A video security monitoring system based on a distributed internet of things end

Abstract

The application discloses a kind of video security monitoring systems based on distributed Internet of Things end, it is related to security monitoring technical field, the present application is deployed by setting several security points several distributed monitoring devices, corresponding deployment area is collected by distributed monitoring device video image, and corresponding video abstract is generated, a communication point is built at each security point, and the communication network of global contact is created based on the communication message of each communication point, the distributed monitoring device of several security points can be intelligently controlled, generate respective security data packet, set discrimination routing to security data packet Execute safe reception, based on the pre-constructed video model, analyze the security data packet after completing safe reception, obtain corresponding video feature information, input all video feature information into security decision library, and match corresponding decision information from security decision library, based on decision information corresponding decision chain is generated for executing final security scheme.

Description

Technical Field

[0001] This invention relates to the field of security monitoring technology, specifically to a video security monitoring system based on a distributed Internet of Things (IoT) terminal. Background Technology

[0002] In the field of existing video security monitoring technology, traditional monitoring systems mostly adopt a centralized deployment architecture, which has technical defects such as poor deployment flexibility and insufficient coordination of various functional modules. Specifically, existing systems lack a communication support mechanism for full-domain linkage, resulting in low management efficiency of monitoring equipment, disconnect between video acquisition and analysis, and inability to effectively guarantee the security of data transmission. Meanwhile, its security decision-making process relies heavily on manual intervention, resulting in delayed decision response and insufficient decision accuracy. It cannot achieve intelligent and collaborative management of the entire security monitoring process, and it is difficult to meet the actual application needs of distributed deployment scenarios with multiple security points. There is an urgent need for a video security monitoring system that can solve the above-mentioned technical defects. Summary of the Invention

[0003] The purpose of this invention is to provide a video security monitoring system based on a distributed Internet of Things (IoT) terminal to address the shortcomings in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a video security monitoring system based on a distributed Internet of Things (IoT) terminal, the system comprising: The distributed deployment module is used to set up several security points, deploy a distributed monitoring device at each security point, and collect video images of the corresponding deployment area through the distributed monitoring device and generate video summaries of the corresponding video images. The communication control module is used to set up a communication point at each security point and create a communication network for global communication based on the communication messages of each communication point. Based on the communication network, it can intelligently control the distributed monitoring devices of several security points and generate their respective security data packets. The video analysis module is used to set up an authentication route to perform secure reception of secure data packets, and to analyze the secure data packets after secure reception based on a pre-built video model to obtain corresponding video feature information. The security decision module is used to input all video feature information into the security decision database, match the corresponding decision information from the security decision database, and generate a corresponding decision chain based on the decision information to execute the final security plan.

[0005] Furthermore, several security points are set up, and a distributed monitoring device is deployed at each security point. The process of collecting video images of the corresponding deployment area through the distributed monitoring device and generating video summaries of the corresponding video images includes: Several security points are set at the location of the security area to obtain the device operation sequence of the distributed monitoring device. Based on the device operation sequence, an operation package of the distributed monitoring device is generated. The operation package is used to store all operation records of the distributed monitoring device. Establish a mapping relationship between security points and distributed monitoring devices, start the distributed monitoring devices at all security points, collect video images of the deployment area where all security points are located, and associate the corresponding video images with the corresponding operation packages at the security points; Set keyframes, process each operation packet into several corresponding sequence frames based on the preset frame length, calculate the similarity between each sequence frame and the keyframe, and extract the sequence frames whose similarity meets the preset threshold as video summaries of the video images of the corresponding security points.

