One-way door control method, system and equipment based on ground sensing grating

Abstract

The invention discloses a one-way door control method, system and equipment based on a ground sensing grating, and belongs to the field of one-way doors. According to the embodiment, vibration spectrum data including position information and frequency domain energy distribution information of a vibration sensor and passing direction data including an object moving direction, a position sequence and ground sensing grating layout information in a passing area are obtained; determining time sequence fluctuation information corresponding to the abnormal behavior based on the frequency domain energy distribution information and preset frequency domain energy distribution characteristics associated with the passing behavior; obtaining space-time correlation information according to the position sequence and the layout information, and dividing the position sequence based on the layout information according to a preset space-time resolution to generate space-time grid distribution; generating judgment parameters based on the space-time correlation information, the object moving direction, the position information, the space-time grid distribution and the time sequence fluctuation information; and comparing the judgment parameter with a preset threshold range to generate a control instruction for the one-way door, so that the one-way door can be accurately controlled under the interference of a complex environment.

Description

Technical Field

[0001] This application belongs to the field of one-way doors, and particularly relates to a one-way door control method, system, device, and computer-readable storage medium based on ground sensor gratings. Background Art

[0002] One-way doors are widely used in public places such as subway stations, high-speed railway stations, and airports to guide one-way traffic and prevent reverse intrusion or abnormal behavior from disturbing the normal order.

[0003] In such scenarios, it is necessary to distinguish normal traffic behavior and abnormal traffic behavior in real time and accurately, and quickly generate control instructions.

[0004] However, there are often complex environmental noises and interferences in the one-way door passage area, so it is necessary to accurately control the one-way door in a complex environment.

[0005] In the prior art, the one-way door is generally controlled based on the trigger of the grating. This solution detects the moving direction of the object through the ground sensor grating to determine the abnormal traffic behavior in the one-way door passage area, and then completes the control of the one-way door based on the determination result.

[0006] The existing technical solutions only rely on a single grating trigger direction for determination, and cannot effectively distinguish complex environmental interferences from real abnormal traffic behaviors, thus leading to misjudgments. At the same time, the fixed determination logic in the existing technical solutions is difficult to adapt to dynamic scenarios such as peak passenger flow, and cannot meet the real-time and accurate control requirements. Therefore, the accuracy of controlling the one-way door under complex environmental interferences in the existing technical solutions is relatively low. Summary of the Invention

[0007] The embodiments of this application provide a one-way door control method, system, device, and computer-readable storage medium based on ground sensor gratings, which can accurately control the one-way door under complex environmental interferences.

[0008] In a first aspect, the embodiments of this application provide a one-way door control method based on ground sensor gratings, and the method includes: Obtain the vibration spectrum data and traffic direction data of the passage area. The vibration spectrum data includes the position information of the vibration sensor and the frequency domain energy distribution information, and the traffic direction data includes the object moving direction, the position sequence, and the layout information of the ground sensor grating for collecting the traffic direction data; Based on the frequency domain energy distribution information and the preset frequency domain energy distribution characteristics associated with normal traffic behavior and abnormal behavior, determine the timing fluctuation information corresponding to the abnormal behavior; According to the position sequence and the layout information, perform the continuity verification of the time interval and the spatial distance to obtain the spatio-temporal correlation information, and based on the layout information, divide the position sequence according to the preset spatio-temporal resolution to generate a spatio-temporal grid distribution; Generate a determination parameter that characterizes the interaction relationship between normal passage behavior and abnormal behavior based on spatio-temporal correlation information, object movement direction, location information, spatio-temporal grid distribution, and temporal fluctuation information; Compare the determination parameter with a preset threshold range to generate a control instruction, and complete the control of the one-way door through the control instruction.

[0009] In an implementable embodiment, generating a determination parameter that characterizes the interaction relationship between normal passage behavior and abnormal behavior based on spatio-temporal correlation information, object movement direction, location information, spatio-temporal grid distribution, and temporal fluctuation information includes: Extract the grating coordinate set corresponding to the reverse movement path in the spatio-temporal grid distribution, determine the movement change trend based on the object movement direction, and generate a spatio-temporal correlation feature vector based on the grating coordinate set and the movement change trend. The spatio-temporal correlation feature vector is used to characterize the distribution characteristics of abnormal behavior in the spatial dimension; Perform time window synchronization processing on the temporal fluctuation information, and align the time window with the time span of the reverse movement path in the spatio-temporal grid distribution; Based on the synchronized temporal fluctuation information and location information, calculate the frequency domain energy distribution weight of each grid cell in the spatio-temporal grid distribution, and generate a determination parameter according to the spatio-temporal correlation feature vector and the frequency domain energy distribution weight. The determination parameter is used to quantify the interaction relationship between normal passage behavior and abnormal behavior in the spatio-temporal and frequency domain dimensions.

[0010] In an implementable embodiment, based on the synchronized temporal fluctuation information and location information, calculate the frequency domain energy distribution weight of each grid cell in the spatio-temporal grid distribution, and generate a determination parameter according to the spatio-temporal correlation feature vector and the frequency domain energy distribution weight. The determination parameter is used to quantify the interaction relationship between normal passage behavior and abnormal behavior in the spatio-temporal and frequency domain dimensions, including: Based on the synchronized temporal fluctuation information and the spatio-temporal grid distribution, calculate the energy distribution stability index of each grid cell within the time window, and determine the sensor coverage integrity feature of each grid cell based on the location information and the spatio-temporal grid distribution; Construct a spatial mapping relationship according to the energy distribution stability index of each grid cell, the sensor coverage integrity feature, and the spatial relationship of each grid cell in the spatio-temporal grid distribution; Perform frequency band decomposition on the synchronized temporal fluctuation information, and extract the energy value of a preset target frequency band associated with abnormal behavior; According to the spatial mapping relationship, allocate the energy value of the target frequency band to the corresponding spatio-temporal grid cell to generate the initial frequency domain energy weight of each grid cell within the time window in the spatio-temporal grid distribution; Perform a product operation on the spatio-temporal correlation feature vector and the initial frequency domain energy weight according to the grid cell to generate a determination parameter.

[0011] In an implementable embodiment, according to the spatial mapping relationship, the energy values of the target frequency band are allocated to the corresponding spatio-temporal grid cells to generate the initial frequency-domain energy weights of each grid cell within the time window in the spatio-temporal grid distribution, including: Based on the spatial mapping relationship, a spatial weight model is constructed, and the spatial weight model includes the energy allocation ratio of each grid cell within the coverage range of each vibration sensor; The energy values of the target frequency band are normalized to obtain the energy contribution values of each vibration sensor within the time window; According to the spatial weight model, the energy contribution values of each vibration sensor are allocated to the grid cells covered by each vibration sensor to generate the initial frequency-domain energy weights of each grid cell.

[0012] In an implementable embodiment, before performing the element-wise multiplication operation on the spatio-temporal correlation feature vector and the initial frequency-domain energy weights according to the grid cells to generate the decision parameter, the method further includes: Based on the reverse trigger frequency of each grid cell in the spatio-temporal correlation feature vector, the initial frequency-domain energy weights are dynamically weighted and adjusted to generate the adjusted frequency-domain energy distribution weights; Performing a multiplication operation on the spatio-temporal correlation feature vector and the initial frequency-domain energy weights according to the grid cells to generate the decision parameter, including: Performing a multiplication operation on the spatio-temporal correlation feature vector and the frequency-domain energy weights according to the grid cells to generate the decision parameter.

[0013] In an implementable embodiment, based on the frequency-domain energy distribution information and the preset frequency-domain energy distribution characteristics associated with normal passage behavior and abnormal behavior, the time-series fluctuation information is determined, including: The frequency-domain energy distribution information is decomposed into energy components of multiple frequency bands according to the preset frequency band division rule, and the preset frequency band division rule is defined according to the frequency-domain energy distribution characteristics of normal passage behavior and abnormal behavior; The target frequency band energy components associated with abnormal behavior are extracted from the energy components of the multiple frequency bands after frequency band division, and the target frequency band energy components are dynamically baseline calibrated to generate the calibrated target frequency band energy sequence; Based on the target frequency band energy sequence, the start time point, end time point and sudden increase amplitude of the energy sudden increase sequence exceeding the preset change rate threshold are determined to generate the energy sudden increase sequence set; According to the time distribution density and the sudden increase amplitude change rate of the sequences in the energy sudden increase sequence, the time-series fluctuation information is generated.

[0014] In an implementable embodiment, before obtaining the vibration spectrum data and the passage direction data of the passage area, the method further includes: Obtain the original vibration signal of the vibration sensor and the position information of the vibration sensor; Decompose the original vibration signal into vibration sub-signals of multiple independent wavelength channels according to a preset wavelength channel division rule; Perform frequency-domain demodulation processing on the vibration sub-signals of each wavelength channel, and extract the frequency-domain energy components of each wavelength channel; According to the position information of the vibration sensor, perform weighting processing on the frequency-domain energy components of each wavelength channel to generate vibration spectrum data.

