A method for identifying and warning abnormal personnel in smart buildings

Abstract

The present invention relates to the technical field of smart building security management, and specifically to a method for identifying and warning abnormal personnel in a smart building, comprising: acquiring image information collected by monitoring equipment in the building and user identity information perceived by access control equipment in real time, setting a cloud database, and using the cloud database to distinguish and store the image information and the user identity information; performing a tracking operation with a designated tracking target in the image information according to the user identity information. The present invention continuously monitors personnel who enter the smart building without regular identity verification through linkage between monitoring equipment and access control equipment deployed in the building, and analyzes the security threat to the smart building based on the movement path monitoring of the personnel entering the smart building, and further provides trigger conditions and logic for scheduling security resources for the smart building based on the security threat analysis result, i.e., the monitoring of the movement path, to ensure that the daily operation and use of the smart building are safer.

Description

Technical Field

[0001] The present invention relates to the technical field of smart building safety management, and in particular to an abnormal personnel identification and alarm method based on a smart building. Background Art

[0002] The security management of smart buildings covers many aspects. In terms of personnel safety, smart access control strictly controls entry and exit in a variety of identification methods, and indoor positioning technology can track the location of personnel. In terms of facility safety, there are fire warning, elevator monitoring and other systems to ensure operation. In terms of environmental safety, air quality is monitored and adjusted in a coordinated manner. Multiple systems work together to create a safe and efficient building environment.

[0003] The invention patent with application number 202111278532.8 discloses a digital safety detection method based on smart buildings, including the following steps: obtaining a detection signal emitted by a sensor deployed in the building; determining whether the detection signal is an abnormal signal, and if so, determining the corresponding sensor area, operating time and corresponding abnormal type based on the abnormal signal: predicting the probability of an accident and the loss caused by the consequences of the accident based on the sensor area, operating time and corresponding abnormal type; determining the safety risk level based on the probability of an accident and the loss caused by the consequences of the accident; and determining a maintenance plan based on the safety risk level.

[0004] The application aims to solve the problem that "for office buildings, commercial complexes, cultural, media, schools, stadiums, hospitals, transportation, industrial buildings, residential communities and other places, the main buildings all contain various sensors for monitoring water, electricity, gas, temperature, humidity, smoke, network and other systems. Although the monitoring functions of similar sensors are the same, the consequences and losses caused by accidents are different due to different installation locations. The safety detection methods in the existing technology cannot define the risk level according to the type and location of the risk, and therefore it is impossible to formulate a reasonable maintenance plan, which brings a lot of inconvenience to the daily maintenance work of the staff and is not conducive to overall control."

[0005] However, existing technologies for operational risk identification and management of smart buildings mostly focus on the equipment, fire, environment and other aspects of smart buildings. The flow management of people in smart buildings still relies heavily on offline management of gate equipment and security personnel. When unauthenticated personnel break in, the safety hazards caused by unauthenticated personnel cannot be quickly controlled, posing a security threat to the daily operation of smart buildings.

[0006] Therefore, a method for abnormal personnel identification and alarm based on smart buildings is proposed. Summary of the invention

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides an abnormal personnel identification and alarm method based on a smart building, which solves the problems raised in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A method for identifying and warning abnormal personnel in a smart building, comprising:

[0010] Acquire the image information collected by the monitoring equipment in the building and the user identity information perceived by the access control equipment in real time, set up a cloud database, and use the cloud database to distinguish and store the image information and user identity information; perform tracking operations with the specified tracking target in the image information according to the user identity information, and in the tracking target tracking operation stage, build a three-dimensional topology of the internal path of the building based on the internal path of the building, and mark the installation positions of the monitoring equipment and access control equipment on the three-dimensional topology of the internal path of the building; build a mobile three-dimensional topology of the tracking target based on the tracking result of the tracking target, and use the mobile three-dimensional topology of the tracking target to analyze the threat level of the tracking target to the building; continuously analyze the threat level of the tracking target to the building, set the risk judgment threshold, and The threat level of the tracked target to the building is continuously compared with the risk determination threshold. When the threat level of the tracked target to the building meets the risk determination threshold, an alarm is triggered. After the alarm is triggered, the latest collected image information containing the tracked target is obtained, the monitoring device of the image information source is identified, and the position of the monitoring device in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices are further identified. According to the position of the monitoring device in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices, the three-dimensional topology of the interception path of the tracked target is determined; the three-dimensional topology of the interception path of the tracked target is obtained, the interception success probability of each section in the three-dimensional topology of the interception path is analyzed, and the building security resources are allocated based on the interception success probability of each section.

