing terminal automatically performs basic facial feature recognition on the facial images of registered users marked as trustworthy and performs complex facial feature recognition on the facial images of registered users marked as risky, respectively acquiring the basic facial feature recognition mode and the user face key feature selection and coordinate positioning submodule. In the complex facial feature recognition mode, facial images of different users are used to extract the nasal root point and nasal tip point of each facial image. The nasal root point is located at the top of the nose between the eyes. The nasal root point and nasal tip point are connected by a straight line, and the midpoint of the line is collected and marked. The midpoint of the line connecting the marked nasal root point and nasal tip point is defined as the facial coordinate base point, i.e., the origin of the coordinate system. Based on the facial coordinate base point of each registered user, the initial two-dimensional coordinate system of the registered user's own facial structure is constructed with the horizontal rightward direction as the positive X-axis and the vertical downward direction as the positive Y-axis.

[0008] Further settings: The facial target feature point selection and coordinate positioning submodule obtains the initial coordinate system of each registered user's face, determines the facial feature recognition mode executed by different registered users, and when a registered user executes the user's basic facial feature recognition mode, selects the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip of the current registered user to construct the basic feature authentication range set of the current registered user. According to the initial coordinate system of the face image of the current registered user, the coordinates of the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip within the basic feature authentication range set of the registered user are located respectively, and the coordinates of the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip of the current registered user are output. The coordinates of the center feature point of the left eye of the current user are set as... The coordinates of the central feature point of the right eye are: The coordinates of the feature point at the vertex of the upper lip are: The system determines whether the number of preprocessed face images of the current registered user is greater than one. If the number is greater than one, an initial two-dimensional coordinate system is constructed based on each face image of the current registered user. Feature positioning coordinates of the basic feature authentication range set within each face image of the current registered user are extracted. The system then screens the feature positioning coordinates of the basic feature authentication range set within each face image of the current registered user. The system analyzes the offset between the positioning coordinates of the same feature point within different face images of the same registered user. If the offset is greater than or equal to a set threshold, the face images are sent to the data processing terminal for manual review. If the offset is less than the set threshold, the average coordinates of the left eye center feature point, right eye center feature point, and upper lip vertex feature point within each face image of the current registered user are calculated. The coordinates of the left eye center feature point within the face image of the current registered user are defined as follows: The coordinates of the central feature point of the right eye are The coordinates of the feature point at the vertex of the upper lip are: This involves summarizing the basic feature location coordinates of different registered users under the basic facial feature recognition mode.

[0009] Further settings: The facial target feature point selection and coordinate positioning submodule determines a registered user and executes the complex facial feature recognition mode. It pre-constructs a basic feature authentication range set for the current registered user, and locates the feature points within this range set according to the initial facial coordinate system of the user's face image. It then performs a secondary selection of the user's facial feature points, including several contour feature points and detail feature points. Contour feature points include, but are not limited to, the center point of the chin and the outer edge of the earlobe. Detail feature points include, but are not limited to, the inner corner of the left eye and the outer corner of the right eye. The specific feature points, such as the inner corner of the eye and the points on both sides of the nose, are set by the data processing terminal backend administrator. Several contour feature points and detail feature points are summarized to construct the complex feature authentication range set of the current registered user. The coordinates of the feature points within the complex feature authentication range set of the current registered user are located to determine whether there are multiple preprocessed face images of the current registered user. Based on the basic feature authentication range set and the complex feature authentication range set of the current user, the average value of a single feature point is taken for each face image of the registered user. The coordinate data of the multi-feature point positioning of different registered users under the complex facial feature recognition mode of the user's face are summarized.

