A community security strategy optimization method and system based on big data

By analyzing event information in the community, identifying potential threats and determining security threat levels, the problem of lagging processing of community security systems is solved, the accuracy and response capabilities of abnormal events are improved, and the comprehensive assessment of community security and reasonable allocation of resources are ensured.

CN118429168BActive Publication Date: 2024-12-06XIAMEN ZHONGMIN JUHAO REAL ESTATE DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410875742.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-02
Publication Date
2024-12-06
Estimated Expiration
2044-07-02

AI Technical Summary

Technical Problem

The existing community security system cannot effectively identify and handle potential threats within the community, resulting in lagging early warning and processing, and lacks accuracy and comprehensiveness of single-dimensional analysis of security threats.

Method used

By obtaining and analyzing event information, extracting event element sets, determining whether there are abnormal events, determining security threat levels, and determining processing strategies based on the level. A community security strategy optimization method and system based on big data is adopted, combining monitoring video information and device status to monitor video information and and platform information, conduct comprehensive analysis to identify and respond to potential threats.

Benefits of technology

Improve the accuracy and rapid response ability of abnormal events, reduce waste of processing resources, ensure a comprehensive and accurate assessment of community security, eliminate potential threats in a timely manner, and avoid negative impacts caused by lag and single-dimensional analysis. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118429168B_ABST
    Figure CN118429168B_ABST
Patent Text Reader

Abstract

The present application relates to the field of community security technology, and in particular to a community security strategy optimization method and system based on big data. The method includes: obtaining event information, analyzing event information, and determining an event element set; analyzing the event element set, and judging whether there is an abnormal event based on the event analysis results; if there is an abnormal event, extracting the event element set and determining the abnormal element information; determining the security threat level based on the abnormal element information; determining the processing strategy based on the security threat level, and obtaining the processing result. The present application solves the problem of lagging processing of existing community security systems by discovering potential threats in the community and eliminating them in the bud stage, comprehensively analyzing a number of abnormal element information, and determining the security threat level of the current abnormal event, thereby avoiding the negative impact of single-dimensional analysis on the accuracy and comprehensiveness of security threat judgment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of community security technology, and in particular to a community security strategy optimization method and system based on big data. Background Art

[0002] With the rapid development of urbanization and the improvement of residents' living standards, community facilities and services are constantly upgraded, and community security systems have become an indispensable part of community management. Community security systems not only play a vital role in protecting the lives and property safety of residents, but also effectively reduce the occurrence of community crimes, improve residents' sense of security, and thus improve the living quality of the entire community.

[0003] However, the existing community security system can only identify and deal with security threats with obvious surface characteristics, and is unable to make accurate analysis and judgment of potential threats within the community. As a result, the existing community security system often has a lag in warning and handling of security threats, and the existing community security system's single-dimensional analysis of security threats lacks accuracy and comprehensiveness in judging security threats. Summary of the invention

[0004] This application proposes a community security strategy optimization method and system based on big data to solve the above technical problems.

[0005] In a first aspect, the present application provides a community security strategy optimization method based on big data, the method comprising:

[0006] Acquire event information, analyze the event information, and determine an event element set;

[0007] Analyze the event element set, and determine whether there is an abnormal event based on the event analysis result;

[0008] If an abnormal event exists, extract the event element set and determine the abnormal element information;

[0009] Determining a security threat level based on the abnormal factor information;

[0010] According to the security threat level, a processing strategy is determined and a processing result is obtained.

[0011] The solution provided by the present application analyzes the event information corresponding to the events occurring in the community, extracts the event element set reflecting the event characteristics, and then determines whether the current event is an abnormal event through the analysis of the event element set, so as to discover potential threats and eliminate potential threats in the bud, thereby solving the problem of lagging processing of the existing community security system, and extracts a number of abnormal element information in the abnormal event through the analysis of the event element set, and comprehensively analyzes a number of abnormal element information to determine the security threat level of the current abnormal event, thereby avoiding the negative impact of a single-dimensional analysis on the accuracy and comprehensiveness of the security threat judgment, so as to accurately reflect the degree of impact of abnormal events on community security, and then determine the processing strategy, adopt different processing strategies for abnormal events with different threat levels, so that the processing strategy is highly matched with the threat level of the corresponding abnormal event, reasonably allocates processing resources for abnormal events, and reduces the waste of processing resources.

[0012] Optionally, the event information includes but is not limited to surveillance video information, the event element set includes but is not limited to the subject involved, the time of occurrence and the place of occurrence, and the analyzing the event information to determine the event element set includes:

[0013] Analyze the surveillance video information, and determine the subject involved according to the video analysis result;

[0014] Extracting a video timestamp according to the monitoring video information, and determining the occurrence time according to the video timestamp;

[0015] The camera position is determined based on the monitoring video information, and the occurrence location is determined based on the camera position.

[0016] The above technical solution, by analyzing the video surveillance information in the community, determines the three core elements corresponding to the events in the community, namely the time of occurrence, the place of occurrence and the subjects involved, so as to preliminarily clarify the overall context of the event, and provide the necessary analytical basis for subsequent in-depth analysis of the event and judgment on whether there is any abnormality in the event.

[0017] Optionally, the event element set further includes specific behaviors, and the analyzing the event element set and judging whether there is an abnormal event according to the event analysis result includes:

[0018] Analyze the subjects involved and determine the specific behaviors;

[0019] Analyze the specific behavior based on the occurrence time, the occurrence location and the involved subjects, and determine whether the specific behavior is abnormal according to the specific behavior analysis result;

[0020] If the specific behavior is abnormal, it is determined that an abnormal event exists.

[0021] The above technical scheme extracts the specific behaviors of the subjects involved in the surveillance video information, comprehensively considers the time and place of occurrence of the event and the subjects involved, and analyzes the specific behaviors to determine whether there are any abnormalities in the specific behaviors of the subjects involved at the time and place of occurrence. If there are any abnormalities, it is determined that an abnormal event has occurred. By analyzing the specific behaviors at the time and place of occurrence and the subjects involved, compared with a single method of analyzing specific behaviors, the accuracy of identifying abnormal events can be significantly improved, the false alarm rate can be reduced, and the ability to respond quickly to abnormal events can be improved.