[0015] In a second aspect, an embodiment of the present application provides a one-way door control system based on a ground induction grating. The system includes: An acquisition module, configured to acquire vibration spectrum data and passage direction data of a passage area. The vibration spectrum data includes the position information and frequency-domain energy distribution information of a vibration sensor, and the passage direction data includes the object movement direction, position sequence, and layout information of the ground induction grating for collecting the passage direction data; A determination module, configured to determine timing fluctuation information based on the frequency-domain energy distribution information and preset frequency-domain energy distribution characteristics associated with normal passage behavior and abnormal behavior; A division module, configured to perform time interval and spatial distance continuity verification according to the position sequence and the layout information to obtain spatio-temporal correlation information, and divide the position sequence according to a preset spatio-temporal resolution based on the layout information to generate a spatio-temporal grid distribution; A generation module, configured to generate a determination parameter characterizing the interaction relationship between normal passage behavior and abnormal behavior based on the spatio-temporal correlation information, object movement direction, position information, spatio-temporal grid distribution, and timing fluctuation information; The determination module is further configured to compare the determination parameter with a preset threshold range to generate a control instruction, and complete the control of the one-way door through the control instruction.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, a one-way door control method based on a ground induction grating in any one of the implementation manners of the first aspect is implemented.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is caused to execute a one-way door control method based on a ground induction grating in any one of the implementation manners of the first aspect.

[0018] A one-way door control method, system, device, and computer-readable storage medium according to an embodiment of the present application. First, by synchronously collecting vibration spectrum data and passage direction data, a multi-dimensional behavioral feature representation ability is constructed to effectively distinguish the physical feature differences between normal passage and abnormal interference. Second, based on the preset frequency-domain energy distribution characteristics, temporal fluctuation information is dynamically extracted to accurately identify abnormal energy surge events, suppress the interference of environmental noise, and significantly reduce the misjudgment rate. Further, by modeling the physical continuity of the reverse movement path through spatio-temporal grid distribution and generating dynamic interaction intensity parameters in combination with parameters such as frequency-domain energy weight and change trend of movement direction, the spatio-temporal and frequency-domain interaction relationships between normal and abnormal behaviors are quantified to adapt to the dynamic change requirements in different scenarios. Finally, based on the real-time comparison between the dynamic determination parameters and the preset threshold, control instructions are generated to ensure real-time and accurate response in high-traffic scenarios. It realizes the precise control of the one-way door under complex environmental interference.

[0019] Further, by combining spatio-temporal correlation feature vectors with synchronous temporal fluctuation information, the frequency-domain energy distribution weight is accurately aligned with the spatial abnormal diffusion path, solving the misjudgment problem caused by the dislocation between the energy surge event and the reverse movement time window. On this basis, an energy distribution stability index and a sensor coverage integrity feature are introduced to dynamically suppress the energy distribution weights in high-fluctuation noise regions and low-density monitoring regions, and optimize the energy distribution ratio through spatial mapping relationships, improving the environmental adaptability of abnormal behavior determination. The two jointly construct a three-dimensional dynamic determination model of space-time, frequency-domain, and time sequence. Through dynamic weight adjustment and path alignment strategies, the accuracy of abnormal behavior recognition and the reliability of control in complex scenarios are significantly enhanced. Brief Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 It is a schematic flowchart of a one-way door control method based on a ground sensor grating provided by an embodiment of the present application; Figure 2 It is a schematic flowchart of a method for determining determination parameters provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of a one-way door control system based on a ground sensor grating provided by an embodiment of the present application; Figure 4 It is a schematic hardware structure diagram of an electronic device provided by an embodiment of the present application. Detailed Embodiments

[0022] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0023] It should be understood that the specific embodiments described herein are only intended to explain the present application and not to limit the present application.

[0024] For those skilled in the art, the present application can be implemented without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0025] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0026] Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0027] Without further limitation, elements defined by the statement "comprising..." do not exclude the presence of additional identical elements in the process, method, article or device comprising the said elements.

[0028] One-way doors are widely used in public places such as subway stations, high-speed railway stations and airports to guide one-way traffic and prevent reverse intrusion or abnormal behavior from disturbing the normal order.

[0029] For example, at the turnstile of a subway station, abnormal behavior may be caused by passengers dragging luggage, suddenly turning back or forcibly ramming; at the security check channel of a high-speed railway station, the collision of large luggage with the door body or the detention of people may lead to mis-triggering; at the international arrival gate of an airport, the frequent passage of trolleys is mixed with reverse intrusion behavior.

[0030] In such scenarios, it is necessary to distinguish normal passage behavior and abnormal passage behavior in real time and accurately, and quickly generate control instructions.

[0031] However, there are often complex environmental noises and interferences in the one-way door passage area. For example, the low-frequency vibration of subway motors in subway stations, the high-frequency friction of luggage wheels in high-speed railway stations, the random collision noise of airport trolleys with the ground, and the energy superposition effect caused by dense crowds during peak hours. Therefore, it is necessary to implement precise control of one-way doors in complex environments.

[0032] In the prior art, the one-way door is generally controlled based on the grating trigger method. This solution uses a ground sensor grating to detect the moving direction of an object to determine abnormal passing behaviors in the passing area of the one-way door, and then completes the control of the one-way door based on the determination result. For example, when it is detected that an object enters from the exit direction, the door body is directly triggered to lock.

[0033] The existing technical solutions rely only on a single grating trigger direction for determination, which has significant limitations: First, the grating trigger logic cannot distinguish sudden interference from real abnormal behaviors. For example, when a luggage wheel quickly passes through the grating, it may generate a short-time reverse signal, and its trigger sequence is highly similar to that of a real reverse intrusion behavior in the time domain; Second, the grating can only provide direction and position information and cannot capture the physical energy characteristics of abnormal behaviors, such as the sudden increase in high-frequency vibration energy when hitting the door body and the low-frequency energy difference from normal pushing of a cart, resulting in the decoupling of energy characteristics and behavior types; In addition, the existing solutions lack the spatio-temporal correlation modeling of the grating trigger sequence. For example, when dragging a luggage in reverse, it may continuously trigger multiple gratings, but due to the lack of a path continuity verification mechanism, isolated grating trigger events are prone to misjudgment.

[0034] Such technical defects are particularly prominent in complex scenarios: For example, in the high-speed railway station scenario, the instantaneous reverse signal generated by a large piece of luggage hitting the grating cannot be distinguished from a real reverse intrusion behavior by a fixed threshold; and the dense passing during the peak passenger flow period leads to the superposition of multi-target energies, the grating trigger signal is distorted and the noise interference is difficult to eliminate, ultimately resulting in missed detections or false triggers. Therefore, the one-way door control method of the existing solutions has core defects of low accuracy and insufficient dynamic adaptability under complex environmental interferences.

[0035] To solve the problems of the prior art, the embodiments of the present application provide a one-way door control method, system, device, and computer-readable storage medium based on a ground sensor grating.

[0036] To address the defect that the grating trigger logic in traditional one-way door control is difficult to distinguish complex environmental noises from real abnormal behaviors, the embodiments of the present application construct a dynamic interaction intensity parameter by fusing the frequency-domain energy characteristics of the vibration spectrum and the multi-dimensional spatio-temporal data of the passing direction, and realize the accurate identification and adaptive control of abnormal behaviors.

[0037] First, the one-way door control method based on a ground sensor grating provided by the embodiments of the present application will be introduced below.

[0038] Figure 1 The flowchart of the one-way door control method based on a ground sensor grating provided by an embodiment of the present application is shown. As Figure 1 shown, the method includes steps S110 to S150.

[0039] S110: Obtain the vibration spectrum data and the traffic direction data of the passage area. The vibration spectrum data includes the position information of the vibration sensor and the frequency-domain energy distribution information. The traffic direction data includes the object movement direction, the position sequence, and the layout information of the ground induction grating for collecting the traffic direction data.

[0040] The vibration spectrum data is the energy distribution characteristic data obtained by frequency-domain analysis of the vibration signals generated by the movement of objects in the passage area collected by the vibration sensor, and includes the position information of the vibration sensor and the frequency-domain energy distribution information.

[0041] The position information of the vibration sensor refers to the installation coordinates of the sensor in the passage area, which is used for subsequent mapping of the space-time grid. The frequency-domain energy distribution information reflects the energy distribution characteristics of the vibration signal in different frequency intervals. For example, low-frequency energy may correspond to normal traffic behavior, and high-frequency energy may be associated with abnormal collisions.

[0042] The traffic direction data is the space-time sequence data triggered by detecting the object movement direction and position through the ground induction grating system, and includes the object movement direction, the position sequence, and the layout information of the ground induction grating.

[0043] The object movement direction is judged by the order and time interval of the grating trigger. For example, forward or reverse movement; the position sequence refers to the set of the time sequence and space coordinates of the ground induction grating being triggered. For example, the time and position records of multiple grating units being triggered in sequence when the object moves. The layout information refers to the physical deployment information of the ground induction grating units, including the installation position and coverage range of each grating.

[0044] Vibration data acquisition captures the vibration signals of the ground or the door structure in real time through the vibration sensors deployed in the passage area, converts the original signals into the frequency-domain energy spectrum, and divides the preset frequency bands to statistically analyze the energy distribution.

