Smart airport boundary safety monitoring method, system and device and storage medium

Abstract

The invention relates to the technical field of safety early warning, and discloses a smart airport boundary safety monitoring method, system and device and a storage medium. The method comprises the following steps: collecting monitoring information through a multi-type sensor array and performing time synchronization processing to obtain multi-source synchronous monitoring data; performing spatial registration and noise reduction processing on the data, and generating situation fusion data according to a sensor reliability configuration weight; extracting spatio-temporal features to construct intrusion feature representation, and obtaining a threat assessment result; determining an early warning level and matching response measures to form a disposal scheme; an optimal prevention and control scheme is obtained through situation deduction; and optimizing the monitoring parameters according to the execution data, and generating an updated monitoring strategy. Through intelligent analysis and fusion processing of the multi-source data, accurate identification and quick response to the security threats of the enclosure are realized.

Description

Technical Field

[0001] The present application relates to the field of security warning, and in particular to a smart airport perimeter security monitoring method, system, equipment and storage medium. Background Art

[0002] With the continuous expansion of airport scale and the increasing security requirements, perimeter security monitoring faces severe challenges. Existing airport perimeter security monitoring methods mainly rely on single technical means such as video surveillance and infrared detection, and adopt a fixed threshold alarm judgment mechanism. In practical applications, existing technologies respond to security threats through manual inspections, fixed sentries and simple linkage disposal, lacking intelligent analysis methods and adaptive disposal mechanisms. At the same time, existing monitoring systems often adopt an independent deployment method, with insufficient information sharing between subsystems, making it difficult to achieve deep data integration and linkage response.

[0003] However, these traditional monitoring methods have obvious shortcomings. First, a single technical means is easily affected by environmental factors, such as weather changes, lighting conditions, etc., resulting in unstable monitoring effects. Second, fixed alarm thresholds cannot adapt to the security needs of different scenarios and time periods, and false alarms and missed alarms often occur. Third, the manual-based disposal method has a slow response speed and is difficult to deal with sudden threats. In addition, the information island problem between subsystems makes it impossible to fully utilize multi-source data for intelligent analysis and decision-making, affecting the overall prevention and control effect. Summary of the invention

[0004] The present application provides a smart airport perimeter security monitoring method, system, device and storage medium, which are used to accurately identify and quickly respond to perimeter security threats through intelligent analysis and fusion processing of multi-source data.

[0005] In the first aspect, the present application provides a smart airport perimeter security monitoring method, which includes: collecting monitoring information of the perimeter area through a multi-type sensor array, performing time synchronization processing on the collected monitoring information through a time synchronization server, and obtaining multi-source synchronous monitoring data; using a spatial registration system to perform spatial registration and noise reduction processing on the multi-source synchronous monitoring data, and dynamically configuring weights according to the reliability of each sensor to generate perimeter situation fusion data; extracting and analyzing spatiotemporal features based on the perimeter situation fusion data, and constructing perimeter intrusion feature representation through multi-level security features , and then analyze it according to the security threat level assessment rules to obtain the perimeter threat assessment results; based on the perimeter threat assessment results, determine the security warning level through the dynamic threshold system, and match the response measures from the plan library based on historical disposal experience to form a perimeter security disposal plan; according to the perimeter security disposal plan, establish a security resource scheduling strategy, and verify the plan through the situation simulation platform to obtain the optimal perimeter prevention and control plan; based on the execution data of the optimal perimeter prevention and control plan, evaluate and screen the perimeter security monitoring parameters, and adjust the monitoring threshold and warning rules according to the screening results to generate an updated perimeter security monitoring strategy.

[0006] On the second aspect, the present application provides a smart airport perimeter security monitoring system, which includes: a processing module for collecting monitoring information of the perimeter area through a multi-type sensor array, and performing time synchronization processing on the collected monitoring information through a time synchronization server to obtain multi-source synchronous monitoring data; a noise reduction module for using a spatial registration system to perform spatial registration and noise reduction processing on the multi-source synchronous monitoring data, and dynamically configure weights according to the reliability of each sensor to generate perimeter situation fusion data; an extraction module for extracting and analyzing spatiotemporal features based on the perimeter situation fusion data, and constructing perimeter intrusion feature representation through multi-level security features , and then analyze it according to the security threat level assessment rules to obtain the perimeter threat assessment results; the matching module is used to determine the security warning level through the dynamic threshold system based on the perimeter threat assessment results, and match the response measures from the plan library based on the historical disposal experience to form a perimeter security disposal plan; the verification module is used to establish a security resource scheduling strategy based on the perimeter security disposal plan, and verify the plan through the situation deduction platform to obtain the optimal perimeter prevention and control plan; the screening module is used to evaluate and screen the perimeter security monitoring parameters based on the execution data of the optimal perimeter prevention and control plan, and adjust the monitoring threshold and warning rules according to the screening results to generate an updated perimeter security monitoring strategy.

[0007] The third aspect of the present application provides a computer device, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate through a bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned smart airport perimeter security monitoring method are performed.

[0008] The fourth aspect of the present application provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the above-mentioned smart airport perimeter security monitoring method.

[0009] In the technical solution provided by the present application, monitoring information of the perimeter area is collected through a multi-type sensor array, and a time synchronization server is used for time synchronization processing, which solves the problem of time consistency of different sensor data and ensures the synchronization and integrity of multi-source monitoring data; a spatial registration system is used to perform spatial registration and noise reduction on multi-source synchronous monitoring data, and dynamic weight configuration is performed according to sensor reliability, effectively improving the accuracy and reliability of data fusion; spatiotemporal feature extraction and analysis are performed on the perimeter situation fusion data, and a perimeter intrusion feature representation is constructed through multi-level security features, thereby achieving accurate identification of intrusion threats; a dynamic threshold system is used to determine the security warning level, and response measures are matched from the plan library based on historical disposal experience, thereby improving the accuracy of the warning and the pertinence of the disposal; by establishing a security resource scheduling strategy and using a situation deduction platform to verify the plan, the optimal configuration of prevention and control resources is achieved; the perimeter security monitoring parameters are evaluated and screened, and the monitoring thresholds and warning rules are dynamically adjusted according to the screening results, and a set of adaptive monitoring strategy update mechanisms are established. The overall solution forms a complete closed-loop monitoring system. From data collection, analysis and processing, threat identification to emergency response, each link is closely connected and data flows smoothly, which greatly improves the intelligence level of airport perimeter security monitoring. The solution makes full use of the complementary advantages of multi-source data and effectively solves the problems existing in traditional monitoring methods through intelligent analysis means and adaptive disposal mechanisms. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0011] Figure 1 This is a schematic diagram of an embodiment of the smart airport perimeter security monitoring method in the embodiment of the present application; Figure 2This is a schematic diagram of the spatial area division of the airport enclosure in the embodiment of the present application; Figure 3 This is a schematic diagram of an embodiment of a smart airport perimeter security monitoring system in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0012] Embodiments of the present application provide a method, system, device and storage medium for intelligent airport perimeter security monitoring. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the smart airport perimeter security monitoring method in the embodiment of the present application includes: Step S101: collecting monitoring information of the enclosed area through a multi-type sensor array, and performing time synchronization processing on the collected monitoring information through a time synchronization server to obtain multi-source synchronous monitoring data; Step S102: Use the spatial registration system to perform spatial registration and noise reduction on the multi-source synchronous monitoring data, and dynamically configure weights according to the reliability of each sensor to generate perimeter situation fusion data; Step S103: extract and analyze the spatiotemporal features based on the perimeter situation fusion data, construct a perimeter intrusion feature representation through multi-level security features, and then analyze it according to the security threat level assessment rules to obtain the perimeter threat assessment result; Step S104: Based on the perimeter threat assessment result, the security warning level is determined through the dynamic threshold system, and response measures are matched from the plan library based on historical disposal experience to form a perimeter security disposal plan; Step S105: Establish a security resource scheduling strategy based on the perimeter security disposal plan, and verify the plan through the situation simulation platform to obtain the optimal perimeter prevention and control plan; Step S106: Evaluate and screen perimeter security monitoring parameters based on the execution data of the optimal perimeter prevention and control plan, adjust monitoring thresholds and warning rules based on the screening results, and generate an updated perimeter security monitoring strategy.