[0010] Further configuration: The user login security authentication coordination module includes a multi-feature recognition coordinate migration range dynamic setting submodule and a user face recognition authentication association database construction submodule. The multi-feature recognition coordinate migration range dynamic setting submodule obtains the multi-feature point positioning coordinate data of different registered users under the basic facial feature recognition mode and the complex facial feature recognition mode. It sets differentiated coordinate migration permission ranges for feature points selected within the basic feature authentication range set and feature points selected within the complex feature authentication range set, and constructs the recognition coordinate authentication range bounding boxes for feature points within the basic feature authentication range set and the complex feature authentication range set, respectively. The user face recognition authentication association database construction submodule obtains the unique identifier, registration information, and face image of different registered users and the positioning coordinate data of each feature point within the corresponding feature authentication range set, as well as the recognition coordinate authentication range bounding box corresponding to each feature point, constructs the face recognition authentication database for different registered users, and uploads it to the data processing terminal.

[0011] Further settings: The data processing terminal's backend administrator pre-constructs different recognition coordinate authentication bounding boxes for feature points selected within the basic feature authentication range set and feature points selected within the complex feature authentication range set, and sets the positioning coordinates of a certain feature point as... The system pre-determines whether a feature point belongs to the basic feature authentication range set or the complex feature authentication range set. The data processing terminal's backend administrator pre-sets the maximum permissible deviation range for feature points selected within both the basic and complex feature authentication range sets. Specifically, the maximum permissible deviation range in the X-axis direction for each feature point selected within the basic feature authentication range set is set to... The maximum permissible deviation range in the Y-axis direction for each feature point selected within the basic feature authentication range set is [value missing]. The maximum permissible deviation range in the X-axis direction for each feature point selected within the complex feature authentication range set is set to... The maximum permissible deviation range in the Y-axis direction for each feature point selected within the complex feature authentication range set is [value missing]. When a certain feature point is within the basic feature authentication range set, a recognition coordinate authentication range bounding box is constructed for that feature point, wherein the upper bound of the recognition coordinate authentication range bounding box for that feature point in the X-axis direction is... The lower bound in the X-axis direction is The upper bound in the Y-axis direction is The lower bound in the X-axis direction is The feature point recognition coordinate authentication bounding box is represented as follows: When a feature point falls within a complex feature authentication range set, a recognition coordinate authentication bounding box is constructed for that feature point. The upper bound of the feature point's recognition coordinate authentication bounding box in the X-axis direction is... The lower bound in the X-axis direction is The upper bound in the Y-axis direction is The lower bound in the X-axis direction is The feature point recognition coordinate authentication bounding box is represented as follows: Further configuration: The user authorization access verification guarantee module includes a user authentication authorization decision submodule and an authentication information local security cache submodule. The user authentication authorization decision submodule collects the registered user information currently undergoing face recognition in real time, calls the positioning coordinate data of each feature point within the corresponding feature authentication range set of the registered user currently undergoing face recognition login, and determines whether the current coordinates of each feature point within the corresponding feature authentication range set of the current registered user fall within the recognition coordinate authentication range set of the user's corresponding feature point in the user's face recognition authentication database. If the current coordinates of all verified feature points of the registered user undergoing face recognition fall within the corresponding recognition coordinate authentication range set, the authentication is successful and login is authorized. If the coordinates of any feature point exceed its specific range, the authentication fails. The authentication information local security cache submodule performs local security caching of the authorization authentication records of different successfully authenticated registered users.

[0012] Compared with existing technologies, the beneficial effects of this invention are as follows: It aims to pre-acquire registration information and facial images of different registered users, perform multi-dimensional verification of registration information and construct initial coordinate systems for different facial images of users, quantify the quality scores of different registered users based on the registration information verification results, classify different registered users according to quality, automatically trigger different facial feature recognition modes, select multi-feature authentication range sets for facial images of different registered users, extract coordinate positioning data of authentication range sets under different facial feature recognition modes, set multi-feature point differentiated permission migration ranges based on multi-feature authentication range sets for facial multi-feature point recognition and authentication, directly link registration quality with facial feature extraction strategies, provide differentiated facial verification for users of different levels, and construct a multi-user intelligent authentication system that intelligently identifies risks and accurately allocates resources. Attached Figure Description

