Intelligent security linkage early warning method and system based on big data service

Abstract

The invention discloses an intelligent security linkage early warning method and system based on big data service, and relates to the field of intelligent security, and the method comprises the steps: constructing a big data security control platform, connecting multiple data sources, and collecting corresponding security data; constructing a double-layer security early warning model to identify security risks and abnormal behavior modes; according to the judgment result of the double-layer security early warning model, early warning information is responded, and the response strategy of the security equipment is adjusted. Through the big data platform and the double-layer early warning model, real-time security data collection and behavior pattern analysis are realized, and the early warning precision and the response speed are improved. Through graded early warning and dynamic response, high-risk behaviors are ensured to be intervened in time, resource allocation is optimized, the efficiency of a security and protection system is improved, potential risks are effectively prevented, and community safety and resident satisfaction are guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of smart security, and in particular to a smart security linkage early warning method and system based on big data services. Background Art

[0002] In the field of smart security, with the rapid development of information technology, big data services have gradually become a key technology to improve the effectiveness of security systems. In recent years, relevant technologies have made significant progress, mainly reflected in the diversity of data collection and the improvement of processing capabilities, as well as the precision of early warning models. However, the existing smart security system still has certain shortcomings in terms of linkage early warning.

[0003] First, although the existing smart security systems cover a variety of data sources in terms of data collection, they often lack effective data integration and analysis, resulting in inaccurate warning information. Secondly, in terms of warning model construction, most systems use a single-level warning model, which makes it difficult to fully identify security risks and abnormal behavior patterns in complex environments. In particular, the warning effect of existing technologies is not ideal in terms of environmental anomalies, equipment failures, and intrusion risks. In addition, existing technologies lack flexibility and pertinence in terms of warning information response and security equipment strategy adjustment, and cannot achieve true linkage warning.

[0004] In view of these problems, the present invention proposes a smart security linkage early warning method based on big data service. The method realizes the effective connection of multiple data sources and the comprehensive collection of security data by building a big data security control platform. Summary of the invention

[0005] In view of the above-mentioned existing problems, the present invention aims to solve the technical problems of low warning accuracy, inflexible response strategy and lack of comprehensive analysis of environment and personnel behavior in existing intelligent security systems. By constructing a two-layer security warning model based on big data services, in-depth analysis of environmental characteristics and personnel behavior characteristics is achieved to improve the accuracy of the warning system; at the same time, the response strategy of the security equipment is intelligently adjusted according to the warning level, the response mechanism of the warning information is optimized, multi-source security data is integrated to comprehensively evaluate the security status, and warning information is issued in advance by predicting and locking potential chain reaction impact areas, thereby improving the comprehensive prevention and control capabilities of the system, reducing the possibility of accidents, and ensuring the safety of people and property.

[0006] In order to solve the above technical problems, a smart security linkage early warning method based on big data service is proposed, including:

[0007] Build a big data security control platform, connect multiple data sources and collect corresponding security data; build a two-layer security early warning model to identify security risks and abnormal behavior patterns; respond to early warning information based on the judgment results of the two-layer security early warning model, and adjust the response strategy of security equipment.

[0008] As a preferred solution of the intelligent security linkage early warning method based on big data service described in the present invention, wherein: the corresponding security data collected includes environmental data, personnel behavior data and conventional data;

[0009] The environmental data include meteorological data, air quality monitoring data, and lighting intensity;

[0010] The personnel behavior data includes human flow monitoring data and visitor registration data;

[0011] The conventional data includes geographic location data, equipment status monitoring data, alarm record data, and community event data.