[0011] Furthermore, when the image information is stored in the cloud database, it is synchronously differentiated and stored based on the source monitoring device. When the user identity information is stored in the cloud database, it is synchronously differentiated and stored based on the source access control device. The user identity information sensed by the access control device includes facial information and identity card information. When the user identity information is acquired, the type of user identity information is synchronously identified. When the identified user identity information is identity card information, the user identity information is stored. When the identified user identity information is facial information, the user identity information is treated as a non-storage target and a deletion operation is performed.

[0012] Furthermore, the access control device stores the user's facial information and identity card information. The access control device releases the user when any of the information is verified. For the user released by the identity card information verification, the access control device simultaneously retrieves the user's facial information corresponding to the verified identity card information;

[0013] The user identity information stored in the cloud database performs a user face information retrieval operation, and all retrieved user face information is used as a designated tracking target;

[0014] During the three-dimensional topology construction stage of the internal paths of the building, the location information of the starting points and end points of each segment of the path inside the building is uploaded synchronously, and the starting points and end points of each segment of the path are adjacently connected to each other based on the location information to construct the three-dimensional topology of the internal paths of the building. The installation locations of the monitoring equipment and access control equipment marked on the three-dimensional topology of the internal paths of the building are manually marked on the topology by the user end.

[0015] Furthermore, the tracking operation logic of the tracking target is expressed as:

[0016] Step 1: Identify the access control device from which the tracking target is sourced and the monitoring device closest to the access control device, obtain the image information of the monitoring device after the access control device verifies the identity of the tracking target, set a retrieval time threshold, and retrieve the earliest image information within a time threshold equal to the retrieval time threshold from the acquired image information based on the retrieval time threshold, and record it as sub-image information;

[0017] Step 2: Obtain the dynamic target face image in the sub-image information, and determine whether there is user face information corresponding to the identity card information used by the tracking target through the access control device in all the acquired dynamic target face images;

[0018] Step 3: If the result is yes, the ID card information corresponding to the tracking target is deleted from the cloud database. If the result is no, the dynamic target face image obtained in the sub-image information is used as the screening target group, and the user face information that exists in the screening target group but does not exist in the access control device is identified as the final tracking target;

[0019] Step 4: Use the facial image of the final tracking target to query the image information collected by all monitoring devices, record the source monitoring device of the image information of the final tracking target, and the number of times each monitoring device monitors the final tracking target.

[0020] Furthermore, the retrieval time threshold set in Step 1 is subject to:

[0021] ;

[0022] Where: To retrieve the time threshold; is the time threshold base; is a constant; It is the total number of dynamic targets in the first frame of the image information after the access control device verifies the identity of the tracking target; For the The moving speed of a dynamic target;

[0023] in, Express The average, constant >0, used for constraints Always in the range (0, 2).

[0024] Furthermore, the tracking result of the tracking target is: recording the source monitoring device that queried the image information of the final tracking target, and the number of times each monitoring device monitored the final tracking target;

[0025] In the stage of constructing the moving three-dimensional topology of the tracking target, the corresponding monitoring device marked positions are picked up in the three-dimensional topology of the internal path of the building based on the recorded monitoring devices and the number of times each monitoring device monitors the final tracking target, and the marked positions are connected based on the time sequence and the number of times each monitoring device monitors the final tracking target to construct the moving three-dimensional topology of the tracking target;

[0026] The analysis logic of the threat level of the tracking target to the building is expressed as:

[0027] ;

[0028] Where: To track the threat level of the target to the building; The total amount of road segments contained in the mobile three-dimensional topology corresponding to the tracking target; is the number of times the target repeatedly passes through the i-th road segment in the target moving three-dimensional topology; For the The total number of rooms on the road segment; For the The security level of each room; To track the target, The average speed of the road section;

[0029] Among them, the threat level of the tracking target to the building The larger the value, the greater the threat level of the tracking target to the building. Conversely, the smaller the threat level of the tracking target to the building. The threat level of each tracking target to the building is calculated based on the above formula. The security level of the rooms distributed on each section in the three-dimensional topology of the internal path of the building is customized by the system end user. The security level is 1 to 4, and level 4 is the highest level of security level.