[0013] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0014] Figure 1 is a schematic diagram of the implementation process of a big data processing terminal based on user identification and authentication according to the present invention; Figure 2 is a specific module diagram of a big data processing terminal based on user identification and authentication according to the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please refer to Figures 1 and 2. In this embodiment of the invention, a big data processing terminal based on user identification and authentication includes a user information registration and raw image acquisition module, a user registration quality grading verification and analysis module, a user facial key feature selection and coordinate positioning module, a user login security authentication coordination module, a user authorized access verification and protection module, and a data processing terminal. The user information registration and raw image acquisition module collects the registration information submitted by the user at the data processing terminal, assigns a unique identifier to the registered user, and simultaneously acquires the registered user's facial image and performs standardized processing on the facial image. Further configuration is required: the user information registration and raw image acquisition module includes a user information registration submission initial identifier submodule and a user facial image acquisition preprocessing submodule. The user information registration submission initial identifier submodule collects user registration information according to a pre-set personal information form in the terminal backend, extracts the name and ID number from each successfully registered user, and concatenates the name (pinyin) with the ID number to form a concatenated string. The system utilizes a cryptographic hash function to calculate a fixed-length hash value for each user's concatenated string. Simultaneously, the data processing terminal randomly assigns a unique serial number to each registered user. This unique serial number is combined with the hash value generated from the user's concatenated string to generate a unique identifier for each registered user. The unique identifiers of all registered users are then aggregated. The user face image acquisition and preprocessing submodule includes a camera invocation unit. This unit is used by the camera on the data processing terminal to acquire face images of different successfully registered users, capturing multiple frames of each registered user's face. Preprocessing of these frames includes, but is not limited to, face region cropping, size normalization, grayscale conversion, illumination compensation, and noise filtering. Face images with quality defects are removed, and the preprocessed face image for each registered user is output. The number of preprocessed face images is less than or equal to a set threshold. Finally, the unique identifier of each registered user is matched and aggregated with the preprocessed face images.

[0017] The user registration quality grading verification and analysis module acquires user registration information forms and performs multi-dimensional verification and quantitative analysis on accuracy and completeness, generating a comprehensive registration quality score for each user for grading. Furthermore, this module includes a multi-dimensional verification and analysis submodule for registration forms and a comprehensive user registration quality grading submodule. The multi-dimensional verification and analysis submodule acquires registration information from different users, validates the format of required fields, and, upon successful validation, connects to an external identity verification database. This external identity verification database is provided by a third-party access service, and the database is used to verify the identity information of each user's registration form. The system sequentially verifies each user's registration information, extracting the number of successfully verified registrations from different registration forms. It calculates the percentage of successfully verified registrations in the total number of registration forms and marks this percentage as the accuracy of each user's registration information. Users with accuracy below a set threshold are removed, and their registration failure data and verification failure information are sent to their clients. The user comprehensive registration quality grading submodule collects registrations with accuracy greater than or equal to a set accuracy threshold. The data processing terminal's backend administrator pre-assigns different weights to each registration information within a registration form, setting a preset weight coefficient for each registration information list within the registration form. A comprehensive quality weight score is calculated for the registration information of each registered user. The list of registration information within each registration form is defined as follows: If the registration information for a certain list has been filled in and is in a valid format, and has been successfully verified through an external identity verification database, then the current list of registration information is considered complete. If the registration information for a certain list is not filled in, is in an invalid format, or fails to be verified through an external identity verification database, then the current registration information list will be set to 1. The score is set to 0, and the registration information of each registered user is weighted with a comprehensive quality score. According to the formula: The comprehensive quality weight score of each registered user's registration information is calculated according to the above formula. This score is then compared with the weight score threshold preset by the data processing terminal backend administrator. Registered users whose comprehensive quality weight score is greater than or equal to the weight score threshold are marked as trustworthy users, while registered users whose comprehensive quality weight score is less than the weight score threshold are marked as risky users.