[0014] It is understandable that the execution subject of the present application may be a smart airport perimeter security monitoring system, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0015] Specifically, the monitoring information of the enclosed area is collected through a multi-type sensor array. The multi-type sensor array includes a video monitoring subsystem, an infrared thermal imaging device, a vibration optical fiber and a millimeter wave radar. These devices are responsible for the collection of visible light image data, thermal imaging data, ground vibration data and airspace detection data respectively. The original monitoring information collected by the sensor is grouped and marked according to the monitoring time, and each group of monitoring information is assigned a unique timestamp code. The timestamp code adopts the UTC time format and is accurate to the millisecond level to ensure the time sequence of the data. The monitoring information with timestamps is divided into time windows, and the length of the time window is set to 100 milliseconds. The monitoring information of different sensors in the same time window is correlated and matched to generate time-correlated monitoring information. The signal quality of the time-correlated monitoring information is evaluated, and the evaluation indicators include signal-to-noise ratio, data integrity and signal stability. The signal quality evaluation will eliminate the monitoring information with a signal-to-noise ratio lower than the preset threshold, and the screened valid monitoring information will then be time-synchronized and calibrated. The time synchronization calibration aligns the monitoring information collected by different sensors to a unified time base. During the calibration process, the missing data points are supplemented by a linear interpolation method to generate multi-source synchronous monitoring data.

[0016] The multi-source synchronous monitoring data is processed by spatial registration and noise reduction. The spatial registration is used to convert the coordinate system, unify the data of each sensor into the WGS84 coordinate system, and then identify the spatial overlapping area of ​​different sensor data, and extract the feature data of the overlapping area. The noise intensity of the feature data in the overlapping area is analyzed, and the signal interference model is established to remove the interference signal. For the feature data after noise reduction, the signal fluctuation coefficient of each sensor is calculated, the signal stability of the sensor is evaluated, and the sensor reliability index is obtained. The data fusion weight is calculated according to the sensor reliability index, and the multi-source data weighted fusion matrix is ​​constructed to generate the perimeter situation fusion data. The perimeter situation fusion data is used for spatiotemporal feature extraction and analysis. The fused data is segmented according to the time series, each time period is 5 minutes, and the data change characteristics in each time period are extracted to form time series feature data. At the same time, the monitoring area is divided into several spatial areas, and the data distribution characteristics in each area are calculated to obtain spatial feature data. The temporal feature data and the spatial feature data are correlated and analyzed, and a feature correlation matrix is ​​established to construct multi-level security features. Abnormal behavior analysis is performed based on multi-level security features, the intrusion behavior feature sequence is extracted, and the perimeter intrusion feature representation is generated. The matching degree between the boundary intrusion feature representation and the security threat level assessment rules is calculated, the threat level is marked, and the boundary threat assessment result is output.

[0017] The results of perimeter threat assessment are processed by the dynamic threshold system to determine the security warning level. The threat assessment results are analyzed in time windows to extract the fluctuation characteristics of the threat level and obtain threat trend data. The threat trend data is compared and analyzed with historical data, and the threat deviation value is calculated to obtain the threat threshold parameter. The warning interval is divided according to the threat threshold parameter, and the warning level is divided into four levels: level 1 (red), level 2 (orange), level 3 (yellow) and level 4 (blue). The similarity between the security warning level data and the historical disposal experience is calculated to extract the disposal elements and form the disposal matching data. The disposal matching data is analyzed for scene adaptability, and appropriate response measures are selected from the plan library to generate the perimeter security disposal plan. The prevention and control task elements are extracted according to the perimeter security disposal plan, and the required human and equipment resources are statistically analyzed to obtain the scheduling demand data. The scheduling demand data is sorted and sorted according to the resource type, and compared with the existing resource status to form the resource matching data. The resource matching data is analyzed for spatiotemporal distribution, the resource scheduling priority is divided, and the security resource scheduling strategy is established. The security resource scheduling strategy is input into the situation simulation platform for scenario simulation analysis to obtain the simulation verification data. Quantitatively evaluate the prevention and control effects in the simulation verification data, extract key indicator parameters, and generate program evaluation data. Select the strategy combination with the best prevention and control effect from the program evaluation data to form the optimal perimeter prevention and control plan.

[0018] The execution data of the optimal perimeter control plan is used for system optimization and update. Extract perimeter security monitoring parameters from the execution data, classify and organize them according to parameter types, and obtain monitoring parameter sets. Perform statistical analysis on the monitoring parameter set, calculate the fluctuation range and distribution characteristics of the parameters, and generate parameter evaluation data. Based on the parameter evaluation data, eliminate abnormal fluctuation parameters, extract key monitoring indicators, and obtain parameter screening results. Analyze the correspondence between each monitoring threshold and the actual threat in the parameter screening results, and construct a monitoring threshold mapping table. Compare the monitoring threshold mapping table with the existing warning rules, mark the rule items that need to be updated, and generate a rule update list. Replace the original rules according to the updated items in the rule update list to generate an updated perimeter security monitoring strategy.

[0019] For example, in a section of the perimeter of an airport, the video surveillance subsystem captured an abnormal moving target, the thermal imaging device detected heat changes, the vibration fiber detected ground vibration signals, and the millimeter wave radar detected flying objects in the air. These raw data are processed by time synchronization to align all data to the same time reference point. The spatial registration system uniformly converts the coordinate information of different sensor data and calculates the sensor signal stability index, with a weight of 0.3 for video data, 0.25 for thermal imaging data, 0.25 for vibration data, and 0.2 for radar data. After the fused data is extracted through spatiotemporal features, it is found that the motion trajectory, thermal features, and vibration features of the abnormal target all point to perimeter intrusion behavior. The system matches the threat level assessment rules based on multi-dimensional features and determines it as a secondary threat. Based on the threat level, the system matches the corresponding disposal plan from the plan library and dispatches the nearest security personnel and equipment for disposal. The data transmission, processing, and analysis of the entire process form a closed loop to ensure the timeliness and accuracy of perimeter security monitoring.

