Intelligent linkage response inspection method integrating traditional security and protection and unmanned aerial vehicle

Abstract

The invention discloses an intelligent linkage response inspection method integrating traditional security and protection and an unmanned aerial vehicle, and belongs to the technical field of security and protection, and the method comprises the steps: generating a security and protection information graph of a traditional security and protection system in a target region; unmanned aerial vehicle inspection points are set in the target area according to the security information graph, and the unmanned aerial vehicle inspection points are correspondingly marked in the security information graph; acquiring a security analysis result of the traditional security system in real time; identifying the event background probability of each security event in the current background in a preset security event table in real time, and supplementing the event background probability into the security event table; according to the security analysis result and the security event table, dynamically updating the routing inspection priority of the unmanned aerial vehicle routing inspection points, according to the routing inspection priority of each unmanned aerial vehicle routing inspection point, generating a routing inspection route of the unmanned aerial vehicle, and performing routing inspection by the unmanned aerial vehicle according to the routing inspection route; the rapid and flexible movement characteristic of the unmanned aerial vehicle is fully utilized, the unmanned aerial vehicle can rapidly arrive at a designated area for routing inspection, the routing inspection time is greatly shortened, and the overall routing inspection efficiency is remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of security technology, and specifically provides an inspection method that integrates traditional security and the intelligent linkage response of unmanned aerial vehicles. Background Art

[0002] With the rapid development of society and the continuous advancement of science and technology, security inspections are playing an increasingly important role in ensuring public safety, maintaining social order, and preventing accidents and disasters. However, traditional security inspection methods, such as manual inspections and fixed camera surveillance, have many limitations.

[0003] Manual inspections are not only inefficient but also limited by labor costs, inspector expertise, and the work environment, making them difficult to achieve around-the-clock, comprehensive monitoring. The difficulty and risk of manual inspections are particularly increased in complex terrain, hazardous areas, or large areas. While fixed camera surveillance has improved monitoring efficiency to a certain extent, its limited coverage makes it difficult to cover all critical areas. Furthermore, fixed cameras have drawbacks such as fixed viewing angles and susceptibility to obstructions, making them unable to meet the needs of dynamic and flexible inspections.

[0004] Based on this, the present invention provides an inspection method that integrates traditional security and the intelligent linkage response of drones. Summary of the Invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an inspection method that integrates traditional security and intelligent linkage response of drones.

[0006] The purpose of the present invention can be achieved through the following technical solutions: The inspection method integrates traditional security and drone intelligent linkage response, including: Generate a security information map of a traditional security system in a target area; set up drone inspection points in the target area according to the security information map, and mark the drone inspection points accordingly in the security information map; Obtain security analysis results of traditional security systems in real time; identify the event background probability of each security event in the preset security event table in the current context in real time, and add the event background probability to the security event table; The inspection priority of the drone inspection points is dynamically updated according to the security analysis results and the security event table. The drone inspection route is generated according to the inspection priority of each drone inspection point, and the drone conducts inspections according to the inspection route.

[0007] Furthermore, the method for generating the security information graph includes: Obtaining a target area map and historical security data of a traditional security system, and identifying a monitoring area and security system information corresponding to the traditional security system based on the historical security data, wherein the security system information includes a monitoring device location and monitoring device information; The target area map is marked accordingly according to the monitoring area and security system information, and the current target area map is marked as a security information map.

[0008] Furthermore, drone inspection points are set up in the target area according to the security information map, including: Conduct security assessments based on security information maps and determine the inspection areas for drones; Set up drone inspection points according to the inspection area.

[0009] Furthermore, a security assessment is conducted based on the security information map, including: Identify security blind spots and monitoring areas of traditional security systems based on security information graphs; set up a security event table, which is used to count corresponding security events; Conduct security simulation analysis based on the historical security data and security event table of the traditional security system to determine the simulation analysis results of the traditional security system for the corresponding security events in the monitoring area. The simulation analysis results include simulation pass and simulation fail. Determine the security failure area based on the simulation analysis results, and mark the security failure area with a corresponding security event label; Integrate unqualified security areas and security blind spots into inspection areas.

