A video portrait capture prevention and control method and system

Abstract

The application discloses a video portrait capturing prevention and control method and system, relates to the technical field of behavior analysis and emergency management, and comprises the following steps: acquiring the current position of a target person, the starting position of deviating from a preset evacuation path and the position of a disaster source; calculating the moving direction of the target person according to the current position and the starting position of deviating from the preset evacuation path; calculating the relative direction of the disaster source according to the current position and the position of the disaster source; calculating the first direction included angle according to the moving direction and the relative direction of the disaster source; determining the personnel deviating behavior property information according to the first direction included angle and a preset reference angle; and performing prevention and control according to the personnel deviating behavior property information. The application realizes prevention and control, and improves the recognition accuracy and safety.

Description

Technical Field

[0001] This invention relates to the fields of behavior analysis and emergency management technology, and in particular to a video human image capture and control method and system. Background Technology

[0002] Video-based security systems play a crucial role in modern community safety management. These systems typically deploy video capture devices at key locations within the community to record and analyze human activity. They can identify individuals entering the community and assign them access permits with time and spatial restrictions. If an individual is detected outside their permitted area or time, an alarm is triggered, notifying security personnel. In the event of sudden disasters such as fires or chemical spills, the existing access permit system must be immediately suspended, switching to emergency mode and planning mandatory evacuation routes based on the location of the disaster source. However, individuals in a state of extreme panic often behave irrationally. Security centers struggle to accurately determine whether individuals are choosing alternative escape routes out of panic or harboring malicious intent, resulting in low accuracy and potentially delaying rescue efforts or causing malicious damage to facilities, thus compromising security.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this invention is to propose a video portrait capture and control method and system that can combine positional relationships and movement directions to judge deviation behavior in order to achieve control, thereby improving recognition accuracy and security.

[0005] On one hand, embodiments of the present invention provide a video portrait capture and control method, comprising the following steps:

[0006] Obtain the current location of the target personnel, their starting point of deviation from the preset evacuation route, and the location of the disaster source;

[0007] Calculate the movement direction of the target personnel based on the current location and the starting position of the deviation from the preset evacuation route;

[0008] Calculate the relative direction of the disaster source based on the current location and the location of the disaster source;

[0009] Calculate the angle between the first direction and the relative direction of the disaster source based on the direction of movement;

[0010] Based on the included angle of the first direction and the preset reference angle, determine the nature of the person's deviation behavior;

[0011] Based on the information regarding the nature of the individuals' deviant behavior, prevention and control measures will be implemented.

[0012] In some embodiments, determining the nature of the person's deviation behavior based on the first directional angle and a preset reference angle includes:

[0013] Obtain the movement trajectory information of the target personnel after they deviate from the preset evacuation route;

[0014] The motion trajectory information is analyzed to obtain the trajectory analysis results;

[0015] If the trajectory analysis result indicates the existence of a preset movement pattern, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance, and the preset movement pattern includes a sudden drop in speed or a reversal of direction;

[0016] If the trajectory analysis result indicates that there is no preset motion pattern, then determine whether the included angle of the first direction is less than the preset reference angle;

[0017] If the included angle of the first direction is less than the preset reference angle, then the nature of the personnel deviation behavior is determined to be a high-risk abnormal behavior;

[0018] If the angle of the first direction is greater than or equal to the preset reference angle, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance.

[0019] In some embodiments, performing trajectory analysis on the motion trajectory information to obtain trajectory analysis results includes:

[0020] Obtain a list of location data from the motion trajectory information, the list of location data including multiple consecutive location coordinates of the target person;

[0021] Based on the location data list, calculate the changes in movement speed and direction of movement within a preset time period during which the deviation occurs;

[0022] If the change in movement speed satisfies the condition of a sudden drop in speed and the change in movement direction satisfies the condition of a reversal of direction, then the trajectory analysis result is determined to indicate the existence of a preset movement pattern; otherwise, the trajectory analysis result is determined to indicate the absence of a preset movement pattern.

[0023] The conditions for the sudden drop in speed are that the moving speed decreases from greater than a first speed threshold to less than a second speed threshold and the deceleration rate is greater than a preset deceleration rate threshold, wherein the first speed threshold is greater than the second speed threshold. The conditions for the reversal of direction are that the reversal angle is greater than a preset angle threshold and the angular velocity is greater than a preset angular velocity threshold. The reversal angle is the angle between the direction before the sudden drop in speed and the direction after the sudden drop in speed.

[0024] In some embodiments, after implementing prevention and control measures based on the information regarding the nature of the individual's deviant behavior, the method further includes:

[0025] Get the updated current position;

[0026] Based on the updated current location, determine the prevention and control compliance status of the target personnel;

[0027] Based on the updated current location and the location of the disaster source, determine the movement trend of the target personnel;

[0028] If the control and prevention status is not followed or the movement trend changes, then the information on the nature of the person's deviation behavior is updated to high-risk abnormal behavior.