[0018] The user facial key feature selection and coordinate positioning module performs tiered judgment on facial feature recognition decisions based on the quality level of each user. Facial feature recognition decisions include basic facial feature recognition modes and complex facial feature recognition modes. Reference points are pre-set for the user's face, establishing relative coordinate systems for different facial features. Based on different facial recognition modes, multiple target feature points on the user's face are selected for coordinate positioning. Furthermore, the user facial key feature selection and coordinate positioning module includes a user face initial coordinate system correspondence construction submodule and a facial target feature point selection and coordinate positioning submodule. The user face initial coordinate system correspondence construction submodule acquires facial images of registered users marked as trustworthy and those marked as risky, respectively. The data processing terminal automatically processes the facial images of registered users marked as trustworthy. For example, the basic facial feature recognition mode is used to perform complex facial feature recognition mode on the facial images of registered users marked as high-risk users. Facial images of different users under both the basic and complex facial feature recognition modes are obtained. The root of the nose and the tip of the nose are extracted from each facial image. The root of the nose is located at the top of the bridge of the nose between the eyes. A straight line is drawn connecting the root of the nose and the tip of the nose, and the midpoint of the line is marked. The midpoint of the line connecting the marked root of the nose and the tip of the nose is defined as the facial coordinate base point, i.e., the origin of the coordinate system. Based on the facial coordinate base point of each registered user, an initial two-dimensional coordinate system for the registered user's own facial structure is constructed with the horizontal rightward direction as the positive X-axis and the vertical downward direction as the positive Y-axis.

[0019] Specifically, the facial target feature point selection and coordinate localization submodule obtains the initial facial coordinate system of each registered user's face, determines the facial feature recognition mode executed by different registered users, and when a registered user executes the user's basic facial feature recognition mode, it selects the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip of the current registered user to construct the basic feature authentication range set of the current registered user. According to the initial facial coordinate system of the current registered user's face image, it performs coordinate localization on the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip within the basic feature authentication range set of the registered user, and outputs the coordinate points of the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip of the current registered user. The coordinate point of the center feature point of the left eye of the current registered user is set as... The coordinates of the central feature point of the right eye are: The coordinates of the feature point at the vertex of the upper lip are: The system determines whether the number of preprocessed face images of the current registered user is greater than one. If the number is greater than one, an initial two-dimensional coordinate system is constructed based on each face image of the current registered user. Feature positioning coordinates of the basic feature authentication range set within each face image of the current registered user are extracted. The system then screens the feature positioning coordinates of the basic feature authentication range set within each face image of the current registered user. The system analyzes the offset between the positioning coordinates of the same feature point within different face images of the same registered user. If the offset is greater than or equal to a set threshold, the face images are sent to the data processing terminal for manual review. If the offset is less than the set threshold, the average coordinates of the left eye center feature point, right eye center feature point, and upper lip vertex feature point within each face image of the current registered user are calculated. The coordinates of the left eye center feature point within the face image of the current registered user are defined as follows: The coordinates of the central feature point of the right eye are The coordinates of the feature point at the vertex of the upper lip are: This involves summarizing the basic feature location coordinates of different registered users under the basic facial feature recognition mode.

[0020] The facial feature point selection and coordinate localization submodule determines a registered user and executes a complex facial feature recognition mode. It pre-constructs a basic feature authentication range set for the current registered user and, according to the initial facial coordinate system of the current registered user's face image, locates the feature points within this basic feature authentication range set. A secondary selection is then performed on the current registered user's facial feature points, including several contour feature points and detail feature points. Contour feature points include, but are not limited to, the center point of the chin and the outer edge of the earlobe. Detail feature points include, but are not limited to, the inner canthus of the left eye and the inner canthus of the right eye. Corner points, points on both sides of the nose, and other specific feature points are set by the data processing terminal backend administrator. Several contour feature points and detail feature points are summarized to construct the complex feature authentication range set of the current registered user. The coordinates of the feature points within the complex feature authentication range set of the current registered user are located to determine whether there are multiple preprocessed face images of the current registered user. Based on the basic feature authentication range set and the complex feature authentication range set of the current user, the average value of a single feature point is taken for each face image of the registered user. The coordinate data of multi-feature point positioning of different registered users under the complex facial feature recognition mode of the user's face are summarized.

