A remote security monitoring system based on image recognition

By building a monitoring model and behavioral space model, and real-time acquisition and analysis of video image information, the problems of monitoring rules formulation and video analysis in the existing technology are solved, monitoring accuracy and reaction speed are improved, and better security monitoring effects are achieved.

CN118429783BActive Publication Date: 2024-11-08SHENZHEN HISTAR TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410654472.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-24
Publication Date
2024-11-08
Estimated Expiration
2044-05-24

AI Technical Summary

Technical Problem

The existing remote security monitoring system based on image recognition is difficult to formulate corresponding monitoring rules for different monitoring areas, and cannot effectively perform video image analysis, resulting in low reaction efficiency and insufficient monitoring accuracy.

Method used

By obtaining the work information of monitoring equipment, building a monitoring model, dividing management space and setting management rules, collecting video image information in real time, extracting key target information, building a behavioral space model, and verifying and judging according to security conditions, locking target information that does not meet the conditions is locked. .

Benefits of technology

The behavioral analysis of key goals has been achieved, monitoring accuracy and response speed have been improved, and the role of remote monitoring and autonomous analysis is good.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118429783B_ABST
    Figure CN118429783B_ABST
Patent Text Reader

Abstract

The present invention discloses a remote safety monitoring system based on image recognition, which relates to the technical field of safety monitoring, an acquisition module, which is used to acquire working information of multiple monitoring devices, a division module, which is connected to the acquisition module, which is used to divide the space based on the monitoring model to obtain multiple management spaces, and to set management rule information based on the multiple management spaces, an information acquisition module, which is connected to the division module, a verification module, which is connected to the information acquisition module, which is used to verify the behavior space model based on the monitoring information to obtain verification result information; a judgment module, which is connected to the verification module, which is used to lock the key target information corresponding to the behavior space model that does not meet the safety conditions and prompt it to the management end. The present invention can better perform behavior analysis on key targets, has a good behavior analysis effect, and is convenient for subsequent safety monitoring.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of security monitoring, and in particular to a remote security monitoring system based on image recognition. Background Art

[0002] With the rapid development of information technology, people's demand for security monitoring is growing day by day; whether it is the internal area of ​​the community, public areas, commercial places or private residences, security monitoring has become an indispensable part; and the remote security monitoring system based on image recognition, by utilizing advanced image recognition technology, can achieve real-time monitoring of the monitored area, but the current monitoring system is difficult to formulate corresponding monitoring rules for different monitoring areas, which is not convenient for analysis with subsequent video images, and cannot effectively carry out precise target monitoring of the monitored environment, which leads to low response efficiency and insufficient monitoring accuracy, and no better security monitoring effect. Summary of the invention

[0003] The purpose of the present invention is to provide a remote safety monitoring system based on image recognition to solve the shortcomings of the background technology.

[0004] In order to achieve the above object, the present invention provides the following technical solution: a remote safety monitoring system based on image recognition, comprising:

[0005] An acquisition module is used to acquire working information of multiple monitoring devices, wherein the working information of the monitoring devices includes monitoring information and corresponding monitoring range information, and to build a monitoring model based on the target monitoring range;

[0006] A partitioning module, connected to the acquisition module, is used to partition the space based on the monitoring model to obtain multiple management spaces, set management rule information based on the multiple management spaces, and bind the management rule information with the corresponding management space as monitoring information;

[0007] An information collection module, connected to the division module, is used to collect video image information in real time based on monitoring information, extract key target information from the video image information, and construct a behavior space model of the key target based on the key target information;

[0008] A verification module, connected to the information collection module, is used to verify the behavior space model based on the monitoring information to obtain verification result information;

[0009] The judgment module is connected to the verification module and is used to lock the key target information corresponding to the behavior space model that does not meet the safety conditions and prompt it to the management end.