[0029] Based on the updated information regarding the nature of the personnel's deviant behavior, prevention and control measures will be implemented.

[0030] In some embodiments, determining the prevention and control compliance status of the target personnel based on the updated current location includes:

[0031] Obtain the safe direction indicated by the prevention and control response;

[0032] Update the movement direction based on the updated current position;

[0033] Based on the safety direction and the updated movement direction, a consistency analysis is performed to obtain the consistency analysis results;

[0034] If the consistency analysis result is consistent, then the control and prevention compliance status is determined to be "followed control and prevention"; otherwise, the control and prevention compliance status is determined to be "not followed control and prevention".

[0035] In some embodiments, determining the prevention and control compliance status of the target personnel based on the updated current location includes:

[0036] Obtain the safe zones indicated by the prevention and control response;

[0037] Based on the updated current location and the aforementioned safe zone, a location relationship analysis is performed to obtain the location relationship analysis results;

[0038] If the location relationship analysis result indicates that the location is safe, then the control and prevention compliance status is determined to be "controlled and prevention measures have been followed"; otherwise, the control and prevention compliance status is determined to be "not controlled and prevention measures have been followed".

[0039] In some embodiments, the step of performing a consistency analysis based on the security direction and the updated movement direction to obtain a consistency analysis result includes:

[0040] Generate a preset safety angle sector based on the stated safety direction;

[0041] If the updated direction of movement is within the preset safety angle sector, the consistency analysis result is determined to be consistent; otherwise, the consistency analysis result is determined to be inconsistent.

[0042] In some embodiments, the step of performing a positional relationship analysis based on the updated current location and the avoidance area to obtain the positional relationship analysis result includes:

[0043] Obtain disaster source information, which includes type, intensity, diffusion characteristics, and environmental topography information;

[0044] Based on the disaster source information, calculate the potential impact range of the disaster source;

[0045] The boundaries of the evacuation area are updated based on the dynamic evolution of the disaster source and the potential impact range.

[0046] The updated evacuation zone is divided into a core safety zone and an edge transition zone.

[0047] If the updated current location is within the core security area, then the location relationship analysis result is determined to be location safe;

[0048] If the updated current position is not within the core security area, then determine whether the updated current position is within the edge transition area;

[0049] If the updated current position is in the edge transition region and the movement trend tends towards the core security region, then the position relationship analysis result is determined to be a safe position; otherwise, the position relationship analysis result is determined to be an unsafe position.

[0050] In some embodiments, after determining the nature of the person's deviation behavior based on the first directional angle and a preset reference angle, the method further includes:

[0051] Get the topological properties of the off-path segments;

[0052] If the topological attribute is a single restricted channel, then obtain the direction of motion before the deviation occurs;

[0053] Calculate the included angle of the second direction based on the direction of movement before the deviation occurred and the direction of movement;

[0054] If the angle of the second direction is greater than the preset reverse angle, then the nature of the personnel deviation behavior is updated to stress-induced risk avoidance; otherwise, the nature of the personnel deviation behavior is updated to high-risk abnormal behavior.

[0055] On the other hand, embodiments of the present invention provide a video portrait capture and control system, comprising:

[0056] The location acquisition module is used to acquire the current location of the target personnel, the starting location of the deviation from the preset evacuation route, and the location of the disaster source;

[0057] The personnel movement direction determination module is used to calculate the movement direction of the target personnel based on the current position and the starting position of the deviation from the preset evacuation path;

[0058] The disaster source relative direction determination module is used to calculate the relative direction of the disaster source based on the current position and the disaster source position;

[0059] Angle calculation module, used to calculate the angle in the first direction based on the moving direction and the relative direction of the disaster source;

[0060] The nature determination module is used to determine the nature information of the person's deviation behavior based on the first directional angle and the preset reference angle.

[0061] The control module is used to perform prevention and control measures based on the information regarding the nature of the personnel's deviation behavior.

[0062] The embodiments of this application include at least the following beneficial effects: The embodiments of this application first obtain the current location of the target personnel, the starting location of the deviation from the preset evacuation route, and the location of the disaster source. Then, based on the current location and the starting location of the deviation from the preset evacuation route, the movement direction of the target personnel is calculated, and based on the current location and the location of the disaster source, the relative direction of the disaster source is calculated. Then, based on the movement direction and the relative direction of the disaster source, the first direction angle is calculated, and based on the first direction angle and the preset reference angle, the nature information of the personnel's deviation behavior is determined. Finally, based on the nature information of the personnel's deviation behavior, prevention and control are carried out. Thus, it is possible to combine the positional relationship and the movement direction to judge the deviation behavior in order to achieve prevention and control, thereby improving the accuracy and safety of identification.