[0021] The user login security authentication coordination module determines the secure migration range of the facial target feature point positioning coordinates selected within the basic feature authentication range set and the complex feature authentication range set, associates them with user registration information, and constructs a user face recognition authentication association database. Further explanation is needed: the user login security authentication coordination module includes a multi-feature recognition coordinate migration range dynamic setting sub-module and a user face recognition authentication association database construction sub-module. The multi-feature recognition coordinate migration range dynamic setting sub-module obtains multi-feature point positioning coordinate data for different registered users under the basic facial feature recognition mode and the complex facial feature recognition mode, and sets differentiated coordinate migration permission ranges for feature points selected within the basic feature authentication range set and feature points selected within the complex feature authentication range set, respectively, constructing recognition coordinate authentication range boxes for feature points within the basic feature authentication range set and the complex feature authentication range set. The user face recognition authentication association database construction sub-module obtains the unique identifier, registration information, and face image of different registered users, along with the positioning coordinate data of each feature point within its corresponding feature authentication range set, and the recognition coordinate authentication range box corresponding to each feature point, constructing a face recognition authentication database for different registered users, and uploading it to the data processing terminal.

[0022] It needs to be explained in detail that the data processing terminal backend administrator pre-constructs different recognition coordinate authentication bounding boxes for feature points selected within the basic feature authentication range set and feature points selected within the complex feature authentication range set, and sets the positioning coordinates of a certain feature point as... The system pre-determines whether a feature point belongs to the basic feature authentication range set or the complex feature authentication range set. The data processing terminal's backend administrator pre-sets the maximum permissible deviation range for feature points selected within both the basic and complex feature authentication range sets. Specifically, the maximum permissible deviation range in the X-axis direction for each feature point selected within the basic feature authentication range set is set to... The maximum permissible deviation range in the Y-axis direction for each feature point selected within the basic feature authentication range set is [value missing]. The maximum permissible deviation range in the X-axis direction for each feature point selected within the complex feature authentication range set is set to... The maximum permissible deviation range in the Y-axis direction for each feature point selected within the complex feature authentication range set is [value missing]. When a certain feature point is within the basic feature authentication range set, a recognition coordinate authentication range bounding box is constructed for that feature point, wherein the upper bound of the recognition coordinate authentication range bounding box for that feature point in the X-axis direction is... The lower bound in the X-axis direction is The upper bound in the Y-axis direction is The lower bound in the X-axis direction is The feature point recognition coordinate authentication bounding box is represented as follows: When a feature point falls within a complex feature authentication range set, a recognition coordinate authentication bounding box is constructed for that feature point. The upper bound of the feature point's recognition coordinate authentication bounding box in the X-axis direction is... The lower bound in the X-axis direction is The upper bound in the Y-axis direction is The lower bound in the X-axis direction is The feature point recognition coordinate authentication bounding box is represented as follows: The user authorization access verification and protection module collects the user's identity identifier and facial target feature points from the face image during login to verify login authorization, and marks recently successfully verified users in a secure cache.

[0023] Further explanation is needed. The user authorization access verification guarantee module includes a user authentication authorization decision submodule and an authentication information local security cache submodule. The user authentication authorization decision submodule collects the registered user information currently undergoing face recognition in real time, calls the positioning coordinate data of each feature point within the corresponding feature authentication range set of the registered user currently undergoing face recognition login, and determines whether the current coordinates of each feature point within the corresponding feature authentication range set of the current registered user fall within the recognition coordinate authentication range set of the user's corresponding feature point in the user's face recognition authentication database. If the current coordinates of all verified feature points of the registered user undergoing face recognition fall within the corresponding recognition coordinate authentication range set, the authentication is successful and login is authorized. If the coordinates of any feature point exceed its specific range, the authentication fails. The authentication information local security cache submodule performs local security caching of the authorization authentication records of different successfully authenticated registered users.