[0020] In the embodiment of the present application, monitoring information of the perimeter area is collected by a multi-type sensor array, and a time synchronization server is used for time synchronization processing, which solves the problem of time consistency of different sensor data and ensures the synchronization and integrity of multi-source monitoring data; a spatial registration system is used to perform spatial registration and noise reduction on multi-source synchronous monitoring data, and dynamic weight configuration is performed according to sensor reliability, effectively improving the accuracy and reliability of data fusion; spatiotemporal feature extraction and analysis are performed on the perimeter situation fusion data, and a perimeter intrusion feature representation is constructed through multi-level security features, thereby achieving accurate identification of intrusion threats; a dynamic threshold system is used to determine the security warning level, and response measures are matched from the plan library based on historical disposal experience, thereby improving the accuracy of the warning and the pertinence of the disposal; by establishing a security resource scheduling strategy and using a situation deduction platform to verify the plan, the optimal configuration of prevention and control resources is achieved; the perimeter security monitoring parameters are evaluated and screened, and the monitoring thresholds and warning rules are dynamically adjusted according to the screening results, and a set of adaptive monitoring strategy update mechanisms are established. The overall solution forms a complete closed-loop monitoring system. From data collection, analysis and processing, threat identification to emergency response, each link is closely connected and data flows smoothly, which greatly improves the intelligence level of airport perimeter security monitoring. The solution makes full use of the complementary advantages of multi-source data and effectively solves the problems existing in traditional monitoring methods through intelligent analysis means and adaptive disposal mechanisms.

[0021] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) A multi-type sensor array scans the perimeter area at multiple angles. The video monitoring subsystem collects visible light image data, the infrared thermal imaging device collects thermal imaging data, the vibration optical fiber collects ground vibration data, and the millimeter wave radar collects airspace detection data to obtain original monitoring information. (2) The original monitoring information is grouped and marked according to the monitoring time, and a unified timestamp code is assigned to each group of original monitoring information to obtain monitoring information with a timestamp; (3) Divide the monitoring information with timestamps into time windows, associate and match the monitoring information in the same time window, and obtain time-correlated monitoring information; (4) Perform signal quality assessment on the time-correlated monitoring information, eliminate monitoring information with a signal-to-noise ratio lower than a preset threshold, and obtain effective monitoring information; (5) Perform time synchronization calibration on the effective monitoring information, align the monitoring information collected by different sensors to a unified time reference, and obtain time-calibrated monitoring information; (6) Time calibration is used to check the data integrity of monitoring information, interpolate the time points of missing data, and generate multi-source synchronous monitoring data.

[0022] Specifically, the multi-type sensor array consists of a video surveillance subsystem, infrared thermal imaging equipment, vibration optical fiber and millimeter wave radar, which are responsible for collecting different types of data. The video surveillance subsystem uses a high-definition camera array, one is installed every 50 meters along the perimeter, and the adjacent cameras overlap by 30% to collect visible light image data. The infrared thermal imaging equipment uses dual-spectrum imaging technology, works in the 8-14μm band, and detects temperature changes of the target. The vibration optical fiber is buried 0.5 meters underground to collect ground vibration signals. The millimeter wave radar works in the 77GHz frequency band and uses phased array technology to achieve 360-degree scanning and detect airspace targets. The time stamp processing of the original monitoring information adopts a unified timestamp encoding mechanism. Each data record contains three basic fields: data source identification, acquisition time, and data content. The data source identification is represented by a four-bit code, the first bit represents the sensor type (1-video, 2-infrared, 3-vibration, 4-radar), and the last three bits are the device number. The acquisition time uses the Unix timestamp format, accurate to milliseconds. The data content field stores the specific data values ​​collected by the sensor. Time stamp processing sorts and groups the data from different sensors according to the acquisition time to ensure the timing of the data.

[0023] The time window division adopts the sliding window method, with the window length set to 100 milliseconds and the sliding step length to 20 milliseconds. In each time window, the same source data is aggregated according to the data source identifier to extract data features. Video data extracts features such as target position, size, and direction of movement; infrared data extracts features such as temperature distribution and thermal anomalies; vibration data extracts amplitude and frequency features; radar data extracts features such as target distance, speed, and azimuth. The features of different source data in the same time window are associated and matched to establish a feature association matrix. Signal quality assessment adopts a multi-index comprehensive evaluation method. Calculate the signal-to-noise ratio, and the ratio of signal strength to background noise must be higher than the preset threshold: the signal-to-noise ratio threshold for video images is set to 30dB, 25dB for infrared images, 20dB for vibration signals, and 15dB for radar signals. Secondly, evaluate data integrity, check whether data frames are lost, and whether data fields are complete. Finally, evaluate signal stability and calculate the amplitude of signal fluctuations in a short period of time. Data with a comprehensive score below the threshold will be eliminated, and the remaining data will constitute effective monitoring information.

[0024] Time synchronization calibration solves the problem of inconsistent sampling rates of different sensors. The video sampling rate is 25 frames / second, the infrared sampling rate is 15 frames / second, the vibration sampling rate is 1000Hz, and the radar sampling rate is 200Hz. Linear interpolation method is used to resample all data to a unified time point. The interpolation calculation adopts the weighted average of two adjacent sampling points, and the weight is inversely proportional to the time interval. The calibrated data has a unified time base and the sampling rate is the highest sampling rate of 1000Hz. The data integrity check adopts a two-way verification mechanism. Check the continuity of the time series and mark the time points where the data is missing. Then check the integrity of the data content, including data format, value range, logical relationship, etc. For missing data points, different interpolation methods are used according to the data type. The state quantity data uses nearest neighbor interpolation to maintain the discreteness of the state; the continuous quantity data uses cubic spline interpolation to maintain the smoothness of the curve. After interpolation and completion, multi-source synchronous monitoring data is formed.

[0025] For example, when an intrusion occurs at the airport perimeter, the video surveillance subsystem captures the moving target at time T, with an image signal-to-noise ratio of 35dB, and extracts the target position coordinates (x, y) and movement speed v. The infrared device at the same location detects an abnormal temperature point with a signal-to-noise ratio of 27dB, and the temperature is 8°C higher than the background. The vibration fiber detects a vibration signal at time T+15ms, with a signal-to-noise ratio of 22dB and an amplitude of 5 times the background. The radar detects the target at time T+30ms, with a signal-to-noise ratio of 18dB, and measures the radial velocity and azimuth. These asynchronously collected data are processed by time synchronization and uniformly aligned to a sampling rate of 1000Hz. The vibration data of a certain sampling point is missing, and it is calculated by cubic spline interpolation. The amplitude of the interpolation point is consistent with the change trend of the adjacent points. After integrity check and interpolation completion, a set of multi-source synchronized target feature data is formed. The data record clearly shows the characteristic correlation of the intrusion target in time and space.

[0026] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Perform spatial coordinate conversion on multi-source synchronous monitoring data, unify the data collected by different sensors into the same spatial coordinate system, and obtain unified coordinate system monitoring data; (2) Identify the spatial overlapping areas of the unified coordinate system monitoring data, extract the data features of the overlapping areas, and obtain the feature data of the overlapping areas; (3) Perform noise intensity analysis on the feature data in the overlapping area, establish a signal interference model, remove the interference signal, and obtain the feature data after noise reduction; (4) Evaluate the signal stability of each sensor in the denoised feature data, calculate the signal fluctuation coefficient, and obtain the sensor reliability index; (5) Calculate the weight of the de-noised feature data based on the sensor reliability index, establish a multi-source data weighted fusion matrix, and obtain weighted fusion data; (6) Organize and sort the weighted fusion data according to the temporal and spatial correlation, perform data consistency verification, and generate boundary situation fusion data.