[0010] Furthermore, security simulation analysis is performed based on historical security data and security event tables, including: Identify each security event corresponding to the security event table, perform feature extraction on historical security data based on the security event, and obtain simulation material data of the security event; A simulation evaluation model is established. The expression of the simulation evaluation model is: ; Where: s is the input data, which represents the material simulation result of the corresponding security event; the output data is the security simulation value MP(s), which is 1 or 0; Analyze the simulation results of the material through the simulation evaluation model to obtain corresponding security simulation values; When the security simulation value is 1, the simulation analysis result is qualified; When the security simulation value is 0, the simulation analysis result is simulation failure.

[0011] Furthermore, drone inspection points are set up according to the inspection area, including: Set inspection point setting standards. The inspection point setting standards require that the drone can meet the user's security inspection requirements for the target area after inspecting each inspection point. Simulate the inspection point setting of the inspection area based on the inspection point setting standards to obtain several candidate setting solutions that meet the inspection point setting standards. Screen the selected setting plans, determine the target setting plan, and set the drone inspection points according to the target setting plan.

[0012] Further, the selected setting schemes are screened, including: Count the share of each security event based on the historical security data of the target area; set the event risk value of each security event, and the value range of the event risk value is [0, 100]; The weight coefficient of the security event is calculated according to the weight formula, which is: ; Where: δ i Represents the weight coefficient of the corresponding security event, i represents the corresponding security event, i=1, 2, ..., n, n is the number of security events; FZ i Indicates the event risk value of the corresponding security event; β i Indicates the share of corresponding security incidents; An inspection simulation is performed on each candidate setting scheme according to historical security data and weight coefficients to determine the priority of the candidate setting scheme, and the candidate setting scheme with the highest priority is marked as the target setting scheme.

[0013] Furthermore, the determination of the background probability of an event includes: Set background factors for each security event, and identify the background factor combination corresponding to historical security data based on the background factors; and calculate the probability of the event corresponding to the corresponding background factor combination in real time based on historical security data; Combine the background factors of security events with the same probability of occurrence to obtain the background classification of the corresponding security events; The current background corresponding to each security event is identified in real time according to background factors; the corresponding event occurrence probability is matched according to the current background, and the event occurrence probability is marked as the event background probability of the security event.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The intelligent linkage response inspection method integrating traditional security and drones provided by the present invention fully utilizes the fast and flexible mobility characteristics of drones compared to traditional manual inspection methods, can quickly reach designated areas for inspections, greatly shortens inspection time, significantly improves overall inspection efficiency, and can respond to various security needs in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 Flow chart of the method of the present invention; Figure 2 This is a flow chart of the UAV low-altitude target recognition process of the present invention. DETAILED DESCRIPTION

[0017] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] like Figures 1 to 2 As shown in the figure, the inspection method that integrates traditional security and drone intelligent linkage response includes: Step 1: Obtain historical security data corresponding to the traditional security system in the target area. The historical security data includes monitoring data, processing result data and other related data. Traditional security system; the target area refers to the area that needs to be protected; set the security information map of the traditional security system for the target area according to the historical security data; the security information map is set according to the target area map, and the monitoring area of ​​the traditional security system and the location of the monitoring equipment, equipment information and other related security system information are marked on the target area map to form a security information map.

[0019] Set up corresponding drone inspection points in the target area according to the security information map, and mark the drone inspection points accordingly in the security information map.

[0020] In one embodiment, drone inspection points can be set in an existing manner, such as manually by personnel with professional experience; or an intelligent model based on deep learning algorithms can be established for intelligent analysis to set corresponding inspection points, that is, the drone inspection points can be dynamically adjusted according to changes in the security information map.