[0006] Furthermore, the process of establishing a communication point at each security checkpoint and creating a communication network for global connectivity based on the communication messages from each communication point includes: Establish a channel point at each security point, establish a communication channel between two adjacent security points, connect the two ends of the communication channel to the channel point through a freeline, construct a corresponding communication point for each channel point, and establish a network channel between the communication points and the channel points; The message records existing in each communication channel are obtained as the communication messages between two communication points. The message sender is regarded as the holder of the communication message. It is determined whether there is any abnormality in the communication message of the holder. Holders with abnormality are marked as control objects, and holders without abnormality are marked as control objects. All communication messages in the same communication channel are summarized as the communication message of the corresponding communication channel. The communication message of a communication channel is distributed to all authorized objects and controlled objects of the communication channel. A grid connection application is initiated for the grid connection channel at the location of the authorized object that receives the communication message. For the controlled object that receives the communication message, the corresponding mobile offline operation is performed to control data entry into the network. All grid connection channels that have initiated grid connection applications are connected based on their location relationship to build a communication network between several communication points.

[0007] Furthermore, the process of intelligently controlling distributed monitoring devices at several security points based on the communication network and generating corresponding security data packets includes: The communication network is used to initiate corresponding control requests to several security points; A virtual device object is virtualized at the location of each security point. When a security point receives a control request, it initiates a control simulation of the virtual device object at that location. The control simulation consists of several categories of simulation items, including environment initialization, control operation drills, and security situation awareness. The control simulation includes qualified simulations and unqualified simulations. All distributed monitoring devices in the qualified simulation virtual object use the relevant simulation data of their respective control simulations as control parameters for intelligent control, and execute the corresponding control operation based on the control parameters. Record the relevant data of each distributed monitoring device to complete the current control and identify it as operation record data. Obtain the original video images collected at each distributed monitoring device and identify them as meta record data. Package the operation record data, video summary and meta record data to generate the security data packet corresponding to each distributed monitoring device.

[0008] Furthermore, the process of setting up authentication routes to perform secure reception of secure data packets includes: Build a cloud data pool based on a cloud server, deploy a routing device for the cloud data pool, set authentication rules, compile the authentication rules into corresponding script files and import them into the routing device, and convert the routing device into an authentication route; The authentication rules are used to record the reception conditions for receiving security data packets, including but not limited to source authentication and integrity verification of security data packets uploaded by each security point. Only security data packets with legitimate sources, that have not been tampered with, and that meet the conditions for complete files are received, and the corresponding reception operation is marked as a secure reception.

[0009] Furthermore, the process of analyzing the secure data packets after secure reception based on a pre-built video model to obtain the corresponding video feature information includes: Obtain historical imagery data, which includes video footage from several scenes; Each video image in a scene is used as a model scene training sample to build a video parsing sub-model for the corresponding scene. Each video parsing sub-model is set up with a corresponding model interface. Establish a relay node and interface integration terminal, connect the model interfaces of several video parsing sub-models through the relay node, obtain several interface sets through the relay node, and establish an information path between the relay node and the interface integration terminal. Several sets of interfaces are uploaded to the interface integration terminal through the information path to build a large video model. The large video model corresponds to a model overall script, and each video parsing sub-model corresponds to a sub-model script. All sub-model scripts are assigned their own file index paths on the model overall script. When security data packets need to be analyzed, each video parsing sub-model under the large video model performs the processing of the security data packets. The video parsing sub-models under the same interface protocol are based on a file index path as the processing path to complete the processing of all sub-model scripts under the corresponding processing path and obtain all video feature information corresponding to the security data packets.

[0010] Furthermore, the process of inputting all video feature information into the security decision database and matching the corresponding decision information from the security decision database includes: Each type of video feature is used to characterize a type of behavioral event; The process involves obtaining historical behavioral feature information from historical behavioral events, identifying historical video footage of these events for extracting historical video feature information, establishing a mapping relationship between the individual historical behavioral feature information and the historical video feature information, and making a decision-making scheme for the abnormal behavior represented by the historical behavioral events based on the historical behavioral feature information. All decision-making schemes are input into the database, which is then converted into a security decision-making database. The obtained video feature information is input into the security decision-making database, and decision-making schemes for resolving the behavioral events corresponding to the current video feature information are matched from the security decision-making database. These decision-making schemes are then used as decision information.