[0030] Furthermore, the continuous analysis frequency of the threat level of the tracking target to the building is customized by the user, the risk judgment threshold is customized by the user, and a speaker is provided in the building management background monitoring room. The speaker stores the designated audio as the alarm audio, and when the threat level of the tracking target to the building meets the risk judgment threshold, the speaker is triggered to continuously play the alarm audio.

[0031] Furthermore, in the stage of determining the three-dimensional topology of the tracking target interception path after the alarm is triggered, after the position of the monitoring device corresponding to the tracking target in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices are identified, the section where the monitoring device is located in the three-dimensional topology of the internal path of the building and its adjacent sections are obtained, the section where the adjacent monitoring devices of the monitoring device are located is obtained, and the set of all obtained sections is used as the three-dimensional topology of the tracking target interception path.

[0032] Furthermore, the building security resources are security personnel currently patrolling in the building, and the interception success probability analysis logic of each road section is expressed as:

[0033] ;

[0034] Where: To track the interception success probability of the first section in the three-dimensional topology of the target interception path; The number of times the tracked target walks the first road segment in the corresponding mobile 3D topology of the tracked target; The sum of the number of times the tracked target walks on each road section in the mobile 3D topology corresponding to the tracked target; The shortest path distance between the first segment in the three-dimensional topology of the intercepted path for the tracking target and the segment with the most number of user trips in the three-dimensional topology corresponding to the tracking target; The average moving speed of the tracking target in the moving three-dimensional topology corresponding to the tracking target; The average moving speed of the target in the first section of the interception path three-dimensional topology;

[0035] in, represents the constraint function, , , , .

[0036] Furthermore, based on the interception success probability analysis logic, the interception success probability of each road section in the three-dimensional topology of the target interception path is tracked, and each road section is arranged in descending order based on the calculated interception success probability. Based on the arrangement result, the allocation of building security resources is performed, so that the road section arranged in the front is configured with more building security resources, and the road section arranged in the back is configured with fewer building security resources;

[0037] Among them, the building security resources configured on the road section shall not be less than 1. During the stage of configuring the building security resources, the user end shall independently decide on the road section target that needs to be configured with building security resources based on the descending order results of the road sections.

[0038] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0039] The present invention provides an abnormal personnel identification and alarm method based on a smart building. During the execution of the method, the monitoring equipment deployed in the building is linked with the access control equipment to continuously monitor the personnel who enter the smart building without conventional identity verification, and analyzes the security threat to the smart building based on the movement path monitoring of the personnel entering the smart building. Further, based on the security threat analysis result, i.e. the movement path monitoring, trigger conditions and logic for the scheduling of security resources for the smart building are provided to ensure that the daily operation and use of the smart building are safer, and to minimize the security threat to the smart building caused by personnel who enter the building through unconventional methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 A flowchart of an abnormal personnel identification and alarm method based on a smart building;

[0042] Figure 2 This is a schematic diagram showing an example of the three-dimensional topology of the internal paths of a building in the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The present invention will be further described below in conjunction with the embodiments.

[0045] Embodiment 1:

[0046] In this embodiment, a method for identifying and warning abnormal personnel in a smart building is provided. Figure 1 As shown, including:

[0047] Obtain the image information collected by the monitoring equipment in the building and the user identity information perceived by the access control equipment in real time, set up a cloud database, and use the cloud database to distinguish and store the image information and user identity information;

[0048] The access control device stores the user's facial information and identity card information. The access control device releases the user when any of the information is verified. For users released through identity card information verification, the access control device simultaneously retrieves the user's facial information corresponding to the verified identity card information;

[0049] The user identity information stored in the cloud database performs the operation of retrieving the user face information, and all retrieved user face information is used as the designated tracking target;

[0050] In the phase of constructing the three-dimensional topology of the internal path of the building, the location information of the starting point and the end point of each segment of the path in the building is uploaded synchronously, and the starting point and the end point of each segment of the path are adjacently connected to each other based on the location information to construct the three-dimensional topology of the internal path of the building. The installation positions of the monitoring equipment and access control equipment marked on the three-dimensional topology of the internal path of the building are manually marked on the topology by the user end;