[0022] Optionally, the event element set further includes a device state, and the analyzing the event element set and determining whether an abnormal event occurs according to the event analysis result includes:

[0023] Determine the device information of the occurrence location according to the occurrence location, and determine the device status according to the device information;

[0024] Analyze the device status, and determine whether the device is abnormal according to the device status analysis result;

[0025] If there is an abnormality, it is determined that an abnormal event exists.

[0026] Through the above technical solution, the device information is analyzed to determine the device status. If the device is abnormal, it can directly reflect the abnormal event at the location where it occurred. By utilizing the strong real-time monitoring characteristics of the device, abnormal events with a high degree of security threat can be discovered in time, thus gaining valuable time for abnormal event processing.

[0027] Optionally, the event element further includes platform information, and the analyzing the event element set and judging whether there is an abnormal event according to the event analysis result includes:

[0028] Obtain the platform information and extract several information tags;

[0029] Determining a timing analysis plan according to the platform information and the plurality of information tags;

[0030] Analyze the monitoring video information according to the timing analysis plan, and determine the information matching value between the monitoring video information and the platform information according to the timing analysis result;

[0031] The information matching value is compared with a preset information matching threshold, and if the information matching value is greater than the preset information matching threshold, it is determined that an abnormal event exists.

[0032] The above technical solution analyzes the platform information related to community safety released by community residents, extracts the label information corresponding to the platform information, and determines a timing analysis plan based on the platform information and the label information. Then, according to the timing analysis plan, the surveillance video information corresponding to the location and time mentioned in the platform information is analyzed to determine whether the surveillance video information is consistent with the description in the platform information, and then determine whether an abnormal event has occurred. This effectively avoids the continued existence and fermentation of certain safety hazards due to problems such as negligence of security personnel on patrol, and at the same time prevents some false platform information from wasting resources for handling abnormal events.

[0033] Optionally, extracting the event element set and extracting abnormal element information includes:

[0034] Analyzing the event element set, screening the event elements in the event element set, and determining a number of abnormal elements;

[0035] Analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements;

[0036] The plurality of abnormal elements and the abnormal correlation information are determined as the abnormal element information.

[0037] The above technical scheme screens the event elements in the event element set corresponding to the abnormal event, extracts several abnormal elements therein, and then obtains the abnormal correlation information between the several abnormal elements by analyzing the several abnormal elements. After integrating the several abnormal elements and their corresponding abnormal correlation information, the abnormal element information is determined to improve the accuracy and comprehensiveness of the security threat level subsequently determined based on the abnormal element information, and prevent the security threat level subsequently obtained from being one-sided.

[0038] Optionally, analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements includes:

[0039] The abnormal elements are matched with the preset correlation data set, and the following abnormal correlation matrix is ​​established according to the matching results:

[0040] ;

[0041] in, is the abnormal correlation matrix, For the The abnormal element and The correlation between the abnormal elements;

[0042] The abnormal correlation matrix is ​​used as the abnormal correlation information.

[0043] The above technical scheme establishes an abnormal correlation matrix according to the matching results between a number of abnormal elements and a preset correlation data set, and uses the abnormal correlation matrix as abnormal correlation information to systematically express the correlation relationship between the abnormal elements, which helps to conduct more detailed, in-depth and efficient mathematical analysis and calculation of the corresponding safety level of abnormal events based on the abnormal correlation information.

[0044] Optionally, determining the security threat level according to the abnormal factor information includes:

[0045] Matching the several abnormal elements with a preset risk value data set, and determining the risk value corresponding to each abnormal element according to the matching result;

[0046] According to the abnormal correlation matrix, the risk value corresponding to each abnormal element is adjusted to determine the adjusted risk value of each abnormal element;

[0047] The adjusted risk value of each abnormal factor is summed up to determine the total risk value;

[0048] The security threat level is determined according to the total risk value.

[0049] The above technical scheme determines the risk value corresponding to each abnormal element according to the matching results between a number of abnormal elements and a preset risk value data set, and then determines the adjusted risk value of each abnormal element according to the abnormal correlation matrix, and determines the sum of the adjusted risk values ​​of all abnormal elements as the total risk value, so that the total risk value accurately reflects the impact of abnormal events on community safety, thereby improving the accuracy of the security threat level determined according to the total risk value.

[0050] Optionally, adjusting the risk value corresponding to each abnormal element according to the abnormal correlation matrix to determine the adjusted risk value of each abnormal element includes:

[0051] ;

[0052] in, For the The adjusted risk value of the abnormal factor is For the The risk value of an abnormal factor, is the correlation adjustment coefficient, For the The abnormal element and The correlation of abnormal elements.

[0053] Through the above technical solution, a calculation formula is designed to accurately calculate the adjusted risk value of each abnormal factor to reflect the impact of the correlation between abnormal factors on the risk value, thereby improving the accuracy and comprehensiveness of the adjusted risk value.

[0054] In a second aspect, the present application provides a community security strategy optimization system based on big data, the system comprising:

[0055] An element analysis module, used to obtain event information, analyze the event information, and determine an event element set;

[0056] An abnormality judgment module is used to analyze the event element set and judge whether there is an abnormal event according to the event analysis result;

[0057] An element extraction module is used to extract the event element set and determine the abnormal element information if an abnormal event exists;

[0058] A level classification module, used to determine the security threat level according to the abnormal element information;

[0059] The processing module is used to determine a processing strategy and obtain a processing result according to the security threat level.

[0060] Optionally, the element analysis module is specifically used to:

[0061] Analyze the surveillance video information, and determine the subject involved according to the video analysis result;

[0062] Extracting a video timestamp according to the monitoring video information, and determining the occurrence time according to the video timestamp;

[0063] The camera position is determined based on the monitoring video information, and the occurrence location is determined based on the camera position.

[0064] Optionally, the abnormality judgment module is specifically used to:

[0065] Analyze the subjects involved and determine the specific behaviors;

[0066] Analyze the specific behavior based on the occurrence time, the occurrence location and the involved subjects, and determine whether the specific behavior is abnormal according to the specific behavior analysis result;

[0067] If the specific behavior is abnormal, it is determined that an abnormal event exists.