[0011] Furthermore, the process of generating a corresponding decision chain based on decision information to execute the final security solution includes: Establish a migration point at each decision information point, establish a migration path between migration points, set a number of information replication points on the migration path, the information replication points are used to replicate the decision information, and the replicated decision information is stored in the replication space of the information replication points. For information replication points that replicate the same decision information on different migration paths, an information chain is constructed. The decision information of one information replication point in the information chain is selected for compilation. The spatial address of the compilation result in the replication space of the corresponding information replication point is obtained. The spatial address is migrated to other information replication points on the chain through the information chain. The spatial address is stored in the other information replication points respectively to construct the decision chain of the corresponding decision information. When performing security processing on a behavioral event of a certain video feature information, the decision information matched from the security decision database is used as the retrieval item. The compilation result is obtained from the decision chain of the decision information through the spatial address, and the corresponding decision scheme is obtained as the final security scheme.

[0012] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. The distributed deployment module of the present invention overcomes the monitoring blind spot problem caused by the traditional centralized deployment architecture by pre-setting several security points and deploying distributed monitoring devices accordingly. The communication control module builds communication points at each security point and constructs a communication network with full-domain linkage based on communication messages to realize intelligent management and control of each distributed monitoring device and generation of security data packets, thus ensuring the coordination and basic security of data transmission.

[0013] 2. The video analysis module of this invention achieves secure reception of security data packets by setting up an authentication route. It analyzes the security data packets in conjunction with a pre-built large video model, achieving efficient and accurate extraction of video feature information and improving video analysis efficiency. The security decision module inputs all video feature information into a preset security decision database. Through decision information matching and decision chain generation, it realizes the automated and intelligent execution of security solutions, effectively improving the timeliness and accuracy of security decisions and achieving closed-loop optimization of the entire security management process. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0015] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 As shown, a video security monitoring system based on a distributed Internet of Things (IoT) terminal is disclosed. The system includes: The distributed deployment module is used to set up several security points, deploy a distributed monitoring device at each security point, and collect video images of the corresponding deployment area through the distributed monitoring device and generate video summaries of the corresponding video images. The communication control module is used to set up a communication point at each security point and create a communication network for global communication based on the communication messages of each communication point. Based on the communication network, it can intelligently control the distributed monitoring devices of several security points and generate their respective security data packets. The video analysis module is used to set up an authentication route to perform secure reception of secure data packets, and to analyze the secure data packets after secure reception based on a pre-built video model to obtain corresponding video feature information. The security decision module is used to input all video feature information into the security decision database, match the corresponding decision information from the security decision database, and generate a corresponding decision chain based on the decision information to execute the final security plan.

[0018] It should be further explained that, in the specific implementation process, several security points are set up, and a distributed monitoring device is deployed at each security point. The process of collecting video images of the corresponding deployment area through the distributed monitoring device and generating video summaries of the corresponding video images includes: Determine the security zone, set up several security points at the location of the security zone, and label the security points. Let the label be i, then i = 1, 2, 3, ..., n, where n is a natural number greater than 0; Debug the distributed monitoring devices, obtain the device operation sequence corresponding to each distributed monitoring device, and encapsulate the device operation sequence to generate the operation package corresponding to the distributed monitoring device. The operation package is used to store all operation records corresponding to the distributed monitoring device. Establish a one-to-one mapping relationship between security points and distributed monitoring devices, thereby completing the deployment of the corresponding distributed monitoring devices at each security point, activating all the distributed monitoring devices at all security points, collecting video images of the deployment area where all security points are located, and associating the corresponding video images with the operation packages at the corresponding security points. Keyframes are set, and each operation packet is processed based on a preset frame length. Each operation packet is processed into several corresponding sequence frames, and the similarity between each sequence frame and the keyframe is calculated. Sequence frames whose similarity meets a preset threshold are extracted as video summaries of the video images at the corresponding security points.

[0019] It should be noted that keyframes are obtained by training a big data model based on feature data from historical images. Keyframes are used to represent the main feature segments in a video image, serving as distinguishing identifiers for the corresponding video images.

