Intelligent building access control system

Abstract

The application relates to the technical field of access control, and discloses an access control optimization control system for intelligent buildings, which comprises a plurality of access units, a plurality of image acquisition modules, a time division module, a plurality of verification information modules and an analysis module. The application acquires image information data of the corresponding access units in real time through the image acquisition modules; divides a target monitoring time period into a plurality of time units through the time division module; stores the verification information data of the corresponding time units through the verification information modules; and analyzes the verification information data and the characteristic images of the characteristic time units of the access units through the analysis module, so that the current traffic level of each access unit at the current time is obtained. The verification strategy is determined based on the dynamically generated current traffic level, the intelligentization and individualization of access control are realized, the dynamic environment changes such as the personnel flow and activity rules in the building can be quickly adapted, and the safety and management efficiency of the access control system are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of access control technology, and more specifically to an optimized access control system for intelligent buildings. Background Technology

[0002] With the rapid advancement of technology, intelligent buildings have been widely applied and developed globally; among them, access control systems, as an important component of intelligent buildings, bear the key responsibility of ensuring the safety of people and property within the building.

[0003] In existing technologies, biometric technology is increasingly being applied to access control systems; common biometric methods include fingerprint recognition, facial recognition, and iris recognition. These technologies all rely on unique human biometric characteristics for identity verification, offering high accuracy and security. Taking facial recognition as an example, this technology analyzes and compares captured facial images to quickly and accurately confirm an individual's identity. Users do not need to carry additional devices or remember passwords, making operation much simpler.

[0004] However, the above technologies still have significant drawbacks. For example, current access control systems often struggle to flexibly adjust the scope and permissions of verification information according to changes over time. This results in the system needing to store a large amount of verification information and compare all the information one by one during the verification process, increasing the system's burden and processing time, and reducing verification efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide an optimized access control system for intelligent buildings, addressing the following technical problems:

[0006] How to make the access control system meet the dynamic changes in personnel flow at different times?

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] An intelligent building access control system includes several access control units, and the control system further includes:

[0009] Several image acquisition modules, each corresponding to an access control unit, are used to collect image information data within the target monitoring area of ​​the corresponding access control unit in real time;

[0010] The time segmentation module is used to divide the target monitoring period into several time units;

[0011] Several verification information modules, each corresponding to a time unit, are used to store verification information data for the corresponding time unit;

[0012] The analysis module analyzes the verification information data of each access control unit's characteristic time units within a preset number of days in the past to obtain the basic flow index of each access control unit's characteristic time units; then, based on the basic flow index and characteristic image of each access control unit's characteristic time units, it analyzes to obtain the current comprehensive flow index of each access control unit; then, based on the current comprehensive flow index of each access control unit, it analyzes to obtain the current flow level of each access control unit at the current time; and finally, it determines the verification strategy based on the current flow level.

[0013] As a further aspect of the present invention: the analysis module includes an identification unit and an analysis unit, wherein the identification unit is a trained convolutional neural network model, used to identify based on feature images and obtain the number of users to be verified in each feature image.

[0014] As a further aspect of the present invention: the analysis process of the analysis unit is as follows:

[0015] S1: By analyzing the verification information data of each access control unit's characteristic time unit within the past preset number of days, the basic flow index of each access control unit's characteristic time unit is obtained;

[0016] S2: Analyze the basic flow index and characteristic image of each access control unit's characteristic time unit to obtain the current comprehensive flow index of each access control unit;

[0017] S3: Analyze the current comprehensive flow index of each access control unit to obtain the current flow level of each access control unit;

[0018] S4: Determine the verification strategy for each access control unit based on the current traffic level of each access control unit.

[0019] As a further aspect of the present invention: In step S1, using formula one:

[0020] ;

[0021] Calculate the basic flow index of any access control unit's characteristic time unit. ;

[0022] in, For any access control unit; For characteristic time units; Preset the number of days in the past. ; The first of the preset days in the past The number of verified users in the access control unit's characteristic time unit; This is the first adjustment coefficient.