[0063] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and the drawings. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is a flowchart of a video portrait capture and control method according to an embodiment of the present invention;

[0066] Figure 2 This is a schematic diagram of the structure of a video human image capture and control system according to an embodiment of the present invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0068] In related technologies, traditional video security and control systems, under emergency evacuation mode, suffer from the problem of being unable to effectively distinguish the nature of deviation behavior when personnel deviate from the preset evacuation route. The system can only identify that personnel have deviated from the designated path, but cannot determine whether it is a reactive avoidance of danger or an abnormal movement with potential safety risks. This lack of identification capability prevents the system from providing differentiated judgment criteria for subsequent prevention and control, affecting the efficiency and accuracy of emergency response.

[0069] For example, suppose a large industrial park has an integrated video surveillance and emergency evacuation system. When the system detects a chemical spill in an area of ​​the park, it immediately activates the emergency evacuation plan and plans and issues mandatory evacuation routes for workers in the area. A worker, while moving along the system-planned route, senses an unusual odor or sound from the direction of the spill source and, out of instinct, changes direction, deviating from the planned evacuation route and turning onto what they perceive as a safer path. At this moment, the video capture device acquires the worker's current location information in real time and identifies the deviation from the preset route. However, based on its inherent logic, the system can only classify this deviation as "path mismatch" and trigger a uniform alarm. The system cannot determine, based on the relationship between the worker's direction of movement and the location of the disaster source, whether the deviation is an reactive avoidance towards a safe area or an abnormal movement that might lead to entering a more dangerous area. This single judgment makes it impossible for the emergency command center to immediately decide whether to provide guidance through the broadcast system or to dispatch on-site personnel to intervene, which may delay the best time for emergency response.

[0070] If the aforementioned problems are not addressed, video-based security and control systems will be unable to finely categorize deviations in personnel behavior during emergency disasters. This will lead the system to adopt a uniform response strategy for all deviations, such as triggering the highest-level interception command for all deviations. This lack of differentiation in control may result in unnecessary intervention for individuals in a state of stress-induced risk aversion, wasting valuable emergency resources and potentially hindering their effective avoidance of danger. Simultaneously, for abnormal movements that genuinely pose a safety risk, the system may fail to promptly identify their high-risk nature, delaying the implementation of stronger control measures and increasing the risk of casualties or property damage. This inefficiency in the response mechanism will severely weaken the system's decision-making support capabilities and overall emergency management effectiveness in emergency situations.

[0071] Faced with the aforementioned problems, this application initially considered analyzing a person's movement trajectory to determine the nature of their deviation behavior. For example, it could attempt to identify pre-defined patterns such as sudden stops or reversals in the trajectory to infer whether the person was in a stress state. However, simple trajectory analysis may not be able to completely distinguish all situations. For instance, a person might continue moving in the wrong direction at a normal speed after deviating from the path, which might not show obvious stress patterns on the trajectory, but their behavior still carries a high risk. Therefore, this application further considers whether, in addition to the trajectory itself, the relationship between the person's current location and the location of the disaster source can be introduced as a basis for judgment. If the person's movement direction is away from the disaster source, even if they deviate from the pre-defined path, their behavior may fall within the scope of risk avoidance; conversely, if their movement direction is towards or parallel to the disaster source, the risk of their deviation behavior will significantly increase. Therefore, this application conceives a method that combines the person's movement direction with the relative direction of the disaster source for angular analysis to more accurately assess the nature of the deviation behavior.

[0072] The embodiments of this application will be explained in detail below with reference to the accompanying drawings:

[0073] Figure 1 This is an optional flowchart of a video portrait capture and control method provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0074] Step S101: Obtain the current location of the target personnel, the starting location of their deviation from the preset evacuation route, and the location of the disaster source;

[0075] Step S102: Calculate the movement direction of the target personnel based on the current location and the starting position deviating from the preset evacuation route;

[0076] Step S103: Calculate the relative direction of the disaster source based on the current location and the location of the disaster source;

[0077] Step S104: Calculate the angle between the first direction and the relative direction of the disaster source based on the direction of movement;

[0078] Step S105: Determine the nature of the personnel deviation behavior based on the first direction angle and the preset reference angle;

[0079] Step S106: Implement prevention and control measures based on information about the nature of personnel deviation behaviors.

[0080] Steps S101 to S106 shown in the embodiments of this application can combine positional relationships and movement directions to determine deviation behavior in order to achieve prevention and control, thereby improving the accuracy and security of identification.