[0068] Optionally, the abnormality judgment module is specifically used to:

[0069] Determine the device information of the occurrence location according to the occurrence location, and determine the device status according to the device information;

[0070] Analyze the device status, and determine whether the device is abnormal according to the device status analysis result;

[0071] If there is an abnormality, it is determined that an abnormal event exists.

[0072] Optionally, the abnormality judgment module is specifically used to:

[0073] Obtain the platform information and extract several information tags;

[0074] Determining a timing analysis plan according to the platform information and the plurality of information tags;

[0075] Analyze the monitoring video information according to the timing analysis plan, and determine the information matching value between the monitoring video information and the platform information according to the timing analysis result;

[0076] The information matching value is compared with a preset information matching threshold, and if the information matching value is greater than the preset information matching threshold, it is determined that an abnormal event exists.

[0077] Optionally, the element extraction module is specifically used to:

[0078] Analyzing the event element set, screening the event elements in the event element set, and determining a number of abnormal elements;

[0079] Analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements;

[0080] The plurality of abnormal elements and the abnormal correlation information are determined as the abnormal element information.

[0081] Optionally, the element extraction module is specifically used to:

[0082] The abnormal elements are matched with the preset correlation data set, and the following abnormal correlation matrix is ​​established according to the matching results:

[0083] ;

[0084] in, is the abnormal correlation matrix, For the The abnormal element and The correlation between the abnormal elements;

[0085] The abnormal correlation matrix is ​​used as the abnormal correlation information.

[0086] Optionally, the level classification module is specifically used to:

[0087] Matching the several abnormal elements with a preset risk value data set, and determining the risk value corresponding to each abnormal element according to the matching result;

[0088] According to the abnormal correlation matrix, the risk value corresponding to each abnormal element is adjusted to determine the adjusted risk value of each abnormal element;

[0089] The adjusted risk value of each abnormal factor is summed up to determine the total risk value;

[0090] The security threat level is determined according to the total risk value.

[0091] Optionally, the level classification module is specifically used to:

[0092] ;

[0093] in, For the The adjusted risk value of the abnormal factor is For the The risk value of an abnormal factor, is the correlation adjustment coefficient, For the The abnormal element and The correlation of abnormal elements. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0095] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0096] Figure 2 A flowchart of a community security strategy optimization method based on big data provided in one embodiment of the present application;

[0097] Figure 3 A schematic diagram of the structure of a community security strategy optimization system based on big data provided in one embodiment of the present application. DETAILED DESCRIPTION

[0098] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0099] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0100] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0101] The existing community security system can only identify and handle security threats with obvious surface characteristics, but cannot make accurate analysis and judgment of potential threats within the community. As a result, the existing community security system often has a lag in warning and handling of security threats, and the existing community security system's single-dimensional analysis of security threats lacks accuracy and comprehensiveness in judging security threats.

[0102] Based on this, the present application provides a community security strategy optimization method and system based on big data, which analyzes the event information corresponding to the events occurring in the community, extracts the event element set reflecting the event characteristics, and then determines whether the current event is an abnormal event through the analysis of the event element set, so as to discover potential threats and eliminate potential threats in the bud, solve the problem of lagging processing of the existing community security system, and extracts some abnormal element information in the abnormal event by analyzing the event element set, and comprehensively analyzes some abnormal element information to determine the security threat level of the current abnormal event, so as to avoid the negative impact of single-dimensional analysis on the accuracy and comprehensiveness of security threat judgment, so as to accurately reflect the degree of impact of abnormal events on community security, and then determine the processing strategy, adopt different processing strategies for abnormal events with different threat levels, make the processing strategy highly matched with the threat level of the corresponding abnormal event, reasonably allocate processing resources for abnormal events, and reduce the waste of processing resources.

[0103] Figure 1 A schematic diagram of an application scenario provided for this application. During the process of community security work, the method of this application is applied to analyze event information, determine a processing strategy that meets the security threat level of the event, and obtain a processing result.

[0104] The method of the present application is applied to any server that communicates with the existing monitoring system of the community. Specifically, event information is obtained through the server, and the event information corresponding to the events occurring in the community is analyzed, and the event element set reflecting the event characteristics is extracted to determine whether the current event is an abnormal event, so as to discover potential threats and eliminate potential threats in the bud, thereby solving the problem of lagging processing of the existing community security system. By analyzing the event element set and comprehensively analyzing a number of abnormal element information, the security threat level of the current abnormal event is determined, and the negative impact of a single-dimensional analysis on the accuracy and comprehensiveness of the security threat judgment is avoided, so as to accurately reflect the degree of impact of abnormal events on community security, and then determine the processing strategy, adopt different processing strategies for abnormal events with different threat levels, so that the processing strategy is highly matched with the threat level of the corresponding abnormal event, and the processing resources for abnormal events are reasonably allocated to reduce the waste of processing resources.

[0105] For specific implementation methods, please refer to the following embodiments.

[0106] Figure 2 This is a flowchart of a community security strategy optimization method based on big data provided by an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes:

[0107] S201: Acquire event information, analyze the event information, and determine an event element set.

[0108] Event information can be considered as relevant information that can reflect events occurring in the community, including but not limited to video surveillance information.

[0109] The event element set can be considered as a series of key information reflecting the characteristics of community events, including but not limited to the time and place of occurrence.

[0110] In a specific implementation method, a series of monitoring equipment is usually installed in the community to monitor the actual conditions of various areas in the community in real time. When an incident occurs in the community or there is a potential threat, it can be reflected in the monitoring video at the first time. Therefore, the monitoring video is analyzed in real time through video analysis technology, and event elements reflecting the characteristics of the event are extracted from it, which can efficiently characterize the event, thereby providing an important data basis for subsequently determining the threat level of the event based on the event elements.

[0111] In some other embodiments, the event information may be conversation information provided between community patrol personnel through intercom equipment. The conversation information of the community patrol personnel is analyzed by natural language analysis tools, and event elements that can reflect the characteristics of the event are extracted to determine the event element set. The events in the blind spots of video surveillance can be analyzed through the conversation information of the patrol personnel.