[0027] Specifically, the multi-source synchronous monitoring data is converted into spatial coordinates, and all sensor data are unified into the WGS84 coordinate system. The pixel coordinates (u, v) of the video monitoring subsystem are converted into world coordinates (x, y, z) through the internal and external parameter matrices; the image coordinates of the infrared thermal imaging device are converted into actual physical coordinates after distortion correction; the linear distance of the vibrating optical fiber is converted into three-dimensional spatial coordinates; the polar coordinates (r, θ, φ) of the millimeter wave radar are converted into rectangular coordinates. After all sensor data are converted to the same coordinate system, subsequent spatial analysis and data fusion are facilitated. In the process of identifying overlapping areas, the following formula is used to calculate the spatial overlap: ; in: Indicates the recognition degree of overlapping area; represents the reference sensor data point set; represents a set of data points from sensors; Represents the total data point set of the target area; M, N, K represent the number of three data point sets respectively; α, β are weight coefficients.

[0028] When extracting data features of overlapping areas, the space is divided into grid units, and the distribution of data points in each grid is counted. The noise intensity of the feature data of the overlapping area is analyzed, and the environmental noise is divided into two categories: background noise and sudden interference. The background noise is Gaussian distributed and suppressed by Gaussian filtering; the sudden interference is removed by combining median filtering and wavelet transform. A signal interference model is established to compare the signal features of different sensors in the same area, identify and eliminate abnormal interference signals. The signal stability of the feature data after noise reduction is evaluated, and the signal fluctuation coefficient of each sensor is calculated. The evaluation indicators include signal amplitude stability, phase stability and delay stability. For video and infrared images, the image quality and target detection stability are mainly evaluated; for vibration signals, the amplitude and frequency stability are evaluated; for radar signals, the stability of distance, speed and azimuth measurement is evaluated. The sensor reliability index is calculated based on the evaluation results as the weight basis for data fusion.

[0029] The data fusion weight is calculated based on the sensor reliability index, and a multi-source data weighted fusion matrix is ​​established. The weight calculation takes into account the sensor's historical performance, current status, and environmental adaptability. The data of each sensor is weighted according to the weight to construct a multi-dimensional feature space. In the feature space, data from different sources are organized through spatiotemporal correlation to form a unified data structure. Data consistency verification is performed to check whether the fused data meets physical constraints and logical relationships, and to eliminate unreasonable data combinations.

[0030] The perimeter situation fusion data is the output of weighted fusion of multi-source data, which contains comprehensive information such as target location, motion state, and characteristic attributes. For example, during perimeter monitoring, a suspicious target is found somewhere: video monitoring captures the image features of the moving target, infrared equipment detects temperature anomalies, vibration optical fiber detects ground vibrations, and millimeter wave radar measures the target speed. After the spatial coordinates are converted, these data are determined to be in the same spatial area. Overlapping area identification finds that the four sensor data all point to the same target. After removing environmental interference through noise analysis, the quality of each sensor data is evaluated: the video image is clear and stable, the infrared imaging is slightly fluctuated by the temperature, the vibration signal is affected by the ground conditions and has a certain amount of noise, and the radar signal is relatively stable. The fusion weight is determined based on the evaluation results, with video and radar data having higher weights and infrared and vibration data having appropriately lower weights. The fused data fully describes the target's location, motion trajectory, temperature characteristics, and vibration characteristics.

[0031] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Segment the boundary situation fusion data into time series, extract the time series change characteristics of the data in each time period, and obtain the time series feature data; (2) Divide the time series feature data into spatial regions, calculate the data distribution characteristics in each region, and obtain spatial feature data; (3) Perform correlation analysis on the temporal feature data and the spatial feature data, establish a feature correlation matrix, and generate multi-level security features; (4) Analyze abnormal behavior based on multi-level security features, extract intrusion behavior feature sequences, and construct boundary intrusion feature representations; (5) Calculate the matching degree between the boundary intrusion feature representation and the security threat level assessment rules, mark the threat level, and obtain the threat level annotation data; (6) Aggregate and analyze the threat level annotation data according to the time and space dimensions, and output the boundary threat assessment results.

[0032] Specifically, the time series segmentation of the perimeter situation fusion data is carried out using the sliding time window method, with the window length set to 5 minutes and the sliding step length set to 1 minute. In each time period, the statistical characteristics, change trends and mutation points of the data are extracted. Statistical characteristics include mean, standard deviation, kurtosis and skewness; the change trend is obtained through linear regression and curve fitting; the mutation point is identified by the cumulative sum test method. These characteristics constitute the time series feature data. When the time series feature data is divided into spatial regions, an adaptive grid division method is adopted. Based on the terrain characteristics and protection requirements of the monitoring area, the perimeter area is divided into several grid units of varying sizes. The grid density is higher near the airport runway and around important facilities; in open areas, the grid density is relatively low. The distribution characteristics of the data in each grid unit are calculated, including data density, aggregation, directionality and correlation. The data density distribution is calculated by the kernel density estimation method, the degree of data aggregation is calculated by spatial autocorrelation analysis, and the main distribution direction of the data is determined by principal component analysis to obtain spatial feature data. As Figure 2 As shown, it is a schematic diagram of the spatial area division of the airport fence in the embodiment of the present application, which shows the intelligent monitoring grid division scheme of the spatial area of ​​the airport fence. According to the different protection levels, the areas are divided into three categories: core protection area (red area): including the area around the runway, using a high-density grid division of 5m×5m, equipped with high-density video monitoring points and infrared detectors to achieve all-round coverage without blind spots. Key protection area (green area): including important facility areas such as cargo areas, using a medium-density grid division of 10m×10m, and video monitoring and infrared detection equipment are evenly distributed. General protection area (blue area): including open areas, using a low-density grid division of 20m×20m, and monitoring equipment is arranged according to the terrain characteristics. The correlation analysis of time series feature data and spatial feature data adopts the tensor decomposition method. The two types of feature data are organized into a third-order tensor, and the three dimensions represent time, space and feature attributes respectively. The association pattern between features is extracted through Tucker decomposition, and a feature association matrix is ​​established. The feature association matrix reflects the interaction relationship between different features and generates multi-level security features. Multi-level security features form a hierarchical feature system, including basic features at the bottom layer, combined features at the middle layer, and semantic features at the high layer. Abnormal behavior analysis is performed based on multi-level security features, and abnormal patterns are identified using deep learning methods. Spatial features are extracted through convolutional neural networks, temporal features are captured through recurrent neural networks, and multi-layer features are fused through attention mechanisms. The determination of abnormal behavior is based on distance measurement and probability distribution in the feature space. The intrusion behavior feature sequence is extracted from the detected abnormal behavior, including features such as intrusion path, intrusion speed, and intrusion method, to construct a boundary intrusion feature representation.