[0010] In a preferred embodiment, the acquisition module includes:

[0011] An information acquisition unit, used to acquire a target monitoring range, and acquire monitoring device type information and corresponding monitoring range information of multiple monitoring devices within the target monitoring range as monitoring device working information;

[0012] An encoding unit, used for numbering the monitoring devices based on the working information of the monitoring devices to obtain the identity information of the monitoring devices, and binding the identity information of the monitoring devices with the working information of the monitoring devices;

[0013] The construction unit is used to obtain the structural information within the target monitoring range, construct a three-dimensional space model based on the structural information within the target monitoring range, construct the monitoring equipment working information in the three-dimensional space model and mark the corresponding monitoring equipment identity information to obtain the monitoring model.

[0014] In a preferred embodiment, the division module includes:

[0015] A division unit, used for obtaining multiple management types within a target monitoring range based on the structural information, dividing the target monitoring range based on the multiple management types to obtain multiple management ranges, and dividing and marking the multiple management ranges in the monitoring model to obtain multiple management spaces;

[0016] A setting unit, used for setting a management rule information base, wherein the management rule information base includes a plurality of general management rule information and a plurality of specific management rule information;

[0017] A setting unit, used for setting corresponding label information based on the general management rule information and the specific management rule information corresponding to the big data, wherein the label information includes a plurality of management type features;

[0018] A matching unit, configured to extract multiple management types within a target monitoring range to obtain multiple management type features, match corresponding tag information in a management rule information base based on the management type features, and obtain one or more of corresponding general management rule information and specific management rule information as management rule information based on the tag information;

[0019] The management unit is used to bind the management rule information with the corresponding management space and use it as the management and control information.

[0020] In a preferred embodiment, the information collection module includes:

[0021] A real-time acquisition unit, used to respectively acquire video image information in real time from multiple monitoring devices based on corresponding monitoring information, and to match the video image information with corresponding management spaces to obtain integrated video image information;

[0022] The information construction unit is used to extract and integrate key target information in the video image information based on the monitoring information, and construct the key target information into the monitoring model to obtain a behavior space model.

[0023] In a preferred embodiment, the information construction unit includes:

[0024] A monitoring unit, used to respectively obtain video image information of a plurality of monitoring devices and put the video image information into a management space;

[0025] A splicing unit, used to obtain common feature information between adjacent contacting management spaces, and splice the video image information corresponding to the adjacent contacting management spaces based on the common feature information and according to the time of video image information acquisition to obtain integrated video image information;

[0026] A data extraction unit, used to extract key target information in the integrated video image information based on the management rule information in the monitoring information, wherein the key target information includes a key target shape and a key target position trajectory;

[0027] The splicing construction unit is used to perform three-dimensional construction based on the key target shape in the key target information to obtain a key target model, and construct the key target model in the monitoring model according to the key target position trajectory to obtain a behavior space model.

[0028] In a preferred embodiment, the verification module includes:

[0029] A security setting unit, used to set security monitoring information based on a behavior space model;

[0030] A verification unit is used to obtain behavior information of key targets in a behavior space model based on security monitoring information, wherein the behavior information includes speed information of the key targets and range information of the key targets, and to verify the behavior information based on the monitoring information to obtain a verification result.

[0031] In a preferred embodiment, the security setting unit includes:

[0032] A layout unit, used for selecting a spatial range in the behavior space model based on the behavior space model, and setting a plurality of evenly distributed monitoring points based on the spatial range;

[0033] A space division unit is used to divide the corresponding monitoring points according to the management space, and bind the monitoring points with the management space;

[0034] The enabling unit is used to enable the monitoring point based on the management rule information corresponding to the management space to obtain the monitoring point function information, wherein the monitoring point function information includes speed monitoring and range monitoring.

[0035] In a preferred embodiment, the judgment module includes:

[0036] A security formulation unit, used to set security conditions based on management rule information corresponding to the management space;

[0037] The judgment unit is used to lock the key target information corresponding to the behavior space model that does not meet the safety conditions, and transmit the behavior space model to the management end for prompting.