[0051] The tracking operation logic of the tracking target is expressed as:

[0052] Step 1: Identify the access control device from which the tracking target is sourced and the monitoring device closest to the access control device, obtain the image information of the monitoring device after the access control device verifies the identity of the tracking target, set a retrieval time threshold, and retrieve the earliest image information within a time threshold equal to the retrieval time threshold from the acquired image information based on the retrieval time threshold, and record it as sub-image information;

[0053] Step 2: Obtain the dynamic target face image in the sub-image information, and determine whether there is user face information corresponding to the identity card information used by the tracking target through the access control device in all the acquired dynamic target face images;

[0054] Step 3: If the result is yes, the ID card information corresponding to the tracking target is deleted from the cloud database. If the result is no, the dynamic target face image obtained in the sub-image information is used as the screening target group, and the user face information that exists in the screening target group but does not exist in the access control device is identified as the final tracking target;

[0055] Step 4: Use the facial image of the final tracking target to query the image information collected by all monitoring devices, record the source monitoring device of the image information of the final tracking target, and the number of times each monitoring device monitors the final tracking target;

[0056] The retrieval time threshold set in Step 1 is subject to:

[0057] ;

[0058] Where: To retrieve the time threshold; is the time threshold base; is a constant; It is the total number of dynamic targets in the first frame of the image information after the access control device verifies the identity of the tracking target; For the The moving speed of a dynamic target;

[0059] in, Express The average, constant >0, used for constraints Always in the range (0, 2);

[0060] Through the above settings, the tracking operation logic of the tracking target is limited, and a setting logic for calling the time threshold is given to provide execution logic support for Step 1 of the tracking operation logic of the tracking target.

[0061] Perform tracking operations with designated tracking targets in the image information according to the user identity information. In the tracking target tracking operation phase, a three-dimensional topology of the internal path of the building is constructed based on the internal path of the building, and the installation locations of the monitoring equipment and the access control equipment are marked on the three-dimensional topology of the internal path of the building;

[0062] The tracking result of the tracking target is: recording the source monitoring device that queried the image information of the final tracking target, and the number of times each monitoring device monitored the final tracking target;

[0063] In the stage of constructing the three-dimensional topology of the moving target, the corresponding monitoring device marked positions are picked up in the three-dimensional topology of the internal path of the building based on the recorded monitoring devices and the number of times each monitoring device monitors the final tracking target. The marked positions are connected based on the time sequence and the number of times each monitoring device monitors the final tracking target to construct the three-dimensional topology of the moving target;

[0064] The analysis logic of the threat level of the tracking target to the building is expressed as follows:

[0065] ;

[0066] Where: To track the threat level of the target to the building; The total amount of road segments contained in the mobile three-dimensional topology corresponding to the tracking target; is the number of times the target repeatedly passes through the i-th road segment in the target moving three-dimensional topology; For the The total number of rooms on the road segment; For the The security level of each room; To track the target, The average speed of the road section;

[0067] Among them, the threat level of the tracking target to the building The larger the value, the greater the threat level of the tracking target to the building. Conversely, the smaller the threat level of the tracking target to the building. The threat level of each tracking target to the building is calculated based on the above formula. The security level of the rooms distributed on each section in the three-dimensional topology of the internal path of the building is customized by the system end user. The security level is 1 to 4, and level 4 is the highest level of security level.

[0068] Through the above logical formula, the threat level of the tracking target to the building is analyzed and calculated in a digital form, further providing the necessary parameter conditions to support the determination of the three-dimensional topology of the interception path of the tracking target.

