Face access control sub-control management and stranger early warning system based on inpatient area
The ward facial recognition access control system, which communicates with the hospital's HIS system, performs multi-factor identity verification and abnormal behavior analysis, solving the problems of inflexible permission configuration and low security in ward access control management, and achieving efficient and intelligent ward security management.
Patent Information
- Application Number
- CN202511642752.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-11
- Publication Date
- 2026-02-13
AI Technical Summary
The existing hospital ward access control system has problems such as inflexible permission configuration, low security, inability to identify abnormal behavior, inability to coordinate response, and lax management of caregivers, resulting in low ward management efficiency and poor security.
A ward-based facial recognition access control and stranger warning system is adopted. Through two-way communication with the hospital's HIS system, medical process data is synchronized, multi-factor identity verification is performed, normal behavior patterns are constructed, abnormal behavior is identified, and warning and handling strategies are output according to risk level, so as to realize dynamic access control and automated verification of the qualifications of caregivers.
It has achieved automated management of ward access control, improved the level of security protection, ensured rapid passage for authorized personnel, enhanced the accuracy of abnormal behavior identification, and provided efficient and intelligent security management support with fast response speed.
Smart Images
Figure CN121528458A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of medical security monitoring, in particular to a face access control management and stranger early warning system based on a ward. BACKGROUND
[0002] Current hospital ward access control management mainly relies on manual registration and traditional access control, such as card swiping, password, face recognition, etc. The current access control management has the following defects:
[0003] 1. The existing technology mostly adopts the mode of manual entry of personnel information and manual configuration of access control permissions, lacks effective association with the basic state of patient admission / discharge in the hospital information management (HIS) system, and cannot associate the dynamic adjustment of medical procedures, such as the need for nurses to manually open permissions for patients to enter the corridor of the operating room before surgery, and the failure to timely recover permissions after surgery, which poses the risk of unauthorized personnel entering by mistake; after the patient is transferred to another department, the original ward permission is not synchronized to cancel, which easily leads to chaos in cross-ward access, seriously affecting the efficiency and safety of ward management.
[0004] 2. Most ward access control uses card swiping or single face recognition, card swiping poses the risk of lost ID cards being used by others; single face recognition is easily broken by photo forgery and video deception, especially for key areas such as ICUs and pharmacies, which cannot meet the high security protection needs, and for special scenarios such as medical staff with injured hands and patients with limited mobility, there is a lack of emergency verification solutions, which prevents legitimate personnel from quickly accessing;
[0005] 3. The existing access control system can only record whether access is allowed or not, and cannot analyze the reasonableness of the behavior in combination with the characteristics of the medical scene in the ward, such as the daily ward rounds by nurses being a normal behavior, but if a patient frequently attempts to enter the pharmacy at night, the system cannot identify it as abnormal, and can only determine whether the permission exists or not, and cannot distinguish between the case of having permission but behaving abnormally, leading to the failure to report abnormal events in advance and prevent risks;
[0006] 4. The management of accompanying personnel relies on paper registration and one-time accompanying certificates, and strangers can use the accompanying identity to enter the ward;
[0007] 5. The existing system can only achieve a single level of early warning from recognition failure to prohibition of access, and cannot distinguish between misoperations such as face angle deviation, expired permissions, and stranger intrusion; when a low-risk event occurs due to face matching errors, frequent triggering of sound and light alarms disturbs the normal order of the ward, and when a high-risk event occurs, local alarms are usually triggered, lacking the ability to link and dispose, with high response delay, and easily missing the best disposal opportunity.
[0008] Therefore, the face access control management and stranger early warning system based on a ward is proposed. SUMMARY
[0009] In view of the deficiencies of the prior art, the present application provides a face access control sub-control management and stranger early warning system based on a ward, which solves the problems raised in the above background art.