[0038] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0039] The present invention constructs the key target model in the monitoring model according to the key target position trajectory to obtain the behavior space model, which can better analyze the behavior of the key target, has a good behavior analysis function, and is convenient for subsequent security monitoring. The monitoring accuracy is good and the response is fast, and it has a good remote monitoring and autonomous analysis function. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1, please refer to Figure 1 As shown, the remote security monitoring system based on image recognition described in this embodiment includes:

[0044] An acquisition module is used to acquire working information of multiple monitoring devices, wherein the working information of the monitoring devices includes type information of the monitoring devices and corresponding monitoring range information, and to build a monitoring model based on the target monitoring range;

[0045] A partitioning module, connected to the acquisition module, is used to partition the space based on the monitoring model to obtain multiple management spaces, set management rule information based on the multiple management spaces, and bind the management rule information with the corresponding management space as monitoring information;

[0046] An information collection module, connected to the division module, is used to collect video image information in real time based on monitoring information, extract key target information from the video image information, and construct a behavior space model of the key target based on the key target information;

[0047] A verification module, connected to the information collection module, is used to verify the behavior space model based on the monitoring information to obtain verification result information;

[0048] The judgment module is connected to the verification module and is used to lock the key target information corresponding to the behavior space model that does not meet the safety conditions and prompt it to the management end;

[0049] It should be noted that with the rapid development of information technology, people's demand for security monitoring is growing; whether it is the internal area of ​​the community, public areas, commercial places or private residences, security monitoring has become an indispensable part; and the remote security monitoring system based on image recognition can realize real-time monitoring of the monitoring area by utilizing advanced image recognition technology, but the current monitoring system is difficult to formulate corresponding monitoring rules for different monitoring areas, which is not convenient for analysis with subsequent video images, and cannot effectively perform accurate target monitoring of the monitored environment, which leads to low response efficiency and insufficient monitoring accuracy, and no better security monitoring effect; and the application constructs the key target model in the monitoring model according to the key target position trajectory to obtain the behavior space model, which can better analyze the behavior of the key target, has a good behavior analysis effect, is convenient for subsequent security monitoring, has good monitoring accuracy, and responds quickly, and has a good remote monitoring autonomous analysis effect;

[0050] In one embodiment, the acquisition module includes:

[0051] An information acquisition unit, used to acquire a target monitoring range, and acquire monitoring device type information and corresponding monitoring range information of multiple monitoring devices within the target monitoring range as monitoring device working information;

[0052] An encoding unit, used for numbering the monitoring devices based on the working information of the monitoring devices to obtain the identity information of the monitoring devices, and binding the identity information of the monitoring devices with the working information of the monitoring devices;

[0053] A construction unit is used to obtain structural information within the target monitoring range, construct a three-dimensional space model based on the structural information within the target monitoring range, construct the monitoring equipment working information in the three-dimensional space model and mark the corresponding monitoring equipment identity information to obtain a monitoring model;

[0054] It should be noted that, first, it is necessary to obtain the range to be monitored, which is taken as the target monitoring range here. The monitoring equipment working information of the multiple monitoring equipment set within the target monitoring range is obtained, including the monitoring equipment type information and the corresponding monitoring range information. Then, the monitoring equipment is numbered to obtain the monitoring equipment identity information. Then, the monitoring equipment identity information is matched with the monitoring equipment working information and the information is bound. The structural information within the target monitoring range is obtained, and a three-dimensional space model is constructed based on the structural information within the target monitoring range. The monitoring equipment working information is constructed in the three-dimensional space model and the corresponding monitoring equipment identity information is marked to obtain the monitoring model. Here, it is equivalent to first obtaining the structural information within the target monitoring range. The structural information here is the object information and geographical environment information within the target monitoring range, wherein the object information is the object type and the object position. The purpose here is to obtain the physical environment information within the target monitoring range, and then a three-dimensional model is constructed according to the obtained structural information. The structural information is constructed within the target monitoring range to obtain a three-dimensional space model, which can more clearly understand the target monitoring range and facilitate the subsequent monitoring of key targets in the target monitoring range.