[0045] The frequency band division is based on the energy characteristic differences between normal traffic and abnormal behaviors in the experimental data. For example, the low-frequency energy accounts for a relatively high proportion during normal walking, while abnormal impacts will cause a sudden increase in high-frequency energy.

[0046] Direction data acquisition relies on the ground induction grating system. When the grating unit is blocked by an object, it triggers the position coordinates and the corresponding time stamps, and the movement direction is deduced from the time difference and space distance between adjacent grating triggers.

[0047] The grating layout information determines the physical position of each unit through a predefined coordinate mapping table.

[0048] Abnormal behaviors include real abnormal behaviors such as reverse intrusion, ramming, and staying, as well as abnormal interferences such as train vibrations and luggage friction. Finally, the vibration data and the direction data are synchronized and aligned according to the time stamps to form a comprehensive data set including space-time positions, energy distributions, and movement directions.

[0049] Taking the one-way door in the airport security check passage as an example, vibration sensors are installed on the ground on both sides of the door body, and the ground sensing grating units are deployed in the entrance and exit areas of the passage path.

[0050] When a passenger passes through normally, the grating units trigger the position sequence from the entrance to the exit in turn, and the moving direction is marked as positive. The vibration sensors detect the energy distribution mainly dominated by low frequency, corresponding to the stable vibration characteristics of walking.

[0051] If someone tries to break in reversely, the grating units will trigger the reverse sequence from the exit to the entrance. At the same time, the vibration sensors capture the sudden increase in high-frequency energy when the door body is impacted, such as the abnormal vibration of forcibly pushing the door or hitting the door body.

[0052] The vibration spectrum data outputs the sensor coordinates and the energy proportion of the corresponding frequency bands, and the passage direction data includes the reverse grating trigger sequence and layout parameters.

[0053] This data provides a basis for subsequent determination of abnormal intrusion behavior. For example, the correlation characteristics between the reverse movement path and the sudden increase in high-frequency energy can distinguish normal passage from abnormal interference.

[0054] S120: Determine the timing fluctuation information based on the frequency domain energy distribution information and the preset frequency domain energy distribution characteristics associated with normal passage behavior and abnormal behavior.

[0055] The frequency domain energy distribution information refers to the distribution characteristics of energy in different frequency intervals obtained by frequency domain analysis of the vibration signals collected by vibration sensors. For example, high-frequency energy may correspond to abnormal impacts, and low-frequency energy corresponds to normal passage.

[0056] The frequency domain energy distribution characteristics associated with normal passage behavior and abnormal behavior are the frequency band energy patterns predefined based on historical data or experiments. For example, the energy of normal walking is concentrated in the low frequency band (such as 0 - 50Hz), and the energy of abnormal behavior may suddenly increase in the medium and high frequency bands (such as 100 - 500Hz).

[0057] The timing fluctuation information refers to the energy change characteristics of abnormal behavior in the time dimension, which can include parameters such as the time point of energy sudden increase, the duration, and the fluctuation amplitude.

[0058] By comparing the real-time frequency domain energy distribution with the preset abnormal characteristic frequency bands, the energy timing characteristics of abnormal behavior are extracted.

[0059] First, decompose the real-time collected frequency domain energy distribution information into energy components of multiple frequency bands according to the preset frequency band division rules. For example, divide 0 - 500Hz into three sub-frequency bands: low frequency, medium frequency, and high frequency.

[0060] Next, according to the preset abnormal behavior associated frequency band (such as the high frequency band), extract the energy components of the target frequency band, and generate the time series fluctuation information characterizing the dynamic characteristics of abnormal behavior by statistically analyzing the fluctuation characteristics of its time series, such as the start time, peak amplitude, and duration of sudden energy increase. The time series fluctuation information is the key time series feature for distinguishing abnormal interference behavior from real abnormal behavior.

[0061] S130: According to the position sequence and layout information, perform the continuity verification of time interval and spatial distance to obtain the spatio-temporal correlation information, and based on the layout information, divide the position sequence according to the preset spatio-temporal resolution to generate the spatio-temporal grid distribution.

[0062] The continuity verification of time interval and spatial distance is a process of verifying whether the movement of an object conforms to the reasonable passage logic by calculating the time difference and spatial distance between adjacent grating triggers.

[0063] The spatio-temporal correlation information is the result obtained after verification, which is used to describe the spatio-temporal rationality of the grating trigger sequence, such as whether there are abnormal jumps or reverse movement paths. The spatio-temporal grid distribution is a set of grid cells obtained by dividing the passage area according to the preset time granularity and spatial resolution, which is used to quantify the distribution characteristics of abnormal behavior in the spatio-temporal dimension.

[0064] First, perform the continuity verification on the position sequence. By calculating the time difference and spatial distance between adjacent grating triggers, deduce whether the moving speed exceeds the normal range, and combine the grating layout information to detect whether the path direction conforms to the preset passage logic.

[0065] For example, if the time difference between adjacent grating triggers is too short and the spatial distance is large, it is determined as abnormal high-speed movement; if the grating trigger order is opposite to the forward passage direction, it is marked as a reverse path. Through the above verification, the spatio-temporal correlation information is generated to eliminate noise interference or isolated events of mis-triggering.

[0066] Subsequently, based on the layout information, divide the passage area into grid cells according to the preset spatio-temporal resolution. For example, divide the space into uniform grids and the time into fixed windows, and map the grating trigger events to the corresponding spatio-temporal grid cells to form the spatio-temporal grid distribution for subsequent analysis of the spatio-temporal aggregation characteristics of abnormal behavior.

[0067] S140: Based on the spatio-temporal correlation information, the object movement direction, the position information, the spatio-temporal grid distribution, and the time series fluctuation information, generate the decision parameters characterizing the interaction relationship between normal passage behavior and abnormal behavior.

[0068] The decision parameters are quantitative indicators generated by fusing the spatio-temporal correlation information, the object movement direction, the spatio-temporal grid distribution, and the time series fluctuation information, which are used to comprehensively evaluate the interaction intensity between normal passage behavior and abnormal behavior in the spatio-temporal and frequency domains.

[0069] This parameter integrates the spatio-temporal aggregation of abnormal paths, the sudden increase in the frequency domain of vibration energy, and the conflict characteristics of the moving direction, and finally outputs a quantifiable judgment basis.

[0070] By integrating the reverse path distribution density (spatio-temporal correlation information) of abnormal behaviors in the spatio-temporal grid, the degree of conflict between the moving direction and the preset passage logic (object moving direction), the temporal fluctuation amplitude of vibration energy in the target frequency band (temporal fluctuation information), and the aggregation characteristics of abnormal paths in the spatio-temporal grid (spatio-temporal grid distribution), a quantitative index for multi-dimensional feature fusion is constructed.

[0071] Exemplarily, first, isolated interference events are removed through spatio-temporal continuity verification, and abnormal movement sequences with path continuity are retained; then the passage area is discretized into spatio-temporal grid cells, and the trigger frequency of reverse movement events and the sudden increase intensity of the corresponding vibration energy in each grid are statistically calculated; finally, the aggregation degree of abnormal paths in the spatio-temporal dimension and the energy abnormality in the frequency domain are weighted and superimposed to generate a judgment parameter comprehensively representing the intensity of abnormal behaviors.

[0072] By quantifying the interaction intensity of abnormal behaviors in three dimensions of spatio-temporal correlation, direction conflict, and energy abnormality, the judgment parameter can dynamically reflect the abnormal risk level of the passage area.

[0073] When the judgment parameter exceeds the first threshold, it indicates the strong spatio-temporal coupling of high-frequency energy sudden increase and reverse path, and the door locking instruction is triggered; when the parameter is within the second threshold interval, it indicates the existence of abnormal characteristics in a single dimension, such as isolated reverse trigger without energy sudden increase, then an alarm is started but not locked; when the parameter is lower than all thresholds, it is determined as normal passage.

[0074] For example, in the high-speed railway station scenario, the isolated reverse grating trigger (low spatio-temporal aggregation degree) caused by dragging a suitcase is accompanied by low-frequency energy (low frequency domain abnormality), and its judgment parameter will be lower than the locking threshold, thus avoiding false triggering; while the real reverse intrusion behavior, due to continuously triggering multiple gratings (high spatio-temporal aggregation degree) and accompanied by a sudden increase in high-frequency energy of door impact (high frequency domain abnormality), the judgment parameter will exceed the locking threshold, accurately triggering the door control.

[0075] S150: Generate a control instruction based on the judgment parameter and the preset judgment threshold, and complete the control of the one-way door through the control instruction.

[0076] The preset threshold range is a critical value set according to historical data and experimental verification, used to distinguish different behavior types, such as normal passage, abnormal interference, and real abnormal behaviors. The control instruction is an execution signal generated by the system according to the comparison result of the judgment parameter and the threshold, such as operation instructions like door locking, alarm, or keeping open.

[0077] First, based on the real-time environmental noise, such as the low-frequency vibration background of subway stations, and historical abnormal behavior characteristics, the sliding window statistical method is used to dynamically adjust the threshold range to ensure a balance between the false touch rate and the missed detection rate during peak hours.

