Dynamic scenario-aware access permission allocation method, system and device

Abstract

The application discloses a dynamic scene-aware access permission allocation method, system and device, and relates to the technical field of data processing. The method comprises the following steps: determining an access gate of a preset area in a target campus; collecting access scene images of the gate in real time; performing access aggregation degree identification based on the access scene images of the gate, and when the access aggregation degree is greater than a preset aggregation threshold, performing group face recognition on the access scene images of the gate to generate first-class personnel and second-class personnel; performing authorized trajectory extraction on the first-class personnel to construct a first authorized trajectory mapping; performing permission verification on the first-class personnel to exempt from access processing, and performing tracking verification after access based on the first authorized trajectory mapping; and performing verification of a preset permission verification strategy on the second-class personnel, and issuing semantic micro-permissions for tracking verification. The technical problem that the access gate in the prior art is difficult to balance fast access and permission security verification is solved, and the technical effect of improving access efficiency and security is achieved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method, system, and device for allocating access permissions based on dynamic scene perception. Background Technology

[0002] With frequent personnel movement in areas such as campuses, businesses, and public places, access control systems face challenges in terms of both security and efficiency. Traditional access gates typically rely on static permission verification methods, such as card swiping and password input. While secure, these methods can easily cause congestion and reduce efficiency during peak hours, especially when many people are queuing. Particularly in time-sensitive situations, how to quickly complete access verification, ensure accurate identification of personnel permissions, and avoid security risks has become a pressing issue. Therefore, balancing access speed with the security of permission verification has become an important research direction in access control systems. Summary of the Invention

[0003] This application provides a method, system, and device for dynamically scene-aware access permission allocation, which solves the technical problem in the prior art that access gates cannot simultaneously achieve fast passage and secure permission verification.

[0004] The first aspect of this application provides a method for dynamically scene-aware access permission allocation, the method comprising:

[0005] The system identifies access gates within a predefined area of ​​the target campus; it then captures real-time images of the access gates using non-authorized verification cameras; based on these images, it identifies the degree of crowding, and when the crowding exceeds a preset threshold, it performs group face recognition to generate a first category of personnel and a second category of personnel. The first category consists of dynamically authorized personnel, and the second category consists of temporary visitors. For the first category, it extracts authorization trajectories and constructs a first authorization trajectory mapping; it performs authorization verification to exempt the first category of personnel from access and performs post-access tracking verification based on the first authorization trajectory mapping; for the second category, it performs verification using a preset authorization verification strategy and issues semantic micro-permissions for tracking verification.

[0006] A second aspect of this application provides a dynamic scene-aware access permission allocation system, the system comprising:

[0007] The system includes a gate positioning module for identifying access gates within a preset area on campus; an image acquisition module for capturing real-time images of the access gate's passage scene using a non-authorized verification camera at the gate; a personnel classification module for identifying passage clustering based on the access gate's passage scene images; and a group face recognition module for performing group face recognition on the access gate's passage scene images when the passage clustering exceeds a preset clustering threshold, generating a first category of personnel and a second category of personnel, wherein the first category of personnel are dynamically authorized personnel and the second category of personnel are temporary visitors; a trajectory extraction module for extracting authorized trajectories for the first category of personnel and constructing a first authorized trajectory mapping; a first processing module for performing authorization verification to exempt the first category of personnel from passage and performing post-passage tracking verification based on the first authorized trajectory mapping; and a second processing module for performing verification using a preset authorization verification strategy for the second category of personnel and issuing semantic micro-authorizations for tracking verification.

[0008] A third aspect of this application provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing the dynamic scene-aware access permission allocation method provided in this application when executing the executable instructions stored in the memory.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] First, the access gates in a predetermined area within the target campus are identified. Next, non-authorized verification cameras at the access gates capture real-time images of the passage scene. Further, based on these images, passage clustering is identified. When the clustering exceeds a preset threshold, group face recognition is performed to generate two categories of people: the first category represents dynamically authorized personnel, and the second category represents temporary visitors. Then, authorization trajectory extraction is performed on the first category of people, constructing a first authorization trajectory mapping. Access is then waived for them, and post-access tracking verification is performed based on the first authorization trajectory mapping. Finally, the second category of people undergoes verification using a preset access control strategy, and semantic micro-permissions are granted for tracking verification. This approach solves the technical problem in existing technologies where access gates struggle to balance rapid passage with secure access control, achieving improved efficiency and security. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic diagram of the dynamic scene-aware access permission allocation method provided in an embodiment of this application;

[0013] Figure 2 A schematic diagram of the structure of a dynamic scene-aware access control allocation system provided in an embodiment of this application;

[0014] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application.