[0023] As a further aspect of the present invention: In step S2, formula two is used:

[0024] ;

[0025] Calculate the current comprehensive traffic index of any access control unit at the current time. ;

[0026] in, This represents the number of feature images for this access control unit. ; For the first access control unit The number of users to be verified in each feature image; This is the first weighting coefficient; This is the second weighting coefficient; This is the first preset constant; This is the second preset constant.

[0027] As a further aspect of the present invention: the verification strategy includes an access control recognition verification method, a verification method combining access control recognition and habit verification matching algorithms, and a verification method combining access control recognition and image feature verification matching algorithms.

[0028] As a further aspect of the present invention: the process for determining the current flow level of each access control unit is as follows:

[0029] The current comprehensive traffic index of any access control unit at the current time. With preset threshold Compare;

[0030] when At that time, the current traffic level of the access control unit is low.

[0031] when At that time, the current traffic level of the access control unit was medium.

[0032] when At that time, the current traffic level of the access control unit is high.

[0033] As a further aspect of the present invention: the process for determining the verification strategy of each access control unit is as follows:

[0034] When the current traffic level of this access control unit is low, the access control identification and verification method is used;

[0035] When the current traffic level of the access control unit is medium, a combined verification method using access control recognition and habit verification matching algorithms is used.

[0036] When the current traffic level of the access control unit is high, a combined verification method using access control recognition and image feature verification matching algorithms is used.

[0037] As a further aspect of the present invention: the habit verification matching algorithm analyzes the verification information data of the feature time unit and the reference time unit to obtain the user habit verification index of each user in any access control unit for each user's feature time unit; and then performs access control identification according to the order of the user habit verification index of each user in the feature time unit.

[0038] As a further aspect of the present invention: the identification unit is also used to identify based on feature images, obtain feature elements of each user to be verified in each feature image; analyze based on the feature elements of each user to be verified, the facial image of each user and the habit verification index, obtain the user feature verification index of each user and each user to be verified; and then perform access control identification according to the order of the user feature verification index of each user and each user to be verified.

[0039] The beneficial effects of this invention are:

[0040] (1) The present invention determines the verification strategy based on the dynamically generated current traffic level, realizing the intelligent and personalized access control verification. At the same time, it can quickly adapt to dynamic environmental changes such as the flow of people and activity patterns in the building, greatly improving the security and management efficiency of the access control system, and creating a safer and more convenient access environment for intelligent buildings.

[0041] (2) By fusing the current flow pattern with the feature image, this invention can more accurately predict the flow status of each access control unit in the short term, providing strong support for the subsequent formulation of verification strategies; then, by analyzing the current comprehensive flow index, the current flow level of each access control unit is obtained; and the verification strategy of each access control unit is determined based on the current flow level of each access control unit; thus, it provides a basis for the optimized configuration of access control equipment, rationally allocates resources, improves the operating efficiency and service life of the equipment, reduces maintenance costs, and thereby comprehensively improves the overall efficiency of access control management, creating a safe, convenient and efficient access environment for intelligent buildings;

[0042] (3) The present invention performs access control identification according to the order of the user feature verification index of each user and each user to be verified, so that the system can prioritize the identification of users with high verification index and high probability of using access control; during high traffic periods, this sequential identification method can avoid indiscriminate verification of all users and reduce the waiting time of users. Attached Figure Description

[0043] The invention will now be further described with reference to the accompanying drawings.

[0044] Figure 1 This is a system module framework diagram of one embodiment of the present invention. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] Please see Figure 1 As shown, in one embodiment, an intelligent building access control optimization control system is provided, the control system including a plurality of access control units, a plurality of image acquisition modules, a time division module, a plurality of verification information modules and an analysis module;

[0047] The image acquisition module corresponds one-to-one with the access control unit and is used to collect image information data within the target monitoring area of ​​the corresponding access control unit in real time. The image acquisition module may include several image acquisition devices. This design achieves precise correspondence in the spatial dimension, which not only avoids blind spots in image acquisition but also provides reliable basic data support for subsequent verification.