[0021] In one embodiment, setting up drone inspection points in a target area based on a security information map includes: Conduct security assessments based on security information maps to identify areas where traditional security systems fall short of standards and require additional drone inspections. Set up drone inspection points according to the inspection area.

[0022] In one embodiment, security assessment is performed based on the security information map. The assessment can be performed based on existing methods, and security blind spots can be used as inspection areas.

[0023] In one embodiment, in actual applications, the monitoring quality of the monitoring areas of some traditional security systems may not meet the standards, which may affect security analysis. Therefore, the monitoring areas are analyzed and areas where the monitoring does not meet the standards are marked as inspection areas. The judgment can be made based on existing methods, such as manual judgment.

[0024] In one embodiment, performing a security assessment based on the security information graph includes: Identify security blind spots and monitoring areas of traditional security systems based on security information maps; obtain various security events in the target area for security analysis, such as intrusion, fire, and other related security events, and organize them into a security event table. The security event table is generally obtained based on relevant departments, etc., and can also be adjusted and set according to the needs of managers in the target area; Conduct security simulation analysis based on historical security data and security event tables to determine whether the traditional security system can meet the analysis requirements of the corresponding security event, that is, whether the monitoring data of the traditional security system can meet the identification analysis of the security event; determine the corresponding simulation analysis results, which include simulation pass and simulation fail; The areas corresponding to the simulation failure results are integrated into security failure areas, and the corresponding security event labels are marked for the security failure areas; this facilitates targeted monitoring during subsequent drone inspections; that is, when it is determined that the traditional security system fails the simulation of the corresponding security event in the monitoring area, it can be determined which monitoring areas have failed the historical security data, and then the security failure areas can be determined.

[0025] Integrate unqualified security areas and security blind spots into inspection areas.

[0026] In one embodiment, security simulation analysis is performed based on historical security data and a security event table. The analysis can be performed based on existing methods, such as identifying and analyzing historical security data according to existing security event identification and analysis methods to determine simulation analysis results.

[0027] In one embodiment, security simulation analysis is performed based on historical security data and a security event table, including: Identify each security event corresponding to the security event table, extract features from historical security data based on the security event, and obtain simulation material data corresponding to the security event, that is, historical security data corresponding to the security event, to facilitate subsequent security event identification simulation; analyze the simulation material data to obtain corresponding material simulation results, such as the accuracy rate of security event identification after multiple simulations, and subsequently determine whether the security meets the requirements of the user's security standards; A simulation evaluation model is established. The expression of the simulation evaluation model is: ; Where: s is the input data, representing the material simulation result of the corresponding security event, and the security standard is set according to user needs; the output data is the security simulation value MP(s), which is 1 or 0; the corresponding training set is set using the corresponding historical security data for training; Analyze the simulation results of the corresponding materials through the simulation evaluation model to obtain the corresponding security simulation value; When the security simulation value is 1, the simulation analysis result is qualified; When the security simulation value is 0, the simulation analysis result is simulation failure.

[0028] In one embodiment, drone inspection points are set according to the inspection area. When the inspection area is determined, drone inspection points can be set based on the existing method, as long as the drone can meet the security inspection requirements after inspecting each inspection point; drone inspection points are generally set according to the security events that need to be monitored in the inspection area and the monitoring range of the drone.

[0029] In one embodiment, setting drone inspection points according to the inspection area includes: Set inspection point setting standards. The inspection point setting standards require that the drone can meet the user's security inspection requirements for the target area after inspecting each inspection point; simulate the inspection point setting of the inspection area according to the inspection point setting standards, obtain several inspection point setting plans that meet the inspection point setting standards, and mark them as candidate setting plans. That is, according to the benchmark setting requirements of the drone inspection point, simulate multiple inspection point setting plans that can be adopted in the inspection area, and then evaluate whether the inspection point setting plan meets the inspection point setting standards. The inspection point setting plan that meets the inspection point setting standards is marked as a candidate setting plan; Screen the selected setting plans, determine the target setting plan, and set the drone inspection points according to the target setting plan.