[0015] Explanation of reference numerals in the attached drawings: Gate positioning module 11, Image acquisition module 12, Personnel classification module 13, Trajectory extraction module 14, First processing module 15, Second processing module 16, Processor 21, Memory 22, Input device 23, Output device 24. Detailed Implementation

[0016] This application solves the technical problem in the prior art that access gates cannot simultaneously achieve fast passage and secure access verification by providing a method, system and device for dynamic scene-aware access permission allocation.

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0019] Example 1, as Figure 1 As shown in the embodiment of this application, a method for dynamically scene-aware access permission allocation is provided, wherein the method includes:

[0020] Identify the access gates for the pre-defined area within the target campus.

[0021] In this embodiment of the application, based on the building layout and personnel flow needs of the target campus, areas that require access control are selected, and access gates are set up in these areas. Then, the access gates are connected to the campus smart management platform so that the status of the gates and the access situation can be updated to the platform in real time.

[0022] The non-authorized verification camera at the access gate captures images of the passage scene in real time.

[0023] In this embodiment of the application, a non-authorization verification camera is installed above each access gate. These non-authorization verification cameras are responsible for monitoring and recording the activities and passage of people around the gate. They are not used to directly verify the passage of people, but to capture images of the passage scene at the gate so as to analyze the crowd gathering situation, identify the queue length, passage speed and other information.

[0024] Based on the gate passage scene image, the passage aggregation degree is identified. When the passage aggregation degree is greater than the preset aggregation threshold, the gate passage scene image is subjected to group face recognition to generate a first type of person and a second type of person. The first type of person is a dynamically authorized person, and the second type of person is a temporary visitor.

[0025] By analyzing the collected images of the toll gate passage scene, the system identifies the density and distribution of people in the images and calculates the passage aggregation degree. When the passage aggregation degree exceeds a preset aggregation threshold, it indicates that congestion or dense crowds have occurred in the toll gate area, and the system immediately initiates the group face recognition process. Group face recognition, based on deep learning algorithms, can accurately detect and match multiple faces appearing simultaneously in an image. By analyzing the facial features of each person, the system identifies the identity of each person passing through. Specifically, the system uses a face detection model to locate all faces in an image, extracting each facial region. A pre-trained face recognition network is then used to extract features from each extracted face, transforming the unique features of each face (such as the position and shape of key facial points like the eyes, nose, and mouth) into high-dimensional feature vectors. These feature vectors reflect the unique identity of each person passing through. The system compares the extracted feature vectors with known personnel information in the database, accurately identifying each person by calculating the similarity between the feature vectors. For example, for dynamically authorized personnel, the system matches them with pre-stored facial features to confirm they are in the first category; if the facial feature matching fails, they are classified as second-category personnel, i.e., temporary visitors. Through group face recognition, personnel are divided into first-category and second-category personnel. The first category consists of dynamically authorized personnel (those with long-term authorization, such as faculty and students); the second category consists of temporary visitors (external personnel who have not been authorized in advance but are temporarily visiting the campus).

[0026] Furthermore, based on the gate passage scene image, the process of identifying passage clustering includes:

[0027] Based on the gate passage scene image, the passage rate and queue length changes within the camera's field of view are identified to generate passage rate index and queue length change index; the passage time is predicted using the passage rate index and queue length change index, and the ratio of the predicted time to the preset non-aggregation discrimination time is calculated to generate the passage aggregation degree.

[0028] The system analyzes captured images of the gate passage scene to identify the passage rate of people and changes in queue length within the camera's field of view. It calculates the number of people passing through the gate per unit time and generates a passage rate index based on this data, thus assessing the speed of people flow. Specifically, the system analyzes the gate passage scene images, identifies each individual in the image, and locates each person in the image using object detection algorithms (such as YOLO or Faster R-CNN). After determining the position of each person, it tracks their movement trajectory by calculating the displacement of the same person between different time frames. During this process, feature-based tracking algorithms (such as KLT tracking or deep learning object tracking methods) can be used to continuously track the positional changes of each person in the video. By analyzing the movement distance of each person within a certain time interval, the system can calculate the number of people passing through the gate per unit time, thereby generating a passage rate index. This index reflects the speed of people flow; a higher passage rate indicates higher efficiency in people passing through the gate, while a lower rate indicates slower passage and potential congestion.