[0048] The time division module is used to divide the target monitoring period into several time units. The target monitoring period can be 24 hours or set to a specific time period (such as 6:00 to 22:00). This function can also flexibly adjust the granularity of the time unit according to actual needs, which can meet the needs of refined management during peak periods and take into account the efficiency optimization during off-peak periods, so as to achieve flexible control in the time dimension.

[0049] The verification information module corresponds one-to-one with the time unit and is used to store the verification information data of the corresponding time unit. This function forms a mapping relationship between the verification information module and the time unit, and each verification information module independently stores the dynamic verification information data of the corresponding time unit. This spatiotemporal related storage architecture enables the verification information to be automatically updated with the time unit, which not only ensures the traceability of historical data, but also ensures the timeliness of real-time verification information, effectively improving the system's adaptability to dynamic environments.

[0050] The analysis module is used to analyze the verification information data of each access control unit's characteristic time unit (i.e., the time unit corresponding to the current time) within a preset number of days in the past to obtain the basic flow index of each access control unit's characteristic time unit; then, based on the basic flow index and characteristic image (i.e., image information data collected at the current time) of each access control unit's characteristic time unit, it analyzes to obtain the current comprehensive flow index of each access control unit; then, based on the current comprehensive flow index of each access control unit, it analyzes to obtain the current flow level of each access control unit at the current time; and finally, it determines the verification strategy based on the current flow level.

[0051] Through the above technical solution, this embodiment determines the verification strategy based on the dynamically generated current traffic level, realizing the intelligence and personalization of access control verification. At the same time, it can quickly adapt to dynamic environmental changes such as the flow and activity patterns of people in the building, greatly improving the security and management efficiency of the access control system, and creating a safer and more convenient entry and exit environment for intelligent buildings.

[0052] As one embodiment of the present invention, the analysis module includes an identification unit and an analysis unit. The identification unit is a trained convolutional neural network model, used to identify based on feature images and obtain the number of users to be verified in each feature image.

[0053] Through the above technical solution, the user to be verified in this embodiment can be a user staying in the target monitoring area or a user moving towards the access control unit from the access control entrance; the trained convolutional neural network model can quickly and accurately identify the number of users to be verified, that is, accurately predict the current traffic volume of the access control unit, so that the access control unit can adjust the verification strategy in a timely manner according to the real-time traffic situation.

[0054] It should be noted that the training process of the convolutional neural network model is existing technology and will not be described in detail here.

[0055] As one embodiment of the present invention, the analysis process of the analysis unit is as follows:

[0056] S1: By analyzing the verification information data of each access control unit's characteristic time unit within the past preset number of days, the basic flow index of each access control unit's characteristic time unit is obtained;

[0057] S2: Analyze the basic flow index and characteristic image of each access control unit's characteristic time unit to obtain the current comprehensive flow index of each access control unit;

[0058] S3: Analyze the current comprehensive flow index of each access control unit to obtain the current flow level of each access control unit;

[0059] S4: Determine the verification strategy for each access control unit based on the current traffic level of each access control unit;

[0060] Through the above technical solution, this embodiment analyzes the verification information data of each access control unit's characteristic time units within a preset number of days in the past, which can uncover the flow patterns of each access control unit in the current time period; providing a reliable benchmark for determining verification strategies; the feature images can reflect the current actual situation of the target monitoring area of ​​the access control unit in real time, such as the gathering of people and the direction of flow; by fusing the flow patterns of the current time with the feature images, the flow status of each access control unit in the short term in the future can be predicted more accurately, providing strong support for the subsequent formulation of verification strategies; further, the current flow level of each access control unit is obtained by analyzing the current comprehensive flow index; and the verification strategy of each access control unit is determined based on the current flow level of each access control unit; thus, it provides a basis for the optimized configuration of access control equipment, rationally allocates resources, improves the operating efficiency and service life of equipment, reduces maintenance costs, and comprehensively improves the overall efficiency of access control management, creating a safe, convenient, and efficient access environment for intelligent buildings.