[0078] When the judgment parameter exceeds the locking threshold, the door locking command is triggered, the electromagnetic lock is controlled to close and the sound and light alarm is linked; if the judgment parameter is in the warning range, the warning light flashes and the voice prompt is started, and the timestamp and location information of the abnormal event are recorded; when the judgment parameter does not reach the threshold, the door maintains a normal open and closed state.

[0079] In one embodiment, in the one-way door control scenario of an airport security channel, the vibration sensor collects vibration signals of the door and the ground in real time, and generates vibration spectrum data of the energy proportion of each frequency band after frequency domain analysis. For example, the low frequency band of 0-50Hz corresponds to normal walking, and the high frequency band of 100-500Hz is associated with impact behavior.

[0080] At the same time, the ground-sensitive grating detects the moving direction of the object and records the position sequence of the triggered grating unit, such as entrance to exit for the forward direction and exit to entrance for the reverse direction, as well as its layout coordinates.

[0081] The system verifies the continuity of movement based on the time interval and spatial distance of the grating triggering. For example, if the time difference between adjacent grating triggering is 0.2 seconds and the interval is 1 meter, the derived speed is 5m / s, which exceeds the normal walking range and is marked as abnormally high-speed movement; if the grating sequence is triggered continuously in the reverse direction, such as from the exit to the middle of the channel and then to the entrance, it is judged as a reverse intrusion path.

[0082] By integrating the spatiotemporal concentration of the reverse path, such as the continuous triggering of three reverse gratings, the sudden increase in high-frequency energy and the conflict of moving direction, that is, the deviation from the preset forward logic, the system generates comprehensive judgment parameters.

[0083] When the parameter exceeds the locking threshold, such as the reverse path length is greater than or equal to 2 meters and the high-frequency energy lasts for 1 second, the door electromagnetic lock is triggered to lock and the alarm is linked; if the parameter is in the warning range, such as isolated reverse triggering and low-frequency energy is dominant, the warning light is activated; during normal passage, the parameter does not reach the threshold and the door remains open.

[0084] For example, when a single reverse grating is mistakenly triggered by dragging a suitcase, the parameters are lower than the threshold due to the low-frequency energy and isolated path characteristics, thus avoiding false locking. However, when a forced reverse intruder triggers a continuous grating accompanied by high-frequency impact energy, the parameters exceed the threshold, and the door is locked within 0.5 seconds, accurately intercepting abnormal behavior.

[0085] This embodiment constructs a multi-dimensional behavior feature characterization capability through the synchronous collection of vibration spectrum data and traffic direction data, and effectively distinguishes the physical feature differences between normal traffic and abnormal interference.

[0086] Secondly, based on the preset frequency-domain energy distribution characteristics, the time-series fluctuation information is dynamically extracted to accurately identify abnormal energy surge events, suppress the interference of environmental noise, and significantly reduce the false positive rate.

[0087] Furthermore, by modeling the physical continuity of the reverse movement path through spatio-temporal grid distribution and combining parameters such as frequency-domain energy weight and change trend of movement direction, dynamic interaction intensity parameters are generated to quantify the spatio-temporal and frequency-domain interaction relationships between normal and abnormal behaviors, adapting to the dynamic change requirements in different scenarios.

[0088] Finally, based on the real-time comparison between the dynamic decision-making parameters and the preset threshold, control instructions are generated to ensure real-time and accurate response in high-traffic scenarios. The accurate control of the one-way door under complex environmental interference is achieved.

[0089] Due to the misjudgment problem caused by the similar spatio-temporal characteristics of sudden interference (such as accidental contact with the light grid by dragging luggage) and real abnormal behaviors (such as reverse intrusion) faced by the one-way door in complex passage scenarios, the existing technology cannot effectively distinguish the two types of events due to the isolated analysis of the triggering direction of the light grid.

[0090] To solve this problem, in this embodiment, the following methods are used to improve the accuracy and anti-interference ability of abnormal behavior recognition, accurately distinguish between accidental triggering of suitcases and real intrusion behaviors, reduce the false locking rate, and improve the passage safety.

[0091] Figure 2 The flowchart of the method for determining the decision-making parameters provided by an embodiment of the present application is shown.

[0092] In a feasible embodiment, as Figure 2 shown, step S140: Based on spatio-temporal correlation information, object movement direction, position information, spatio-temporal grid distribution, and time-series fluctuation information, decision-making parameters representing the interaction relationship between normal passage behavior and abnormal behavior are generated, including steps S141 to S143.

[0093] S141: Extract the light grid coordinate set corresponding to the reverse movement path in the spatio-temporal grid distribution, determine the movement change trend based on the object movement direction, and generate a spatio-temporal correlation feature vector based on the light grid coordinate set and the movement change trend. The spatio-temporal correlation feature vector is used to represent the distribution characteristics of abnormal behavior in the spatial dimension.

[0094] S142: Perform time window synchronization processing on the time-series fluctuation information, and align the time window with the time span of the reverse movement path in the spatio-temporal grid distribution.

[0095] S143: Based on the synchronized timing fluctuation information and position information, calculate the frequency-domain energy distribution weights of each grid cell in the spatio-temporal grid distribution, and generate a determination parameter according to the spatio-temporal correlation feature vector and the frequency-domain energy distribution weights. The determination parameter is used to quantify the interaction relationship between normal passing behaviors and abnormal behaviors in the spatio-temporal and frequency-domain dimensions.

[0096] The spatio-temporal correlation feature vector is a multi-dimensional vector generated by fusing the raster coordinate set of the reverse movement path and the movement change trend, and is used to quantify the aggregation and direction deviation degree of abnormal behaviors in the spatial dimension. The raster coordinate set is the position set of the reverse-triggered raster, and the movement change trend is deduced from the time difference between adjacent raster triggers and the direction sequence; the time window synchronization process is an operation that precisely aligns the time axis of the timing fluctuation information with the time span of the reverse movement path, ensuring that the frequency-domain energy sudden increase event and the reverse movement path are completely matched in the time dimension; the frequency-domain energy distribution weight is the grid cell energy contribution intensity value calculated based on the synchronized timing fluctuation information and the vibration sensor position, reflecting the frequency-domain energy distribution characteristics of abnormal behaviors in a specific spatio-temporal grid.

[0097] First, extract the raster coordinate set corresponding to the reverse movement path from the spatio-temporal grid distribution. By screening the direction markers of raster trigger events (such as encoding reverse movement as a specific identifier) and recording their position coordinates in chronological order, form the raster coordinate set of the reverse path. Based on the time difference and spatial distance between adjacent raster triggers, calculate the instantaneous speed and acceleration change trends. For example, deduce the sudden change of the movement direction or the speed gradient through the difference method, and combine the raster layout information to judge whether the path conforms to the preset abnormal behavior logic (such as continuous reverse trigger and increasing speed).

[0098] Subsequently, encode the raster coordinate set and movement trend parameters (such as speed change rate, direction deviation angle) into a multi-dimensional spatio-temporal correlation feature vector, which is used to characterize the aggregation and direction deviation intensity of abnormal behaviors in the spatial dimension. Next, the system performs time window synchronization processing on the timing fluctuation information, determines the time window range according to the start and end timestamps of the reverse movement path, intercepts the high-frequency energy timing data within this window, and aligns the time axis through linear interpolation to eliminate the timing offset error.

[0099] Based on the position information of the vibration sensor, use a spatial interpolation algorithm (such as inverse distance weighted interpolation) to allocate the synchronized timing fluctuation energy values to the corresponding spatio-temporal grid cells, and calculate the frequency-domain energy distribution weights of each grid cell in combination with the sensor coverage range model (such as distance attenuation function), reflecting the distribution intensity of abnormal energy in space.

[0100] Finally, perform a multiplication operation on the spatio-temporal correlation feature vector and the frequency-domain energy distribution weight for each grid cell. Among them, the reverse trigger frequency or path density in the spatio-temporal correlation feature vector is multiplied by the energy weight to generate the local decision parameter component of each grid cell, and the comprehensive decision parameter is obtained by accumulating the components of all grid cells.

[0101] This parameter integrates the spatio-temporal aggregation of abnormal paths, the conflict of moving directions, and the sudden increase in frequency-domain energy. For example, when the reverse path continuously covers multiple grid cells and is accompanied by concentrated high-frequency energy, the decision parameter increases significantly.

[0102] In one embodiment, in the one-way door control scenario of the high-speed railway station security check channel, the system detects a reverse moving path through the ground sensor grating: the passenger sequentially triggers the grating units G1, G2, G3 from the exit direction, forming a reverse coordinate set.

[0103] Based on the time difference and spatial distance between adjacent grating triggers, the system calculates the instantaneous speed and acceleration, and finds that the moving speed increases from 0.5 m / s to 1.2 m / s (abnormal acceleration), generating a spatio-temporal correlation feature vector [G1, G2, G3, +0.7] (+0.7 indicates an accelerating trend).

[0104] At the same time, the vibration sensor captures a sudden increase in the high-frequency energy of the door body (100 - 500 Hz). The system intercepts the high-frequency energy time series data of the corresponding period according to the time window of the reverse path (such as 2 seconds), and distributes the energy values to the G1 - G3 grid cells through the inverse distance weighted interpolation method, and calculates that the frequency-domain energy weights of each cell are 0.85, 0.9, and 0.78 respectively.