[0069] According to the tracking results of the tracking target, a moving three-dimensional topology of the tracking target is constructed, and the threat level of the tracking target to the building is analyzed by using the moving three-dimensional topology of the tracking target;

[0070] Continuously analyze the threat level of the tracking target to the building, set the risk determination threshold, and continuously compare the threat level of the tracking target to the building based on the continuous analysis with the risk determination threshold. When the threat level of the tracking target to the building meets the risk determination threshold, trigger an alarm;

[0071] After the alarm is triggered, the latest collected image information containing the tracking target is obtained, the monitoring device of the image information source is identified, and the position of the monitoring device in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices are further identified. According to the position of the monitoring device in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices, the three-dimensional topology of the interception path of the tracking target is determined;

[0072] Obtain the three-dimensional topology of the interception path of the tracking target, analyze the interception success probability of each section in the three-dimensional topology of the interception path, and allocate building security resources based on the interception success probability of each section;

[0073] Building security resources refer to the security personnel currently patrolling the building. The logic of interception success probability analysis of each section is expressed as:

[0074] ;

[0075] Where: To track the interception success probability of the first section in the three-dimensional topology of the target interception path; The number of times the tracked target walks the first road segment in the corresponding mobile 3D topology of the tracked target; The sum of the number of times the tracked target walks on each road section in the mobile 3D topology corresponding to the tracked target; The shortest path distance between the first segment in the three-dimensional topology of the intercepted path for the tracking target and the segment with the most number of user trips in the three-dimensional topology corresponding to the tracking target; The average moving speed of the tracking target in the moving three-dimensional topology corresponding to the tracking target; The average moving speed of the target in the first section of the interception path three-dimensional topology;

[0076] in, represents the constraint function, , , , ;

[0077] Through the above logic formula, the interception success probability of each section in the three-dimensional topology of the tracking target interception path is calculated, so as to reasonably schedule the configuration of building security resources for the method in the above embodiment.

[0078] Based on the interception success probability analysis logic, the interception success probability of each road section in the three-dimensional topology of the target interception path is tracked, and each road section is arranged in descending order based on the calculated interception success probability. Based on the arrangement result, the allocation of building security resources is performed, so that the road sections arranged in the front are configured with more building security resources, and the road sections arranged in the back are configured with fewer building security resources;

[0079] Among them, the building security resources configured on the road section shall not be less than 1. During the stage of configuring the building security resources, the user end shall independently decide on the road section target that needs to be configured with building security resources based on the descending order results of the road sections.

[0080] In this embodiment, by executing the method in the above embodiment, risk identification and alarm management services at the personnel management level are provided for smart buildings, which effectively ensures the safety of daily operation of smart buildings and reduces the safety risks caused by human factors to the daily operation of smart buildings;

[0081] See also Figure 2 As shown in the figure, the construction result of the three-dimensional topology of the internal path of the building is illustrated, which makes it easier for users who implement the solution to understand the specific structure of the three-dimensional topology of the internal path of the building applied by the method in the above embodiment.

[0082] Embodiment 2:

[0083] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The following is a further detailed description of an abnormal personnel identification and alarm method based on a smart building in Example 1:

[0084] When image information is stored in a cloud database, it is differentiated and stored synchronously based on the source monitoring device. When user identity information is stored in a cloud database, it is differentiated and stored synchronously based on the source access control device. The user identity information perceived by the access control device includes facial information and identity card information. When the user identity information is obtained, the type of user identity information is identified. When the identified user identity information is identity card information, the user identity information is stored. When the identified user identity information is facial information, the user identity information is treated as a non-storage target and a deletion operation is performed.

[0085] Through the above settings, further execution logic support is provided for the execution of the method in Example 1, ensuring that the method in Example 1 continuously and stably performs safety management of the flow of people in smart buildings.

[0086] like Figure 1 As shown, the continuous analysis frequency of the threat level of the tracking target to the building is customized by the user, the risk judgment threshold is customized by the user, and a speaker is set in the monitoring room of the building management background. The speaker stores the specified audio as the alarm audio. When the threat level of the tracking target to the building meets the risk judgment threshold, the speaker is triggered to run and continuously play the alarm audio.

[0087] Through the above settings, the alarm mode and alarm logic of the method in the above embodiment 1 for smart buildings are further limited.

[0088] like Figure 1 As shown, in the stage of determining the three-dimensional topology of the interception path of the tracking target after the alarm is triggered, after the position of the monitoring device corresponding to the tracking target in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices are identified, the section where the monitoring device is located in the three-dimensional topology of the internal path of the building and its adjacent sections are obtained, the section where the adjacent monitoring devices of the monitoring device are located is obtained, and the set of all obtained sections is used as the three-dimensional topology of the interception path of the tracking target.

[0089] Through the above settings, the construction logic of the three-dimensional topology of the tracking target interception path is further limited.