[0061] As one embodiment of the present invention, in step S1, formula one is used:

[0062] ;

[0063] Calculate the basic flow index of any access control unit's characteristic time unit. ;

[0064] in, For any access control unit; For characteristic time units; Preset the number of days in the past. ; The first of the preset days in the past The number of verified users in the access control unit's characteristic time unit; This is the first adjustment coefficient;

[0065] Formula 1 Explanation: The average number of verified users in the characteristic time unit of this access control unit over the past preset number of days; the larger the average number of verified users, the more frequent the flow of people and the higher the overall activity level of the access control unit in the characteristic time unit over the past preset number of days. The standard deviation of the number of verified users in this access control unit's characteristic time unit over a preset number of days is used. A smaller standard deviation indicates less fluctuation in the number of verified users within that time unit over the preset number of days, and more stable traffic flow. When the average number of verified users is the same, more stable traffic flow suggests that stable personnel flow in this access control unit's characteristic time unit tends to have higher controllability and predictability, making it more valuable for access control management in intelligent buildings. Therefore, the basic traffic flow index... The larger the value, the lower the stability of the pedestrian flow in the characteristic time unit of the access control unit, and the lower its reference value for access control management in intelligent buildings; therefore, the basic flow index... The smaller;

[0066] Through the above technical solution, this embodiment, by comprehensively considering the average number of verified users and the standard deviation of the number of verified users in the characteristic time unit of the access control unit over a preset number of days, can more accurately and comprehensively reflect the personnel flow characteristics of each access control unit in the characteristic time unit. The average number of verified users reflects the frequency and overall activity of personnel flow from a macro perspective, while the standard deviation describes the stability of personnel flow from a micro perspective. The basic flow index obtained by organically combining the two can not only provide key information for access control management departments to help them to reasonably and dynamically adjust the access control verification strategy, but also minimize personnel waiting time and improve access convenience while ensuring security, thereby improving the operational efficiency and user experience of the building site.

[0067] It should be noted that the first adjustment coefficient These are preset values, set based on empirical fitting, and will not be detailed here.

[0068] As one embodiment of the present invention, in step S2, formula two is used:

[0069] ;

[0070] Calculate the current comprehensive traffic index of any access control unit at the current time. ;

[0071] in, This refers to the number of image acquisition devices in the image acquisition module corresponding to the access control unit (i.e., the number of feature images of the access control unit). ; For the first access control unit The number of users to be verified in each feature image; This is the first weighting coefficient; This is the second weighting coefficient; This is the first preset constant; This is the second preset constant;

[0072] Explanation of Formula 2: The cumulative number of users to be verified in all feature images of this access control unit reflects the actual population density and traffic demand within the target monitoring area of ​​the access control unit at the current moment. A larger cumulative number of users to be verified indicates a greater number of users needing to pass through the access control unit, thus increasing the current comprehensive traffic flow index. The larger the value; the more characteristic the time unit, the more basic the flow index of the access control unit. The higher the overall traffic index, the more frequently a large number of people have passed through the access control unit within the same time period of the preset number of days. This indicates that the access control unit has a high user activity level during that time period, hence the higher current overall traffic index. The larger;

[0073] Through the above technical solution, this embodiment uses Formula 2 to quantify the current comprehensive traffic index, enabling access control managers to obtain a specific and measurable indicator. This indicator can intuitively predict the traffic situation of each access control unit in the short term, thereby more accurately grasping the operating status of the access control system and providing a strong basis for subsequent dynamic adjustment and verification strategies.

[0074] It should be noted that the first weighting coefficient Second weighting coefficient First preset constant Second preset constant These are preset values, obtained based on experience, and will not be detailed here.