[0105] Multiply the reverse trigger frequency (each of G1 - G3 is triggered once) in the spatio-temporal correlation feature vector by the energy weight for each grid to obtain the local decision parameter components 0.85, 0.9, and 0.78. After accumulation, the comprehensive decision parameter is 2.53.

[0106] This embodiment generates a dynamic decision parameter by integrating the spatio-temporal correlation features of the reverse moving path and the frequency-domain weight distribution of the sudden increase in high-frequency energy, effectively solving the misjudgment problem of one-way doors in complex scenarios.

[0107] At the same time, based on the dynamic weight distribution mechanism of time window synchronization and grid calculation, environmental noise interference is suppressed to ensure real-time response and stable control in high-density traffic scenarios.

[0108] Due to the misjudgment problem caused by the coupling of spatio-temporal and frequency-domain characteristics between sudden interferences (such as luggage dragging) and real abnormal behaviors (such as reverse intrusion) of one-way doors in complex scenarios, existing technical solutions cannot accurately quantify the interaction intensity of abnormal behaviors because they analyze grating triggers or vibration energy in isolation.

[0109] To solve this problem, in this embodiment, a spatial mapping relationship is constructed to dynamically associate the time-series fluctuation energy with the sensor coverage characteristics, so as to improve the spatial resolution and anti-interference ability of abnormal behavior determination.

[0110] In an implementable embodiment, based on the synchronized time-series fluctuation information and position information, calculate the frequency-domain energy distribution weights of each grid cell in the spatio-temporal grid distribution, and generate a determination parameter according to the spatio-temporal correlation feature vector and the frequency-domain energy distribution weights. The determination parameter is used to quantify the interaction relationship between normal passage behavior and abnormal behavior in the spatio-temporal and frequency-domain dimensions, including: Based on the synchronized time-series fluctuation information and the spatio-temporal grid distribution, calculate the energy distribution stability index of each grid cell within the time window, and based on the position information and the spatio-temporal grid distribution, determine the sensor coverage integrity feature of each grid cell; according to the energy distribution stability index and the sensor coverage integrity feature of each grid cell, as well as the spatial relationship of each grid cell in the spatio-temporal grid distribution, construct a spatial mapping relationship; perform frequency band decomposition on the synchronized time-series fluctuation information, and extract the energy value of a preset target frequency band associated with abnormal behavior; according to the spatial mapping relationship, allocate the energy value of the target frequency band to the corresponding spatio-temporal grid cell, and generate the initial frequency-domain energy weights of each grid cell within the time window in the spatio-temporal grid distribution; perform a product operation on the spatio-temporal correlation feature vector and the initial frequency-domain energy weights according to the grid cell to generate a determination parameter.

[0111] The energy distribution stability index is used to quantify the severity of the frequency-domain energy fluctuation over time within the spatio-temporal grid cell, and is achieved by calculating the variance or standard deviation of the energy values within the time window. For example, when the high-frequency energy suddenly increases, the variance increases significantly, indicating the occurrence of abnormal behavior. The sensor coverage integrity feature describes the detection ability of the vibration sensor for the grid cell, and is calculated according to the proportion of the number of grid cells within the sensor coverage radius. For example, 80% of the grid cells within the coverage range are regarded as high-integrity areas.

[0112] The spatial mapping relationship is a topological model constructed based on the spatial adjacency of grid cells (such as adjacent grids) and the sensor coverage characteristics, and is used to define the energy distribution rule between grids. For example, adjacent grids share the sensor energy contribution.

[0113] The target frequency band energy value is the energy integral value of a preset frequency band (such as 100 - 500 Hz) associated with abnormal behavior extracted through frequency band decomposition (such as Fourier transform), and is used to characterize the frequency-domain characteristics of abnormal behavior.

[0114] First, based on the synchronized timing fluctuation information, calculate the energy distribution stability index of each spatio-temporal grid cell within the time window through sliding window variance analysis to identify the abnormal fluctuation area corresponding to the sudden increase in high-frequency energy; at the same time, according to the physical position of the vibration sensor and the preset coverage radius, use the distance attenuation function to calculate the sensor coverage integrity feature of the grid cell, and quantify the attenuation law of the sensor's energy detection ability for the grid. Then, use the spatial adjacency matrix to construct the topological relationship between the grids, and fuse the energy distribution stability (such as high variance marked as the abnormal area) and the coverage integrity feature (such as higher weight for the high-coverage area), and define the dynamic energy transfer rule. For example, a higher energy transfer ratio is assigned to the adjacent grids of the abnormal area.

[0115] Subsequently, the system performs a fast Fourier transform (FFT) on the timing fluctuation information, extracts the energy integral value of the preset target frequency band (such as the high-frequency band associated with abnormal behavior), and combines the dynamic transfer rule in the spatial mapping relationship to allocate the energy value to the spatio-temporal grid cell through inverse distance weighted interpolation (IDW) to generate the initial frequency domain energy weight. Finally, perform a grid-by-grid product operation on the spatio-temporal correlation feature vector (such as the reverse trigger frequency, path acceleration trend) and the initial energy weight of each grid cell, and accumulate all grid results to obtain the comprehensive judgment parameter. For example, the parameter value is significantly increased after multiplying the energy weight of the high-integrity grid by the spatio-temporal feature of dense reverse triggering, while the interference energy in the low-coverage area is suppressed due to the low weight, thus realizing the accurate quantitative determination of abnormal behavior.

[0116] In this embodiment, by dynamically fusing multi-dimensional features of space-time and frequency domain, a precise quantitative model for abnormal behavior is constructed: the energy distribution stability index identifies the abnormal fluctuation area of sudden increase in high-frequency energy, the sensor coverage integrity feature suppresses the noise interference in the long-distance or edge area, and the spatial mapping relationship dynamically allocates the energy weight by combining grid adjacency and coverage integrity to ensure the accurate spatial positioning of abnormal energy; extract the energy value of the target frequency band through frequency band decomposition and allocate it to the grid cell to generate the initial frequency domain energy weight, and then multiply it grid by grid with the spatio-temporal correlation feature vector (such as reverse path density, moving acceleration trend), so that the true abnormal behavior (space-time continuous trigger + high-frequency energy concentration) significantly improves the judgment parameter due to the strong coupling of spatio-temporal and frequency domain features, while the normal passage or isolated interference (such as luggage dragging) results in a lower parameter value due to low spatio-temporal correlation or low-frequency energy distribution, thus achieving high-precision discrimination in complex scenarios.

[0117] In an implementable embodiment, in order to further improve the spatial resolution and anti-interference ability of abnormal behavior determination, according to the spatial mapping relationship, allocate the energy value of the target frequency band to the corresponding spatio-temporal grid cell to generate the initial frequency domain energy weight of each grid cell within the time window in the spatio-temporal grid distribution, including: Based on the spatial mapping relationship, a spatial weight model is constructed. The spatial weight model includes the energy allocation ratio of each grid cell within the coverage range of each vibration sensor. Normalize the energy values in the target frequency band to obtain the energy contribution value of each vibration sensor within the time window. According to the spatial weight model, allocate the energy contribution value of each vibration sensor to the grid cells covered by each vibration sensor to generate the initial frequency-domain energy weight of each grid cell.

[0118] The spatial weight model refers to a mathematical model based on the spatial relationship between the coverage range of vibration sensors and grid cells, and defines the allocation ratio of sensor energy among grids through the distance attenuation rule. For example, grid cells closer to the sensor are assigned higher weights due to higher detection sensitivity, while the weights of grid cells at a greater distance decrease. Normalization processing is an operation used to scale the original energy values collected by different sensors to the same dimension according to a unified standard, aiming to eliminate the influence of sensor sensitivity differences on the evaluation of energy contribution. The energy contribution value refers to the quantization result of the energy detection of a single sensor in the target frequency band within the time window, reflecting its ability to capture abnormal energy within a specific spatio-temporal range.

[0119] First, based on the physical positions of the sensors and the preset coverage radius, use the inverse distance weighting method to construct a spatial weight model, and define the energy allocation ratio of grid cells within the coverage range of each sensor. For example, the closer the distance, the higher the weight. Subsequently, perform maximum-minimum normalization on the energy values in the target frequency band to eliminate the sensitivity differences of sensors and generate standardized energy contribution values. Finally, according to the spatial weight model, allocate the energy contribution value of each sensor to the grid cells according to the distance weights. If multiple sensors cover the same grid, accumulate the allocation values of each sensor to generate the initial frequency-domain energy weight. For example, if sensor A covers grid G1 (weight 0.8) and G2 (weight 0.5), and its normalized energy contribution value is 0.9, then G1 is allocated 0.72 (0.9×0.8), and G2 is allocated 0.45 (0.9×0.5).