[0112] S202: Analyze the event element set and determine whether there is an abnormal event based on the event analysis result.

[0113] Abnormal events can be considered as events that occur in the community that are abnormal compared to normal circumstances.

[0114] In a specific implementation method, when the current event is an abnormal event, the event elements that reflect the characteristics of the event are concentrated, and some event elements will produce obvious abnormalities. The determination of abnormal events can improve the efficiency of event analysis. Abnormal events do not necessarily mean that there is a danger, but rather that the current event has a trend of developing into a dangerous event. The judgment of abnormal events can serve as a starting point for potential threat analysis.

[0115] Therefore, in the process of judging whether there is an abnormal event, it is necessary to comprehensively analyze each event element in the event element set, compare each event element with the normal situation, determine the event elements that are abnormal, and judge whether there is an abnormal event.

[0116] S203: If an abnormal event exists, extract the event element set and determine the abnormal element information.

[0117] Abnormal element information can be considered as all abnormal event elements in the event element set.

[0118] In a specific implementation, an event element set is usually composed of several independent event elements and several interrelated event elements. Independent event elements can judge their abnormal conditions more accurately, while several interrelated event elements often require the integration of all related event elements to reach an accurate abnormal judgment. Therefore, after determining the existence of an abnormal event, a comprehensive analysis of the event element set can accurately extract the abnormal element information in the event element set.

[0119] S204: Determine the security threat level based on the abnormal factor information.

[0120] The security threat level may be information obtained after analyzing abnormal element information to express the degree of impact of the corresponding abnormal event on community security, including but not limited to low threat, medium threat and high threat.

[0121] In a specific implementation method, after determining the abnormal event and the abnormal element information reflecting the abnormal characteristics in the abnormal event, the security threat level caused by the abnormal event in the community is judged based on the abnormal element information, which can intuitively reflect the impact of the current abnormal event on community security, so that the subsequent handling of the abnormal event matches the impact caused by the abnormal event.

[0122] S205: Determine a processing strategy based on the security threat level and obtain a processing result.

[0123] The processing strategy can be considered as the corresponding processing method that matches the security threat level of the current abnormal event.

[0124] In a specific implementation method, according to the security threat level corresponding to the abnormal event, a corresponding processing method, that is, a processing strategy, that matches the security threat level of the current abnormal event is determined, so that the processing strategy is highly matched with the threat level of the corresponding abnormal event, and processing resources are reasonably allocated, so that the processing strategy can focus on processing high-threat abnormal events while avoiding wasting too many processing resources in low-threat abnormal events.

[0125] Through the scheme provided by this embodiment, the event information corresponding to the events occurring in the community is analyzed, and the event element set reflecting the event characteristics is extracted. Then, through the analysis of the event element set, it is determined whether the current event is an abnormal event, so as to discover potential threats and eliminate potential threats in the bud stage, thereby solving the problem of lagging processing of the existing community security system. By analyzing the event element set, several abnormal element information in the abnormal event is extracted, and the several abnormal element information is comprehensively analyzed to determine the security threat level of the current abnormal event, so as to avoid the negative impact of a single-dimensional analysis on the accuracy and comprehensiveness of the security threat judgment, so as to accurately reflect the degree of impact of abnormal events on community security, and then determine the processing strategy, adopt different processing strategies for abnormal events with different threat levels, so that the processing strategy is highly matched with the threat level of the corresponding abnormal event, reasonably allocate processing resources for abnormal events, and reduce the waste of processing resources.

[0126] In some embodiments, the monitoring video information is analyzed, and the subject involved is determined based on the video analysis results; the video timestamp is extracted based on the monitoring video information, and the time of occurrence is determined based on the video timestamp; the camera position is determined based on the monitoring video information, and the location of occurrence is determined based on the camera position.

[0127] The monitoring video information may be video information captured by cameras in various areas of the community, and the monitoring video information may be obtained by communicating with an existing monitoring system in the community.

[0128] The video timestamp may be the time information of a frame in the monitoring video information.

[0129] The occurrence time may be the occurrence time of the event in the current monitoring video information.

[0130] The location of occurrence may be a location within the community corresponding to the event in the current surveillance video information.

[0131] The involved subjects may be related subjects involved in the event in the current monitoring video information, such as persons and vehicles.

[0132] In a specific implementation method, the subject involved is identified through the target tracking algorithm in the image analysis technology, and then the monitoring video information is analyzed through video codecs, such as FFmpeg and OpenCV, to mark the timestamp of the frame in the video monitoring information, and use the timestamp when the subject involved enters the video screen as the time of occurrence of the event. By communicating with the existing monitoring system in the community, the control position of each camera in the community can be obtained, and then the specific position of the camera corresponding to the current video monitoring information can be determined, thereby determining the location of the event.

[0133] Through the solution provided in this embodiment, by analyzing the video surveillance information in the community, the three core elements corresponding to the events in the community, namely the time of occurrence, the place of occurrence and the subjects involved, are determined, so as to preliminarily clarify the overall context of the event, and provide the necessary analysis basis for subsequent in-depth analysis of the event and judgment of whether there is any abnormality in the event.

[0134] In some embodiments, the subjects involved are analyzed to determine the specific behaviors; the specific behaviors are analyzed based on the time of occurrence, the place of occurrence, and the subjects involved, and based on the results of the specific behavior analysis, it is determined whether the specific behaviors are abnormal; if the specific behaviors are abnormal, it is determined that an abnormal event has occurred.

[0135] Specific behavior can be considered as the specific actions taken by the subject in the video surveillance information.

[0136] The results of specific behavior analysis can be considered as the degree of abnormality of the specific behavior at the current time, place and subject involved after analyzing the specific behavior.

[0137] In a specific implementation, the target tracking algorithm in the video analysis technology is used to track the subject involved, and then the behavior of the subject involved is analyzed and extracted through a behavior analysis algorithm, and the extracted behavior is identified through deep learning technology, such as convolutional neural networks, to determine the specific behavior type currently performed by the subject involved.