[0012] As a preferred solution of the intelligent security linkage warning method based on big data service described in the present invention, wherein: the two-layer security warning model includes: the first layer security warning model is aimed at environmental features, and the environmental data in the collected security data is subjected to high-level feature learning and pattern recognition through a deep belief network, and the input original environmental data set E is preprocessed to generate a feature vector The feature vector generated by the deep belief network Perform high-level feature learning to identify potential security risks:

[0013]

[0014] Among them, R is the potential security risk, m is the number of features after feature extraction, j is the variable index, is the value of the jth eigenvector, ω j is the weight of the jth feature obtained by model adaptive learning, and dati is the dynamic adjustment term for correcting the overall deviation of risk;

[0015] The security risk level is mapped to three types of security risks, including environmental abnormality risk, equipment failure risk, and intrusion risk. The environmental abnormality risk is fire and flood, the equipment failure risk is the failure of surveillance cameras, and the intrusion risk is the entry of unauthorized personnel.

[0016] As a preferred solution of the intelligent security linkage early warning method based on big data service described in the present invention, the first-level security early warning model includes predicting the probability value of the possible chain reaction impact area according to the identified security risk:

[0017]

[0018] Among them, P(Z|R) is the probability of predicting the chain reaction Z under the given risk degree R; α is the weight factor for adjusting the reaction intensity, and k is the control baseline, a constant to ensure that the probability value is within a reasonable range;

[0019] Lock the identified influence area of the chain reaction and issue early warning information of the first-layer model.

[0020] As a preferred solution of a smart security linkage early warning method based on big data services according to the present invention, wherein: the issuance of early warning information of the first-layer model includes setting an early warning level according to the size of the probability value P(Z|R):

[0021] When P(Z|R) > 0.65, it is marked as a highly dangerous area, a red early warning is issued, and the second-layer security early warning model is notified to issue a red early warning simultaneously;

[0022] When 0.4 < P(Z|R) ≤ 0.65, it is marked as a medium-danger area, an orange early warning is issued, and the second-layer security early warning model is notified to monitor whether the behavior of personnel is affected;

[0023] When P(Z|R) ≤ 0.4, it is marked as a low-danger area, a yellow early warning is issued, and the behavior of personnel in the second-layer security early warning model is monitored with emphasis.

[0024] As a preferred solution of a smart security linkage early warning method based on big data services according to the present invention, wherein: the double-layer security early warning model further includes that the second-layer security early warning model is for personnel behavior type features, and by analyzing historical security data, it identifies the personnel behavior pattern BE and predicts the abnormal risk trend of the personnel behavior law according to the personnel behavior pattern:

[0025]

[0026] Among them, T is the judgment value of the abnormal risk trend, D is the number of identified behavior patterns, BE d is the value under the d-th behavior pattern, and β is an adjustment constant to ensure a reasonable assessment of the risk trend;

[0027] When the second-layer model identifies an abnormal trend, it triggers the issuance of early warning information.

[0028] As a preferred solution of a smart security linkage early warning method based on big data services according to the present invention, wherein: the triggering of the issuance of early warning information includes judging whether it belongs to a high-risk behavior pattern according to the result and category of T, including continuous irregular behaviors and strangers frequently appearing in specific areas;

[0029] When T>0.8TH T When the person's behavior is obviously abnormal and belongs to a high-risk behavior pattern, a red warning message is sent to the big data security control platform, including the current abnormal behavior description and corresponding data, and the surrounding surveillance cameras are activated. At the same time, the sensor activates the sound and light alarm to warn the surrounding personnel and automatically dispatch tasks to the patrol unit;

[0030] When 0.4TH T <T≤0.8TH T When the behavior pattern is potentially abnormal and has a tendency towards high-risk behavior patterns, an orange warning message is sent to the big data security control platform, describing the abnormal behavior detected, adjusting the sensitivity of the surveillance camera, increasing the capture rate of suspicious activities, starting crowd flow monitoring, recording abnormal crowd gathering data, and generating a trend analysis report;

[0031] When T≤0.4TH T When a yellow warning message is sent to the big data security control platform, it indicates the normal status of the current monitoring area, retains the current monitoring data, maintains regular dynamic updates of the current area, and reports the monitoring results regularly.

[0032] Another object of the present invention is to provide a smart security linkage warning system based on big data services. The present invention aims to achieve real-time monitoring, warning and response to security risks, improve the intelligence, linkage and warning accuracy of the security system, thereby effectively preventing and preventing the occurrence of various security incidents and protecting the lives and property of the people.