[0020] It should be further explained that, in the specific implementation process, the process of setting up a communication point at each security point and creating a communication network for global connectivity based on the communication messages of each communication point includes: A channel point is established at the location of each security point. A communication channel is established between two adjacent security points. The two ends of the communication channel are connected to the channel point through a free-route. The free-route is a wireless transmission channel generated by a wireless communication device. The wireless transmission channel is used to transmit information between two target objects. Based on digital twin technology, a corresponding communication point is constructed for each channel point. The communication point at each location and the corresponding security point at the same location maintain the same data structure. A network channel is established between the communication point and the channel point. The message records existing in each communication channel are obtained as the mutual messages between the corresponding two communication points. There are several mutual messages. The mutual messages are used to record the message exchanges between different communication points. For any mutual message, the message sender is regarded as the holder of the mutual message. The system sequentially checks whether there are any anomalies in the inter-communication messages of each holder. Holders with anomalies are marked as controlled objects, and holders without anomalies are marked as authorized objects. All inter-communication messages within the same communication channel are aggregated as communication messages for the corresponding communication channel. The communication message of a communication channel is distributed to all authorized objects and control objects associated with the corresponding communication channel. For the authorized object that receives the communication message, a network connection application is initiated for the network connection channel at the location of the authorized object. For the control object that receives the communication message, the data entry control is performed on the corresponding mobile offline. All network connection channels that have initiated network connection applications are connected based on their location relationship, thereby constructing a communication network that enables full-domain communication between several communication points. It should be noted that for the controlled object that receives the communication message, the operation of the corresponding location of the mobile offline to perform data entry control is as follows: the mobile offline is used to connect the receiving end of the communication channel and the channel point. By default, the mobile offline is in the connected state. When it is determined that the communication message transmitted in the communication channel does not meet the corresponding conditions, that is, there is an abnormal intercommunication message in the communication message, the mobile offline between the location of the receiving end of the communication channel and the channel point is disconnected to prevent the data information at the location of the channel point from flowing into the channel network that conducts global communication and causing interference to the normal data information in the channel network.

[0021] In this context, "full-domain connection" means that any two communication points can establish a corresponding communication path through a network channel, and complete the information exchange between the two communication points on the communication path. The communication channel established by the security point is used to acquire communication messages, and based on the communication messages, it is determined whether the two communication points can directly complete the corresponding information exchange under safe conditions, thereby ensuring the communication security between several communication points.

[0022] It should be further explained that, in the specific implementation process, the process of intelligently controlling distributed monitoring devices at several security points based on the communication network and generating corresponding security data packets includes: The communication network is used to initiate corresponding control requests to several security points; Based on digital twin technology, a virtual device object is virtualized at the location of each security point. When a security point receives a control request, it initiates a control simulation of the virtual device object at that location. The control simulation consists of several categories of simulation items, including environment initialization, control operation drills, and security situation awareness. Based on environment initialization, a simulation environment is built to execute the control simulation for each virtual device object. The control operation drills correspond to several control commands when the virtual device object is controlled, as well as the control actions corresponding to each control command. The security situation awareness is used to perceive the situation of the simulated environment in which the virtual object of the device is located, as well as every control action performed in the simulated environment. The situation awareness is executed through preset perception rules, which are created by big data analysis of historical device data of distributed monitoring devices. The perception rules include several rule files simulating the environment in a normal state, as well as several rule files simulating each control action in a normal execution state; If all rule files do not meet the corresponding conditions, the corresponding control simulation is marked as a qualified simulation; if rule files do not meet the corresponding conditions, the corresponding control simulation is marked as an unqualified simulation. All distributed monitoring devices at the corresponding locations of the virtual objects of the devices under qualified simulation are used as control parameters for intelligent control by using the relevant simulation data of the control simulation completed at their respective locations, and then the corresponding control is executed through the control parameters. Record the relevant data corresponding to the current control completed by each distributed monitoring device and identify it as operation record data. Obtain the original video images collected by each distributed monitoring device and identify them as meta record data. Package the operation record data, video summary and meta record data to generate the security data packet corresponding to each distributed monitoring device.