[0033] The formula for calculating the matching degree between the boundary intrusion feature representation and the security threat level assessment rule is as follows: ; in: Indicates the threat level assessment value; represents the intrusion feature vector; represents the threat rule weight; Indicates the detection time interval; Indicates scene complexity; represents the prior risk coefficient; P and Q represent the number of features and the number of rules respectively; γ represents the threat level category.

[0034] The threat level annotation data is aggregated and analyzed according to the time dimension and space dimension. In the time dimension, the change trend and duration of the threat level are calculated; in the space dimension, the spatial aggregation and diffusion of the threat distribution are analyzed. The high-risk areas and high-incidence periods are identified through the spatiotemporal clustering method, and the perimeter threat assessment results are generated.

[0035] For example, in the airport perimeter monitoring, the monitoring data of a certain perimeter area showed that the time series feature data recorded abnormal activities within 30 consecutive minutes: sporadic vibration signals appeared in the first 10 minutes, continuous heat changes were detected in the middle 10 minutes, and obvious displacement characteristics were found in the last 10 minutes. Spatial feature data showed that these abnormal activities were concentrated in an area of ​​50 meters by 50 meters, showing a trend of gradually advancing from the outside to the inside. The temporal and spatial feature correlation analysis found that the vibration signal, heat change and displacement characteristics showed an obvious progressive relationship in time and had a strong concentration in space. Abnormal behavior analysis identified this series of features as a typical deliberate intrusion behavior pattern. According to the duration, activity range and behavior characteristics of the feature sequence, the matching degree was calculated against the threat level assessment rules. The calculation results show that the intrusion behavior is obviously premeditated and dangerous, and meets the characteristics of a secondary threat. The threat assessment results clearly describe the development process, degree of danger and scope of impact of the intrusion event.

[0036] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Perform time window analysis on the perimeter threat assessment results, extract threat fluctuation characteristics, and obtain threat trend data; (2) Compare and analyze the threat trend data with the historical data, calculate the threat deviation value, and obtain the threat threshold parameter; (3) The threat threshold parameters are graded according to the threat degree, and the warning intervals are divided to obtain the security warning level data; (4) Calculate the similarity between the safety warning level data and the historical disposal experience, extract the disposal elements, and obtain the disposal matching data; (5) Conduct scenario adaptability analysis on the disposal matching data, select response measures from the plan library, and obtain disposal response data; (6) Prioritize and combine response data to form a perimeter security response plan.

[0037] Specifically, the time window analysis of the perimeter threat assessment results is carried out using a sliding time window method, with the window size set to 30 minutes and the sliding step length set to 5 minutes. In each time window, the change rate, fluctuation amplitude and duration of the threat level are calculated. The change rate reflects the speed of the rise or fall of the threat level; the fluctuation amplitude indicates the range of change in the threat level; and the duration records the duration of the threat. Through the combination of these characteristic parameters, the development trend of the threat is extracted to form threat trend data. The comparative analysis of threat trend data and historical data adopts the time series pattern matching method. Threat records in similar scenarios are extracted from the historical database, including threat type, occurrence period, meteorological conditions and other related factors. The similarity between the current threat trend and historical cases is calculated by the dynamic time warping algorithm to find the closest historical event. The deviation between the current threat trend and the historical average level is calculated to obtain the threat deviation value. The larger the deviation value, the higher the abnormality of the current threat. According to the distribution characteristics of the deviation value, the threat threshold parameter is determined.

[0038] The classification of threat threshold parameters adopts a multi-level threshold method, dividing the threat level into four warning intervals: blue (observation), yellow (concern), orange (alert) and red (emergency). The range of each interval is determined based on historical statistical data to ensure that the classification results are consistent with the actual threat level. When determining the security warning level data, not only the current threat threshold parameters are considered, but also the spatial distribution and time evolution characteristics of the threat are combined to avoid frequent fluctuations in the warning level. The similarity calculation between the security warning level data and historical disposal experience adopts a case reasoning method. Retrieve historical disposal cases that match the current warning level from the plan library, and extract the disposal elements of each case, including response level, personnel scheduling, equipment deployment, disposal process, etc. By constructing a feature vector, calculate the similarity between each case and the current situation, and select the case with the highest similarity as a reference. Extract key disposal elements from historical cases to form disposal matching data.

[0039] The scenario adaptability analysis of the disposal matching data focuses on five aspects: geographical environment conditions, current resource status, weather factors, time constraints and coordination difficulty. For each disposal element, evaluate its applicability in the current scenario. For example, some disposal plans are difficult to implement at night or in bad weather conditions and need to be adjusted. The adaptability score of each disposal element is calculated by the linear weighted method, and the response measures with higher adaptability are screened from the plan library to generate disposal response data. The priority sorting of disposal response data considers three dimensions: disposal efficiency, resource consumption and implementation risk. Disposal efficiency reflects the timeliness of response measures; resource consumption includes the input of manpower and material resources; and implementation risk assessment potential uncertainties. The weights of each response measure are determined by the hierarchical analysis method and sorted according to the comprehensive score. The response measures with higher priority are combined to form a disposal process and generate a perimeter security disposal plan.

[0040] For example, during the airport perimeter monitoring process, a certain perimeter area has a persistent intrusion threat. Time window analysis shows that the threat level has been on the rise in the past 20 minutes, from the initial abnormal signal to a clear intrusion behavior. Comparison with historical data shows that similar intrusion incidents have occurred in the area, and the current threat characteristics are highly similar to historical cases. The threat deviation value exceeds the preset threshold, triggering an orange warning. The system retrieves similar historical disposal cases from the plan library and extracts disposal elements such as rapid response team dispatch, infrared imaging enhanced monitoring, and ground patrol route optimization. Considering that it is night time, the priority of night vision equipment deployment is increased. The disposal plan organizes response measures according to the principle of "encirclement and control first, verification later, and disposal again", and clarifies the execution order and coordination mechanism of various measures. The entire disposal process forms a set of response plans, including emergency force dispatch, technical means support, disposal process control and other links.

[0041] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Extract the prevention and control task elements from the perimeter security disposal plan, conduct statistical analysis on the demand for human and equipment resources, and obtain the scheduling demand data; (2) Classify and sort the scheduling demand data according to resource type, compare the existing resource status, and form resource matching data; (3) Analyze the spatiotemporal distribution of resource matching data, divide resource scheduling priorities, and establish security resource scheduling strategies; (4) Input the security resource scheduling strategy into the situation simulation platform, conduct scenario simulation and analysis, and obtain simulation verification data; (5) Quantitatively evaluate the prevention and control effects in the simulation verification data, extract key indicator parameters, and generate program evaluation data; (6) Combine the strategies with the best prevention and control effects from the program evaluation data to form the optimal perimeter prevention and control plan.

[0042] Specifically, the elements of the prevention and control tasks are extracted from the perimeter security disposal plan, including the personnel deployment points, equipment deployment locations, patrol route arrangements, etc. Specific resource demand parameters are extracted for each task element, such as the number of personnel, professional categories, equipment models, quantity specifications, etc. The task decomposition method is adopted to split the complex prevention and control tasks into basic task units, and the human and equipment resources required for each task unit are counted. Human resources are counted according to professional skills, duty hours, and job requirements; equipment resources are classified according to functional types, performance parameters, and deployment conditions to form scheduling demand data. The classification and collation of scheduling demand data adopts the resource hierarchical management method. Resource types are divided into three categories: mobile forces, fixed sentries, and technical equipment. Mobile forces include rapid response teams and patrol teams; fixed sentries include observation posts and guard points; technical equipment includes mobile monitoring equipment, communication equipment, night vision equipment, etc. A resource status table is established for each type of resource to record the location, status, task load and other information of the resources. Compare the scheduling requirements with the existing resource status, calculate the resource gap and surplus, focus on the contradiction between resource supply and demand during the peak period of the task, and form resource matching data.