[0010] To achieve the above object, the present application is implemented by the following technical solutions: a face access control sub-control management and stranger early warning system based on a ward, comprising:
[0011] A medical process synchronization module, which is in bidirectional communication with a hospital HIS system, is used to collect and synchronize medical process data and personnel basic data, and transmit them to an authority control module and a behavior pattern analysis and verification module;
[0012] An authority control module, which generates target authority parameters based on the medical process data and the personnel basic data, and uploads them to a multi-factor identity verification module;
[0013] A multi-factor identity verification module, which performs identity verification on a passing person based on access control authority parameters using a verification method of facial features and auxiliary identity media, and outputs verification result data to a behavior pattern analysis and verification module and a multi-level early warning module;
[0014] A behavior pattern analysis and verification module, which constructs a normal behavior pattern of a target person based on the medical process data, the access control authority parameters and the verification structure, identifies abnormal behaviors deviating from the normal behavior pattern, and outputs abnormal determination results to a multi-level early warning module;
[0015] A multi-level early warning module, which outputs early warning disposal instructions and associated processing strategies according to preset risk levels based on the verification result data and the abnormal determination results, and synchronizes them to a ward management terminal and a management personnel terminal;
[0016] An accompanying qualification control module, which is used to collect accompanying qualification data, generate accompanying exclusive access control authority parameters, and synchronize them to the authority control module.
[0017] The present application is further provided with: the medical process synchronization module comprises an information interaction unit, an information classification transmission unit and a trajectory association unit;
[0018] The information interaction unit is used to collect medical process data and personnel basic data from the hospital HIS system;
[0019] The information classification transmission unit classifies the medical process data into surgery, examination and transfer, and classifies the personnel basic data according to personnel types; the personnel types include medical staff, patients and accompanying persons;
[0020] The trajectory association unit associates corresponding medical process data based on the verification result data to generate associated trajectory data with medical scene labels, and transmits the associated trajectory data to the behavior pattern analysis verification module; the verification result data includes a pass-through location, a time, and an identity label.
[0021] The medical process data includes a time, a location, and associated information of a medical event.
[0022] The personnel basic data includes a target personnel identity label, a belonging role, and associated medical attributes.
[0023] The permission control module includes a basic permission control unit, a dynamic permission adjustment unit, and a permission synchronization unit.
[0024] The basic permission control unit divides basic access permission ranges for different types of personnel based on the personnel basic data and a preset role-permission mapping rule.
[0025] The dynamic permission adjustment unit adaptively adjusts the basic access permission ranges based on the medical process data to generate temporary access permission parameters.
[0026] The permission synchronization unit is used to combine the basic access permission ranges and the temporary access permission parameters to generate a target permission parameter, and synchronizes the target permission parameter to the multi-factor identity verification module and the behavior pattern analysis verification module.
[0027] The multi-factor identity verification module includes a verification rule configuration unit, an auxiliary medium verification unit, a face feature verification unit, an emergency verification unit, and a verification result data output unit.
[0028] The verification rule configuration unit configures a combined verification rule of a face feature and an auxiliary identity medium for different personnel types based on the target permission parameter.
[0029] The auxiliary medium verification unit is used to collect auxiliary identity medium information of a pass-through personnel, verify the validity of the auxiliary identity medium information, and obtain auxiliary identity medium verification result data.
[0030] The face feature verification unit is used to collect face feature information of a pass-through personnel and perform matching verification with a target face feature to obtain face feature verification result data.
[0031] The emergency verification unit is used to verify the identity validity of a pass-through personnel in a scenario where an auxiliary identity medium is missing or invalid by using a voice instruction combined with an identity verification code verification method to obtain an emergency verification result.
[0032] The verification result data output unit generates verification result data based on the auxiliary identity medium verification result data, the face feature verification result data and the emergency verification result, and transmits the verification result data to the behavior mode analysis verification module and the multi-level early warning module; the verification result data includes pass / fail, failure reason.
[0033] The application further provides that the behavior mode analysis verification module includes a behavior mode construction unit, a behavior deviation determination unit, a secondary verification triggering unit and a determination result transmission unit.
[0034] The behavior mode construction unit constructs a normal behavior mode adapted to the ward scene based on medical process data, target permission parameters and associated trajectory data according to the type of personnel.
[0035] The behavior deviation determination unit determines whether the current access behavior is abnormal behavior based on the deviation degree of the verification result data and the normal behavior mode, combined with the emergency weight of the medical scene.
[0036] The secondary verification triggering unit generates a secondary verification item related to the medical scene when it is determined that the current access behavior is abnormal behavior, and pushes the secondary verification item to the access personnel interaction terminal and generates a behavior determination result after receiving the feedback; the behavior determination result includes normal / abnormal, abnormal type.