[0090] In summary, during the execution of the method in the above embodiment, the monitoring equipment deployed in the building is linked with the access control equipment to continuously monitor the people who enter the smart building without conventional identity verification, and analyze the security threats to the smart building based on the movement paths of the people entering the smart building. Further, based on the security threat analysis results, i.e., the monitoring of the movement paths, trigger conditions and logic for the scheduling of security resources for smart buildings are provided to ensure that the daily operation and use of smart buildings are safer, and to minimize the security threats to smart buildings caused by people who enter the building through unconventional methods.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying and warning abnormal personnel in a smart building, characterized in that: include: Obtain the image information collected by the monitoring equipment in the building and the user identity information perceived by the access control equipment in real time, set up a cloud database, and use the cloud database to distinguish and store the image information and user identity information; The access control device stores the user's facial information and identity card information. The access control device releases the user when any of the information is verified. For users released through identity card information verification, the access control device simultaneously retrieves the user's facial information corresponding to the verified identity card information; The user identity information stored in the cloud database performs a user face information retrieval operation, and all retrieved user face information is used as a designated tracking target; In the stage of constructing the three-dimensional topology of the internal path of the building, the location information of the starting point and the end point of each segment of the path in the building is uploaded synchronously, and the starting point and the end point of each segment of the path are adjacently connected to each other based on the location information of the starting point and the end point of each segment of the path to construct the three-dimensional topology of the internal path of the building. The installation positions of the monitoring equipment and the access control equipment marked on the three-dimensional topology of the internal path of the building are manually marked on the topology by the user end; The tracking operation logic of the tracking target is expressed as: Step 1: Identify the access control device from which the tracking target is sourced and the monitoring device closest to the access control device, obtain the image information of the monitoring device after the access control device verifies the identity of the tracking target, set a retrieval time threshold, and retrieve the earliest image information within a time threshold equal to the retrieval time threshold from the acquired image information based on the retrieval time threshold, and record it as sub-image information; Step 2: Obtain the dynamic target face image in the sub-image information, and determine whether there is user face information corresponding to the identity card information used by the tracking target through the access control device in all the acquired dynamic target face images; Step 3: If the result is yes, the ID card information corresponding to the tracking target is deleted from the cloud database. If the result is no, the dynamic target face image obtained in the sub-image information is used as the screening target group, and the user face information that exists in the screening target group but does not exist in the access control device is identified as the final tracking target; Step 4: Use the facial image of the final tracking target to query the image information collected by all monitoring devices, record the source monitoring device of the image information of the final tracking target, and the number of times each monitoring device monitors the final tracking target; The retrieval time threshold set in Step 1 is subject to: ； Where: To retrieve the time threshold; is the time threshold base; is a constant; It is the total number of dynamic targets in the first frame of the image information after the access control device verifies the identity of the tracking target; For the The moving speed of a dynamic target; in, Express The average, constant >0, used for constraints Always in the range (0, 2); Perform tracking operations with designated tracking targets in the image information according to the user identity information. In the tracking target tracking operation phase, a three-dimensional topology of the internal path of the building is constructed based on the internal path of the building, and the installation locations of the monitoring equipment and the access control equipment are marked on the three-dimensional topology of the internal path of the building; According to the tracking results of the tracking target, a three-dimensional mobile topology of the tracking target is constructed, and the threat level of the tracking target to the building is analyzed by using the three-dimensional mobile topology of the tracking target; Continuously analyze the threat level of the tracking target to the building, set the risk determination threshold, and continuously compare the threat level of the tracking target to the building based on the continuous analysis with the risk determination threshold. When the threat level of the tracking target to the building meets the risk determination threshold, trigger an alarm; After the alarm is triggered, the latest collected image information containing the tracking target is obtained, the monitoring device of the image information source is identified, and the position of the monitoring device in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices are further identified. According to the position of the monitoring device in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices, the three-dimensional topology of the interception path of the tracking target is determined; Obtain the three-dimensional topology of the interception path of the tracking target, analyze the interception success probability of each section in the three-dimensional topology of the interception path, and allocate building security resources based on the interception success probability of each section; The tracking result of the tracking target is: recording the source monitoring device that queried the image information of the final tracking target, and the number of times each monitoring device monitored the final tracking target; In the stage of constructing the moving three-dimensional topology of the tracking target, the corresponding monitoring device marked positions are picked up in the three-dimensional topology of the internal path of the building based on the recorded monitoring devices and the number of times each monitoring device monitors the final tracking target, and the marked positions are connected based on the time sequence and the number of times each monitoring device monitors the final tracking target to construct the moving three-dimensional topology of the tracking target; The analysis logic of the threat level of the tracking target to the building is expressed as: ； Where: To track the threat level of the target to the building; The total amount of road segments contained in the mobile three-dimensional topology corresponding to the tracking target; is the number of times the target repeatedly passes through the i-th road segment in the target moving three-dimensional topology; For the The total number of rooms on the road segment; For the The security level of each room; To track the target, The average speed of the road section; Among them, the threat level of the tracking target to the building The larger the value, the greater the threat level of the tracking target to the building. Conversely, the smaller the threat level of the tracking target to the building. The threat level of each tracking target to the building is calculated based on the above formula. The security level of the rooms distributed on each section in the three-dimensional topology of the internal path of the building is customized by the system end user. The security level is 1 to 4, and level 4 is the highest level of security level.