[0033] As a preferred solution of the intelligent security linkage early warning system based on big data service described in the present invention, it is characterized by comprising a big data security control platform and a double-layer security early warning module;

[0034] The big data security control platform is responsible for connecting multiple data sources, collecting environmental data, personnel behavior data, and security data of conventional data, and passing the data to the double-layer security early warning module for early warning;

[0035] The two-layer security warning module includes a first-layer security warning unit and a second-layer security warning unit. The first-layer security warning unit performs high-level feature learning and pattern recognition on environmental data through a deep belief network for environmental features, identifies potential security risks, and maps the risks to three types of security risks. The first-layer security warning unit is responsible for issuing warning information and notifying the second-layer security warning unit according to the warning level. The second-layer security warning unit analyzes historical security data for personnel behavior features, identifies personnel behavior patterns, and predicts abnormal risk trends of personnel behavior rules. When abnormal trends are identified, the warning information is triggered.

[0036] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the intelligent security linkage warning method based on big data service are implemented.

[0037] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the intelligent security linkage warning method based on big data services are implemented.

[0038] Beneficial effects of the present invention: By constructing a big data security control platform and connecting multiple data sources, the present invention realizes comprehensive and real-time security data collection, ensures that the information basis of the early warning system is broad and accurate, and improves the comprehensiveness and accuracy of the early warning system information. On this basis, the first-level security early warning model uses a deep belief network to perform high-level feature learning and pattern recognition on environmental data, predicts the probability value of the chain reaction impact area, and implements graded early warning, which improves the pertinence and effectiveness of the early warning information, and at the same time realizes dynamic adjustment and multi-level response of the early warning information, and improves the flexibility and response speed of the early warning system. The second-level security early warning model analyzes historical security data, identifies personnel behavior patterns, and predicts abnormal risk trends, realizing rapid response and effective intervention to high-risk behavior patterns, improving the overall security system's ability to respond to emergencies, ensuring timely prevention of potential dangerous behaviors, continuous monitoring and prevention of potential risks, and stable operation of the security system and reasonable allocation of resources.

[0039] Furthermore, during the process of issuing warning information, the present invention sends red, orange or yellow warning information to the security control center based on the judgment results of abnormal behavior patterns, and takes corresponding response measures, such as initiating emergency response, adjusting monitoring strategies or maintaining routine monitoring. These measures work together to achieve the beneficial effects of improving the response accuracy of the warning system, optimizing resource allocation, and ensuring the efficient operation of the security system, thereby ensuring the safety of the community while also enhancing the sense of security and satisfaction of residents. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0041] Figure 1 An overall flow chart of a smart security linkage warning method based on big data services provided by an embodiment of the present invention.

[0042] Figure 2 A system solution module diagram of a smart security linkage warning system based on big data services provided by an embodiment of the present invention.

[0043] In the figure: 10, big data security control platform; 20, double-layer security warning module; 201, first-layer security warning unit; 202, second-layer security warning unit. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0045] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0046] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.

[0047] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0048] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0049] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a smart security linkage warning method based on big data services, including:

[0051] S1: Build a big data security control platform, connect multiple data sources and collect corresponding security data.

[0052] Collect corresponding security data including environmental data, personnel behavior data and general data;

[0053] The environmental data include meteorological data, air quality monitoring data, and lighting intensity;

[0054] Among them, meteorological data: temperature, humidity, wind speed, precipitation, etc. are real-time data collected by weather stations or sensors.

[0055] Air quality monitoring data: indicators such as PM2.5, PM10, CO2, VOC, etc. can be collected through air quality monitoring instruments.

[0056] Lighting intensity: Use light sensors to monitor ambient light intensity in real time to determine safety at night.

[0057] The personnel behavior data includes human flow monitoring data and visitor registration data;

[0058] Among them, crowd flow monitoring: monitoring the flow of people and movement paths in a specific area through video analysis or infrared sensors.

[0059] Visitor registration data: records visitor entry and exit times, identity information, destination, etc., collected through the access control system.