[0043] The resource matching data is analyzed for spatiotemporal distribution, and the resource demand density in each region and time period is calculated. The resource distribution characteristics are displayed in the form of heat maps to identify the hot spots and key time periods of resource demand. According to the degree of resource matching and the urgency of the task, resource scheduling is divided into three priority levels: emergency scheduling, priority scheduling and regular scheduling. Emergency scheduling is aimed at sudden threats and requires resources to be in place quickly; priority scheduling is aimed at expected threats and sufficient scheduling time is reserved; regular scheduling is used for daily prevention and control and is implemented according to the established plan. Based on the priority classification results, a detailed resource allocation plan is formulated to clarify the scheduling sequence, path selection and coordination mechanism of various resources, and establish a security resource scheduling strategy. The situation simulation platform adopts a combination of three-dimensional scene modeling and simulation simulation. A three-dimensional scene model of the airport perimeter is constructed, including topography, building facilities, monitoring equipment and other elements. The security resource scheduling strategy is imported into the simulation platform, and various threat scenarios are set to simulate the resource scheduling process. The simulation content includes personnel movement trajectory, equipment coverage, communication contact effect, etc. Through multiple simulations, data such as response time, coverage effect, and coordination efficiency in various scenarios are collected to form simulation verification data.

[0044] A multi-dimensional indicator system is used for the quantitative evaluation of simulation verification data. The evaluation indicators include response timeliness, regional coverage, resource utilization, and coordination. Response timeliness calculates the time from threat discovery to the arrival of response forces; regional coverage statistics the spatial distribution effect of prevention and control resources; resource utilization evaluates the efficiency of the use of various resources; and coordination measures the degree of cooperation between different resources. The weight of each indicator is determined by the hierarchical analysis method, the comprehensive score of the prevention and control plan is calculated, the key indicator parameters are extracted, and the plan evaluation data is generated. The formation process of the optimal perimeter prevention and control plan adopts a step-by-step optimization strategy. The plan evaluation data is processed by dimensionality reduction to extract the main influencing factors. According to the performance of each plan on key indicators, the candidate plan with the highest comprehensive score is selected. The candidate plan is optimized in detail, the resource allocation parameters are adjusted, the scheduling sequence is optimized, and the coordination mechanism is improved. The optimized plan is re-simulated and verified, and the strategy combination with the best prevention and control effect is selected to form the optimal perimeter prevention and control plan.

[0045] For example, when an intrusion alarm occurs in a certain section of the airport perimeter, the task elements extracted from the perimeter security disposal plan include: rapid response team dispatch, infrared monitoring equipment enhancement, and ground patrol route optimization. Resource demand statistics show that it is necessary to mobilize 4 groups of patrol personnel, 2 groups of mobile team members, and 2 sets of infrared monitoring equipment. Comparison of resource status found that there are only 2 groups of patrol forces in the vicinity, and additional forces need to be deployed from adjacent areas. The spatiotemporal distribution analysis shows that the location of the alarm is in a key monitoring area and it is at night, so the resource dispatch priority is set at the emergency level. The situation simulation results show that the use of a two-way encirclement method to dispatch patrol forces, combined with high-point monitoring and ground blockade, can form an effective blockade in the shortest time. Evaluation data shows that the plan performs best in terms of response timeliness and regional blockade effect, while also having a high resource utilization rate. The determined prevention and control plan specifies in detail the specific division of labor, arrival time, action route, and communication contact method of each force.

[0046] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Extract perimeter security monitoring parameters from the execution data of the optimal perimeter prevention and control plan, classify and sort them according to parameter types, and obtain a monitoring parameter set; (2) Perform statistical analysis on the monitoring parameter set, calculate the fluctuation range and distribution characteristics of each parameter, and form parameter evaluation data; (3) Based on the parameter evaluation data, eliminate abnormal fluctuation parameters, extract key monitoring indicators, and obtain parameter screening results; (4) Based on the parameter screening results, analyze the corresponding relationship between each monitoring threshold and the actual threat, and build a monitoring threshold mapping table; (5) Compare the monitoring threshold mapping table with the existing warning rules, mark the rule items that need to be updated, and generate a rule update list; (6) Replace the original rules with the updated items in the rule update list to generate an updated perimeter security monitoring strategy.

[0047] Specifically, the perimeter security monitoring parameters are extracted from the execution data of the optimal perimeter prevention and control plan. These parameters include three categories: equipment monitoring parameters, environmental threshold parameters, and behavioral characteristic parameters. Equipment monitoring parameters include sensitivity settings, detection ranges, and alarm thresholds of various sensors; environmental threshold parameters include judgment criteria for environmental factors such as light intensity, temperature changes, and wind speed; behavioral characteristic parameters include behavioral judgment indicators such as target movement speed, residence time, and activity range. A hierarchical parameter structure is established according to parameter types, and related parameters are classified and sorted to form a monitoring parameter set. When statistically analyzing the monitoring parameter set, a multi-dimensional data analysis method is used. The actual value range of each parameter during the execution process is calculated, including the maximum value, minimum value, mean, and standard deviation. Then, the time distribution characteristics of the parameters are analyzed, the parameter change laws in different time periods are statistically counted, and the periodic change patterns of the parameters are identified. Then, the spatial distribution characteristics of the parameters are analyzed, the parameter differences in different regions are statistically counted, and the spatial correlation model of the parameters is established. Through these statistical analyses, the fluctuation laws and distribution characteristics of each parameter are mastered to form parameter evaluation data.

[0048] Based on the parameter evaluation data, perform anomaly analysis and screening of parameters. Identify the abnormal fluctuation points of parameters through the 3σ principle and calculate the stability index of parameters. Focus on analyzing the parameters with frequent abnormalities to determine whether the cause of the abnormality is equipment failure or environmental interference. At the same time, identify the monitoring indicators that play a key role in threat judgment through the principal component analysis method. Retain parameters with strong discrimination ability as key monitoring indicators, eliminate redundant or unstable parameters, and obtain parameter screening results. For the parameter screening results, establish a mapping relationship between monitoring thresholds and actual threats. By analyzing historical alarm data, statistically analyze the corresponding relationship between different parameter values ​​and threat levels. Use the decision tree method to build a multi-level judgment model for parameter thresholds. Consider the interaction between parameters and set combined judgment conditions. Organize these judgment rules into a monitoring threshold mapping table to clarify the alarm threshold and judgment logic of each parameter.

[0049] Compare the newly constructed monitoring threshold mapping table with the existing warning rules item by item. Check the judgment conditions, threshold settings, and trigger logic of each rule. Mark the items where the threshold has changed, the judgment logic needs to be adjusted, and the rules need to be supplemented or deleted. Categorize and organize the rule items that need to be updated to form a rule update list. For each update item, record the content, reason, and scope of impact of the update in detail. Update the original rules according to the rule update list. Use incremental update to replace the rules that need to be modified item by item. Pay attention to maintaining the consistency of the rules during the update process to avoid conflicts between rules. Perform logical verification on the updated rules to ensure the integrity and rationality of the rules. Form an updated perimeter security monitoring strategy, including parameter configuration, judgment rules, and disposal procedures.