[0029] By analyzing the density and distribution of people in images of a gate passage scene, a queue length change index is generated to measure queue trends and congestion levels. Specifically, object detection algorithms (such as YOLO or Faster R-CNN) are used to identify all people in the image and extract the location of each individual. Based on this location information, the image is divided into multiple virtual grids, and the number of people in each grid is analyzed to determine the distribution of people. According to the relative positions and arrangement of people, the start and end areas of the queue are automatically identified, and the length of the queue area is calculated by measuring the farthest distance between people in the queue. Based on this, the queue length is monitored in real time over time, and the rate of change is calculated to generate a queue length change index. If the queue length increases rapidly in a short period, it indicates a slowdown in passage speed and potential congestion; conversely, if the queue length is relatively stable, it indicates smooth passage.

[0030] The system predicts the total passage time required to pass through the gate by real-time calculations of the passage rate and queue length change indicators. The passage time prediction is based on the current passage rate (i.e., the number of people passing through per unit time) and the queue length change (i.e., the rate of change of the queue area). By combining these two indicators, the system calculates the total time required for all people to pass through the gate at the current passage rate. The prediction formula can be expressed as: Predicted Time = Queue Length / Passage Rate, where the queue length is the total length of all people in the current queue area, and the passage rate is the number of people passing through the gate per unit time.

[0031] The system calculates the ratio between the predicted duration and the preset non-clustering discrimination duration to obtain the traffic clustering degree. The non-clustering discrimination duration is based on the average passage time of people under normal circumstances when there is no congestion in the traffic flow. When the ratio of the actual predicted duration to the preset duration is too high, it indicates that there is congestion or clustering in the current traffic situation.

[0032] Furthermore, the method of identifying traffic clustering based on the gate passage scene image also includes:

[0033] Connect to the campus smart management platform and read the authorized access characteristics of the preset area; predict the spatial distribution of access concentration based on the authorized access characteristics and historical access characteristics, and construct a concentration constraint interval; verify the access concentration using the concentration constraint interval, and if the verification shows an anomaly, issue an abnormal concentration reminder to the access control personnel of the preset area.

[0034] The system connects to the campus smart management platform and reads the authorized access characteristics of the area from the platform. The authorized access characteristics include the access policy within the time period, personnel access restrictions (such as the area open within a specific time period, the categories of personnel allowed to enter, etc.), and the normal access patterns within the area (such as the frequency of access during peak hours, the mobility within the open time period, etc.).

[0035] Historical traffic characteristics refer to traffic data in the area during the same time period in the past, including historical traffic patterns such as pedestrian flow, queue length, and transit time, as well as traffic speed and concentration calculated based on this data.

[0036] Based on authorized access characteristics and historical traffic characteristics, data mining or machine learning algorithms (such as regression analysis and time series forecasting) are used to predict future traffic aggregation. By analyzing the relationship between historical traffic characteristics and authorized access characteristics, the system can predict the possible distribution of traffic aggregation in a region at a future time or period. For example, if a large number of people typically pass through a certain area during a certain period, and that area is open for specific activities, the system will estimate the traffic aggregation of that area based on historical data. Based on the predicted aggregation data, the system will construct an aggregation constraint interval, which is an allowable aggregation range reflecting the maximum and minimum population density that the area should have under normal traffic conditions.

[0037] The system uses a clustering constraint range to verify the currently calculated traffic clustering. If the verification result shows that the current traffic clustering exceeds the preset normal range, the system will determine it as an anomaly, indicating a potential risk of excessive crowding in the area. At this time, the system will send an anomaly alert to the access control administrators of the area, allowing them to take timely measures, such as managing pedestrian flow, adding temporary open lanes, or adjusting other access strategies, thereby effectively avoiding congestion or security risks.

[0038] Furthermore, if the verification shows an anomaly, it also includes:

[0039] Real-time spatiotemporal event acquisition is performed to generate spatiotemporal event information; the surge index of the access clustering degree relative to the clustering degree constraint interval is calculated; the surge impact analysis is performed using the spatiotemporal event information, and the surge index is matched and verified; if the matching verification is successful, permission sensitivity analysis is performed based on the regional attributes of the preset area; if the permission sensitivity level is less than the preset threshold, third-category personnel without access permissions are allowed to pass and their activity areas are restricted and tracked, and the information is reported to the permission management personnel.