[0023] It should be further explained that, in the specific implementation process, the process of setting up an authentication route to perform secure reception of secure data packets, and then analyzing the secure data packets after secure reception based on a pre-built video model to obtain the corresponding video feature information includes: A cloud data pool is built based on a cloud server, and a routing device is deployed for the cloud data pool. Authentication rules are set, and the authentication rules are compiled into corresponding script files and imported into the corresponding routing device, thereby converting the routing device into an authentication route. The authentication rules are used to record the reception conditions for receiving security data packets, including but not limited to source authentication and integrity verification of security data packets uploaded by each security point. Only security data packets with legitimate sources, that have not been tampered with, and that meet the conditions for complete files are received, and the corresponding reception operation is marked as a secure reception.

[0024] Obtain historical imagery data, which includes video footage from several scenes; Big data analysis based on historical image data is used to obtain several feature information fragments corresponding to video images in each scene. The feature information fragments are used to characterize a key video feature corresponding to a video image in a scene. The key video feature is used as event-related information of behavioral events in the monitoring screen recorded by the corresponding video image when the distributed monitoring device performs monitoring. Each video image in a scene is used as a model scene training sample. A video parsing sub-model for the corresponding scene is built based on each model scene training sample. Each video parsing sub-model is set up with a corresponding model interface. Establish a relay node and interface integration terminal; The model interfaces corresponding to several video parsing sub-models are connected through relay nodes, and the interface protocol corresponding to each model interface is obtained at the relay node. The model interfaces under the same interface protocol are grouped into an interface set, and the interface protocol of the interface set is serialized and used as the group identifier of the corresponding interface set. Then, several interface sets corresponding to several group identifiers are obtained at the relay node. Establish an information path between relay nodes and interface integration terminals; Several interface sets are uploaded to the interface integration terminal through the information path. The interface integration terminal selects all types of video parsing sub-models included in each interface set and uses the video parsing sub-models as the targets to be fused. All targets to be fused are fused into a model to construct a large video model for performing parsing operations on video images. The large video model corresponds to an overall model script, and each video parsing sub-model corresponds to a sub-model script. Create separate file index paths for each of the sub-model scripts within the overall model script; Each video parsing sub-model is used to parse video images in a specific scenario and obtain video feature information for that scenario. When it is necessary to analyze the security data packets received, the processing of the security data packets is performed by each video parsing sub-model under the large video model. Specifically, video parsing sub-models with the same interface protocol are based on a file index path as the processing path to complete the processing of all sub-model scripts under the corresponding processing path, thereby obtaining all video feature information corresponding to the security data packets.

[0025] It should be further explained that, in the specific implementation process, the process of inputting all video feature information into the security decision database, matching corresponding decision information from the security decision database, and generating a corresponding decision chain based on the decision information to execute the final security solution includes: Each type of video feature is used to characterize a type of behavioral event; Obtain a certain amount of historical behavioral events, extract the corresponding historical behavioral feature information based on the historical behavioral events, determine the historical video images corresponding to each historical behavioral event, and extract the historical video feature information corresponding to the historical video images. For the same historical behavioral event and its corresponding historical video footage, establish a mapping relationship between the respective historical behavioral feature information and the historical video feature information, and make a decision-making scheme for the abnormal behavior represented by the historical behavioral event based on the historical behavioral feature information. A database is constructed, and all decision-making schemes are input into the database. The database is then converted into a security decision database. The video feature information obtained from all distributed monitoring devices is input into the security decision database. All decision-making schemes for resolving the behavioral events corresponding to the current video feature information are then matched from the security decision database, and the decision-making schemes are used as decision information. Obtain the matching decision information corresponding to several security data packets, and establish a migration point at the location of each decision information; Establish migration paths between different migration points, and set a number of information replication points on the migration paths. Information replication points are used to replicate decision information. Each information replication point is used to replicate only one type of decision information. The replicated decision information is stored in the replication space corresponding to the information replication point. An information chain is constructed between information replication points that replicate the same decision information on different migration paths. The decision information of one of the information replication points included in the information chain is compiled, and the compilation result is obtained from the spatial address of the replication space of the corresponding information replication point. The spatial address is migrated to other information replication points on the chain through the information chain, and the spatial address is stored at the other information replication points to construct the decision chain corresponding to the decision information. The above construction method for different decision information is repeated to construct several decision chains corresponding to several decision information. When it is necessary to perform corresponding security processing on the behavioral event corresponding to a certain video feature information, the decision information matched from the security decision database is used as the retrieval item. Based on the retrieval item, the compilation result stored in the decision chain corresponding to the decision information is obtained through the spatial address to complete the acquisition of the decision scheme corresponding to the corresponding decision information, and it is used as the final security scheme for execution.