[0120] In this embodiment, by constructing a spatial weight model, the energy allocation ratio of each grid cell within the coverage range of the sensor is accurately quantified, eliminating the detection blind area and improving the spatial positioning accuracy of abnormal energy; the normalization processing effectively eliminates the sensitivity differences of different sensors, ensuring the fairness and consistency of the energy contribution value evaluation; the dynamic energy allocation mechanism allocates the energy contribution value of the sensor to the grid cells according to the distance attenuation rule based on the spatial weight model. Combined with the superposition of multi-sensor coverage, it significantly enhances the energy aggregation effect of abnormal events and suppresses isolated interference, thereby achieving high-precision abnormal behavior determination in complex scenarios.

[0121] In order to further improve the accuracy of abnormal behavior determination in complex scenarios, in an implementable embodiment, before generating a determination parameter by performing an element-level multiplication operation on the spatio-temporal correlation feature vector and the initial frequency-domain energy weight according to grid cells, the method further includes: Based on the reverse trigger frequency of each grid cell in the spatio-temporal correlation feature vector, perform dynamic weighted adjustment on the initial frequency-domain energy weight to generate an adjusted frequency-domain energy distribution weight; perform a multiplication operation on the spatio-temporal correlation feature vector and the initial frequency-domain energy weight according to grid cells to generate a determination parameter, including: performing a multiplication operation on the spatio-temporal correlation feature vector and the frequency-domain energy weight according to grid cells to generate a determination parameter.

[0122] Dynamic weighted adjustment refers to the operation of dynamically correcting the initial frequency-domain energy weight according to the reverse trigger frequency of grid cells in the spatio-temporal correlation feature vector. For grid cells with a higher trigger frequency, their energy weights are non-linearly amplified to enhance the determination priority of abnormal behaviors. Element-level multiplication operation refers to the mathematical operation of multiplying the spatio-temporal correlation feature vector and the adjusted frequency-domain energy weight one by one according to grid cells, which is used to quantify the coupling strength of spatio-temporal features and frequency-domain energy in each grid cell.

[0123] First, extract the reverse trigger frequency of each grid cell from the spatio-temporal correlation feature vector, and calculate the dynamic adjustment coefficient based on a preset non-linear weighting function (such as an exponential function). For example, grid cells with high-frequency triggers are given a higher adjustment coefficient, and the initial frequency-domain energy weight is amplified accordingly. Subsequently, perform an element-level multiplication operation on the adjusted frequency-domain energy weight and the corresponding feature values in the spatio-temporal correlation feature vector, such as the reverse trigger count and acceleration trend, according to grid cells to generate local determination parameter components, and finally accumulate all grid components to obtain a comprehensive determination parameter. This process achieves millisecond-level response through matrix-based parallel computing to ensure real-time performance.

[0124] This embodiment significantly improves the accuracy and anti-interference ability of abnormal behavior determination through a dynamic weighted adjustment mechanism. Based on the reverse trigger frequency in the spatio-temporal correlation feature vector, the system non-linearly corrects the initial frequency-domain energy weight, amplifying the energy weight of abnormal regions with high-frequency triggers, thereby strengthening the coupling strength of spatio-temporal features and frequency-domain energy.

[0125] In an implementable embodiment, to solve the problem of misjudgment caused by environmental noise interference and the overlap of abnormal behavior frequency-domain features in the prior art, and further improve the timing analysis accuracy of abnormal energy sudden increase events. This embodiment is as follows, step S120: Based on the frequency-domain energy distribution information and the preset frequency-domain energy distribution characteristics associated with normal passing behaviors and abnormal behaviors, determine the timing fluctuation information, including: Decompose the frequency-domain energy distribution information into energy components of multiple frequency bands according to a preset frequency-band division rule, where the preset frequency-band division rule is defined based on the frequency-domain energy distribution characteristics of normal passage behaviors and abnormal behaviors; extract the target frequency-band energy components associated with abnormal behaviors from the energy components of multiple frequency bands after frequency-band division, and perform dynamic baseline calibration on the target frequency-band energy components to generate a calibrated target frequency-band energy sequence; based on the target frequency-band energy sequence, determine the start time point, end time point, and sudden increase amplitude of the energy sudden increase sequence exceeding a preset change rate threshold, and generate an energy sudden increase sequence set; generate time-series fluctuation information according to the time distribution density and sudden increase amplitude change rate of the sequences in the energy sudden increase sequence.

[0126] The frequency-band division rule refers to a pre-defined segmentation strategy based on the frequency-domain energy difference between normal passage behaviors and abnormal behaviors. For example, normal walking mainly excites low-frequency energy (such as 0 - 50 Hz), while abnormal impacts are manifested as sudden increases in high-frequency energy (such as 100 - 500 Hz). The vibration signal is decomposed into discrete frequency bands through the Fast Fourier Transform (FFT) to separate different behavior characteristics. Dynamic baseline calibration eliminates the baseline drift of environmental noise (such as the low-frequency vibration interference of subway motors) through sliding window mean filtering, and retains the real abnormal high-frequency energy sudden increase. The energy sudden increase sequence refers to a set of events exceeding the change rate threshold extracted from the calibrated energy sequence, including the sudden increase time, amplitude, and duration, which are used to quantify abnormal dynamic characteristics.

[0127] First, based on a preset frequency-band division rule (such as low frequency, medium frequency, high frequency), perform FFT decomposition on the vibration signal to extract the target frequency band (such as the high-frequency band) associated with abnormal behaviors. Subsequently, use sliding window mean filtering to perform dynamic baseline calibration on the target frequency-band energy sequence to suppress slow-varying noise interference. For example, in the subway station scenario, calculate the baseline mean through a 1-second window, and subtract the baseline from the real-time energy value to highlight the high-frequency sudden increase of impact events. Then, by calculating the first-order difference slope of the energy sequence, detect sudden increase events exceeding a preset change rate threshold (such as slope > 5 dB / s), record their start and end times, peak amplitude, and duration, and form an energy sudden increase sequence set. Finally, statistically calculate the time distribution density (such as the trigger frequency per unit time) and amplitude change rate (such as the mean and variance of the slope) of sudden increase events to generate time-series fluctuation information. For example, a real intrusion is manifested as a continuous sudden increase with high density and high slope mean, while interference events are isolated sudden increases with low density and low slope.

[0128] Based on a preset frequency band division rule, for example, the low frequency band of normal walking is defined as 0 - 50 Hz, and the high frequency band of abnormal impact is defined as 100 - 500 Hz. The time domain signal of the vibration sensor is decomposed into energy components of multiple frequency bands through the Fast Fourier Transform (FFT), and the energy components of the target frequency band are extracted. Subsequently, a dynamic baseline calibration technique, such as moving window mean filtering, is adopted to eliminate the baseline drift caused by environmental noise, such as the low frequency interference caused by the operation of subway motors, so as to highlight the energy sudden increase feature of real abnormal behaviors. By calculating the first-order difference slope and combining it with a preset slope change rate threshold, energy sudden increase events are detected, such as events with a slope exceeding 5 decibels per second and a duration of more than 0.5 seconds, and their start time, peak amplitude, and duration are recorded to form an energy sudden increase sequence set. Finally, the time series fluctuation information is generated by combining the time distribution density and the amplitude change rate. For example, in the high-speed railway station scenario, the system detects an energy sudden increase in the high frequency band range of 200 - 500 Hz. After calibration, two continuous sudden increase events are identified, with an average slope of 12 decibels per second and a time density of 1.5 times per second, which are determined as real intrusion behaviors and trigger the door locking; while the isolated low slope sudden increase events are determined as interference and ignored because the time density is lower than the threshold, thus achieving accurate time series analysis and feature extraction of abnormal behaviors in a complex noise background, significantly improving the determination accuracy and anti-interference ability.

[0129] When processing multi-source data, due to the superposition of multi-source vibration signals, spectrum aliasing occurs. For example, in the subway station scenario, the low frequency vibration of the train and the high frequency impact signal of the door overlap in the frequency domain, masking the energy characteristics of abnormal behaviors with noise, resulting in misjudgment or missed detection. At the same time, the failure to distinguish the spatial distribution characteristics of the vibration sources (such as impacts near the door and distant interferences) may cause distortion in the energy weight distribution, affecting the accuracy of the determination parameters.

[0130] To solve the above problems, in a feasible implementation, before step S110: obtaining the vibration spectrum data and the passage direction data of the passage area, the method further includes: Obtaining the original vibration signal of the vibration sensor and the position information of the vibration sensor; decomposing the original vibration signal into vibration sub-signals of multiple independent wavelength channels according to a preset wavelength channel division rule; performing frequency domain demodulation processing on the vibration sub-signals of each wavelength channel to extract the frequency domain energy components of each wavelength channel; and performing weighting processing on the frequency domain energy components of each wavelength channel according to the position information of the vibration sensor to generate vibration spectrum data.

[0131] The wavelength channel division rule refers to a processing strategy that decomposes the original vibration signal into multiple independent sub-signals according to the physical characteristics of the vibration signal in different frequency bands (for example, high-frequency impacts correspond to short wavelengths, and low-frequency interferences correspond to long wavelengths). For example, the range of 0 - 500 Hz is divided into a short-wavelength channel (200 - 500 Hz) and a long-wavelength channel (0 - 200 Hz) to separate high-frequency abnormal impacts and low-frequency environmental noise; frequency-domain demodulation processing is a process of performing a fast Fourier transform (FFT) on the sub-signals of each wavelength channel to extract the frequency-domain energy components. For example, calculating the power spectral density of the short-wavelength channel to quantify the high-frequency impact energy; position weighting processing is to perform spatial weight assignment on the energy components of different channels based on the installation position and coverage range of the vibration sensors. For example, higher weights are assigned to the energy of the short-wavelength channel of the sensors near the door body to enhance the sensitivity of abnormal signals in key areas.