[0138] Furthermore, since the judgment of whether some behaviors are abnormal cannot be made simply by the specific behavior type itself, it is also necessary to combine the subjects involved in the behavior, the time and place of occurrence. For example, at 10 a.m. in a community parking lot, a person walked past a vehicle, stopped to check his phone and looked around. At this time, the person may just be resting, looking for information or waiting for a friend, which generally does not constitute abnormal behavior.

[0139] However, if the same behavior of "a person walking past a vehicle, stopping to check his phone and looking around" occurs in a corner of a parking lot with few people in the community at 11 o'clock in the evening, there is a possibility of theft and it can be judged as abnormal behavior.

[0140] The above-mentioned judgment on whether a specific behavior is abnormal can be made through a large amount of community security historical data to establish a normal behavior judgment model at typical time and place in the community, such as a Gaussian distribution model. The specific behavior is analyzed through this model, and the difference between the specific behavior and normal behavior is quantified to determine whether the specific behavior is abnormal.

[0141] Through the scheme provided by this embodiment, the specific behaviors of the subjects involved in the monitoring video information are extracted, and the specific behaviors are analyzed based on the time of occurrence, the place of occurrence and the subjects involved in the event to determine whether the specific behaviors of the subjects involved are abnormal at the time and place of occurrence. If there is an abnormality, it is determined that an abnormal event has occurred. By analyzing the specific behaviors at the time of occurrence, the place of occurrence and the subjects involved, compared with a single method of analyzing specific behaviors, the recognition accuracy of abnormal events can be significantly improved, the false alarm rate can be reduced, and the ability to respond quickly to abnormal events can be improved.

[0142] In some embodiments, the device information of the occurrence location is determined based on the occurrence location, and the device status is determined based on the device information; the device status is analyzed, and based on the device status analysis results, it is determined whether the device has an abnormality; if there is an abnormality, it is determined that an abnormal event has occurred.

[0143] The device information can be considered as the relevant information of the community's supporting security equipment within a preset range around the location of the incident, such as surveillance cameras, smoke alarms, and access control devices.

[0144] The device status can be considered as the working status of the security device obtained after communicating with the relevant security equipment, including but not limited to the monitoring status and triggering status. It can also be status information expressing whether the relevant security device itself is in an available state, such as available state or unavailable state.

[0145] In a specific implementation method, the status of security equipment deployed throughout the community can directly reflect whether man-made or non-man-made abnormal events have occurred in the community. If the smoke alarm at the occurrence location is in a triggered state, it means that the occurrence location is likely to have a fire hazard. When an abnormal event occurs, the surveillance camera at the occurrence location may be blocked by humans, or the access control device may have a high-frequency error verification in a short period of time. The above situations can directly reflect the existence of abnormal events at the occurrence location, and equipment abnormalities are usually accompanied by abnormal events with a higher degree of security threat. By utilizing the strong real-time characteristics of equipment monitoring, abnormal events with a higher degree of security threat can be discovered in a timely manner, thereby gaining valuable time for abnormal event processing.

[0146] Through the solution provided in this embodiment, the device information is analyzed to determine the device status. If the device is abnormal, it can directly reflect the abnormal event at the location where it occurred. By utilizing the strong real-time monitoring characteristics of the device, abnormal events with a high degree of security threat can be discovered in time, thus gaining valuable time for abnormal event processing.

[0147] In some embodiments, platform information is obtained and a number of information tags are extracted; a timing analysis plan is determined based on the platform information and the number of information tags; according to the timing analysis plan, the monitoring video information is analyzed, and based on the timing analysis results, an information matching value between the monitoring video information and the platform information is determined; the information matching value is compared with a preset information matching threshold, and if the information matching value is greater than the preset information matching threshold, it is determined that an abnormal event exists.

[0148] Platform information can be considered as public information related to community safety published by community residents on the community’s dedicated information platform, including but not limited to text information, image information and video information.

[0149] The scheduled analysis plan can be considered as a monitoring and analysis plan for a specific time period set according to the information tag content of the platform information.

[0150] The information matching value can be considered as a quantitative value used to express the similarity between the information obtained after analyzing the actual surveillance video information and the platform information.

[0151] The preset information matching degree threshold can be considered as the threshold that the information matching degree needs to reach when judging that the actual situation in the monitoring video information meets the platform information description.

[0152] In a specific implementation method, the community can allow every community resident to participate in community security work by establishing a community security information platform. Community residents can publish public information related to community security on the community security information platform and reflect the community security problems they have discovered in their daily lives. This can effectively avoid the continued existence and fermentation of certain safety hazards due to blind spots in monitoring, negligence of security personnel on patrol, and other problems.

[0153] By obtaining the platform information posted by users on the community security information platform and extracting the label information added by users, such as "XXX unit building" and "late night", we can determine the key information in the platform information, and then use natural language analysis tools such as LTP (Language Technology Platform, Harbin Institute of Technology Language Technology Platform) and NLPIR (National Language Processing and Intelligent Retrieval, Peking University Unnamed Chinese Information Processing System) to analyze the platform information and extract the text feature vector information about security issues in the platform information.

[0154] Furthermore, in order to verify whether the platform information posted by the user is true, a timing analysis plan is determined through the platform information and label information, and the surveillance video information corresponding to the location and time mentioned in the platform information is analyzed according to the timing analysis plan. The feature information in the surveillance video information is extracted and converted into video vector feature information through machine learning models, such as convolutional neural networks. The matching degree between the video vector feature information and the above-mentioned text feature vector information is quantitatively calculated through matching algorithms, such as Euclidean distance calculation formula and Manhattan distance calculation formula, to determine the information matching degree value. The matching degree value can reflect whether the event characteristics in the video surveillance information are consistent with the description in the platform information, so as to prevent some residents from deliberately spreading rumors in the community safety platform and wasting abnormal event processing resources.

[0155] Through the solution provided in this embodiment, the platform information related to community safety released by community residents is analyzed, the label information corresponding to the platform information is extracted, and a timing analysis plan is determined based on the platform information and the label information. Then, according to the timing analysis plan, the monitoring video information corresponding to the location and time mentioned in the platform information is analyzed to determine whether the monitoring video information is consistent with the description in the platform information, and then determine whether an abnormal event has occurred, thereby effectively avoiding the continued existence and fermentation of certain safety hazards due to problems such as negligence of security personnel on patrol, and at the same time preventing some false platform information from wasting abnormal event processing resources.