[0024] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A big data processing terminal based on user identification and authentication, characterized in that: The terminal includes a user information registration and raw image acquisition module, a user registration quality grading verification and analysis module, a user facial key feature selection and coordinate positioning module, a user login security authentication coordination module, a user authorized access verification and protection module, and a data processing terminal; the user information registration and raw image acquisition module acquires the registration information submitted by the user at the data processing terminal, assigns a unique identifier to the registered user, and simultaneously acquires the registered user's facial image and performs standardized processing on the facial image; The user registration quality grading verification and analysis module acquires user registration information forms and performs multi-dimensional verification and quantitative analysis on accuracy and completeness, generating a comprehensive registration quality score for each user for grading. The user facial key feature selection and coordinate positioning module performs facial feature recognition decision grading based on each user's quality level. Facial feature recognition decisions include basic facial recognition feature recognition modes and complex facial feature recognition modes. It pre-sets reference points for the user's face, establishes a relative coordinate system for different facial features, and selects multiple target feature points on the user's face for coordinate positioning according to different facial recognition modes. The user login security authentication coordination module determines the secure permission migration range of the facial target feature point positioning coordinates selected within the basic feature authentication range set and the complex feature authentication range set, associates them with user registration information, and constructs a user facial recognition authentication association database. The user authorized access verification and assurance module collects the user's identity identifier and facial target feature points from the face image during login to verify login authorization, and marks recently successfully verified users in a secure cache.

2. A big data processing terminal based on user identification and authentication according to claim 1, characterized in that... The user information registration and original image acquisition module includes a user information registration submission initial identification submodule and a user face image acquisition preprocessing submodule. The user information registration submission initial identification submodule collects user registration information according to the personal information form preset in the terminal backend, extracts the name and ID number from the collected successful registration information of each user, concatenates the name pinyin and ID number to form a concatenated string, calculates the hash value of each user's concatenated string using an encrypted hash function, and generates a hash value of fixed length. At the same time, the data processing terminal backend randomly assigns a unique serial number to each registered user, combines the unique serial number assigned to each registered user with the hash value generated by the user's concatenated string, generates a unique identifier for each registered user, and summarizes the unique identifiers of different registered users. The user face image acquisition and preprocessing submodule includes a camera invocation unit. The camera invocation unit is used to acquire face images of different successfully registered users from the camera on the data processing terminal, capture multiple frames of face images of each registered user, preprocess the multiple frames of face images of each registered user, including but not limited to face region cropping, size normalization, grayscale conversion, illumination compensation and noise filtering, filter out face images with quality defects, and output the preprocessed face image of each registered user. The number of preprocessed face images is less than or equal to a set threshold. The unique identifier of each registered user is matched and summarized with the preprocessed face images.

3. A big data processing terminal based on user identification and authentication according to claim 1, characterized in that... The user registration quality grading verification and analysis module includes a multi-dimensional verification and analysis submodule for registration forms and a comprehensive user registration quality grading submodule. The multi-dimensional verification and analysis submodule for registration forms obtains registration information from different users, verifies the format of required registration information fields, and connects to an external identity verification database after successful verification. The external identity verification database is accessed by a third-party access service. The external identity verification database is used to verify each piece of identity information under different user registration forms in turn, extracts the number of successfully verified registration information for each user under different registration forms, calculates the proportion of successfully verified registration information to the total number of registration forms, marks it as the accuracy of different user registration information, and removes registered users whose accuracy is below a set threshold, sending registration failure data and verification failure registration information data to the user client.

4. A big data processing terminal based on user identification and authentication according to claim 3, characterized in that... The user registration quality grading submodule collects registration information with an accuracy greater than or equal to a set accuracy threshold. For registered users, the data processing terminal's backend administrator pre-assigns different weights to each piece of registration information within the registration form. The preset weight coefficient for each piece of registration information within the registration form is set to... A comprehensive quality weight score is calculated for the registration information of each registered user. The list of registration information within each registration form is defined as follows: If the registration information for a certain list has been filled in and is in a valid format, and has been successfully verified through an external identity verification database, then the current list of registration information is considered complete. If the registration information for a certain list is not filled in, is in an invalid format, or fails to be verified through an external identity verification database, then the current registration information list will be set to 1. The score is set to 0, and the registration information of each registered user is weighted with a comprehensive quality score. According to the formula: The comprehensive quality weight score of each registered user's registration information is calculated according to the above formula. This score is then compared with the weight score threshold preset by the data processing terminal backend administrator. Registered users whose comprehensive quality weight score is greater than or equal to the weight score threshold are marked as trustworthy users, while registered users whose comprehensive quality weight score is less than the weight score threshold are marked as risky users.