[0132] First, the original vibration signal is decomposed into multiple sub-signals according to the preset wavelength channel rule through a fast Fourier transform (FFT). For example, separating the high-frequency impacts of the door body and the low-frequency vibrations of the train in the subway station scenario. Then, frequency-domain demodulation processing is performed on the sub-signals of each wavelength channel to generate the frequency-domain energy components of each channel, and low-frequency slowly varying noise is eliminated through sliding window mean filtering, retaining the characteristics of high-frequency sudden increases. Subsequently, the multi-channel energy components are weighted using the inverse distance weighting method based on the sensor position information. For example, the energy weight of the short-wavelength channel of the sensors on both sides of the door body is set to 0.9, and the weight of the remote sensors is 0.2 to highlight the abnormal signals in the key area. Finally, the weighted energy components are integrated to generate vibration spectrum data for subsequent abnormal behavior determination.

[0133] Through the steps of wavelength channel division, frequency-domain demodulation, and position weighting processing in this embodiment, high-frequency abnormal impact signals and low-frequency environmental noise can be effectively separated, avoiding feature confusion caused by spectral aliasing; combined with the spatial weight assignment of the sensor positions, accurately focusing on the energy sudden increase characteristics in the key area near the door body, significantly improving the signal-to-noise ratio and the accuracy of abnormal behavior recognition.

[0134] Based on the same concept, an embodiment of the present application provides a one-way door control system based on a ground sensor grating. The following is combined with Figure 3 to provide a detailed description of the one-way door control system based on the ground sensor grating provided in the embodiment of the present application.

[0135] Figure 3 It is a structural block diagram of a one-way door control system based on a ground sensor grating shown in the embodiment of the present application.

[0136] As Figure 3 shown, the one-way door control system based on the ground sensor grating may include: An acquisition module 310 is configured to acquire vibration spectrum data and passage direction data of a passage area, where the vibration spectrum data includes position information of vibration sensors and frequency domain energy distribution information, and the passage direction data includes an object moving direction, a position sequence, and layout information of ground induction gratings for acquiring the passage direction data.

[0137] A determination module 320 is configured to determine time series fluctuation information based on the frequency domain energy distribution information and preset frequency domain energy distribution characteristics associated with normal passage behaviors and abnormal behaviors.

[0138] A division module 330 is configured to perform time interval and spatial distance continuity verification according to the position sequence and the layout information to obtain spatio-temporal association information, and divide the position sequence according to a preset spatio-temporal resolution based on the layout information to generate a spatio-temporal grid distribution.

[0139] A generation module 340 is configured to generate a determination parameter characterizing the interaction relationship between normal passage behaviors and abnormal behaviors based on the spatio-temporal association information, the object moving direction, the position information, the spatio-temporal grid distribution, and the time series fluctuation information.

[0140] The determination module 320 is further configured to compare the determination parameter with a preset threshold range to generate a control instruction, and complete the control of the one-way door through the control instruction.

[0141] In one embodiment, the generation module 340 is specifically configured to extract a grating coordinate set corresponding to a reverse movement path in the spatio-temporal grid distribution, determine a movement change trend based on the object moving direction, generate a spatio-temporal association feature vector based on the grating coordinate set and the movement change trend, and the spatio-temporal association feature vector is used to characterize the distribution characteristics of abnormal behaviors in the spatial dimension; perform time window synchronization processing on the time series fluctuation information, and align the time window with the time span of the reverse movement path in the spatio-temporal grid distribution; calculate the frequency domain energy distribution weight of each grid unit in the spatio-temporal grid distribution based on the synchronized time series fluctuation information and the position information, and generate a determination parameter according to the spatio-temporal association feature vector and the frequency domain energy distribution weight, and the determination parameter is used to quantify the interaction relationship between normal passage behaviors and abnormal behaviors in the spatio-temporal and frequency domain dimensions.

[0142] In one embodiment, the generating module 340 is specifically configured to calculate the energy distribution stability index of each grid cell within a time window based on the synchronized timing fluctuation information and the spatio-temporal grid distribution, and determine the sensor coverage integrity feature of each grid cell based on the location information and the spatio-temporal grid distribution; construct a spatial mapping relationship according to the energy distribution stability index and the sensor coverage integrity feature of each grid cell, and the spatial relationship of each grid cell in the spatio-temporal grid distribution; perform frequency band decomposition on the synchronized timing fluctuation information, and extract the energy value of a preset target frequency band associated with the abnormal behavior; according to the spatial mapping relationship, allocate the energy value of the target frequency band to the corresponding spatio-temporal grid cell, and generate the initial frequency domain energy weight of each grid cell within the time window in the spatio-temporal grid distribution; perform an element-wise multiplication operation on the spatio-temporal correlation feature vector and the initial frequency domain energy weight according to the grid cell to generate a determination parameter.

[0143] In one embodiment, the generating module 340 is specifically configured to construct a spatial weight model based on the spatial mapping relationship, where the spatial weight model includes the energy allocation ratio of each grid cell within the coverage range of each vibration sensor; perform normalization processing on the energy value of the target frequency band to obtain the energy contribution value of each vibration sensor within the time window; according to the spatial weight model, allocate the energy contribution value of each vibration sensor to the grid cells covered by each vibration sensor to generate the initial frequency domain energy weight of each grid cell.

[0144] In one embodiment, the generating module 340 is further configured to, before performing an element-wise multiplication operation on the spatio-temporal correlation feature vector and the initial frequency domain energy weight according to the grid cell to generate a determination parameter, perform dynamic weighted adjustment on the initial frequency domain energy weight based on the reverse trigger frequency of each grid cell in the spatio-temporal correlation feature vector to generate an adjusted frequency domain energy distribution weight; perform a multiplication operation on the spatio-temporal correlation feature vector and the frequency domain energy weight according to the grid cell to generate a determination parameter.

[0145] In one embodiment, the determining module 320 is specifically configured to decompose the frequency domain energy distribution information into energy components of multiple frequency bands according to a preset frequency band division rule, where the preset frequency band division rule is defined according to the frequency domain energy distribution characteristics of normal passage behavior and abnormal behavior; extract the target frequency band energy component associated with the abnormal behavior from the energy components of multiple frequency bands after frequency band division, and perform dynamic baseline calibration on the target frequency band energy component to generate a calibrated target frequency band energy sequence; based on the target frequency band energy sequence, determine the start time point, end time point and sudden increase amplitude of the energy sudden increase sequence exceeding a preset change rate threshold to generate an energy sudden increase sequence set; generate timing fluctuation information according to the time distribution density and the sudden increase amplitude change rate of the sequences in the energy sudden increase sequence.

[0146] In one embodiment, the obtaining module 310 is further configured to, before obtaining the vibration spectrum data and the passing direction data of the passing area, obtain the original vibration signal of the vibration sensor and the position information of the vibration sensor; decompose the original vibration signal into vibration sub-signals of multiple independent wavelength channels according to a preset wavelength channel division rule; perform frequency-domain demodulation processing on the vibration sub-signals of each wavelength channel, and extract the frequency-domain energy components of each wavelength channel; and perform weighting processing on the frequency-domain energy components of each wavelength channel according to the position information of the vibration sensor to generate vibration spectrum data.

[0147] Figure 3 Each module in the system shown has the functions of implementing Figure 1 and Figure 2 each step in, and can achieve its corresponding technical effects. For the sake of brevity, they will not be described herein again.

[0148] Figure 4 FIG. shows a schematic hardware structure diagram of an electronic device provided by an embodiment of the present application.

[0149] The electronic device may include a processor 410 and a memory 420 storing computer program instructions.

[0150] Specifically, the processor 410 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0151] The memory 420 may include a mass storage for data or instructions. By way of example and not limitation, the memory 420 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In a suitable case, the memory 420 may include a removable or non-removable (or fixed) medium. In a suitable case, the memory 420 may be internal or external to the integrated gateway disaster recovery device. In a specific embodiment, the memory 420 is a non-volatile solid state memory.

[0152] The memory may include a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of the present disclosure.

[0153] The processor 410 reads and executes the computer program instructions stored in the memory 420 to implement any one of the above-described ground-sensing grating-based one-way door control methods in the embodiments.

[0154] In one example, the electronic device may further include a communication interface 430 and a bus 440. Among them, as Figure 4 shown, the processor 410, the memory 420, and the communication interface 430 are connected through the bus 440 and complete communication with each other.

[0155] The communication interface 430 is mainly used to implement communication between each module, device, unit, and / or device in the embodiments of the present application.

[0156] The bus 440 includes hardware, software, or both, and couples the components of the online data flow metering device to each other. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front-Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses or a combination of two or more of these. In a suitable case, the bus 440 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0157] The electronic device can execute the ground-sensing grating-based one-way door control method in the embodiments of the present application, thereby implementing the ground-sensing grating-based one-way door control method described in combination with Figure 1 and Figure 2 described.