[0075] As one embodiment of the present invention, the process for determining the current flow level of each access control unit is as follows:

[0076] The current comprehensive traffic index of any access control unit at the current time. With preset threshold Compare;

[0077] when At that time, the current traffic level of the access control unit is low.

[0078] when At that time, the current traffic level of the access control unit was medium.

[0079] when At that time, the current traffic level of the access control unit is high.

[0080] Through the above technical solution, this embodiment determines the current traffic level by comparing the current comprehensive traffic index with a preset threshold, which can accurately classify the traffic status of the access control unit. This accurate classification allows the system to adopt the most appropriate verification strategy according to different current traffic levels, avoiding the inconvenience and inefficiency caused by "one-size-fits-all" approach, and improving the pertinence and effectiveness of access control system verification.

[0081] It should be noted that the preset threshold These are preset values, set based on empirical fitting, and will not be detailed here.

[0082] As one embodiment of the present invention, the process for determining the verification strategy of each access control unit is as follows:

[0083] When the current traffic level of the access control unit is low, access control recognition verification (such as facial recognition) is used; for example, at night or during non-working hours, fewer people enter and exit the office, and a unified and simple verification method is both convenient and practical. When the current traffic level of the access control unit is medium, a combination of access control recognition and habitual verification matching algorithm is used; the habitual verification matching algorithm is used in access control recognition to optimize the verification method and improve the verification speed to a certain extent. When the current traffic level of the access control unit is high, a combination of access control recognition and image feature verification matching algorithm is used; image features (such as gait) can be identified through feature images within the target monitoring area of ​​the access control unit, and image feature verification matching algorithm is used to reduce the workload of subsequent access control recognition, thereby improving the overall throughput speed; at the same time, the combined verification method also enhances security and prevents unauthorized personnel from entering.

[0084] Through the above technical solution, this embodiment achieves the cooperation of verification strategies under different traffic conditions, which not only ensures economical operation under low traffic conditions, but also takes into account efficiency and security under medium and high traffic conditions. It provides a comprehensive, efficient and reliable solution for access control management in various places and has wide applicability.

[0085] As one embodiment of the present invention, the habit verification matching algorithm analyzes the verification information data of the feature time unit and the reference time unit to obtain the user habit verification index of each user in each user feature time unit in any access control unit; and then performs access control identification according to the order of the user habit verification index of each user in the feature time unit.

[0086] Specifically, the reference time unit is at least two time units adjacent to the feature time unit;

[0087] Through the above technical solution, this embodiment sets the reference time unit to at least two time units adjacent to the feature time unit, which can fully consider the continuity and correlation of user behavior in the time dimension. User access control usage behavior in similar time periods often has certain regularity and similarity. By analyzing the verification information data of these adjacent time units to calculate the user habit verification index of the feature time unit, the user's real usage habits can be captured more accurately. Access control recognition is performed according to the order of the user habit verification index of each user in the feature time unit, so that the system can prioritize the recognition of users with high verification indices and high probability of using access control. This recognition method can avoid indiscriminate verification of all users, greatly reduce unnecessary verification steps, and significantly improve the overall speed of access control recognition.

[0088] As one embodiment of the present invention, formula three is used:

[0089] ;

[0090] Calculate users User habit verification index in any access control unit's characteristic time unit ;

[0091] in, Number the user; For users The number of times the preset number of days was verified in the feature time unit and the reference time unit in the past; For users The number of times a preset number of days was verified in this access control unit; Preset constants based on habits;

[0092] Specifically, the verification information data includes verification time, access control unit number, and user number;

[0093] Through the above technical solution, this embodiment achieves... Reflecting users The probability of verification for a previously preset number of days in this access control unit; the higher the verification probability, the more likely the user... The higher the frequency of use of the access control unit and the more fixed the behavioral patterns during this period, the stronger the stability and predictability of user habits; therefore, users... User habit verification index in the characteristic time unit of the access control unit The larger the value, the better; during access control identification, the user habit verification index of each user is compared in the order of the user habit verification index in the characteristic time unit of the access control unit. The larger the value, the higher the priority.