[0055] In one embodiment, the partitioning module comprises:

[0056] A division unit, used for obtaining multiple management types within a target monitoring range based on the structural information, dividing the target monitoring range based on the multiple management types to obtain multiple management ranges, and dividing and marking the multiple management ranges in the monitoring model to obtain multiple management spaces;

[0057] A setting unit, used for setting a management rule information base, wherein the management rule information base includes a plurality of general management rule information and a plurality of specific management rule information;

[0058] A setting unit, used for setting corresponding label information based on the general management rule information and the specific management rule information corresponding to the big data, wherein the label information includes a plurality of management type features;

[0059] A matching unit, configured to extract multiple management types within a target monitoring range to obtain multiple management type features, match corresponding tag information in a management rule information base based on the management type features, and obtain one or more of corresponding general management rule information and specific management rule information as management rule information based on the tag information;

[0060] A management unit, used to bind the management rule information with the corresponding management space and use it as control information;

[0061] It should be noted that there are multiple management types within a target monitoring range. For example, in a residential area, the driving lanes, green areas and power room control locations are different management types. The driving lanes, green areas and power room control locations here are also multiple different structural information (places) within the target monitoring range. Different places represent different management types. Then, the target monitoring range is divided according to the management type within the target monitoring range to obtain multiple management ranges. Based on the multiple management ranges, they are divided and marked in the monitoring model to obtain multiple management spaces. Here, after obtaining the management range, it is necessary to divide the management space in the monitoring model. Different Places are different types of management, so the management rules corresponding to different types of management are also different. For example, on the driving lane, what needs to be monitored is the safe driving of vehicles, such as the speed of vehicles and the relationship between vehicles and pedestrians. In the green area, what needs to be monitored is the behavior of pedestrians or whether vehicles enter the green area. In the power room control, what needs to be monitored is whether any movable objects enter the safe range (here is the specific management rule information) to avoid damage to the power cabinet or the moving objects in contact. Therefore, it is necessary to set up a management rule information library, which includes multiple general management rule information and multiple specific management rule information. The general management rule information is that all vehicles in the community are managed objects, which is general. The specific management rule information represents the information of the management target that needs to be monitored in some characteristic places. Then, based on big data, the corresponding label information is set for the general management rule information and the specific management rule information. The label information here represents the place. For example, the label information set for the place where no movable objects are allowed to approach is a dangerous place such as a power cabinet. Then, the specific management rule information that does not allow any movable objects to approach here is the specific management rule information. The label information corresponding to the rule is the dangerous place such as the power cabinet. The control can be cancelled when the professional personnel repair it. In daily monitoring, the place is not allowed to be touched. After setting the management rule information base and the label information, multiple management types within the target monitoring range are extracted to obtain multiple management type features. Based on the management type features, the corresponding label information is matched in the management rule information base. Based on the label information, one or more of the corresponding general management rule information and the specific management rule information are obtained as the management rule information. Finally, the management rule information is bound to the corresponding management space and used as the control information. Here, the environment within the target monitoring range can be accurately analyzed, so that subsequent accurate video image acquisition and analysis can be performed, with accurate and clear monitoring and control effects;

[0062] In one embodiment, the information collection module includes:

[0063] A real-time acquisition unit, used to respectively acquire video image information in real time from multiple monitoring devices based on corresponding monitoring information, and to match the video image information with corresponding management spaces to obtain integrated video image information;

[0064] An information construction unit, used for extracting and integrating key target information in the video image information based on the monitoring information, and constructing the key target information into the monitoring model to obtain a behavior space model;