[0037] The determination result transmission unit transmits the behavior determination result to the multi-level early warning module and the medical process synchronization module.
[0038] The application further provides that the multi-level early warning module includes a risk level determination unit, a disposal strategy generation unit and an instruction pushing unit.
[0039] The risk level determination unit determines the early warning level according to the preset risk division rule based on the verification result data and the behavior determination result; the early warning level includes low risk level, medium risk level and high risk level.
[0040] The disposal strategy generation unit generates an adaptive disposal strategy for different early warning levels.
[0041] The low risk level only stores the verification result data.
[0042] The medium risk level starts a local sound and light prompt and pushes information to the ward management terminal.
[0043] The high risk level pushes the early warning information to the management personnel terminal, links the ward associated equipment to execute control measures, and generates a time-limited disposal work order.
[0044] The instruction pushing unit pushes the early warning instruction and the disposal strategy to the corresponding execution terminal, receives the disposal completion feedback and updates the work order status.
[0045] The application is further provided as follows: the accompanying care qualification management module comprises a qualification verification unit, an accompanying care permission generation unit and an abnormality disposal unit;
[0046] The qualification verification unit is used for collecting accompanying care personnel's accompanying care qualification data and generating qualification verification result data; the accompanying care qualification data comprises identity information, family relationship, health status and accompanying care chest card ID;
[0047] The accompanying care permission generation unit generates accompanying care exclusive access permission parameters associated with the patient's hospitalization period and ward range based on the qualification verification result data and the patient's medical process data, and synchronizes to the permission management module;
[0048] The abnormality disposal unit is used for generating accompanying care exclusive access permission parameter freezing instructions and synchronizing to the pass person interactive terminal when the accompanying care personnel's qualification is invalid or the health status is abnormal, and pushing abnormality notifications to ward managers, and updating the accompanying care exclusive access permission parameters after the qualification is restored to be valid.
[0049] The application provides a face access control management and stranger early warning system based on a ward. The application has the following beneficial effects:
[0050] (1) The application synchronizes the patient's medical process data through bidirectional communication with the hospital HIS system, realizes the association of medical events by using trajectory association data, realizes the automatic management of access control by dynamically adjusting permissions, does not need manual intervention, has fast response speed, uses face and auxiliary media as the main verification, configures emergency voice as an emergency verification method, performs multi-factor identity verification, improves the security protection level, ensures the fast passage of legal personnel, effectively improves the identity fraud identification rate, constructs a scenario-based normal behavior mode, improves the abnormal behavior identification accuracy, simultaneously outputs differentiated disposal strategies according to the risk level, improves the response speed, realizes the linkage optimization of the alarm mode.
[0051] (2) The application realizes automatic management by automatically verifying the accompanying care personnel and the patient's family relationship through qualification verification, permission generation and dynamic monitoring, eliminating stranger fraud, freezing the accompanying care exclusive access permission and pushing notifications when the health is abnormal through the automatic synchronization of the health code, providing efficient and intelligent support for the safety management of the ward. BRIEF DESCRIPTION OF DRAWINGS
[0052] Figure 1 The figure is a system framework schematic diagram of the application;
[0053] Figure 2 The figure is a system framework schematic diagram of the medical process synchronization module of the application;
[0054] Figure 3A system framework schematic diagram of the permission control module of the application;
[0055] Figure 4 A system framework schematic diagram of the multi-factor identity verification module of the application;
[0056] Figure 5 A system framework schematic diagram of the behavior pattern analysis verification module of the application;
[0057] Figure 6 A system framework schematic diagram of the multi-level early warning module of the application;
[0058] Figure 7 A system framework schematic diagram of the accompanying qualification control module of the application. DETAILED DESCRIPTION
[0059] The technical solutions in the embodiments of the application will be clearly and completely described in connection with the drawings in the embodiments of the application.