[0094] It should be noted that the user In the past, the number of verifications and users for the preset number of days in the characteristic time unit and reference time unit. In the past, the number of verifications for a preset number of days characteristic time unit in the access control unit was obtained based on the verification information data of the characteristic time unit and the reference time unit. The method of obtaining this information is existing technology and will not be described in detail here.

[0095] It should be noted that the default constant is conventional. These are preset values, set based on empirical fitting, and will not be detailed here.

[0096] In one embodiment of the present invention, the identification unit is further configured to identify based on feature images, obtain feature elements of each user to be verified in each feature image; analyze based on the feature elements of each user to be verified, the facial images of each user, and the habit verification index, to obtain the user feature verification index of each user and each user to be verified; and then perform access control identification according to the order of the user feature verification index of each user and each user to be verified.

[0097] Specifically, the feature elements may include clothing color, gait, and several facial features;

[0098] Through the above technical solution, this embodiment can track people within the target monitoring area of ​​the access control unit by clothing color, and analyze gait or facial features and habit verification index to obtain the user feature verification index of each user to be verified in each access control unit's characteristic time unit. When each user to be verified performs access control identification, the access control identification is performed according to the order of the user feature verification index of each user and each user to be verified, so that the system can prioritize the identification of users with high verification indices and a high probability of using access control. During high-traffic periods, this sequential identification method can avoid indiscriminate verification of all users and reduce user waiting time.

[0099] It should be noted that identifying human gait and facial features in images based on a trained convolutional neural network model is an existing technology and will not be described in detail here.

[0100] As one embodiment of the present invention, formula four is used:

[0101] ;

[0102] Calculate users The user to be verified in the feature image User feature verification index ;

[0103] in, The number of feature elements, ; User ID to be verified; For users The The degree of matching between each feature element and the user to be verified; The first weighting coefficient; This is the second weighting coefficient; This is the first preset constant; This is the second preset constant;

[0104] Specifically, the verification information data also includes user gait and facial image information data;

[0105] Explanation of Formula 4: For users The user to be verified in the feature image The cumulative value of each feature element; the higher the cumulative value, the better the user... The user to be verified in the feature image The greater the similarity in appearance or behavior, the higher the feature matching degree, thus providing strong support for user identity matching at the feature level; User habit verification index The larger the value, the more likely the user is to be affected. The higher the probability of a user appearing in a feature time unit, the more closely the user's behavior pattern matches the expected behavior pattern in the scenario to be verified. This further increases the credibility of the user's identity matching with the user to be verified in the feature image. Therefore, the user... The user to be verified in the feature image User feature verification index The larger;

[0106] Through the above technical solution, this embodiment comprehensively considers the matching degree between the user and the user to be verified on multiple feature elements and the user's habit verification index. Formula 4 can comprehensively evaluate the user's identity from the two key dimensions of feature similarity and behavioral habits. Compared with identity verification methods that rely on only a single feature or a single verification method, this multi-dimensional comprehensive verification strategy can not only significantly reduce the possibility of misjudgment, but also has the ability to pre-match feature elements before the user to be verified performs access control identification. In this process, the system will prioritize matching items with higher user feature verification indices, thereby effectively improving verification efficiency.

[0107] It should be noted that the process of obtaining the matching degree between the user's facial image information data and the feature elements of the image of the user to be characterized in the feature image is existing technology and will not be described in detail here;

[0108] It should be noted that the first weight coefficient Weighting coefficient No. 2 Preset constant No. 1 and the second preset constant These are preset values, obtained based on experience, and will not be detailed here.