[0065] It should be noted that after the monitoring equipment and the monitoring management rule information are arranged, it is necessary to respectively collect video image information in real time based on the corresponding monitoring information for multiple monitoring devices, and correspond the video image information with the corresponding management space to obtain integrated video image information, so as to obtain the video dynamic information within the target monitoring range, and then extract and integrate the key target information in the video image information based on the monitoring information, and build the key target information into the monitoring model to obtain the behavior space model, which is the combination of the key target information in the video image information and the monitoring model, and can better analyze the key target information through the three-dimensional monitoring model, and then can be used for the subsequent judgment and prompt of the behavior of the key target information through the management rule information, and has a good key target information analysis function;

[0066] In one embodiment, the information construction unit includes:

[0067] A monitoring unit, used to respectively obtain video image information of a plurality of monitoring devices and put the video image information into a management space;

[0068] A splicing unit, used to obtain common feature information between adjacent contacting management spaces, and splice the video image information corresponding to the adjacent contacting management spaces based on the common feature information and according to the time of video image information acquisition to obtain integrated video image information;

[0069] A data extraction unit, used to extract key target information in the integrated video image information based on the management rule information in the monitoring information, wherein the key target information includes a key target shape and a key target position trajectory;

[0070] A splicing construction unit is used to perform three-dimensional construction based on the key target shape in the key target information to obtain a key target model, and construct the key target model in the monitoring model according to the key target position trajectory to obtain a behavior space model;

[0071] It should be noted that after the monitoring equipment and the corresponding monitoring range information and management rule information are set up, the monitoring equipment is put into use for monitoring. During the monitoring process, the monitoring equipment collects video images based on the monitoring information to obtain video image information, and the video image information is put into the corresponding management space. At this time, the corresponding video image information between adjacent management spaces will have overlapping collection parts. Therefore, based on the common feature information between adjacent contacting management spaces, the video image information corresponding to the adjacent contacting management spaces is spliced ​​based on the common feature information and according to the time of video image information collection to obtain integrated video image information. Then, based on the management rule information in the monitoring information, the key target information in the integrated video image information is extracted, where the key target information includes the key target shape and the key target position trajectory. The key target here is based on the corresponding monitoring information. The key target information is extracted from the management rule information. For example, the key targets extracted from the driving lane are vehicles, pedestrians or animals, and the key targets extracted from the green area are pedestrians, vehicles and other information. Acquiring the key target information is to obtain the key target shape and key target position trajectory. The key target position trajectory is the acquisition of the real-time trajectory position. While collecting the video image information, it is spliced ​​to obtain the integrated video image information, and the key target information is also extracted. Here, the splicing unit, the data extraction unit and the splicing construction unit are operated while collecting. The operation in the splicing construction unit is to perform three-dimensional construction based on the key target shape in the key target information to obtain a key target model. The key target model is constructed in the monitoring model according to the key target position trajectory to obtain a behavior space model, which can better analyze the behavior of the key target, has a good behavior analysis effect, and is convenient for subsequent safety monitoring.

[0072] In one embodiment, the verification module includes:

[0073] A security setting unit, used to set security monitoring information based on a behavior space model;

[0074] A verification unit, used to obtain behavior information of a key target in a behavior space model based on security monitoring information, wherein the behavior information includes speed information of the key target and range information of the key target, and to verify the behavior information based on the monitoring information to obtain a verification result;

[0075] It should be noted that after determining the behavior space model, it is necessary to set security monitoring information for the space model. The security monitoring information here is used to obtain behavior information after monitoring the behavior of the key targets in the behavior space model. The behavior information includes the speed information of the key targets and the range information of the key targets. The behavior information is verified based on the monitoring information to obtain a verification result. The verification result is the position of the key target in the security monitoring information. It is then used to judge whether the behavior information of the key targets meets the monitoring information, and can perform security warnings, which is convenient for subsequent security monitoring management.