[0040] The system dynamically monitors the area surrounding the access gate, collecting real-time spatiotemporal event information during the passage of people. This information includes the time and location of the event, the people involved, the type of event (such as queuing, changes in passage speed, etc.), and possible abnormal situations. For example, a surge in people entering a library during rainy weather could be observed.

[0041] The system compares the current traffic density with a preset density constraint range and calculates a surge index: Surge Index = (Traffic Density - Upper Limit of Density Constraint Range) / Upper Limit of Density Constraint Range. The surge index reflects the degree of sudden increase in the current density compared to the normal range. For example, if the population density rises rapidly at a certain moment and significantly exceeds the normal traffic flow range, the surge index value will become higher.

[0042] The system combines spatiotemporal event information with surge indicators to perform surge impact analysis, in order to determine whether abnormal fluctuations in surge indicators are related to certain specific events (such as special activities, peak population flows, etc.). By matching surge indicators with related event information, the system can verify whether there are traffic anomalies caused by special factors, thereby determining whether further intervention or adjustments to traffic strategies are needed.

[0043] If the matching verification passes, the system will perform a permission sensitivity analysis based on the preset area's attributes. Area attributes include the area's accessibility (e.g., whether it's a high-risk or restricted area), its security level (e.g., whether it contains critical infrastructure or important equipment), and its access control requirements (e.g., whether strict permission management is needed). Through analysis, the system can determine the area's permission sensitivity level, which reflects the area's security requirements and the need to control personnel movement. If the permission sensitivity level is lower than a preset security threshold, indicating relatively low security requirements for the area, the system will determine whether to allow access to third-party personnel (e.g., temporary visitors or unauthorized personnel) without proper permissions. In this case, the system will grant temporary access, but only within a specific area. To ensure security, the system will explicitly restrict the activity areas of these personnel, ensuring they only move or operate within permitted areas and avoid entering high-risk or restricted areas. Furthermore, the system will monitor the activity trajectories of these third-party personnel in real time, ensuring their actions are under surveillance; any unauthorized or abnormal behavior will be recorded and trigger an alarm. At the same time, the system will record the release information and activity trajectory of the third category of personnel in detail and report them to the access control administrator.

[0044] Furthermore, when the passage aggregation degree exceeds a preset aggregation threshold, group face recognition is performed on the gate passage scene image to generate a first category of people and a second category of people, including:

[0045] Perform group face recognition on the gate passage scene image to generate group face recognition results; based on the group face recognition results, dynamically classify authorized personnel and temporary visitors to generate the first type of personnel and the second type of personnel.

[0046] When the system detects excessive crowding in the gate area, it automatically triggers the group face recognition function. This function uses advanced face detection algorithms (such as Convolutional Neural Networks (CNNs) or deep learning models) to detect multiple faces in an image. Based on the group face recognition results, the system compares the identified facial features with personnel information stored in the database to classify dynamically authorized personnel and temporary visitors. By comparing identity information, the system can automatically distinguish between two categories of personnel: the first category is dynamically authorized personnel who have already obtained authorization (such as staff or students holding long-term valid passes), and the second category is temporary visitors who require temporary authorization (such as outsiders or participants in short-term events).

[0047] Furthermore, after generating the first type of personnel and the second type of personnel, the process also includes:

[0048] The second category of personnel is retrieved to perform a preset permission verification strategy to identify authorized personnel and related permissions, thereby determining the fourth category of personnel; the fourth category of personnel is then classified into the first category of personnel and bound and tracked with related authorized personnel.

[0049] After identifying the second category of personnel (temporary visitors), the system will further verify their identities according to preset permission verification policies. These policies may include temporary identity verification, pre-registered visitor information, and visitor activity permissions. The system will then determine whether to grant the temporary visitor additional permissions—specifically, permissions granted to authorized personnel. For example, if a visitor is traveling with an authorized person (such as faculty or students), the system may grant that visitor specific temporary access permissions or allow them to share certain access permissions with the authorized person. Once the second category of personnel (temporary visitors) has passed permission verification and is confirmed to have additional permissions, the system will classify them as a fourth category of personnel and categorize them under the first category. Simultaneously, the system will track these fourth category personnel in conjunction with the authorized person (such as accompanying faculty or students), continuously monitoring their movements and ensuring they adhere to restricted areas or access rules. For instance, the system will monitor the fourth category's activity range in real time, ensuring they remain confined to authorized areas or activities and do not cross permission boundaries.