[0050] For example, in the practice of perimeter monitoring, false alarms were found to occur frequently at night. By analyzing the execution data, key parameters such as the temperature threshold of the infrared sensor, the contrast parameter of the video monitoring, and the sensitivity of the vibration sensor were extracted. Statistical analysis shows that the change patterns of these parameters are significantly different in different time periods and under different weather conditions. Through data screening, it was found that the setting of the temperature threshold was too fixed and did not take into account the impact of seasonal changes; the sensitivity of the vibration sensor was obviously not applicable on rainy days. Based on this, a new threshold mapping relationship was constructed, and the temperature threshold was changed to dynamically adjust with the season, and the vibration sensitivity added the judgment condition of weather factors. By comparing the existing rules, it was found that many rule items needed to be updated, such as changing the fixed temperature difference threshold to relative temperature difference judgment and adding special processing rules for rainy days. The updated monitoring strategy effectively reduced false alarms and improved the accuracy of monitoring in actual applications.

[0051] The above describes the smart airport perimeter security monitoring method in the embodiment of the present application. The following describes the smart airport perimeter security monitoring system in the embodiment of the present application. Figure 3 In the embodiment of the present application, an embodiment of the smart airport perimeter security monitoring system includes: The processing module 201 is used to collect monitoring information of the enclosed area through a multi-type sensor array, and perform time synchronization processing on the collected monitoring information through a time synchronization server to obtain multi-source synchronous monitoring data; The noise reduction module 202 is used to perform spatial registration and noise reduction processing on the multi-source synchronous monitoring data using the spatial registration system, and dynamically configure the weights according to the reliability of each sensor to generate perimeter situation fusion data; The extraction module 203 is used to extract and analyze the spatiotemporal features according to the perimeter situation fusion data, and construct the perimeter intrusion feature representation through multi-level security features, and then analyze it according to the security threat level assessment rules to obtain the perimeter threat assessment result; A matching module 204 is used to determine the security warning level through a dynamic threshold system based on the perimeter threat assessment result, and match the response measures from the plan library according to historical handling experience to form a perimeter security handling plan; Verification module 205, used to establish a security resource scheduling strategy according to the perimeter security disposal plan, and verify the plan through the situation deduction platform to obtain the optimal perimeter prevention and control plan; The screening module 206 is used to evaluate and screen perimeter security monitoring parameters based on the execution data of the optimal perimeter prevention and control plan, and adjust the monitoring thresholds and warning rules according to the screening results to generate an updated perimeter security monitoring strategy.

[0052] Through the coordinated cooperation of the above-mentioned components, the monitoring information of the perimeter area is collected through a multi-type sensor array, and the time synchronization server is used for time synchronization processing, which solves the problem of time consistency of different sensor data and ensures the synchronization and integrity of multi-source monitoring data; the spatial registration system is used to perform spatial registration and noise reduction on the multi-source synchronous monitoring data, and the dynamic weight configuration is performed according to the reliability of the sensor, which effectively improves the accuracy and reliability of data fusion; the spatiotemporal features of the perimeter situation fusion data are extracted and analyzed, and the perimeter intrusion feature representation is constructed through multi-level security features, which realizes the accurate identification of intrusion threats; the dynamic threshold system is used to determine the security warning level, and the response measures are matched from the plan library based on historical disposal experience, which improves the accuracy of the warning and the pertinence of the disposal; by establishing a security resource scheduling strategy and using the situation deduction platform to verify the plan, the optimal configuration of prevention and control resources is achieved; the perimeter security monitoring parameters are evaluated and screened, and the monitoring threshold and warning rules are dynamically adjusted according to the screening results, and a set of adaptive monitoring strategy update mechanism is established. The overall solution forms a complete closed-loop monitoring system. From data collection, analysis and processing, threat identification to emergency response, each link is closely connected and data flows smoothly, which greatly improves the intelligence level of airport perimeter security monitoring. The solution makes full use of the complementary advantages of multi-source data and effectively solves the problems existing in traditional monitoring methods through intelligent analysis means and adaptive disposal mechanisms.

[0053] Based on the same technical concept, the embodiment of the present application also provides an electronic device. Figure 4As shown, it is a schematic diagram of the structure of the electronic device 300 provided in the embodiment of the present application, including a processor 301, a memory 302, and a bus 303. Among them, the memory 302 is used to store execution instructions, including a memory 3021 and an external memory 3022; the memory 3021 here is also called an internal memory, which is used to temporarily store the operation data in the processor 301, and the data exchanged with the external memory 3022 such as a hard disk. The processor 301 exchanges data with the external memory 3022 through the memory 3021. When the electronic device 300 is running, the processor 301 communicates with the memory 302 through the bus 303.

[0054] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the smart airport perimeter security monitoring method.

[0055] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0056] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0057] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A smart airport perimeter security monitoring method, characterized in that: The smart airport perimeter security monitoring method comprises: The monitoring information of the enclosed area is collected through a multi-type sensor array, and the collected monitoring information is time synchronized through a time synchronization server to obtain multi-source synchronous monitoring data; The spatial registration system is used to perform spatial registration and noise reduction on multi-source synchronous monitoring data, and dynamic weight configuration is performed according to the reliability of each sensor to generate perimeter situation fusion data; According to the perimeter situation fusion data, the spatiotemporal features are extracted and analyzed, and the perimeter intrusion feature representation is constructed through multi-level security features. Then, the perimeter threat assessment result is obtained by analyzing it according to the security threat level assessment rules; Based on the perimeter threat assessment results, the security warning level is determined through the dynamic threshold system, and response measures are matched from the plan library based on historical disposal experience to form a perimeter security disposal plan; According to the perimeter security disposal plan, establish a security resource scheduling strategy, and verify the plan through the situation simulation platform to obtain the optimal perimeter prevention and control plan; Based on the execution data of the optimal perimeter prevention and control plan, the perimeter security monitoring parameters are evaluated and screened, and the monitoring thresholds and warning rules are adjusted according to the screening results to generate an updated perimeter security monitoring strategy.

2. The smart airport perimeter security monitoring method according to claim 1 is characterized in that: The monitoring information of the enclosed area is collected through a multi-type sensor array, and the collected monitoring information is time synchronized by a time synchronization server to obtain multi-source synchronous monitoring data, including: The multi-type sensor array performs multi-angle scanning on the enclosure area, the video monitoring subsystem collects visible light image data, the infrared thermal imaging device collects thermal imaging data, the vibration optical fiber collects ground vibration data, and the millimeter wave radar collects airspace detection data to obtain original monitoring information; The original monitoring information is grouped and marked according to the monitoring time, and a unified timestamp code is assigned to each group of original monitoring information to obtain monitoring information with a timestamp; Dividing the monitoring information with the timestamp into time windows, correlating and matching the monitoring information in the same time window, and obtaining time-correlated monitoring information; Performing signal quality evaluation on the time-correlated monitoring information, eliminating monitoring information with a signal-to-noise ratio lower than a preset threshold, and obtaining valid monitoring information; Performing time synchronization calibration on the effective monitoring information, aligning the monitoring information collected by different sensors to a unified time reference, and obtaining time-calibrated monitoring information; The time calibration monitoring information is subjected to a data integrity check, and the time points of the missing data are interpolated and supplemented to generate the multi-source synchronous monitoring data.