[0156] In some embodiments, an event element set is analyzed, event elements in the event element set are screened, and a number of abnormal elements are determined; the number of abnormal elements are analyzed to determine abnormal correlation information between the number of abnormal elements; and the number of abnormal elements and the abnormal correlation information are determined as abnormal element information.

[0157] The abnormal elements can be considered as the corresponding event elements that have abnormalities in the event element set.

[0158] The abnormal correlation information can be considered as information used to express the correlation relationship between various abnormal elements.

[0159] The abnormal element information may be an information set including several abnormal elements and abnormal correlation information between them.

[0160] In the process of analyzing the event element set, screening the event elements in the event element set, and determining a number of abnormal elements, each event element in the event element set can be matched with the corresponding event element in the preset normal event element set. When the corresponding difference value between the current event element and the preset normal event element set exceeds a preset difference range, the current event element is judged to be an abnormal element. The preset normal event element set can be obtained by analyzing a large amount of historical data, and the preset difference range can be set according to the actual situation, and this embodiment does not make any specific limitation.

[0161] Specifically, after determining the existence of an abnormal event through the aforementioned embodiments, since the corresponding event elements in the abnormal event are usually not all in an abnormal state, in order to improve the analysis efficiency of abnormal events and improve the representativeness of subsequent analysis results, the event elements in the event element set are screened to determine several abnormal elements with abnormal situations. Furthermore, considering that there are close correlations between some abnormal elements, if several abnormal elements are viewed independently, it is easy to cause the accuracy and comprehensiveness of the security threat level subsequently determined based on the abnormal element information to be reduced. Therefore, several abnormal elements and their corresponding abnormal correlation information are determined as abnormal element information to prevent the security threat level subsequently derived from being one-sided.

[0162] Through the scheme provided by this embodiment, the event elements in the event element set corresponding to the abnormal event are screened, and several abnormal elements are extracted therein. Then, by analyzing the several abnormal elements, abnormal correlation information between the several abnormal elements is obtained. After combining the several abnormal elements and their corresponding abnormal correlation information, the abnormal element information is determined to improve the accuracy and comprehensiveness of the security threat level subsequently determined based on the abnormal element information, and prevent the security threat level subsequently obtained from being one-sided.

[0163] In some embodiments, a number of abnormal elements are matched with a preset correlation data set, and the following abnormal correlation matrix is ​​established based on the matching results:

[0164] ;

[0165] in, is the abnormal correlation matrix, For the The abnormal element and The correlation between the abnormal elements; the abnormal correlation matrix is ​​used as the abnormal correlation information.

[0166] The preset correlation data set may be a data set used to store abnormal correlation values ​​between event elements, and the preset correlation data set may be obtained by analyzing a large amount of historical data.

[0167] In a specific implementation, by matching abnormal elements with preset correlation data sets, all abnormal correlation values ​​between several abnormal elements are obtained, and then an abnormal correlation matrix is ​​established to systematically express the correlation relationship between the abnormal elements. The abnormal correlation matrix is ​​used as abnormal correlation information, which helps to perform more detailed, in-depth and efficient mathematical analysis and calculation of the corresponding safety level of abnormal events based on the abnormal correlation information.

[0168] Through the solution provided in this embodiment, an abnormal correlation matrix is ​​established according to the matching results between a number of abnormal elements and a preset correlation data set, and the abnormal correlation matrix is ​​used as abnormal correlation information to systematically express the correlation relationship between the various abnormal elements, which is helpful for subsequent more detailed, in-depth and efficient mathematical analysis and calculation of the corresponding safety level of abnormal events based on the abnormal correlation information.

[0169] In some embodiments, a number of abnormal elements are matched with a preset risk value data set, and the risk value corresponding to each abnormal element is determined based on the matching result; the risk value corresponding to each abnormal element is adjusted according to the abnormal correlation matrix to determine the adjusted risk value of each abnormal element; the adjusted risk value of each abnormal element is summed to determine the total risk value; and the security threat level is determined based on the total risk value.

[0170] The preset risk value data set can be considered as a data set used to store the risk value that will be generated for the overall risk of the event when an abnormality occurs in an event element. The preset risk value data set can be obtained by analyzing a large amount of historical data on abnormal events.

[0171] The risk value corresponding to each abnormal element can be considered as the risk value of each abnormal element in the abnormal event to community safety.

[0172] The adjusted risk value of each abnormal factor can be considered as a risk value that can more accurately reflect the impact of abnormal factors on community safety after adjusting the risk value of each abnormal factor based on the abnormal correlation between several abnormal events that are related in the abnormal event.

[0173] The total risk value may be the total risk value to community safety after the current abnormal event occurs.

[0174] In the specific implementation method, the assessment of the security threat level of abnormal events needs to consider the security risks that abnormal events pose to the community. The greater the security risk, the higher the corresponding security threat level. Therefore, it is necessary to quantify the overall community security risk caused by abnormal events. The overall community security risk caused by abnormal events is related to each abnormal element in the abnormal events. The risk value corresponding to each abnormal element is determined by matching the results between several abnormal elements and the preset risk value data set.

[0175] Under ideal circumstances, the total risk value corresponding to the overall community safety risk caused by an abnormal event should be the sum of the risk values ​​corresponding to all abnormal elements. However, in actual situations, some abnormal elements are correlated with each other. This correlation will affect the risk value of the abnormal elements, resulting in the accuracy of the risk values ​​corresponding to some abnormal elements obtained by the aforementioned method being difficult to guarantee. Therefore, according to the abnormal correlation matrix, the risk value corresponding to each abnormal element is adjusted to obtain the adjusted risk value of each abnormal element, and the sum of the adjusted risk values ​​of all abnormal elements is determined as the total risk value to reflect the community safety risk caused by the entire abnormal event.

[0176] Furthermore, the security threat level is determined based on the total risk value. The security threat level can be divided into three levels according to the value range of the total risk value, namely, low threat level, medium threat level and high threat level. The security threat level corresponding to the current abnormal event is determined based on the obtained total risk value. The security threat level can also be divided into more detailed levels based on actual conditions, which is not specifically limited in this application.