[0060] The conventional data includes geographic location data, equipment status monitoring data, alarm record data, and community event data;

[0061] Among them, geographic location data: use GPS or base station positioning technology to obtain the location information and movement trajectory of people in a specific area.

[0062] Equipment status monitoring data: such as the working status and fault records of fire alarm equipment and surveillance cameras to ensure the normal operation of the equipment.

[0063] Alarm record data: historical alarm records collected from the security system, including alarm time, alarm type, processing results and other information.

[0064] Community event data: including historical crime records, community event arrangements, etc., to assess security risks and potential threats.

[0065] S2: Construct a two-layer security warning model to identify security risks and abnormal behavior patterns.

[0066] Furthermore, the first-level security warning model is based on environmental features. It uses a deep belief network to perform high-level feature learning and pattern recognition on the environmental data in the collected security data, preprocesses the input original environmental data set E, and generates a feature vector:

[0067]

[0068] in, is the feature vector set obtained after deep belief network processing, Φ is a nonlinear activation function used to introduce nonlinear features; n is the number of features in the environmental data, i is the variable index, and a i is the i-th environmental feature e i The adaptive weight, e i is the ith original environment characteristic value, b i is the i-th environmental feature e i The modulation factor controls the importance of the feature; x is a constant used to adjust the overall offset;

[0069] The feature vector generated by the deep belief network Perform high-level feature learning to identify potential security risks:

[0070]

[0071] Among them, R is the potential security risk, m is the number of features after feature extraction, j is the variable index, is the value of the jth eigenvector, ω j is the weight of the jth feature obtained by model adaptive learning, and dati is the dynamic adjustment term for correcting the overall deviation of risk;

[0072] The security risk level is mapped to three types of security risks, including environmental abnormality risk, equipment failure risk, and intrusion risk. The environmental abnormality risk is fire and flood, the equipment failure risk is the failure of surveillance cameras, and the intrusion risk is the entry of unauthorized personnel.

[0073] It should be noted that based on the identified security risk, the probability values ​​of the areas affected by possible chain reactions are predicted:

[0074]

[0075] Among them, P(Z|R) is the probability of predicting the chain reaction Z under the given risk level R; α is the weight factor for adjusting the reaction intensity, and k is the control baseline, a constant to ensure that the probability value is within a reasonable range;

[0076] Lock the identified impact area of the chain reaction and issue early warning information for the first-layer model.

[0077] Set the early warning level according to the magnitude of the probability value P(Z|R):

[0078] When P(Z|R) > 0.65, mark it as a highly dangerous area, issue a red early warning, and notify the second-layer security early warning model to issue a red early warning simultaneously;

[0079] When 0.4 < P(Z|R) ≤ 0.65, mark it as a moderately dangerous area, issue an orange early warning, and notify the second-layer security early warning model to monitor whether the behavior of personnel is affected;

[0080] When P(Z|R) ≤ 0.4, mark it as a low-danger area, issue a yellow early warning, and focus on monitoring the behavior of personnel in the second-layer security early warning model.

[0081] S3: Respond to the early warning information according to the judgment result of the double-layer security early warning model, and adjust the response strategy of the security equipment.

[0082] Furthermore, the second-layer security early warning model is for personnel behavior characteristics. By analyzing historical security data, it identifies personnel behavior patterns:

[0083]

[0084] Among them, BE is the identified behavior pattern, Θ is a new normalization function to ensure that the output result is within a reasonable range; L is the number of behavior patterns in the behavior data, ξ l is the weight of the l-th behavior pattern h l of, h l is the l-th historical behavior data value, θ is the non-linear exponent to enhance the sensitivity of the model, and ρ is the offset to ensure the applicability of the output adjustment of pattern recognition;

[0085] Predict the abnormal risk trend of personnel behavior patterns according to the personnel behavior patterns:

[0086]

[0087] Among them, T is the judgment value of the abnormal risk trend, D is the number of identified behavior patterns, BE dis the value in the dth behavior mode, and β is the adjustment constant to ensure a reasonable assessment of the risk trend;

[0088] When the second-layer model identifies an abnormal trend, it triggers the release of early warning information.