[0030] In one embodiment, the selection of setting schemes can be screened based on existing methods, such as determining the priority of the corresponding setting schemes by inspection duration, number of inspection points, etc., and then determining the target setting scheme; other priority algorithms can also be applied to determine the priority.

[0031] In one embodiment, screening the configuration options to be selected includes: The share of corresponding security events is calculated based on the historical security data of the target area; and the corresponding event risk value is set according to the risk size of the corresponding security event. The event risk value is determined according to the maximum economic loss. The value range of the event risk value is [0, 100]. The event risk value corresponding to the maximum economic loss is 100, and the event risk value of no economic loss or no risk is 0. The maximum economic loss is determined based on various potential losses in the target area, and is generally determined by comparing with the historical maximum economic losses of similar areas. That is, the platform conducts real-time statistics of the corresponding maximum economic losses based on the target area and marks its event risk value as 100; the highest economic loss corresponding to the corresponding security event is identified based on the historical security data, and compared with the maximum economic loss corresponding to the event risk value of 100 in turn to determine the event risk value of the security event; the event risk value of the security event can also be determined in other ways.

[0032] Calculate the weight coefficient of the corresponding security event according to the weight formula. The weight formula is: ; Where: δ i Represents the weight coefficient of the corresponding security event, i represents the corresponding security event, i=1, 2, ..., n, n is the number of security events; FZ i Indicates the event risk value of the corresponding security event; β i Indicates the share of corresponding security incidents; According to the historical security data and weight coefficients, inspection simulation is carried out on each candidate setting scheme to determine the priority of the corresponding candidate setting scheme, and the candidate setting scheme with the highest priority is marked as the target setting scheme; that is, the security events corresponding to the historical security data are used to simulate the inspection efficiency under the corresponding candidate setting scheme, and the priority is determined in combination with the weight coefficient; after the weight coefficient of each security event is clarified, the priority of the candidate setting scheme can be determined based on various existing methods; for example, the efficiency of completing the inspection task according to the candidate setting scheme and inspection requirements can be estimated based on the occurrence of security events in each area, and then the priority can be determined.

[0033] Step 2: Obtain the security analysis results of the traditional security system in real time. The security analysis results indicate whether the traditional security system has identified abnormal situations, security events, etc.; based on historical security data, calculate the probability of each security event in the security event table in the current context in real time, mark it as the event background probability, and add the event background probability to the security event table; The inspection priority of the drone inspection points is dynamically updated according to the security analysis results and the security event table. The drone inspection route is generated according to the inspection priority of each drone inspection point, and the drone conducts inspections according to the inspection route.

[0034] In one embodiment, the event background probability of each security event in the security event table in the current background is counted in real time based on historical security data. The background factors applicable to the event probability of each security event, such as weather, time, etc., are first determined. This is because the event probability of different security events in different situations is different. For example, the event probability of sneak attacks in nighttime and daytime periods is different. Therefore, statistics can be performed based on historical security data. If the target area does not have sufficient historical security data, that is, the data volume is small and the representativeness is low, it can be supplemented by collecting historical security data similar to the target area. Subsequently, the time background probability of the corresponding security event in the current background can be determined according to the existing method.

[0035] In one embodiment, determining the background probability of an event includes: Set the background factors corresponding to each security event, and identify the background factor combination corresponding to the historical security data based on the background factors, that is, the historical background; and calculate the probability of the event corresponding to the corresponding background factor combination in real time based on the historical security data; Combining the background factors of security events with the same probability of occurrence to obtain the corresponding background classification, that is, the probability of occurrence of events within this background classification is the same; The background of each security event at the current time is determined in real time based on the background factors corresponding to each security event; the corresponding event occurrence probability is matched according to the background, and the event occurrence probability is marked as the event background probability of the security event.