[0026] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A video security monitoring system based on a distributed Internet of Things (IoT) terminal, characterized in that, The system includes: The distributed deployment module is used to set up several security points, deploy a distributed monitoring device at each security point, and collect video images of the corresponding deployment area through the distributed monitoring device and generate video summaries of the corresponding video images. The communication control module is used to set up a communication point at each security point and create a communication network for global communication based on the communication messages of each communication point. Based on the communication network, it can intelligently control the distributed monitoring devices of several security points and generate their respective security data packets. The video analysis module is used to set up an authentication route to perform secure reception of secure data packets, and to analyze the secure data packets after secure reception based on a pre-built video model to obtain corresponding video feature information. The security decision module is used to input all video feature information into the security decision database, match the corresponding decision information from the security decision database, and generate a corresponding decision chain based on the decision information to execute the final security plan.

2. The video security monitoring system based on a distributed Internet of Things terminal according to claim 1, characterized in that, The process of setting up several security points, deploying a distributed monitoring device at each security point, and collecting video images of the corresponding deployment area through the distributed monitoring device to generate video summaries includes: Several security points are set at the location of the security area to obtain the device operation sequence of the distributed monitoring device. Based on the device operation sequence, an operation package of the distributed monitoring device is generated. The operation package is used to store all operation records of the distributed monitoring device. Establish a mapping relationship between security points and distributed monitoring devices, start the distributed monitoring devices at all security points, collect video images of the deployment area where all security points are located, and associate the corresponding video images with the corresponding operation packages at the security points; Set keyframes, process each operation packet into several corresponding sequence frames based on the preset frame length, calculate the similarity between each sequence frame and the keyframe, and extract the sequence frames whose similarity meets the preset threshold as video summaries of the video images of the corresponding security points.

3. A video security monitoring system based on a distributed Internet of Things terminal according to claim 2, characterized in that, The process of setting up a communication point at each security checkpoint and creating a communication network for global connectivity based on the communication messages from each communication point includes: Establish a channel point at each security point, establish a communication channel between two adjacent security points, connect the two ends of the communication channel to the channel point through a freeline, construct a corresponding communication point for each channel point, and establish a network channel between the communication points and the channel points; The message records existing in each communication channel are obtained as the communication messages between two communication points. The message sender is regarded as the holder of the communication message. It is determined whether there is any abnormality in the communication message of the holder. Holders with abnormality are marked as control objects, and holders without abnormality are marked as control objects. All communication messages in the same communication channel are summarized as the communication message of the corresponding communication channel. The communication message of a communication channel is distributed to all authorized objects and controlled objects of the communication channel. A grid connection application is initiated for the grid connection channel at the location of the authorized object that receives the communication message. For the controlled object that receives the communication message, the corresponding mobile offline operation is performed to control data entry into the network. All grid connection channels that have initiated grid connection applications are connected based on their location relationship to build a communication network between several communication points.

4. A video security monitoring system based on a distributed Internet of Things terminal according to claim 3, characterized in that, The process of intelligently controlling distributed monitoring devices at several security points based on a communication network and generating corresponding security data packets includes: The communication network is used to initiate corresponding control requests to several security points; A virtual device object is virtualized at the location of each security point. When a security point receives a control request, it initiates a control simulation of the virtual device object at that location. The control simulation consists of several categories of simulation items, including environment initialization, control operation drills, and security situation awareness. The control simulation includes qualified simulations and unqualified simulations. All distributed monitoring devices in the qualified simulation virtual object use the relevant simulation data of their respective control simulations as control parameters for intelligent control, and execute the corresponding control operation based on the control parameters. Record the relevant data of each distributed monitoring device to complete the current control and identify it as operation record data. Obtain the original video images collected at each distributed monitoring device and identify them as meta record data. Package the operation record data, video summary and meta record data to generate the security data packet corresponding to each distributed monitoring device.