[0089] It should be noted that the results and categories of T are used to determine whether it belongs to a high-risk behavior pattern, including continuous irregular behavior and frequent strangers in a specific area;

[0090] When T>0.8TH T When the person's behavior is obviously abnormal and belongs to a high-risk behavior pattern, a red warning message is sent to the big data security control platform, including the current abnormal behavior description and corresponding data, and the surrounding surveillance cameras are activated. At the same time, the sensor activates the sound and light alarm to warn the surrounding personnel and automatically dispatch tasks to the patrol unit;

[0091] When 0.4TH T <T≤0.8TH T When the behavior pattern is potentially abnormal and has a tendency towards high-risk behavior patterns, an orange warning message is sent to the big data security control platform, describing the abnormal behavior detected, adjusting the sensitivity of the surveillance camera, increasing the capture rate of suspicious activities, starting crowd flow monitoring, recording abnormal crowd gathering data, and generating a trend analysis report;

[0092] When T≤0.4TH T When a yellow warning message is sent to the big data security control platform, it indicates the normal status of the current monitoring area, retains the current monitoring data, maintains regular dynamic updates of the current area, and reports the monitoring results regularly.

[0093] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0094] Embodiment 2, the second embodiment of the present invention, is different from the first two embodiments in that:

[0095] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0096] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0097] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0098] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0099] Example 3, reference Figure 2 , which is the third embodiment of the present invention, and which provides a smart security linkage warning system based on big data services, including a big data security control platform 10 and a double-layer security warning module 20;

[0100] The big data security control platform 10 is responsible for connecting multiple data sources, collecting security data of environmental data, personnel behavior data and conventional data, and passing the data to the double-layer security warning module 10 for warning;

[0101] The double-layer security warning module 20 includes a first-layer security warning unit 201 and a second-layer security warning unit 202. With respect to environmental features, the first-layer security warning unit 201 performs high-level feature learning and pattern recognition on environmental data through a deep belief network, identifies potential security risks, and maps the risks to three types of security risks. The first-layer security warning unit 201 is responsible for issuing warning information and notifying the second-layer security warning unit 202 according to the warning level. With respect to personnel behavior features, the second-layer security warning unit 202 identifies personnel behavior patterns by analyzing historical security data, predicts abnormal risk trends of personnel behavior rules, and triggers the issuance of warning information when abnormal trends are identified.

[0102] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A smart security linkage early warning method based on big data services, characterized by: including Construct a big data security control platform to connect multiple data sources and collect corresponding security data; Construct a two-layer security warning model to identify security risks and abnormal behavior patterns; Respond to warning information according to the judgment results of the two-layer security warning model and adjust the response strategies of security devices.

2. The intelligent security linkage early warning method based on big data service as claimed in claim 1, characterized in that: The collection of corresponding security data includes environmental data, personnel behavior data, and conventional data; The environmental data includes meteorological data, air quality monitoring data, and lighting intensity; The personnel behavior data includes pedestrian flow monitoring data and visitor registration data; The conventional data includes geographical location data, equipment status monitoring data, alarm record data, and community event data.

3. The intelligent security linkage early warning method based on big data service as claimed in claim 2, characterized in that: The two-layer security warning model includes: the first layer of the security warning model is aimed at environmental features, and the environmental data in the collected security data are subjected to high-level feature learning and pattern recognition through a deep belief network, and the input original environmental data set E is preprocessed to generate a feature vector The feature vector generated by the deep belief network Perform high-level feature learning to identify potential security risks: Among them, R is the potential security risk, m is the number of features after feature extraction, j is the variable index, is the value of the jth eigenvector, ω j is the weight of the jth feature obtained by model adaptive learning, and dati is the dynamic adjustment term for correcting the overall deviation of risk; Map the security risk level to three types of security risks, including environmental anomaly risk, equipment failure risk, and intrusion risk. The environmental anomaly risk is fire and flood, the equipment failure risk is the failure of surveillance cameras, and the intrusion risk is the entry of unauthorized personnel.