[0036] In one embodiment, the inspection priority of the drone inspection point is dynamically updated according to the security analysis results and the security event table, and can be determined according to the existing priority determination method and the drone path planning method; for example, when the security analysis result is that an abnormal event or a security event is identified, the inspection priority of the drone inspection point at that location is raised to the highest level, and when the security analysis result is normal, the priority of the drone inspection point is dynamically evaluated according to the event background probability and the inspection interval of each security event; for example, each security event corresponding to the drone inspection point is identified, the product between the event background probability and the event risk value is calculated, marked as the event value, and each event value is accumulated to obtain a fixed event value for the drone inspection point; the priority value is subsequently adjusted in combination with the inspection interval time, that is, the initial priority value is a fixed event value, and the priority value is subsequently dynamically increased according to the inspection interval time, and the priority value can also be adjusted considering the inspection efficiency; there are multiple ways to determine the inspection priority.

[0037] In this embodiment, the present invention proposes a UAV low-altitude target recognition method of "dynamic and static area collaboration + waypoint triggering multi-algorithm", which specifically includes: Regional planning and triggering mechanisms: Static detection area: Before the drone takes flight, a predefined detection area (such as a 1km area around a school wall) is delineated on a 2D / 3D map through a ground station or the cloud. When the drone enters this area, basic detection algorithms (such as infrared thermal imaging or AI vision) are automatically activated.

[0038] Dynamic ROI area: Set a dynamically adjustable detection area at each preset waypoint. Based on the real-time position, altitude, and sensor data (such as LiDAR point cloud), the size and shape of the ROI are dynamically adjusted (such as rectangle, sector), and bound to a specific algorithm (such as enabling "rooftop person detection" at waypoint A and "fire smoke recognition" at waypoint B).

[0039] Flexible configuration and switching of multiple algorithms: Modular algorithm library: Build a lightweight algorithm library (such as YOLO for target detection and UNet for semantic segmentation), supporting the calling of different algorithms by waypoints and target types.

[0040] Priority logic: When a drone enters both a static region and a dynamic ROI, the high-precision algorithm bound to the dynamic ROI is prioritized, while retaining low-power monitoring of the static region (such as detecting abnormal motion through background subtraction).

[0041] Collaborative optimization of dynamic and static detection: Data fusion mechanism: The global detection results of static areas (such as crowd distribution heat maps) and the high-precision recognition results of dynamic ROIs (such as personnel posture analysis) are temporally and spatially aligned and feature-fused, and the target state is comprehensively determined through a decision tree or Bayesian network.

[0042] Adaptive parameter adjustment: Dynamically adjust detection thresholds and algorithm parameters for dynamic and static areas based on environmental factors (such as light intensity and wind speed) and mission requirements (such as increased sensitivity for emergency fire detection).

[0043] System Architecture: Hardware layer: The drone is equipped with multimodal sensors (visible light camera, infrared sensor, LiDAR) and edge computing units (such as the Jetson series).

[0044] Control layer: The flight control system (such as PX4) is linked with the algorithm scheduling module to realize waypoint triggering algorithm switching.

[0045] Software layer: algorithm library, dynamic ROI generation module, dynamic and static data fusion engine.

[0046] Through the collaboration of dynamic and static areas, ineffective calculations are reduced and the power consumption of edge devices is lowered (the measured computing amount is reduced by 40%). The complementary mechanism of dynamic ROI and static area can improve the target recognition accuracy in complex scenarios to more than 95% (an increase of 20% compared to a single method). It supports the on-demand activation of multiple algorithms on the same route (such as simultaneous detection of fires, climbing over walls, and crowds in school scenarios), reducing the cost of repeated flights. Through the fusion of dynamic and static data, it effectively copes with lighting changes and occlusion problems during low-altitude flight, and reduces the false alarm rate by 30%.