5. A video security monitoring system based on a distributed Internet of Things terminal according to claim 4, characterized in that, The process of setting up an authentication route to perform secure reception of secure data packets includes: Build a cloud data pool based on a cloud server, deploy a routing device for the cloud data pool, set authentication rules, compile the authentication rules into corresponding script files and import them into the routing device, and convert the routing device into an authentication route; The authentication rules are used to record the reception conditions for receiving security data packets, including but not limited to source authentication and integrity verification of security data packets uploaded by each security point. Only security data packets with legitimate sources, that have not been tampered with, and that meet the conditions for complete files are received, and the corresponding reception operation is marked as a secure reception.

6. A video security monitoring system based on a distributed Internet of Things terminal according to claim 5, characterized in that, The process of analyzing the secure data packets after secure reception based on a pre-built video model to obtain the corresponding video feature information includes: Obtain historical imagery data, which includes video footage from several scenes; Each video image in a scene is used as a model scene training sample to build a video parsing sub-model for the corresponding scene. Each video parsing sub-model is set up with a corresponding model interface. Establish a relay node and interface integration terminal, connect the model interfaces of several video parsing sub-models through the relay node, obtain several interface sets through the relay node, and establish an information path between the relay node and the interface integration terminal. Several sets of interfaces are uploaded to the interface integration terminal through the information path to build a large video model. The large video model corresponds to a model overall script, and each video parsing sub-model corresponds to a sub-model script. All sub-model scripts are assigned their own file index paths on the model overall script. When security data packets need to be analyzed, each video parsing sub-model under the large video model performs the processing of the security data packets. The video parsing sub-models under the same interface protocol are based on a file index path as the processing path to complete the processing of all sub-model scripts under the corresponding processing path and obtain all video feature information corresponding to the security data packets.

7. A video security monitoring system based on a distributed Internet of Things terminal according to claim 6, characterized in that, The process of inputting all video feature information into the security decision database and matching the corresponding decision information from the security decision database includes: Each type of video feature is used to characterize a type of behavioral event; The process involves obtaining historical behavioral feature information from historical behavioral events, identifying historical video footage of these events for extracting historical video feature information, establishing a mapping relationship between the individual historical behavioral feature information and the historical video feature information, and making a decision-making scheme for the abnormal behavior represented by the historical behavioral events based on the historical behavioral feature information. All decision-making schemes are input into the database, which is then converted into a security decision-making database. The obtained video feature information is input into the security decision-making database, and decision-making schemes for resolving the behavioral events corresponding to the current video feature information are matched from the security decision-making database. These decision-making schemes are then used as decision information.

8. A video security monitoring system based on a distributed Internet of Things terminal according to claim 7, characterized in that, The process of generating a corresponding decision chain based on decision information to execute the final security solution includes: Establish a migration point at each decision information point, establish a migration path between migration points, set a number of information replication points on the migration path, the information replication points are used to replicate the decision information, and the replicated decision information is stored in the replication space of the information replication points. For information replication points that replicate the same decision information on different migration paths, an information chain is constructed. The decision information of one information replication point in the information chain is selected for compilation. The spatial address of the compilation result in the replication space of the corresponding information replication point is obtained. The spatial address is migrated to other information replication points on the chain through the information chain. The spatial address is stored in the other information replication points respectively to construct the decision chain of the corresponding decision information. When performing security processing on a behavioral event of a certain video feature information, the decision information matched from the security decision database is used as the retrieval item. The compilation result is obtained from the decision chain of the decision information through the spatial address, and the corresponding decision scheme is obtained as the final security scheme.

Images