5. A big data processing terminal based on user identification and authentication according to claim 1, characterized in that... The user face key feature selection and coordinate positioning module includes a user face initial coordinate system correspondence construction submodule and a facial target feature point selection and coordinate positioning submodule. The user face initial coordinate system correspondence construction submodule acquires the face images of registered users marked as trustworthy and those marked as risky. The data processing terminal automatically performs basic facial feature recognition on the face images of registered users marked as trustworthy and performs complex facial feature recognition on the face images of registered users marked as risky, respectively acquiring the basic facial feature recognition mode and the complex facial feature recognition mode. In the complex facial feature recognition mode, facial images of different users are used to extract the nasal root point and nasal tip point of each facial image. The nasal root point is located at the top of the nose between the eyes. The nasal root point and nasal tip point are connected by a straight line, and the midpoint of the line is marked. The midpoint of the line connecting the marked nasal root point and nasal tip point is defined as the facial coordinate base point, i.e., the origin of the coordinate system. Based on the facial coordinate base point of each registered user, an initial two-dimensional coordinate system of the registered user's own facial structure is constructed with the horizontal rightward direction as the positive X-axis and the vertical downward direction as the positive Y-axis.

6. A big data processing terminal based on user identification and authentication according to claim 5, characterized in that... The facial target feature point selection and coordinate positioning submodule obtains the initial coordinate system of each registered user's face, determines the facial feature recognition mode executed by different registered users, and when a registered user executes the user's basic facial feature recognition mode, selects the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip of the current registered user to construct the basic feature authentication range set of the current registered user. According to the initial coordinate system of the face image of the current registered user, the coordinates of the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip within the basic feature authentication range set of the registered user are located respectively, and the coordinates of the center feature point of the left eye, the center feature point of the right eye, and the vertex feature point of the upper lip of the current registered user are output. The coordinate point of the center feature point of the left eye of the current registered user is set as... The coordinates of the central feature point of the right eye are: The coordinates of the feature point at the vertex of the upper lip are: The system determines whether the number of preprocessed face images of the current registered user is greater than one. If the number is greater than one, an initial two-dimensional coordinate system is constructed based on each face image of the current registered user. Feature positioning coordinates of the basic feature authentication range set within each face image of the current registered user are extracted. The system then screens the feature positioning coordinates of the basic feature authentication range set within each face image of the current registered user. The system analyzes the offset between the positioning coordinates of the same feature point within different face images of the same registered user. If the offset is greater than or equal to a set threshold, the face images are sent to the data processing terminal for manual review. If the offset is less than the set threshold, the average coordinates of the left eye center feature point, right eye center feature point, and upper lip vertex feature point within each face image of the current registered user are calculated. The coordinates of the left eye center feature point within the face image of the current registered user are defined as follows: The coordinates of the central feature point of the right eye are The coordinates of the feature point at the vertex of the upper lip are: This involves summarizing the basic feature location coordinates of different registered users under the basic facial feature recognition mode.

7. A big data processing terminal based on user identification and authentication according to claim 5, characterized in that... The facial target feature point selection and coordinate positioning submodule determines the complex facial feature recognition mode for a registered user. It pre-constructs a basic feature authentication range set for the current registered user, and locates the feature points within this range set according to the initial facial coordinate system of the user's face image. A secondary selection is then performed on the user's facial feature points, including several contour feature points and detail feature points. Contour feature points include, but are not limited to, the center point of the chin and the outer edge of the earlobe. Detail feature points include, but are not limited to, the inner canthus of the left eye and the inner canthus of the right eye. The specific feature points, such as the corners of the eyes and the sides of the nose, are set by the backend administrator of the data processing terminal. Several contour feature points and detail feature points are summarized to construct the complex feature authentication range set of the current registered user. The coordinates of the feature points within the complex feature authentication range set of the current registered user are located to determine whether there are multiple preprocessed face images of the current registered user. Based on the basic feature authentication range set and the complex feature authentication range set of the current user, the average value of a single feature point is taken for each face image of the registered user. The coordinate data of the multi-feature point positioning of different registered users under the complex facial feature recognition mode of the user's face are summarized.