[0177] Through the solution provided by this embodiment, the risk value corresponding to each abnormal element is determined according to the matching results between a number of abnormal elements and a preset risk value data set, and then the adjusted risk value of each abnormal element is determined according to the abnormal correlation matrix, and the sum of the adjusted risk values ​​of all abnormal elements is determined as the total risk value, so that the total risk value accurately reflects the impact of abnormal events on community safety, thereby improving the accuracy of the security threat level determined according to the total risk value.

[0178] In some embodiments, the risk value corresponding to each abnormal element is adjusted according to the abnormal correlation matrix to determine the adjusted risk value of each abnormal element, referring to formula (1):

[0179] (1)

[0180] in, For the The adjusted risk value of the abnormal factor is For the The risk value of an abnormal factor, is the correlation adjustment coefficient, For the The abnormal element and The correlation of abnormal elements.

[0181] The correlation adjustment coefficient can be considered as the influence coefficient of the correlation between abnormal factors on the risk value of abnormal factors. The correlation adjustment coefficient can be obtained by statistically analyzing historical data.

[0182] In the specific implementation method, through the formula (1) The first part expresses the influence of the correlation between the current abnormal factor and all other abnormal factors associated with it on the risk value. Then, based on the risk value before adjustment, the risk value is adjusted through formula (1) to accurately obtain the adjusted risk value of each abnormal factor, thus ensuring the comprehensiveness of the adjusted risk value.

[0183] Through the solution provided in this embodiment, a calculation formula is designed to accurately calculate the adjusted risk value of each abnormal factor to reflect the impact of the correlation between abnormal factors on the risk value, thereby improving the accuracy and comprehensiveness of the adjusted risk value.

[0184] Figure 3 A schematic diagram of a community security strategy optimization system based on big data is provided in an embodiment of the present application, such as Figure 3 As shown, the community security strategy optimization system 300 based on big data of this embodiment includes: an element analysis module 301, an abnormality judgment module 302, an element extraction module 303, a level classification module 304 and a processing module 305.

[0185] The element analysis module 301 is used to obtain event information, analyze the event information, and determine an event element set;

[0186] The abnormality judgment module 302 is used to analyze the event element set and judge whether there is an abnormal event according to the event analysis result;

[0187] The element extraction module 303 is used to extract the event element set and determine the abnormal element information if an abnormal event exists;

[0188] A level classification module 304 is used to determine the security threat level according to the abnormal element information;

[0189] The processing module 305 is used to determine a processing strategy and obtain a processing result according to the security threat level.

[0190] Optionally, the element analysis module 301 is specifically used for:

[0191] Analyze the surveillance video information, and determine the subject involved according to the video analysis result;

[0192] Extracting a video timestamp according to the monitoring video information, and determining the occurrence time according to the video timestamp;

[0193] The camera position is determined based on the monitoring video information, and the occurrence location is determined based on the camera position.

[0194] Optionally, the abnormality determination module 302 is specifically configured to:

[0195] Analyze the subjects involved and determine the specific behaviors;

[0196] Analyze the specific behavior based on the occurrence time, the occurrence location and the involved subjects, and determine whether the specific behavior is abnormal according to the specific behavior analysis result;

[0197] If the specific behavior is abnormal, it is determined that an abnormal event exists.

[0198] Optionally, the abnormality determination module 302 is specifically configured to:

[0199] Determine the device information of the occurrence location according to the occurrence location, and determine the device status according to the device information;

[0200] Analyze the device status, and determine whether the device is abnormal according to the device status analysis result;

[0201] If there is an abnormality, it is determined that an abnormal event exists.

[0202] Optionally, the abnormality determination module 302 is specifically configured to:

[0203] Obtain the platform information and extract several information tags;

[0204] Determining a timing analysis plan according to the platform information and the plurality of information tags;

[0205] Analyze the monitoring video information according to the timing analysis plan, and determine the information matching value between the monitoring video information and the platform information according to the timing analysis result;

[0206] The information matching value is compared with a preset information matching threshold, and if the information matching value is greater than the preset information matching threshold, it is determined that an abnormal event exists.

[0207] Optionally, the element extraction module 303 is specifically used to:

[0208] Analyzing the event element set, screening the event elements in the event element set, and determining a number of abnormal elements;

[0209] Analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements;

[0210] The plurality of abnormal elements and the abnormal correlation information are determined as the abnormal element information.

[0211] Optionally, the element extraction module 303 is specifically used to:

[0212] The abnormal elements are matched with the preset correlation data set, and the following abnormal correlation matrix is ​​established according to the matching results:

[0213] ;

[0214] in, is the abnormal correlation matrix, For the The abnormal element and The correlation between the abnormal elements;

[0215] The abnormal correlation matrix is ​​used as the abnormal correlation information.

[0216] Optionally, the level classification module 304 is specifically configured to:

[0217] Matching the several abnormal elements with a preset risk value data set, and determining the risk value corresponding to each abnormal element according to the matching result;

[0218] According to the abnormal correlation matrix, the risk value corresponding to each abnormal element is adjusted to determine the adjusted risk value of each abnormal element;

[0219] The adjusted risk value of each abnormal factor is summed up to determine the total risk value;

[0220] The security threat level is determined according to the total risk value.

[0221] Optionally, the level classification module 304 is specifically configured to:

[0222] ;

[0223] in, For the The adjusted risk value of the abnormal factor is For the The risk value of an abnormal factor, is the correlation adjustment coefficient, For the The abnormal element and The correlation of abnormal elements.