[0060] Referring to Figures 1-7 The embodiments of the application provide a face access control and stranger early warning system based on a ward, which comprises a medical process synchronization module, a permission control module, a multi-factor identity verification module, a behavior pattern analysis verification module, a multi-level early warning module and an accompanying qualification control module. The medical process synchronization module communicates with a hospital HIS system bidirectionally, is used for collecting and synchronizing medical process data and personnel basic data, and transmits the medical process data and the personnel basic data to the permission control module and the behavior pattern analysis verification module. The permission control module generates target permission parameters based on the medical process data and the personnel basic data, and uploads the target permission parameters to the multi-factor identity verification module. The multi-factor identity verification module performs identity verification on a passing person by using a face feature and an auxiliary identity medium based on access permission parameters, and outputs verification result data to the behavior pattern analysis verification module and the multi-level early warning module. The behavior pattern analysis verification module constructs a normal behavior pattern of a target person based on the medical process data, the access permission parameters and verification structure, identifies an abnormal behavior deviating from the normal behavior pattern, and outputs an abnormal judgment result to the multi-level early warning module. The multi-level early warning module outputs early warning disposal instructions and associated processing strategies according to a preset risk level based on the verification result data and the abnormal judgment result, and synchronizes the early warning disposal instructions and the associated processing strategies to a ward management terminal and a management personnel terminal. The accompanying qualification control module is used for collecting accompanying qualification data, generating accompanying exclusive access permission parameters, and synchronizing the accompanying exclusive access permission parameters to the permission control module.
[0061] In an exemplary embodiment, as shown in Figure 2 The medical process synchronization module comprises an information interaction unit, an information classification transmission unit and a track association unit. The medical process synchronization module collects information from the hospital HIS system as a data collector. Specifically,
[0062] The information interaction unit adopts the HL7 FHIR V4.0 protocol to connect the hospital HIS system, and collects medical process data and personnel basic data from the hospital HIS system. The medical process data includes time, location and associated information of medical events; and the personnel basic data includes target personnel identity, associated role and associated medical attribute.
[0063] The information classification transmission unit classifies the medical process data into surgery, inspection and transfer, and classifies the personnel basic data according to personnel types. The personnel types include medical staff, patients and accompanying persons.
[0064] The trajectory association unit associates the corresponding medical process data based on the verification result data to generate associated trajectory data with medical scene markers, and transmits the associated trajectory data to the behavior pattern analysis and verification module. The verification result data includes passage location, time and identity.
[0065] As a detailed description, for the personnel basic data, the identity of the target personnel includes medical staff ID, patient bracelet number and accompanying person temporary ID; the associated role is medical staff, patient and accompanying person; and the associated medical attribute includes hospitalization period and post responsibility.
[0066] In an example embodiment, as shown in Figure 3 The permission control module includes a basic permission control unit, a dynamic permission adjustment unit and a permission synchronization unit. The permission control module serves as a permission maker, and sets where different personnel can go and where they cannot go according to the collected information, which specifically includes:
[0067] The basic permission control unit divides the basic access permission range for different types of personnel based on the personnel basic data according to a preset role-permission mapping rule;
[0068] The dynamic permission adjustment unit adaptively adjusts the basic access permission range based on the medical process data to generate temporary access permission parameters;
[0069] The permission synchronization unit is used to combine the basic access permission range and the temporary access permission parameters to generate a target permission parameter, and synchronizes the target permission parameter to the multi-factor identity verification module and the behavior pattern analysis and verification module.
[0070] In an example embodiment, as shown in Figure 4 The multi-factor identity verification module includes a verification rule configuration unit, an auxiliary medium verification unit, a face feature verification unit, an emergency verification unit and a verification result data output unit.
[0071] The verification rule configuration unit configures the combined verification rule of face features and auxiliary identity media for different personnel types based on the target permission parameter;
[0072] The auxiliary identity verification unit is used to collect auxiliary identity media information of passers-by, verify the validity of the auxiliary identity media information, and obtain auxiliary identity media verification result data; the auxiliary identity media is a physical / digital carrier used to assist in facial verification, such as medical staff name tags, patient wristbands, and temporary caregiver identification tags;
[0073] The facial feature verification unit is used to collect facial feature information of people passing through and match and verify it with the target facial features to obtain facial feature verification result data;
[0074] The emergency verification unit is used to verify the identity of passers-by by combining voice commands with identity verification codes in scenarios where auxiliary identity media is missing or invalid, and to obtain emergency verification results. The identity verification code is a unique identifier of the target personnel stored in the HIS system, such as the last 4 digits of a patient's hospital number or the last 4 digits of a medical staff's employee number.