8. A big data processing terminal based on user identification and authentication according to claim 1, characterized in that... The user login security authentication coordination module includes a multi-feature recognition coordinate migration range dynamic setting submodule and a user face recognition authentication association database construction submodule. The multi-feature recognition coordinate migration range dynamic setting submodule obtains the multi-feature point positioning coordinate data of different registered users under the basic facial feature recognition mode and the complex facial feature recognition mode. It sets differentiated coordinate migration permission ranges for feature points selected within the basic feature authentication range set and feature points selected within the complex feature authentication range set, and constructs the recognition coordinate authentication range bounding boxes for feature points within the basic feature authentication range set and the complex feature authentication range set, respectively. The user face recognition authentication association database construction submodule obtains the unique identifier, registration information, and face image of different registered users and the positioning coordinate data of each feature point within the corresponding feature authentication range set, as well as the recognition coordinate authentication range bounding box corresponding to each feature point, constructs the face recognition authentication database for different registered users, and uploads it to the data processing terminal.

9. A big data processing terminal based on user identification and authentication according to claim 8, characterized in that... The data processing terminal's backend administrator pre-constructs different recognition coordinate authentication bounding boxes for feature points selected within the basic feature authentication range set and feature points selected within the complex feature authentication range set, and sets the positioning coordinates of a certain feature point as... The system pre-determines whether a feature point belongs to the basic feature authentication range set or the complex feature authentication range set. The data processing terminal's backend administrator pre-sets the maximum permissible deviation range for feature points selected within both the basic and complex feature authentication range sets. Specifically, the maximum permissible deviation range in the X-axis direction for each feature point selected within the basic feature authentication range set is set to... The maximum permissible deviation range in the Y-axis direction for each feature point selected within the basic feature authentication range set is [value missing]. The maximum permissible deviation range in the X-axis direction for each feature point selected within the complex feature authentication range set is set to... The maximum permissible deviation range in the Y-axis direction for each feature point selected within the complex feature authentication range set is [value missing]. When a certain feature point is within the basic feature authentication range set, a recognition coordinate authentication range bounding box is constructed for that feature point, wherein the upper bound of the recognition coordinate authentication range bounding box for that feature point in the X-axis direction is... The lower bound in the X-axis direction is The upper bound in the Y-axis direction is The lower bound in the X-axis direction is The feature point recognition coordinate authentication bounding box is represented as follows: When a feature point falls within a complex feature authentication range set, a recognition coordinate authentication bounding box is constructed for that feature point. The upper bound of the feature point's recognition coordinate authentication bounding box in the X-axis direction is... The lower bound in the X-axis direction is The upper bound in the Y-axis direction is The lower bound in the X-axis direction is The feature point recognition coordinate authentication bounding box is represented as follows:

10. A big data processing terminal based on user identification and authentication according to claim 1, characterized in that... The user authorization access verification guarantee module includes a user authentication authorization decision submodule and an authentication information local security cache submodule. The user authentication authorization decision submodule collects the registered user information currently undergoing face recognition in real time, calls the positioning coordinate data of each feature point within the corresponding feature authentication range set of the registered user currently undergoing face recognition login, and determines whether the current coordinates of each feature point within the corresponding feature authentication range set of the registered user fall within the recognition coordinate authentication range set of the user's corresponding feature point in the user's face recognition authentication database. If the current coordinates of all verified feature points of the registered user undergoing face recognition fall within the corresponding recognition coordinate authentication range set, the authentication is successful and login is authorized. If the coordinates of any feature point exceed its specific range, the authentication fails. The authentication information local security cache submodule performs local security caching of the authorization authentication records of different successfully authenticated registered users.

Images