[0050] For the first type of personnel, the authorization trajectory is extracted, and the first authorization trajectory mapping is constructed.

[0051] The system monitors the passage behavior of Category I personnel, collecting their activity trajectories over a specified time period. The system acquires entry and exit data of these personnel through cameras and sensors (such as RFID and QR code scanners) installed in the passage area. Whenever Category I personnel pass through a gate or enter a specific area, the system records their identity information, timestamp, and location coordinates, tracking their movement path in real time. The system then performs authorized trajectory extraction on this real-time data, extracting the personnel's movement path and behavioral patterns from the collected multi-dimensional data; based on the extracted data, it constructs a first authorized trajectory mapping, i.e., a route map of the personnel's actions within a preset area.

[0052] For the first type of personnel, permission verification is performed to exempt them from access, and post-access tracking verification is performed based on the first authorized trajectory mapping.

[0053] Once the system identifies the first type of person and confirms their valid access rights, it automatically waives secondary access verification. This means that when passing through a gate or entering an authorized area, the person does not need to swipe their card or enter a password again, thus improving access efficiency. However, despite waiving access verification, the system continues to monitor the person's subsequent activities based on the first authorized trajectory mapping. By tracking the person's movement trajectory in real time and comparing it with the pre-established authorized trajectory, the system ensures that the person does not deviate from the predetermined path or enter unauthorized areas during passage. If the person's movement trajectory deviates abnormally, the system will immediately issue an alarm and record the abnormal behavior information for further processing.

[0054] Furthermore, the process of exempting the first type of personnel from access control through authorization verification, and performing post-access tracking verification based on the first authorized trajectory mapping, includes:

[0055] The system uses cameras within the preset area to monitor in real time and determine a first monitoring trajectory; it compares the first monitoring trajectory with the first authorized trajectory to determine the trajectory deviation; and it issues a trajectory permission verification reminder based on the trajectory deviation to verify the trajectory deviation.

[0056] The system uses cameras within a pre-defined area to monitor, track, and record the real-time movement paths of Category I personnel. The cameras capture activity data of the personnel within the area and generate a first monitoring trajectory based on their location information—the personnel's actual movement path within the area. The system compares this first monitoring trajectory with a pre-built first authorized trajectory mapping, which is the personnel's pre-defined path within the authorized time period. By comparing the differences, the system can determine if there is a trajectory deviation—that is, whether the personnel have deviated from the authorized route or entered an unauthorized area. If the system detects a trajectory deviation, i.e., a difference between the personnel's actual path and the authorized trajectory, it will immediately issue a trajectory permission verification alert, notifying the permission administrator to notice the personnel's abnormal behavior. At this point, the system will further verify the trajectory deviation, possibly by re-confirming the personnel's identity or implementing other security control measures to ensure that the personnel's behavior remains within the authorized scope.

[0057] The second type of personnel are subject to a preset permission verification strategy, and semantic micro-permissions are issued for tracking and verification.

[0058] For the second category of personnel, the system implements preset permission verification strategies, such as temporary identity verification or activity participation authentication, to ensure the legitimacy of their access rights. Once verification is successful, the system grants them semantic micro-permissions based on their access needs, limiting their executable operations and behavioral scope within specific times and areas. These micro-permissions allow for fine-grained control over visitors' activities, such as access to specific areas, dwell time, and interactive objects. Subsequently, the system tracks and verifies the personnel's activities through real-time monitoring devices to ensure their behavior remains within the authorized scope. If their behavior deviates from the authorized scope, the system immediately issues an alert and intervenes promptly. Through this fine-grained permission management and continuous monitoring, the system effectively ensures the efficiency of temporary visitor access while maintaining area security.

[0059] Furthermore, the second type of personnel is subject to a preset permission verification strategy, and semantic micro-permissions are granted for tracking and verification, including:

[0060] Based on the preset permission verification strategy, the temporary tasks and associated sub-areas of the second type of personnel are determined; semantic micro-permissions are configured using the temporary tasks and associated sub-areas, including time limits, path limits, area limits, and interaction object limits; permission tracking and verification are performed after the second type of personnel have passed through using the semantic micro-permissions, and an overreach warning is issued when the permissions deviate from the semantic micro-permissions.