[0076] In one embodiment, the security setting unit includes:

[0077] A layout unit, used for selecting a spatial range in the behavior space model based on the behavior space model, and setting a plurality of evenly distributed monitoring points based on the spatial range;

[0078] A space division unit is used to divide the corresponding monitoring points according to the management space, and bind the monitoring points with the management space;

[0079] An enabling unit, used for enabling the monitoring point based on the management rule information corresponding to the management space to obtain the monitoring point function information, wherein the monitoring point function information includes speed monitoring and range monitoring;

[0080] It should be noted that in order to better obtain the behavioral information of key targets, it is necessary to prepare in the behavioral space model. Here, it is necessary to select the spatial range in the behavioral space model, and set multiple evenly distributed monitoring points based on the spatial range. Here, the spatial range in the spatial model is selected as the height. Generally, there should not be too much monitoring at the height. In order to improve the monitoring efficiency, the height of the key target analysis is limited, and then multiple evenly distributed monitoring points are set in the spatial range. There are different management rule information in different spatial ranges. The corresponding monitoring points are divided according to the management space, and the monitoring points are bound to the management space. Based on the management rule information corresponding to the management space, the monitoring points are enabled to obtain the monitoring point function information. The functions of the monitoring points: some need to monitor the location of the personnel, and some need to monitor the speed of the vehicle. Therefore, different monitoring ranges need to be monitored. The monitoring points within the range are empowered. Empowerment here means the granting of functions. The speed of the vehicle can be obtained through the monitoring points. The method for obtaining the speed is to obtain the speed of the covered monitoring points in real time based on time, and to obtain the vehicle speed according to the distance between the monitoring points in the behavior space model, or according to the relationship between vehicles and pedestrians based on the management rule information. If there are overlapping monitoring points between pedestrians and vehicles, there may be a collision. If there is a need to monitor the safety of the surrounding environment of the equipment in the management space, for example, the power room or the power cabinet needs to monitor the environment around it to avoid unrelated personnel approaching. Set the safety range of the equipment that needs to be monitored in the management space, and use the monitoring points within the safety range as the monitoring of the monitoring equipment. If there is a moving object entering the safety range and covering the monitoring point, an alarm is issued. The monitoring accuracy is good and the response is fast, and it has a good remote monitoring and autonomous analysis function.

[0081] In one embodiment, the determination module includes:

[0082] A security formulation unit, used to set security conditions based on management rule information corresponding to the management space;

[0083] A judgment unit, used to lock key target information corresponding to the behavior space model that does not meet the safety conditions, and transmit the behavior space model to the management end for prompting;

[0084] It should be noted that the safety conditions here are safety conditions set in the corresponding management rule information. For example, the speed of the vehicle in the driving lane is equal to or less than V, which means it is safe; the absence of vehicles in the green area means it is safe; the absence of moving objects in the safety range of the power room means it is safe. The key target information corresponding to the behavior space model that does not meet the safety conditions is locked, and the behavior space model is transmitted to the management end for prompting. When the key target information that does not meet the safety conditions is found, the behavior space model is transmitted to the corresponding management end and an early warning is given to the management end. The monitoring video can be analyzed and abnormal situations (when the safety conditions are not met) are transmitted and prompted. When no abnormal situation is found in normal monitoring, the monitoring video is transmitted and stored in real time, and no safety prompts are given to the management end. It has good safety analysis capabilities and reaction rates, can quickly warn of safety situations, and the data analysis is more accurate.