[0047] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0048] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The inspection method integrates traditional security and drone intelligent linkage response, which is characterized by: include: Generate security information map of traditional security system in target area; Setting up drone inspection points in the target area according to the security information map, and marking the drone inspection points accordingly in the security information map; Obtain security analysis results of traditional security systems in real time; Real-time identification of the event background probability of each security event in the preset security event table in the current context, and adding the event background probability to the security event table; The inspection priority of the drone inspection points is dynamically updated according to the security analysis results and the security event table. The drone inspection route is generated according to the inspection priority of each drone inspection point, and the drone conducts inspections according to the inspection route.

2. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 1 is characterized in that: The generation method of the security information graph includes: Obtaining a target area map and historical security data of a traditional security system, and identifying a monitoring area and security system information corresponding to the traditional security system based on the historical security data, wherein the security system information includes a monitoring device location and monitoring device information; The target area map is marked accordingly according to the monitoring area and security system information, and the current target area map is marked as a security information map.

3. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 1 is characterized in that: Set up drone inspection points in the target area according to the security information map, including: Conduct security assessments based on security information maps and determine the inspection areas for drones; Set up drone inspection points according to the inspection area.

4. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 3 is characterized in that: Conduct a security assessment based on the security infographic, including: Identify security blind spots and monitoring areas of traditional security systems based on security information graphs; set up a security event table, which is used to count corresponding security events; Conduct security simulation analysis based on the historical security data and security event table of the traditional security system to determine the simulation analysis results of the traditional security system for the corresponding security events in the monitoring area. The simulation analysis results include simulation pass and simulation fail. Determine the security failure area based on the simulation analysis results, and mark the security failure area with a corresponding security event label; Integrate unqualified security areas and security blind spots into inspection areas.

5. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 4 is characterized in that: Conduct security simulation analysis based on historical security data and security event tables, including: Identify each security event corresponding to the security event table, perform feature extraction on historical security data based on the security event, and obtain simulation material data of the security event; A simulation evaluation model is established. The expression of the simulation evaluation model is: ; Where: s is the input data, which represents the material simulation result of the corresponding security event; the output data is the security simulation value MP(s), which is 1 or 0; Analyze the simulation results of the material through the simulation evaluation model to obtain corresponding security simulation values; When the security simulation value is 1, the simulation analysis result is qualified; When the security simulation value is 0, the simulation analysis result is simulation failure.

6. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 3 is characterized in that: Set up drone inspection points according to the inspection area, including: Set inspection point setting standards. The inspection point setting standards require that the drone can meet the user's security inspection requirements for the target area after inspecting each inspection point. Simulate the inspection point setting of the inspection area based on the inspection point setting standards to obtain several candidate setting solutions that meet the inspection point setting standards. Screen the selected setting plans, determine the target setting plan, and set the drone inspection points according to the target setting plan.

7. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 6 is characterized in that: Filter the available configuration options, including: Count the share of each security event based on the historical security data of the target area; set the event risk value of each security event, and the value range of the event risk value is [0, 100]; The weight coefficient of the security event is calculated according to the weight formula, which is: ; Where: δ i Represents the weight coefficient of the corresponding security event, i represents the corresponding security event, i=1, 2, ..., n, n is the number of security events; FZ i Indicates the event risk value of the corresponding security event; β i Indicates the share of corresponding security incidents; An inspection simulation is performed on each candidate setting scheme according to historical security data and weight coefficients to determine the priority of the candidate setting scheme, and the candidate setting scheme with the highest priority is marked as the target setting scheme.

8. The inspection method integrating traditional security and intelligent linkage response of drones according to claim 1 is characterized in that: Determination of the background probability of an event, including: Set background factors for each security event, and identify the background factor combination corresponding to historical security data based on the background factors; and calculate the probability of the event corresponding to the corresponding background factor combination in real time based on historical security data; Combine the background factors of security events with the same probability of occurrence to obtain the background classification of the corresponding security events; The current background corresponding to each security event is identified in real time according to background factors; the corresponding event occurrence probability is matched according to the current background, and the event occurrence probability is marked as the event background probability of the security event.

Images