[0061] Based on a preset permission verification policy, the system determines the temporary tasks of Category II personnel and the associated sub-areas they need to access. Temporary tasks may include participating in a specific activity, meeting, or other temporarily authorized tasks, while the associated sub-areas are the areas that the personnel are allowed to enter.

[0062] The system configures semantic micro-permissions for the second type of personnel based on temporary tasks and associated sub-regions. Semantic micro-permissions include restrictions on the personnel's access time, path, area, and interaction objects. For example, the system may allow the personnel to access a specified area within a specific time period, limited to a predetermined path, or stipulate that the personnel can only interact with certain specific individuals.

[0063] The system tracks and verifies visitor permissions after access is granted. By monitoring the visitor's activities in real time, the system verifies whether their behavior meets predetermined conditions against semantic micro-permissions. If the visitor deviates from the preset permission range, such as entering an unauthorized area or exceeding the permitted activity time, the system will immediately trigger an unauthorized access warning and take emergency measures, such as issuing a warning, restricting further access, or notifying the access control administrator for intervention.

[0064] In summary, the embodiments of this application have at least the following technical effects:

[0065] First, the access gates in a predetermined area within the target campus are identified. Next, non-authorized verification cameras at the access gates capture real-time images of the passage scene. Further, based on these images, passage clustering is identified. When the clustering exceeds a preset threshold, group face recognition is performed to generate two categories of people: the first category represents dynamically authorized personnel, and the second category represents temporary visitors. Then, authorization trajectory extraction is performed on the first category of people, constructing a first authorization trajectory mapping. Access is then waived for them, and post-access tracking verification is performed based on the first authorization trajectory mapping. Finally, the second category of people undergoes verification using a preset access control strategy, and semantic micro-permissions are granted for tracking verification. This approach solves the technical problem in existing technologies where access gates struggle to balance rapid passage with secure access control, achieving improved efficiency and security.

[0066] Example 2, based on the same inventive concept as the dynamic scene perception access permission allocation method in the foregoing examples, such as... Figure 2 As shown, this application provides a dynamic scene-aware access control system, wherein the system includes:

[0067] The gate positioning module 11 is used to determine the access gate in a preset area within the target campus; the image acquisition module 12 is used to acquire real-time images of the gate passage scene through the non-authorization verification camera of the access gate; the personnel classification module 13 is used to identify the passage aggregation degree based on the gate passage scene image, and when the passage aggregation degree is greater than a preset aggregation threshold, to perform group face recognition on the gate passage scene image to generate a first type of personnel and a second type of personnel, wherein the first type of personnel is dynamically authorized personnel and the second type of personnel is temporary visitors; the trajectory extraction module 14 is used to extract the authorization trajectory for the first type of personnel and construct a first authorization trajectory mapping; the first processing module 15 is used to perform authorization verification exemption processing for the first type of personnel and perform post-passage tracking verification based on the first authorization trajectory mapping; the second processing module 16 is used to perform verification of the preset authorization verification strategy for the second type of personnel and issue semantic micro-authorization for tracking verification.

[0068] Furthermore, the personnel classification module 13 is used to perform the following methods:

[0069] Connect to the campus smart management platform and read the authorized access characteristics of the preset area; predict the spatial distribution of access concentration based on the authorized access characteristics and historical access characteristics, and construct a concentration constraint interval; verify the access concentration using the concentration constraint interval, and if the verification shows an anomaly, issue an abnormal concentration reminder to the access control personnel of the preset area.

[0070] Furthermore, the personnel classification module 13 is used to perform the following methods:

[0071] Real-time spatiotemporal event acquisition is performed to generate spatiotemporal event information; the surge index of the access clustering degree relative to the clustering degree constraint interval is calculated; the surge impact analysis is performed using the spatiotemporal event information, and the surge index is matched and verified; if the matching verification is successful, permission sensitivity analysis is performed based on the regional attributes of the preset area; if the permission sensitivity level is less than the preset threshold, third-category personnel without access permissions are allowed to pass and their activity areas are restricted and tracked, and the information is reported to the permission management personnel.

[0072] Furthermore, the personnel classification module 13 is used to perform the following methods:

[0073] Based on the gate passage scene image, the passage rate and queue length changes within the camera's field of view are identified to generate passage rate index and queue length change index; the passage time is predicted using the passage rate index and queue length change index, and the ratio of the predicted time to the preset non-aggregation discrimination time is calculated to generate the passage aggregation degree.