[0085] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A remote security monitoring system based on image recognition, characterized in that: include: An acquisition module is used to acquire working information of multiple monitoring devices, wherein the working information of the monitoring devices includes type information of the monitoring devices and corresponding monitoring range information, and to build a monitoring model based on the target monitoring range; A partitioning module, connected to the acquisition module, is used to partition the space based on the monitoring model to obtain multiple management spaces, set management rule information based on the multiple management spaces, and bind the management rule information with the corresponding management space as monitoring information; An information collection module, connected to the division module, is used to collect video image information in real time based on monitoring information, extract key target information from the video image information, and construct a behavior space model of the key target based on the key target information; A verification module, connected to the information collection module, is used to verify the behavior space model based on the monitoring information to obtain verification result information; The judgment module is connected to the verification module and is used to lock the key target information corresponding to the behavior space model that does not meet the safety conditions and prompt it to the management end; The acquisition module comprises: An information acquisition unit, used to acquire a target monitoring range, and acquire monitoring device type information and corresponding monitoring range information of multiple monitoring devices within the target monitoring range as monitoring device working information; An encoding unit, used for numbering the monitoring devices based on the working information of the monitoring devices to obtain the identity information of the monitoring devices, and binding the identity information of the monitoring devices with the working information of the monitoring devices; A construction unit, used to obtain structural information within the target monitoring range, construct a three-dimensional space model based on the structural information within the target monitoring range, construct the monitoring equipment working information in the three-dimensional space model and mark the corresponding monitoring equipment identity information to obtain a monitoring model; The division module comprises: A division unit, used for obtaining multiple management types within a target monitoring range based on the structural information, dividing the target monitoring range based on the multiple management types to obtain multiple management ranges, and dividing and marking the multiple management ranges in the monitoring model to obtain multiple management spaces; A setting unit, used for setting a management rule information base, wherein the management rule information base includes a plurality of general management rule information and a plurality of specific management rule information; A setting unit, used for setting corresponding label information based on the general management rule information and the specific management rule information corresponding to the big data, wherein the label information includes a plurality of management type features; A matching unit, configured to extract multiple management types within a target monitoring range to obtain multiple management type features, match corresponding tag information in a management rule information base based on the management type features, and obtain one or more of corresponding general management rule information and specific management rule information as management rule information based on the tag information; A management unit, used to bind the management rule information with the corresponding management space and use it as control information; The information collection module includes: A real-time acquisition unit, used to respectively acquire video image information in real time from multiple monitoring devices based on corresponding monitoring information, and to match the video image information with corresponding management spaces to obtain integrated video image information; An information construction unit, used for extracting and integrating key target information in the video image information based on the monitoring information, and constructing the key target information into the monitoring model to obtain a behavior space model; The information construction unit includes: A monitoring unit, used to respectively obtain video image information of a plurality of monitoring devices and put the video image information into a management space; A splicing unit, used to obtain common feature information between adjacent contacting management spaces, and splice the video image information corresponding to the adjacent contacting management spaces based on the common feature information and according to the time of video image information acquisition to obtain integrated video image information; A data extraction unit, used to extract key target information in the integrated video image information based on the management rule information in the monitoring information, wherein the key target information includes a key target shape and a key target position trajectory; The splicing construction unit is used to perform three-dimensional construction based on the key target shape in the key target information to obtain a key target model, and construct the key target model in the monitoring model according to the key target position trajectory to obtain a behavior space model.

2. The remote security monitoring system based on image recognition according to claim 1, characterized in that: The verification module comprises: A security setting unit, used to set security monitoring information based on a behavior space model; A verification unit is used to obtain behavior information of key targets in a behavior space model based on security monitoring information, wherein the behavior information includes speed information of the key targets and range information of the key targets, and to verify the behavior information based on the monitoring information to obtain a verification result.

3. The remote security monitoring system based on image recognition according to claim 2 is characterized in that: The security setting unit comprises: A layout unit, used for selecting a spatial range in the behavior space model based on the behavior space model, and setting a plurality of evenly distributed monitoring points based on the spatial range; A space division unit is used to divide the corresponding monitoring points according to the management space, and bind the monitoring points with the management space; The enabling unit is used to enable the monitoring point based on the management rule information corresponding to the management space to obtain the monitoring point function information, wherein the monitoring point function information includes speed monitoring and range monitoring.

4. The remote security monitoring system based on image recognition according to claim 1, characterized in that: The judging module comprises: A security formulation unit, used to set security conditions based on management rule information corresponding to the management space; The judgment unit is used to lock the key target information corresponding to the behavior space model that does not meet the safety conditions, and transmit the behavior space model to the management end for prompting.