[0158] In addition, in combination with the one-way door control method based on the ground sensor grating in the above embodiments, an embodiment of the present application can provide a computer-readable storage medium to implement. Computer program instructions are stored on the computer-readable storage medium; when the computer program instructions are executed by a processor, any one of the one-way door control methods based on the ground sensor grating in the above embodiments is implemented.

[0159] It should be clear that the present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0160] The functional blocks shown in the above structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, etc. When implemented in software, the elements of the present application are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted via a data signal carried in a carrier wave on a transmission medium or a communication link. A "machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical discs, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet or an intranet.

[0161] It should also be noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, can be different from the order in the embodiments, or several steps can be executed simultaneously.

[0162] As described above with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each block in the flowchart and / or block diagram, and the combination of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to generate a machine, such that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the functions / actions specified in one or more blocks of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It should also be understood that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can also be implemented by dedicated hardware that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0163] As described above, the foregoing is only the specific implementation manner of the present application. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, modules, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered by the protection scope of the present application.

Claims

1. A one-way door control method based on a ground sensor grating, characterized in that Including: Obtain the vibration spectrum data and the passage direction data of the passage area, where the vibration spectrum data includes the position information and the frequency domain energy distribution information of the vibration sensor, and the passage direction data includes the object movement direction, the position sequence, and the layout information of the ground loop grating for collecting the passage direction data; Based on the frequency domain energy distribution information and the preset frequency domain energy distribution characteristics associated with normal passage behaviors and abnormal behaviors, determine the timing fluctuation information corresponding to the abnormal behavior; According to the position sequence and the layout information, perform time interval and spatial distance continuity verification to obtain spatio-temporal correlation information, and based on the layout information, divide the position sequence according to a preset spatio-temporal resolution to generate a spatio-temporal grid distribution; Based on the spatio-temporal correlation information, the object movement direction, the position information, the spatio-temporal grid distribution, and the timing fluctuation information, generate a determination parameter characterizing the interaction relationship between normal passage behaviors and abnormal behaviors; Compare the determination parameter with a preset threshold range to generate a control instruction, and complete the control of the one-way door through the control instruction.

2. The method according to claim 1, wherein The generating a determination parameter characterizing the interaction relationship between normal passage behaviors and abnormal behaviors based on the spatio-temporal correlation information, the object movement direction, the position information, the spatio-temporal grid distribution, and the timing fluctuation information includes: Extract the grating coordinate set corresponding to the reverse movement path in the spatio-temporal grid distribution, determine the movement change trend based on the object movement direction, and generate a spatio-temporal correlation feature vector based on the grating coordinate set and the movement change trend, where the spatio-temporal correlation feature vector is used to characterize the distribution characteristics of the abnormal behavior in the spatial dimension; Perform time window synchronization processing on the timing fluctuation information, where the time window is aligned with the time span of the reverse movement path in the spatio-temporal grid distribution; Based on the synchronized timing fluctuation information and the position information, calculate the frequency domain energy distribution weights of each grid unit in the spatio-temporal grid distribution, and generate a determination parameter according to the spatio-temporal correlation feature vector and the frequency domain energy distribution weights, where the determination parameter is used to quantify the interaction relationship between normal passage behaviors and abnormal behaviors in the spatio-temporal and frequency domain dimensions.

3. The method according to claim 2, characterized in that, The calculating the frequency domain energy distribution weights of each grid unit in the spatio-temporal grid distribution based on the synchronized timing fluctuation information and the position information, and generating a determination parameter according to the spatio-temporal correlation feature vector and the frequency domain energy distribution weights, where the determination parameter is used to quantify the interaction relationship between normal passage behaviors and abnormal behaviors in the spatio-temporal and frequency domain dimensions includes: Based on the synchronized timing fluctuation information and the spatio-temporal grid distribution, calculate the energy distribution stability index of each grid unit within the time window, and based on the position information and the spatio-temporal grid distribution, determine the sensor coverage integrity feature of each grid unit; According to the energy distribution stability index and the sensor coverage integrity feature of each grid unit, and the spatial relationship of each grid unit in the spatio-temporal grid distribution, construct a spatial mapping relationship; Perform frequency band decomposition on the synchronized timing fluctuation information, and extract the energy value of a preset target frequency band associated with the abnormal behavior; According to the spatial mapping relationship, allocate the energy value of the target frequency band to the corresponding spatio-temporal grid cells, and generate the initial frequency domain energy weights of each grid cell within the time window in the spatio-temporal grid distribution; Perform a multiplication operation on the spatio-temporal correlation feature vector and the initial frequency domain energy weights according to the grid cells to generate the determination parameter.

4. The method according to claim 3, wherein The step of allocating the energy value of the target frequency band to the corresponding spatio-temporal grid cells according to the spatial mapping relationship to generate the initial frequency domain energy weights of each grid cell within the time window in the spatio-temporal grid distribution includes: Based on the spatial mapping relationship, construct a spatial weight model, where the spatial weight model includes the energy allocation ratio of each grid cell within the coverage range of each vibration sensor; Perform normalization processing on the energy value of the target frequency band to obtain the energy contribution value of each vibration sensor within the time window; According to the spatial weight model, allocate the energy contribution value of each vibration sensor to the grid cells covered by each vibration sensor to generate the initial frequency domain energy weights of each grid cell.

5. The method according to claim 3 or 4, characterized in that Before performing the multiplication operation on the spatio-temporal correlation feature vector and the initial frequency domain energy weights according to the grid cells to generate the determination parameter, the method further includes: Based on the reverse trigger frequency of each grid cell in the spatio-temporal correlation feature vector, perform dynamic weighted adjustment on the initial frequency domain energy weights to generate adjusted frequency domain energy distribution weights; The step of performing a multiplication operation on the spatio-temporal correlation feature vector and the initial frequency domain energy weights according to the grid cells to generate the determination parameter includes: Perform a multiplication operation on the spatio-temporal correlation feature vector and the frequency domain energy weights according to the grid cells to generate the determination parameter.

6. The method according to claim 1, characterized in that, The step of determining the timing fluctuation information corresponding to the abnormal behavior based on the frequency domain energy distribution information and the preset frequency domain energy distribution characteristics associated with normal passage behavior and abnormal behavior includes: Decompose the frequency domain energy distribution information into energy components of multiple frequency bands according to a preset frequency band division rule, where the preset frequency band division rule is defined according to the frequency domain energy distribution characteristics of the normal passage behavior and the abnormal behavior; Extract the target frequency band energy components associated with the abnormal behavior from the energy components of the multiple frequency bands after frequency band division, and perform dynamic baseline calibration on the target frequency band energy components to generate a calibrated target frequency band energy sequence; Based on the target frequency band energy sequence, determine the start time point, end time point and sudden increase amplitude of the energy sudden increase sequence exceeding the preset change rate threshold, and generate an energy sudden increase sequence set; Generate the timing fluctuation information according to the time distribution density of the sequences in the energy sudden increase sequence and the change rate of the sudden increase amplitude.

7. The method according to claim 1, wherein Before obtaining the vibration spectrum data and the passage direction data of the passage area, the method further includes: Obtain the original vibration signal of the vibration sensor and the position information of the vibration sensor; Decompose the original vibration signal into vibration sub-signals of multiple independent wavelength channels according to a preset wavelength channel division rule; Perform frequency-domain demodulation processing on the vibration sub-signals of each wavelength channel to extract the frequency-domain energy components of each wavelength channel; According to the position information of the vibration sensor, perform weighting processing on the frequency-domain energy components of each wavelength channel to generate the vibration spectrum data.

8. A one-way door control system based on a ground sensor grating, characterized in that, The system includes: An acquisition module, configured to acquire vibration spectrum data and traffic direction data of a traffic area, where the vibration spectrum data includes the position information and frequency-domain energy distribution information of a vibration sensor, and the traffic direction data includes the object movement direction, position sequence, and layout information of a ground induction grating for collecting the traffic direction data; A determination module, configured to determine the timing fluctuation information corresponding to the abnormal behavior based on the frequency-domain energy distribution information and preset frequency-domain energy distribution characteristics associated with normal traffic behaviors and abnormal behaviors; A division module, configured to perform time interval and spatial distance continuity verification according to the position sequence and the layout information to obtain spatio-temporal correlation information, and divide the position sequence according to a preset spatio-temporal resolution based on the layout information to generate a spatio-temporal grid distribution; A generation module, configured to generate a determination parameter characterizing the interaction relationship between normal traffic behaviors and abnormal behaviors based on the spatio-temporal correlation information, the object movement direction, the position information, the spatio-temporal grid distribution, and the timing fluctuation information; The determination module is further configured to compare the determination parameter with a preset threshold range to generate a control instruction, and complete the control of the one-way gate through the control instruction.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the one-way gate control method based on a ground induction grating according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Computer program instructions are stored on a computer-readable storage medium, and when the computer program instructions are executed by a processor, they implement the one-way gate control method based on a ground induction grating according to any one of claims 1-7.

Images