[0074] Furthermore, the personnel classification module 13 is used to perform the following methods:

[0075] Perform group face recognition on the gate passage scene image to generate group face recognition results; based on the group face recognition results, dynamically classify authorized personnel and temporary visitors to generate the first type of personnel and the second type of personnel.

[0076] Furthermore, the personnel classification module 13 is used to perform the following methods:

[0077] The second category of personnel is retrieved to perform a preset permission verification strategy to identify authorized personnel and related permissions, thereby determining the fourth category of personnel; the fourth category of personnel is then classified into the first category of personnel and bound and tracked with related authorized personnel.

[0078] Furthermore, the first processing module 15 is used to execute the following method:

[0079] The system uses cameras within the preset area to monitor in real time and determine a first monitoring trajectory; it compares the first monitoring trajectory with the first authorized trajectory to determine the trajectory deviation; and it issues a trajectory permission verification reminder based on the trajectory deviation to verify the trajectory deviation.

[0080] Furthermore, the second processing module 16 is used to perform the following method:

[0081] Based on the preset permission verification strategy, the temporary tasks and associated sub-areas of the second type of personnel are determined; semantic micro-permissions are configured using the temporary tasks and associated sub-areas, including time limits, path limits, area limits, and interaction object limits; permission tracking and verification are performed after the second type of personnel have passed through using the semantic micro-permissions, and an overreach warning is issued when the permissions deviate from the semantic micro-permissions.

[0082] Example 3, Figure 3 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality or scope of the embodiments of the present invention. Figure 3 As shown, the electronic device includes a processor 21, a memory 22, an input device 23, and an output device 24; the number of processors 21 in the electronic device can be one or more. Figure 3 Taking a processor 21 as an example, the processor 21, memory 22, input device 23, and output device 24 in an electronic device can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0083] The memory 22, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the dynamic scene-aware access permission allocation method in this embodiment of the invention. The processor 21 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 22, thereby implementing the aforementioned dynamic scene-aware access permission allocation method.

[0084] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0085] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0086] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A dynamic scene-aware access permission allocation method, characterized in that, The method includes: Identify the access gates for the pre-defined area within the target campus; The non-authorized verification camera at the access gate captures images of the passage scene in real time. Based on the gate passage scene image, the passage aggregation degree is identified. When the passage aggregation degree is greater than the preset aggregation threshold, the gate passage scene image is subjected to group face recognition to generate a first type of person and a second type of person. The first type of person is a dynamically authorized person, and the second type of person is a temporary visitor who needs temporary authorization. For the first type of personnel, authorization trajectory extraction is performed to construct a first authorization trajectory mapping; For the first type of personnel, permission verification is performed to exempt them from access, and post-access tracking verification is performed based on the first authorized trajectory mapping; The second type of personnel are subject to a preset permission verification strategy, and semantic micro-permissions are issued for tracking and verification. Semantic micro-permissions include time-limited, path-limited, area-limited, and interaction-limited permissions. The method of identifying traffic clustering based on the gate passage scene image also includes: Connect to the campus smart management platform and read the authorized access characteristics of the preset area; Based on the authorized openness characteristics and historical access characteristics, spatial prediction of access clustering degree distribution is performed, and a clustering degree constraint interval is constructed. The access clustering degree is verified using the clustering degree constraint interval. If the verification shows an abnormality, an abnormal clustering reminder is sent to the permission administrator of the preset area. If the verification shows an error, it also includes: Real-time spatiotemporal event acquisition is performed to generate spatiotemporal event information; Calculate the surge index of the traffic clustering degree relative to the clustering degree constraint interval; The spatiotemporal event information is used to perform a sudden increase impact analysis, and the results are matched and verified with the sudden increase indicators. If the matching verification is successful, a permission sensitivity analysis is performed based on the regional attributes of the preset area. If the permission sensitivity level is less than the preset threshold, the third type of personnel without access permissions are allowed to pass and their activity areas are restricted and tracked, and the information is reported to the permission management personnel.

2. The dynamic scene-aware access permission allocation method as described in claim 1, characterized in that, Based on the gate passage scene image, the passage clustering recognition includes: Based on the gate passage scene image, the passage rate of people and the change in queue length within the camera's field of view are identified, and passage rate index and queue length change index are generated. The passage time is predicted using the passage rate index and the queue length change index. The ratio of the predicted time to the preset non-aggregation discrimination time is calculated to generate the passage aggregation degree.