4. The intelligent security linkage early warning method based on big data service as claimed in claim 3, characterized in that: The first-layer security warning model includes predicting the probability value of the possible chain reaction impact area according to the identified security risk level: Among them, P(Z|R) is the probability of predicting the chain reaction Z under the given risk level R; α is the weight factor for adjusting the reaction intensity, and k is the control baseline, a constant to ensure that the probability value is within a reasonable range; Lock the identified chain reaction impact area and issue warning information for the first-layer model.

5. The intelligent security linkage early warning method based on big data service as claimed in claim 4, characterized in that: The issuance of warning information for the first-layer model includes setting the warning level according to the size of the probability value P(Z|R): When P(Z|R)>0.65, it is marked as a high-risk area, a red warning is issued, and the second-layer security warning model is notified to issue a red warning simultaneously; When 0.4<P(Z|R)≤0.65, it is marked as a medium-risk area, an orange warning is issued, and the second-layer security warning model is notified to monitor whether the personnel behavior is affected; When P(Z|R)≤0.4, it is marked as a low-risk area, a yellow warning is issued, and the personnel behavior in the second-layer security warning model is monitored key.

6. The intelligent security linkage early warning method based on big data service as claimed in claim 5, characterized in that: The two-layer security warning model also includes that the second-layer security warning model is for personnel behavior characteristics. By analyzing historical security data, it identifies the personnel behavior pattern BE and predicts the abnormal risk trend of personnel behavior rules according to the personnel behavior pattern; Among them, T is the judgment value of abnormal risk trend, D is the number of identified behavior patterns, BE d is the value in the dth behavior mode, and β is the adjustment constant to ensure a reasonable assessment of the risk trend; When the second-layer model identifies an abnormal trend, it triggers the issuance of warning information.

7. The intelligent security linkage early warning method based on big data service as claimed in claim 6, characterized in that: The triggering of the issuance of warning information includes judging whether it belongs to a high-risk behavior pattern according to the result and category of T, including continuous irregular behavior and strangers frequently appearing in specific areas; When T>0.8TH T When the person's behavior is obviously abnormal and belongs to a high-risk behavior pattern, a red warning message is sent to the big data security control platform, including the current abnormal behavior description and corresponding data, and the surrounding surveillance cameras are activated. At the same time, the sensor activates the sound and light alarm to warn the surrounding personnel and automatically dispatch tasks to the patrol unit; When 0.4TH T <T≤0.8TH T When the behavior pattern is potentially abnormal and has a tendency towards high-risk behavior patterns, an orange warning message is sent to the big data security control platform, describing the abnormal behavior detected, adjusting the sensitivity of the surveillance camera, increasing the capture rate of suspicious activities, starting crowd flow monitoring, recording abnormal crowd gathering data, and generating a trend analysis report; When T≤0.4TH T When a yellow warning message is sent to the big data security control platform, it indicates the normal status of the current monitoring area, retains the current monitoring data, maintains regular dynamic updates of the current area, and reports the monitoring results regularly.

8. A system using a smart security linkage early warning method based on big data services as claimed in any one of claims 1 to 7, characterized in that: including a big data security control platform and a two-layer security warning module; The big data security control platform is responsible for connecting multiple data sources, collecting security data such as environmental data, personnel behavior data, and conventional data, and transmitting the data to the two-layer security warning module for warning; The two-layer security warning module includes a first-layer security warning unit and a second-layer security warning unit. The first-layer security warning unit performs high-level feature learning and pattern recognition on environmental data through a deep belief network for environmental features, identifies potential security risks, and maps the risks to three types of security risks. The first-layer security warning unit is responsible for issuing warning information and notifying the second-layer security warning unit according to the warning level. The second-layer security warning unit analyzes historical security data for personnel behavior features, identifies personnel behavior patterns, and predicts abnormal risk trends of personnel behavior rules. When abnormal trends are identified, the warning information is triggered.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a smart security linkage warning method based on big data services described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of a smart security linkage warning method based on big data service described in any one of claims 1 to 7 are implemented.

Images