3. The smart airport perimeter security monitoring method according to claim 1 is characterized in that: The spatial registration system is used to perform spatial registration and noise reduction on multi-source synchronous monitoring data, and dynamic weight configuration is performed according to the reliability of each sensor to generate perimeter situation fusion data, including: Performing spatial coordinate conversion on the multi-source synchronous monitoring data, unifying the data collected by different sensors into the same spatial coordinate system, and obtaining unified coordinate system monitoring data; Identifying spatial overlapping regions of the unified coordinate system monitoring data, extracting data features of the overlapping regions, and obtaining overlapping region feature data; Performing noise intensity analysis on the feature data of the overlapping area, establishing a signal interference model, eliminating interference signals, and obtaining feature data after noise reduction; Evaluate the signal stability of each sensor in the noise-reduced feature data, calculate the signal fluctuation coefficient, and obtain the sensor reliability index; The weight of the de-noised feature data is calculated according to the sensor reliability index, a multi-source data weighted fusion matrix is ​​established, and weighted fusion data is obtained; The weighted fusion data is organized and sorted according to the temporal and spatial correlation, and the data consistency is checked to generate the boundary situation fusion data.

4. The smart airport perimeter security monitoring method according to claim 1 is characterized in that: The above method extracts and analyzes the spatiotemporal features based on the perimeter situation fusion data, constructs the perimeter intrusion feature representation through multi-level security features, and then analyzes it according to the security threat level assessment rules to obtain the perimeter threat assessment results, including: The boundary situation fusion data is segmented into time series, and the time series change characteristics are extracted for the data in each time period to obtain time series characteristic data; Dividing the time series feature data into spatial regions, calculating the data distribution characteristics in each region, and obtaining spatial feature data; Perform correlation analysis on the time series feature data and the spatial feature data, establish a feature correlation matrix, and generate the multi-level security features; Perform abnormal behavior analysis based on the multi-level security features, extract intrusion behavior feature sequences, and construct the boundary intrusion feature representation; Calculating the matching degree between the boundary intrusion feature representation and the security threat level assessment rule, marking the threat level, and obtaining threat level marking data; The threat level annotation data is aggregated and analyzed according to the time and space dimensions, and the perimeter threat assessment result is output.

5. The smart airport perimeter security monitoring method according to claim 1 is characterized in that: Based on the perimeter threat assessment results, the security warning level is determined through the dynamic threshold system, and the response measures are matched from the plan library based on historical disposal experience to form a perimeter security disposal plan, including: Performing time window analysis on the perimeter threat assessment results, extracting threat fluctuation characteristics, and obtaining threat trend data; Compare and analyze the threat trend data with historical data, calculate the threat deviation value, and obtain the threat threshold parameter; The threat threshold parameters are graded according to the threat degree, and the warning intervals are divided to obtain security warning level data; Calculate the similarity between the safety warning level data and historical disposal experience, extract disposal elements, and obtain disposal matching data; Performing scenario adaptability analysis on the disposal matching data, screening response measures from the plan library, and obtaining disposal response data; The handling response data are prioritized and combined to form the perimeter security handling plan.

6. The smart airport perimeter security monitoring method according to claim 1 is characterized in that: According to the perimeter security disposal plan, a security resource scheduling strategy is established, and the plan is verified through the situation deduction platform to obtain the optimal perimeter prevention and control plan, including: Extract the prevention and control task elements from the perimeter security disposal plan, conduct statistical analysis on the demand for human and equipment resources, and obtain scheduling demand data; Classify and sort the scheduling demand data according to resource types, compare with existing resource status, and form resource matching data; Performing spatiotemporal distribution analysis on the resource matching data, dividing resource scheduling priorities, and establishing the security resource scheduling strategy; Input the security resource scheduling strategy into the situation simulation platform, perform scenario simulation and analysis, and obtain simulation verification data; Conduct quantitative evaluation of the prevention and control effects in the simulation verification data, extract key indicator parameters, and generate program evaluation data; The strategies with the best prevention and control effects in the scheme evaluation data are combined and integrated to form the optimal perimeter prevention and control scheme.

7. The smart airport perimeter security monitoring method according to claim 1 is characterized in that: The execution data of the optimal perimeter prevention and control plan is used to evaluate and screen the perimeter security monitoring parameters, and the monitoring thresholds and warning rules are adjusted according to the screening results to generate an updated perimeter security monitoring strategy, including: Extracting perimeter security monitoring parameters from the execution data of the optimal perimeter prevention and control plan, classifying and arranging them according to parameter types, and obtaining a monitoring parameter set; Performing statistical analysis on the monitoring parameter set, calculating the fluctuation range and distribution characteristics of each parameter, and forming parameter evaluation data; Taking the parameter evaluation data as a benchmark, eliminating abnormal fluctuation parameters, extracting key monitoring indicators, and obtaining parameter screening results; According to the parameter screening results, the corresponding relationship between each monitoring threshold and the actual threat is analyzed, and a monitoring threshold mapping table is constructed; Compare the monitoring threshold mapping table with the existing warning rules, mark the rule items that need to be updated, and generate a rule update list; The updated items in the rule update list replace the original rules to generate the updated perimeter security monitoring strategy.

8. A smart airport perimeter security monitoring system, used to implement the smart airport perimeter security monitoring method according to any one of claims 1 to 7, characterized in that: The smart airport perimeter security monitoring system includes: A processing module, used to collect monitoring information of the enclosed area through a multi-type sensor array, perform time synchronization processing on the collected monitoring information through a time synchronization server, and obtain multi-source synchronous monitoring data; The noise reduction module is used to perform spatial registration and noise reduction on multi-source synchronous monitoring data using the spatial registration system, and dynamically configure weights according to the reliability of each sensor to generate perimeter situation fusion data; The extraction module is used to extract and analyze the spatiotemporal features according to the perimeter situation fusion data, and construct the perimeter intrusion feature representation through multi-level security features, and then analyze it according to the security threat level assessment rules to obtain the perimeter threat assessment results; The matching module is used to determine the security warning level through the dynamic threshold system based on the perimeter threat assessment results, and match the response measures from the plan library based on historical disposal experience to form a perimeter security disposal plan; The verification module is used to establish a security resource scheduling strategy based on the perimeter security disposal plan, and verify the plan through the situation simulation platform to obtain the optimal perimeter prevention and control plan; The screening module is used to evaluate and screen perimeter security monitoring parameters based on the execution data of the optimal perimeter prevention and control plan, and adjust the monitoring thresholds and warning rules according to the screening results to generate an updated perimeter security monitoring strategy.

9. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus. When the machine-readable instructions are executed by the processor, the steps of the smart airport perimeter security monitoring method as described in any one of claims 1 to 7 are performed.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the smart airport perimeter security monitoring method as described in any one of claims 1-7 is implemented.

Images