[0224] The system of this embodiment can be used to execute the method of any of the above embodiments. The implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. A community security strategy optimization method based on big data, characterized in that: include: Acquire event information, analyze the event information, and determine an event element set; Analyze the event element set, and determine whether there is an abnormal event based on the event analysis result; If an abnormal event exists, extract the event element set and determine the abnormal element information; Determining a security threat level based on the abnormal factor information; Determine a processing strategy based on the security threat level and obtain a processing result; The event information includes but is not limited to surveillance video information, the event element set includes but is not limited to the subject involved, the time of occurrence and the location of occurrence, and the analyzing of the event information to determine the event element set includes: Analyze the surveillance video information, and determine the subject involved according to the video analysis result; Extracting a video timestamp according to the monitoring video information, and determining the occurrence time according to the video timestamp; Determine the camera position according to the surveillance video information, and determine the occurrence location according to the camera position; The event element set also includes specific behaviors. The analyzing the event element set and judging whether there is an abnormal event according to the event analysis result include: Analyze the subjects involved and determine the specific behaviors; Analyze the specific behavior based on the occurrence time, the occurrence location and the involved subjects, and determine whether the specific behavior is abnormal according to the specific behavior analysis result; If the specific behavior is abnormal, it is determined that an abnormal event exists; The event element set also includes a device state. The analyzing the event element set and judging whether there is an abnormal event according to the event analysis result include: Determine the device information of the occurrence location according to the occurrence location, and determine the device status according to the device information; Analyze the device status, and determine whether the device is abnormal according to the device status analysis result; If there is an abnormality, it is determined that an abnormal event exists; The event element set also includes platform information. The analyzing the event element set and judging whether there is an abnormal event according to the event analysis result include: The platform information is public information related to community safety published by community residents on the community-specific information platform, including text information, image information and video information; Obtain the platform information and extract several information tags; Determining a timing analysis plan according to the platform information and the plurality of information tags; Analyze the monitoring video information according to the timing analysis plan, and determine the information matching value between the monitoring video information and the platform information according to the timing analysis result; Comparing the information matching value with a preset information matching threshold, and if the information matching value is greater than the preset information matching threshold, determining that an abnormal event exists; The extracting the event element set and determining abnormal element information includes: Analyzing the event element set, screening the event elements in the event element set, and determining a number of abnormal elements; Analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements; Determining the plurality of abnormal elements and the abnormal correlation information as the abnormal element information; The analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements includes: The abnormal elements are matched with the preset correlation data set, and the following abnormal correlation matrix is ​​established according to the matching results: ; in, is the abnormal correlation matrix, For the The abnormal element and The correlation between the abnormal elements; Using the abnormal correlation matrix as the abnormal correlation information; Determining the security threat level according to the abnormal factor information includes: Matching the several abnormal elements with a preset risk value data set, and determining the risk value corresponding to each abnormal element according to the matching result; According to the abnormal correlation matrix, the risk value corresponding to each abnormal element is adjusted to determine the adjusted risk value of each abnormal element; The adjusted risk value of each abnormal factor is summed up to determine the total risk value; The security threat level is determined according to the total risk value.

2. The method according to claim 1, characterized in that The step of adjusting the risk value corresponding to each abnormal element according to the abnormal correlation matrix to determine the adjusted risk value of each abnormal element includes: ; in, For the The adjusted risk value of the abnormal factor is For the The risk value of an abnormal factor, is the correlation adjustment coefficient, For the The abnormal element and The correlation of abnormal elements.

3. A community security strategy optimization system based on big data, characterized in that: include: An element analysis module, used to obtain event information, analyze the event information, and determine an event element set; An abnormality judgment module is used to analyze the event element set and judge whether there is an abnormal event according to the event analysis result; An element extraction module is used to extract the event element set and determine the abnormal element information if an abnormal event exists; A level classification module, used to determine the security threat level according to the abnormal element information; A processing module, used to determine a processing strategy and obtain a processing result according to the security threat level; The event information includes but is not limited to surveillance video information, the event element set includes but is not limited to the subject involved, the time of occurrence and the place of occurrence, and the element analysis module is specifically used to: Analyze the surveillance video information, and determine the subject involved according to the video analysis result; Extracting a video timestamp according to the monitoring video information, and determining the occurrence time according to the video timestamp; Determine the camera position according to the surveillance video information, and determine the occurrence location according to the camera position; The event element set also includes specific behaviors, and the abnormality judgment module is specifically used to: Analyze the subjects involved and determine the specific behaviors; Analyze the specific behavior based on the occurrence time, the occurrence location and the involved subjects, and determine whether the specific behavior is abnormal according to the specific behavior analysis result; If the specific behavior is abnormal, it is determined that an abnormal event exists; The event element set also includes a device state, and the abnormality judgment module is specifically used to: Determine the device information of the occurrence location according to the occurrence location, and determine the device status according to the device information; Analyze the device status, and determine whether the device is abnormal according to the device status analysis result; If there is an abnormality, it is determined that an abnormal event exists; The event element set also includes platform information, and the abnormality judgment module is specifically used to: Obtain the platform information and extract several information tags; Determining a timing analysis plan according to the platform information and the plurality of information tags; Analyze the monitoring video information according to the timing analysis plan, and determine the information matching value between the monitoring video information and the platform information according to the timing analysis result; Comparing the information matching value with a preset information matching threshold, and if the information matching value is greater than the preset information matching threshold, determining that an abnormal event exists; When extracting the event element set and determining abnormal element information, the element extraction module is used to: Analyzing the event element set, screening the event elements in the event element set, and determining a number of abnormal elements; Analyzing the plurality of abnormal elements to determine abnormal correlation information between the plurality of abnormal elements; Determining the plurality of abnormal elements and the abnormal correlation information as the abnormal element information; When the element extraction module analyzes the plurality of abnormal elements and determines the abnormal correlation information between the plurality of abnormal elements, it is used to: The abnormal elements are matched with the preset correlation data set, and the following abnormal correlation matrix is ​​established according to the matching results: ; in, is the abnormal correlation matrix, For the The abnormal element and The correlation between the abnormal elements; Using the abnormal correlation matrix as the abnormal correlation information; When the level classification module determines the security threat level according to the abnormal factor information, it includes: Matching the several abnormal elements with a preset risk value data set, and determining the risk value corresponding to each abnormal element according to the matching result; According to the abnormal correlation matrix, the risk value corresponding to each abnormal element is adjusted to determine the adjusted risk value of each abnormal element; The adjusted risk value of each abnormal factor is summed up to determine the total risk value; The security threat level is determined according to the total risk value.