[0075] The verification result data output unit generates verification result data based on the auxiliary identity medium verification result data, facial feature verification result data, and emergency verification result data, and transmits it to the behavior pattern analysis and verification module and the multi-level early warning module; the verification result data includes pass / fail and the reason for failure.
[0076] In one exemplary embodiment, such as Figure 5 As shown, the behavior pattern analysis and verification module includes a behavior pattern construction unit, a behavior deviation judgment unit, a secondary verification trigger unit, and a judgment result transmission unit. Acting as a behavior monitor, the behavior pattern analysis and verification module records personnel movement patterns and becomes alert upon detecting anomalies. Specifically:
[0077] The behavior pattern construction unit constructs normal behavior patterns adapted to ward scenarios based on medical process data, target permission parameters, and related trajectory data, according to personnel type.
[0078] The behavioral deviation judgment unit determines whether the current passage behavior is abnormal based on the deviation degree between the verification result data and the normal behavior pattern, combined with the urgency weight of the medical scenario. This includes:
[0079] Let the behavioral parameter vector under the normal behavior pattern be: Let the behavior parameter vector of the current passage behavior be: Calculate the deviation:
[0080]
[0081] In the formula, For deviation degree, The total number of behavioral parameters. For the first A vector of behavioral parameters under a normal behavioral pattern For the first a behavior parameter vector of a current passing behavior;
[0082]
[0083] wherein, is an emergency weight of a current ward medical scene, is a threshold value;
[0084] The secondary verification triggering unit is configured to generate a secondary verification item related to the medical scene when it is determined that the current passing behavior belongs to an abnormal behavior, and to generate a behavior determination result after pushing the secondary verification item to a passing person interactive terminal and receiving feedback; the behavior determination result includes normal / abnormal, abnormal type;
[0085] The determination result transmission unit is configured to transmit the behavior determination result to a multi-level early warning module and a medical process synchronization module.
[0086] In an exemplary embodiment, as shown in Figure 6 the multi-level early warning module includes a risk level determination unit, a disposal strategy generation unit, and an instruction pushing unit;
[0087] The risk level determination unit determines an early warning level based on the verification result data and the behavior determination result according to a preset risk division rule; the early warning level includes a low risk level, a medium risk level, and a high risk level, and the preset risk division rule includes:
[0088] STEP 1, matching degree risk rule formulation:
[0089] When the face feature matching degree is ≥ 90% and the auxiliary identity medium is valid, it is determined as a low risk level;
[0090] When the face feature matching degree is in the interval [70%, 90%) and the auxiliary identity medium is valid, it is determined as a medium risk level;
[0091] When the face feature matching degree is < 70% or the auxiliary identity medium is invalid, it is determined as a high risk level;
[0092] STEP 2, behavior determination risk rule formulation:
[0093] When the behavior is abnormal and the emergency weight of the current ward medical scene is higher than the preset threshold value, it is determined as a high risk level;
[0094] When the behavior is abnormal and the emergency weight of the current ward medical scene is lower than the preset threshold value, it is determined as a medium risk level;
[0095] STEP 3, comprehensive risk determination: when there is no behavior abnormality, the matching degree risk level is taken as the early warning level; when the matching degree risk level and the behavior determination risk level are inconsistent, the highest level is taken as the early warning level.
[0096] The treatment policy generation unit is used to generate an adaptive treatment policy for different early warning levels:
[0097] The low-risk level only stores the verification result data;
[0098] The medium-risk level starts a local sound and light prompt and pushes information to the ward management terminal;
[0099] The high-risk level pushes early warning information to the management personnel terminal, links the ward associated equipment to execute control measures, and generates a time-limited treatment work order;
[0100] The instruction pushing unit is used to push the early warning instructions and treatment policies to the corresponding execution terminal, receive the treatment completion feedback and update the work order status.
[0101] In an exemplary embodiment, as shown in Figure 7 The accompanying care qualification control module includes a qualification verification unit, a care permission generation unit, and an abnormality treatment unit;
[0102] The qualification verification unit is used to collect the care qualification data of the accompanying personnel and generate qualification verification result data; the care qualification data includes identity information, family relationship, health status, and accompanying chest card ID;
[0103] The accompanying permission generation unit generates accompanying exclusive access control permission parameters associated with the patient's hospitalization period and ward range based on the qualification verification result data and the patient's medical process data, and synchronizes them to the permission control module;
[0104] The abnormality treatment unit is used to generate an accompanying exclusive access control permission parameter freezing instruction and synchronize it to the access personnel interactive terminal when the accompanying personnel's qualification is invalid or the health status is abnormal, and at the same time, push an abnormality notification to the ward management personnel, and update the accompanying exclusive access control permission parameters after the qualification is restored to be valid.