[0109] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An intelligent building access control optimization system, comprising a plurality of access control units, characterized in that, The control system further includes: Several image acquisition modules, each corresponding to an access control unit, are used to collect image information data within the target monitoring area of ​​the corresponding access control unit in real time; The time segmentation module is used to divide the target monitoring period into several time units; Several verification information modules, each corresponding to a time unit, are used to store verification information data for the corresponding time unit; The analysis module is used to analyze the verification information data and feature images of each access control unit's characteristic time units to obtain the current traffic level of each access control unit at the current time; and to determine the verification strategy based on the current traffic level. The process for determining the verification strategy for each access control unit is as follows: When the current traffic level of this access control unit is low, the access control identification and verification method is used; When the current traffic level of the access control unit is medium, a combined verification method using access control recognition and habit verification matching algorithms is used. When the current traffic level of the access control unit is high, a combined verification method using access control recognition and image feature verification matching algorithms is used. The habit verification matching algorithm analyzes the verification information data of the feature time unit and the reference time unit to obtain the user habit verification index of each user in each user feature time unit in any access control unit; and then performs access control identification according to the order of the user habit verification index of each user in the feature time unit. Through formula three: ； Calculate users User habit verification index in any access control unit's characteristic time unit ; in, Number the user; For users The number of times the preset number of days was verified in the feature time unit and the reference time unit in the past; For users The number of times a preset number of days was verified in this access control unit; Preset constants based on habits; The recognition unit is also used to perform recognition based on feature images, obtain feature elements of each user to be verified in each feature image; analyze the feature elements of each user to be verified, the facial images of each user and the habit verification index to obtain the user feature verification index of each user and each user to be verified; and then perform access control recognition according to the order of the user feature verification index of each user and each user to be verified. Through formula four: ； Calculate users The user to be verified in the feature image User feature verification index ; in, The number of feature elements, ; User ID to be verified; For users The The degree of matching between each feature element and the user to be verified; This is the first weighting coefficient; This is the second weighting coefficient; This is the first preset constant; This is the second preset constant.

2. The intelligent building access control system according to claim 1, characterized in that, The analysis module includes an identification unit and an analysis unit. The identification unit is a trained convolutional neural network model used to identify users to be verified in each feature image.

3. The intelligent building access control optimization system according to claim 2, characterized in that, The analysis process of the analysis unit is as follows: S1: By analyzing the verification information data of each access control unit's characteristic time unit within the past preset number of days, the basic flow index of each access control unit's characteristic time unit is obtained; S2: Analyze the basic flow index and characteristic image of each access control unit's characteristic time unit to obtain the current comprehensive flow index of each access control unit; S3: Analyze the current comprehensive flow index of each access control unit to obtain the current flow level of each access control unit; S4: Determine the verification strategy for each access control unit based on the current traffic level of each access control unit.

4. The intelligent building access control system according to claim 3, characterized in that, In step S1, according to formula one: ； Calculate the basic flow index of any access control unit's characteristic time unit. ; in, For any access control unit; For characteristic time units; Preset the number of days in the past. ; The first of the preset days in the past The number of verified users in the access control unit's characteristic time unit; This is the first adjustment coefficient.

5. The intelligent building access control optimization system according to claim 4, characterized in that, In step S2, according to formula two: ； Calculate the current comprehensive traffic index of any access control unit at the current time. ; in, This represents the number of feature images for this access control unit. ; For the first access control unit The number of users to be verified in each feature image; This is the first weighting coefficient; This is the second weighting coefficient; This is the first preset constant; This is the second preset constant.

6. The intelligent building access control optimization control system according to claim 1, characterized in that, The verification strategies include access control recognition verification methods, access control recognition and habit verification matching algorithm combination verification methods, and access control recognition and image feature verification matching algorithm combination verification methods.

7. The intelligent building access control optimization system according to claim 5, characterized in that, The process for determining the current flow level of each access control unit is as follows: The current comprehensive traffic index of any access control unit at the current time. With preset threshold Compare; when At that time, the current traffic level of the access control unit is low. when At that time, the current traffic level of the access control unit was medium. when At that time, the current traffic level of the access control unit is high.

Images