3. The dynamic scene-aware access permission allocation method as described in claim 2, characterized in that, When the passage density exceeds a preset density threshold, group face recognition is performed on the gate passage scene image to generate a first category of people and a second category of people, including: Perform group face recognition on the gate passage scene image to generate group face recognition results; Based on the group facial recognition results, dynamic authorized personnel and temporary visitors are divided into the first category of personnel and the second category of personnel.

4. The dynamic scene-aware access permission allocation method as described in claim 3, characterized in that, After generating the first type of personnel and the second type of personnel, the process also includes: The second category of personnel is retrieved to perform a preset permission verification strategy to identify authorized personnel and their associated permissions, thereby determining the fourth category of personnel. The fourth category of personnel is categorized into the first category of personnel and linked to authorized personnel for tracking.

5. The dynamic scene-aware access permission allocation method as described in claim 1, characterized in that, For the first type of personnel, permission verification is performed to exempt them from access, and post-access tracking verification is performed based on the first authorized trajectory mapping, including: The first monitoring trajectory is determined by real-time monitoring using cameras within the preset area; By comparing the mapping between the first monitoring trajectory and the first authorized trajectory, the trajectory deviation is determined; A trajectory permission verification reminder is issued based on the trajectory deviation, and trajectory deviation verification is performed.

6. The dynamic scene-aware access permission allocation method as described in claim 1, characterized in that, For the second category of personnel, a preset permission verification strategy is implemented, and semantic micro-permissions are issued for tracking and verification, including: The temporary tasks and associated sub-areas of the second type of personnel are determined based on the preset permission verification strategy; Semantic micro-permissions are configured using the aforementioned temporary tasks and associated sub-regions; After the second type of personnel are granted access using the aforementioned semantic micro-permissions, permission tracking and verification are performed, and an overreach warning is issued when the permission deviates from the aforementioned semantic micro-permissions.

7. A dynamic scene-aware access control system, characterized in that, The system is used to implement the dynamic scene perception access permission allocation method according to any one of claims 1-6, the system comprising: The gate positioning module is used to determine the access gates in a preset area within the target campus. The image acquisition module is used to acquire images of the passage scene at the gate in real time through the non-authorization verification camera at the passage gate; The personnel classification module is used to identify the passage aggregation degree based on the gate passage scene image. When the passage aggregation degree is greater than a preset aggregation threshold, the module performs group face recognition on the gate passage scene image to generate a first type of personnel and a second type of personnel. The first type of personnel are dynamically authorized personnel, and the second type of personnel are temporary visitors who need temporary authorization. The trajectory extraction module is used to extract authorized trajectories for the first type of personnel and construct a first authorized trajectory mapping; The first processing module is used to perform permission verification and exemption from passage for the first type of personnel, and to perform post-passage tracking verification based on the first authorized trajectory mapping; The second processing module is used to perform preset permission verification strategy verification on the second type of personnel and issue semantic micro-permissions for tracking and verification. Semantic micro-permissions include time-limited, path-limited, area-limited and interaction-limited permissions. Furthermore, the personnel classification module is used to perform the following methods: Connect to the campus smart management platform and read the authorized access characteristics of the preset area; Based on the authorized openness characteristics and historical access characteristics, spatial prediction of access clustering degree distribution is performed, and a clustering degree constraint interval is constructed. The access clustering degree is verified using the clustering degree constraint interval. If the verification shows an abnormality, an abnormal clustering reminder is sent to the permission administrator of the preset area. Furthermore, the personnel classification module is used to perform the following methods: Real-time spatiotemporal event acquisition is performed to generate spatiotemporal event information; Calculate the surge index of the traffic clustering degree relative to the clustering degree constraint interval; The spatiotemporal event information is used to perform a sudden increase impact analysis, and the results are matched and verified with the sudden increase indicators. If the matching verification is successful, a permission sensitivity analysis is performed based on the regional attributes of the preset area. If the permission sensitivity level is less than the preset threshold, the third type of personnel without access permissions are allowed to pass and their activity areas are restricted and tracked, and the information is reported to the permission management personnel.

8. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the access permission allocation method for dynamic scene perception as described in any one of claims 1-6.

Images