[0105] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and changes can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.
Claims
1. A facial recognition access control and stranger early warning system for wards, characterized in that: include: The medical process synchronization module communicates bidirectionally with the hospital's HIS system to collect and synchronize medical process data and basic personnel data, and transmit them to the access control module and the behavior pattern analysis and verification module. The access control module generates target access parameters based on medical process data and personnel basic data, and uploads them to the multi-factor identity verification module. The multi-factor identity verification module verifies the identity of personnel based on access control parameters using facial features and auxiliary identity media, and outputs the verification result data to the behavior pattern analysis and verification module and the multi-level early warning module. The behavior pattern analysis and verification module constructs the normal behavior pattern of the target personnel based on medical process data, access control parameters and verification structure, identifies abnormal behaviors that deviate from the normal behavior pattern and outputs the abnormal judgment result to the multi-level early warning module. A multi-level early warning module, which outputs early warning and handling instructions and related processing strategies according to the verification result data and anomaly judgment results and preset risk levels, and synchronizes them to the ward management terminal and the management personnel terminal. The caregiver qualification management module is used to collect caregiver qualification data, generate caregiver-specific access control parameters, and synchronize them to the access control module.
2. The ward-based facial recognition access control and stranger early warning system according to claim 1, characterized in that, The medical process synchronization module includes an information interaction unit, an information classification and transmission unit, and a trajectory association unit. The information interaction unit is used to collect medical process data and basic personnel data from the hospital HIS system. The information classification and transmission unit categorizes medical process data into surgical, laboratory, and departmental transfer categories, and classifies basic personnel data according to personnel type; the personnel types include medical staff, patients, and caregivers. The trajectory association unit generates associated trajectory data with medical scenario markers based on the verification result data and associates it with the corresponding medical process data, and transmits it to the behavior pattern analysis and verification module; the verification result data includes the travel location, time and identity identifier.
3. The ward-based facial recognition access control and stranger early warning system according to claim 2, characterized in that, The medical process data includes the time, location, and related information of the medical events; The basic personnel data includes the target personnel's identity identifier, role, and associated medical attributes.
4. The ward-based facial recognition access control and stranger early warning system according to claim 3, characterized in that, The permission control module includes a basic permission control unit, a dynamic permission adjustment unit, and a permission synchronization unit. The basic access control unit divides the basic access control permission range for different types of personnel based on basic personnel data and according to preset role-permission mapping rules; The dynamic permission adjustment unit adaptively adjusts the basic access control permission range based on medical process data and generates temporary access control permission parameters; The permission synchronization unit is used to combine the basic access control permission range and the temporary access control permission parameters as the target permission parameters, and synchronize them to the multi-factor identity verification module and the behavior pattern analysis and verification module.
5. The ward-based facial recognition access control and stranger early warning system according to claim 4, characterized in that, The multi-factor identity verification module includes a verification rule configuration unit, an auxiliary medium verification unit, a facial feature verification unit, an emergency verification unit, and a verification result data output unit. The verification rule configuration unit configures combined verification rules for facial features and auxiliary identity media for different personnel types based on the target permission parameters; The auxiliary medium verification unit is used to collect auxiliary identity medium information of the passers-in personnel, verify the validity of the auxiliary identity medium information, and obtain auxiliary identity medium verification result data; The facial feature verification unit is used to collect facial feature information of passing personnel and match and verify it with the target facial features to obtain facial feature verification result data; The emergency verification unit is used to verify the identity validity of passers-by and obtain emergency verification results in scenarios where auxiliary identity media is missing or invalid, by using a verification method that combines voice commands with identity verification codes. The verification result data output unit generates verification result data based on the auxiliary identity medium verification result data, facial feature verification result data, and emergency verification result data, and transmits it to the behavior pattern analysis and verification module and the multi-level early warning module; the verification result data includes pass / fail and the reason for failure.