2. According to claim 1, a method for identifying and warning abnormal personnel in a smart building is characterized in that: When the image information is stored in the cloud database, it is synchronously differentiated and stored based on the source monitoring device. When the user identity information is stored in the cloud database, it is synchronously differentiated and stored based on the source access control device. The user identity information sensed by the access control device includes facial information and identity card information. When the user identity information is acquired, the type of user identity information is synchronously identified. When the identified user identity information is identity card information, the user identity information is stored. When the identified user identity information is facial information, the user identity information is treated as a non-storage target and a deletion operation is performed.

3. According to claim 1, a method for identifying and warning abnormal personnel in a smart building is characterized in that: The continuous analysis frequency of the threat level of the tracking target to the building is customized by the user end, the risk judgment threshold is customized by the user end, and a speaker is provided in the building management background monitoring room, and the speaker stores the specified audio as the alarm audio. When the threat level of the tracking target to the building meets the risk judgment threshold, the speaker is triggered to run and continuously play the alarm audio.

4. The abnormal personnel identification and alarm method based on a smart building according to claim 1 is characterized in that: In the stage of determining the three-dimensional topology of the interception path of the tracking target after the alarm is triggered, after the position of the monitoring device corresponding to the tracking target in the three-dimensional topology of the internal path of the building and its adjacent monitoring devices are identified, the section where the monitoring device is located in the three-dimensional topology of the internal path of the building and its adjacent sections are obtained, and the section where the adjacent monitoring devices of the monitoring device are located is obtained, and the set of all obtained sections is used as the three-dimensional topology of the interception path of the tracking target.

5. The abnormal personnel identification and alarm method based on a smart building according to claim 1 is characterized in that: The building security resources are the security personnel currently patrolling the building. The interception success probability analysis logic of each road section is expressed as: ； Where: To track the interception success probability of the first section in the three-dimensional topology of the target interception path; The number of times the tracked target walks the first road segment in the corresponding mobile 3D topology of the tracked target; The sum of the number of times the tracked target walks on each road section in the mobile 3D topology corresponding to the tracked target; The shortest path distance between the first segment in the three-dimensional topology of the intercepted path for the tracking target and the segment with the most number of user trips in the three-dimensional topology corresponding to the tracking target; The average moving speed of the tracking target in the moving three-dimensional topology corresponding to the tracking target; The average moving speed of the target in the first section of the interception path three-dimensional topology; in, represents the constraint function, , , , .

6. The abnormal personnel identification and alarm method based on a smart building according to claim 5 is characterized in that: Based on the interception success probability analysis logic, the interception success probability of each road section in the three-dimensional topology of the target interception path is tracked, and each road section is arranged in descending order based on the calculated interception success probability. Based on the arrangement result, the allocation of building security resources is performed, so that the road sections arranged in the front are configured with more building security resources, and the road sections arranged in the back are configured with fewer building security resources; Among them, the building security resources configured on the road section shall not be less than 1. During the stage of configuring the building security resources, the user end shall independently decide on the road section target that needs to be configured with building security resources based on the descending order results of the road sections.

Images