6. The ward-based facial recognition access control and stranger early warning system according to claim 5, characterized in that, The behavior pattern analysis and verification module includes a behavior pattern construction unit, a behavior deviation judgment unit, a secondary verification trigger unit, and a judgment result transmission unit. The behavior pattern construction unit constructs normal behavior patterns adapted to ward scenarios based on medical process data, target permission parameters, and associated trajectory data, according to personnel type. The behavior deviation determination unit determines whether the current passage behavior is abnormal based on the deviation degree between the verification result data and the normal behavior pattern, combined with the urgency weight of the medical scenario. The secondary verification triggering unit is used to generate secondary verification items related to the medical scenario when it is determined that the current passage behavior is an abnormal behavior, push them to the interaction terminal of the passage personnel, and generate a behavior judgment result after receiving feedback; the behavior judgment result includes normal / abnormal and abnormal type; The judgment result transmission unit is used to transmit the behavior judgment result to the multi-level early warning module and the medical process synchronization module.
7. The ward-based facial recognition access control and stranger early warning system according to claim 6, characterized in that, The methods for determining whether the current passage behavior is abnormal include: Let the behavioral parameter vector under the normal behavior pattern be: Let the behavior parameter vector of the current passage behavior be: Calculate the deviation: In the formula, For deviation degree, The total number of behavioral parameters. For the first A vector of behavioral parameters under a normal behavioral pattern For the first A vector of behavior parameters for each current passage behavior; In the formula, Assigning a weight to the urgency of the current medical situation in the ward. The threshold value is used.
8. The ward-based facial recognition access control and stranger early warning system according to claim 7, characterized in that, The multi-level early warning module includes a risk level determination unit, a response strategy generation unit, and an instruction push unit. The risk level determination unit determines the warning level based on the verification result data and behavior determination results according to the preset risk classification rules; the warning level includes low risk level, medium risk level and high risk level; The handling strategy generation unit is used to generate appropriate handling strategies for different warning levels: Low-risk levels only store verification result data; For medium-risk areas, local audio-visual alerts will be activated and information will be pushed to the ward management terminal. High-risk levels trigger early warning information to management personnel's terminals, which then coordinate with related equipment in the ward to implement control measures and generate time-limited disposal work orders. The instruction push unit is used to push warning instructions and handling strategies to the corresponding execution terminal, receive handling completion feedback, and update the work order status.
9. The ward-based facial recognition access control and stranger early warning system according to claim 8, characterized in that, The preset risk classification rules include: STEP 1: Formulate matching degree risk rules: When the facial feature matching rate is ≥90% and the auxiliary identity medium is valid, it is judged as a low-risk level; When the facial feature matching rate is in the range of [70%, 90%) and the auxiliary identity medium is valid, it is judged as a medium risk level; When the facial feature matching rate is less than 70% or the auxiliary identity medium is invalid, it is judged as a high-risk level; STEP 2, Drafting of Behavioral Risk Assessment Rules: When abnormal behavior occurs and the urgency weight of the current ward's medical scenario is higher than a preset threshold, it is judged as a high-risk level. When the behavior is abnormal and the urgency weight of the current ward's medical scenario is lower than a preset threshold, it is judged as a medium-risk level; STEP3, Comprehensive Risk Assessment: When there is no abnormal behavior, the matching degree risk level is used as the warning level; when the matching degree risk level and the behavior assessment risk level are inconsistent, the highest level is taken as the warning level.
10. The ward-based facial recognition access control and stranger early warning system according to claim 9, characterized in that, The caregiver qualification management module includes a qualification verification unit, a caregiver permission generation unit, and an anomaly handling unit. The qualification verification unit is used to collect the qualification data of caregivers and generate qualification verification result data; the caregiver qualification data includes identity information, kinship, health status and caregiver badge ID; The caregiver permission generation unit generates caregiver-specific access control parameters associated with the patient's hospitalization cycle and ward area based on the qualification verification results data and the patient's medical process data, and synchronizes them to the permission management module. The anomaly handling unit is used to generate a freeze command for the exclusive access control parameters of the caregiver when the caregiver's qualifications expire or his / her health status is abnormal, and synchronize it to the interactive terminal of the access personnel. At the same time, it pushes an anomaly notification to the ward management personnel. After the qualifications are restored to be valid, the exclusive access control parameters of the caregiver are updated.