[0081] In some embodiments, steps S101-S106 aim to determine the reasons for a person's deviation from a preset evacuation route by analyzing their movement behavior in an emergency, thereby enabling more precise prevention and control measures. First, the current location of the target person, the starting location of the deviation from the preset evacuation route, and the location of the hazard source can be obtained. This location information provides spatial references for the person, the deviation point, and the hazard source. Then, based on the current location and the starting location of the deviation from the preset evacuation route, the movement direction of the target person is calculated, reflecting the immediate movement trend after the deviation occurs. Based on the current location and the location of the hazard source, the relative direction of the hazard source is calculated, clarifying the orientation of the hazard source relative to the target person. Then, based on the movement direction and the relative direction of the hazard source, a first directional angle is calculated. The first directional angle reflects the degree of consistency between the person's movement direction and the direction away from the hazard source. The smaller the angle, the more the person tends to move away from the danger; the larger the angle, the more likely the person is moving towards the hazard source or engaging in other non-avoidance behaviors. Based on the first directional angle and a preset reference angle, the nature of the person's deviation behavior is determined, classifying it as stress-induced avoidance or high-risk abnormal behavior. This classification mechanism allows the system to go beyond simple "violation" judgments and delve deeper into the underlying motivations behind personnel behavior. Finally, based on the nature of the deviation from the expected behavior, control measures are implemented. For behaviors judged as reactive risk-avoidance, gentle guidance measures can be taken, such as reassurance and guidance to safe areas via broadcasts. For behaviors judged as high-risk abnormalities, more stringent intervention measures may be necessary, such as dispatching security personnel to intercept or issuing higher-level alarms. This differentiated control strategy avoids the one-size-fits-all approach of traditional systems, improves response efficiency and resource utilization in emergency situations, and ensures personnel safety.

[0082] Understandably, the direction of movement refers to the vector direction of the target person from their initial position to their current position. This can be determined by calculating the displacement vector between two consecutive points, such as by the coordinate difference between the current position and the previous position. Its primary purpose is to reflect the instantaneous movement trend of the person. The relative direction to the disaster source refers to the vector direction from the target person's current position to the disaster source's location. This can be determined by calculating the direction vector between the target person's current position and the disaster source's location, such as by the coordinate difference between the current position and the disaster source. Its primary purpose is to determine the location of the disaster source relative to the person. The preset reference angle refers to the threshold angle used to judge the nature of the person's deviation behavior. This can be set using expert experience, historical data analysis, or simulation. For example, it can be a preset safe angle range based on different disaster types and scenarios. Its primary purpose is to provide an objective judgment standard to distinguish between reactive risk avoidance and high-risk abnormal behavior. Information on the nature of the person's deviation behavior refers to the classification judgment result of the target person's deviation from the preset evacuation route. This can include categories such as "reactive risk avoidance" or "high-risk abnormal behavior." Its primary purpose is to provide a basis for subsequent differentiated prevention and control measures. Prevention and control refers to the corresponding intervention measures taken based on information about the nature of personnel's deviation behavior. These measures may include voice broadcast guidance, SMS notification, security personnel intervention, area lockdown or route replanning, etc. For example, voice reassurance and route guidance are provided for stress-induced risk avoidance behavior, and real-time interception or alarm is conducted for high-risk abnormal behavior. The main purpose is to ensure personnel safety and maintain on-site order.

[0083] To illustrate this technical solution more clearly, a specific example is provided below. First, video surveillance cameras deployed throughout the community, combined with image recognition and target tracking algorithms, acquire the real-time location of individuals, represented by geographic coordinates or indoor grid coordinates. When the system detects a person deviating from a preset evacuation route, it records the starting point of this deviation. Simultaneously, disaster source location information can be received from the community's fire alarm system or environmental monitoring system, such as the specific floor and room number where the fire occurred. Next, to calculate the movement direction of the individuals, the system continuously acquires multiple current location points and uses the time-series data of these consecutive location points to calculate their instantaneous movement direction using vector calculations. For example, if a person is at point P1 at time t1 and at point P2 at time t2, their movement direction can be represented by the vector P1P2. Simultaneously, the relative direction of the disaster source can be determined by calculating the vector from the individual's current location to the disaster source location. Subsequently, the system uses the calculated movement direction vector and the relative disaster source direction vector to calculate the first directional angle through geometric operations such as vector dot product or cross product. For example, if two vectors are close in direction, the included angle will approach zero; if they are opposite in direction, the included angle will approach 180 degrees. Based on this, the system compares the calculated first directional included angle with a preset reference angle. This preset reference angle can be set according to the actual evacuation scenario and safety requirements, for example, 90 degrees. If the first directional included angle is less than the preset reference angle, it is determined that the person's movement direction is basically consistent with the direction away from the disaster source, thus determining the nature of the person's deviation behavior as reactive avoidance. Conversely, if the first directional included angle is greater than or equal to the preset reference angle, it is determined as high-risk abnormal behavior. Finally, based on the determined nature of the behavior, the system executes corresponding prevention and control measures. For example, if it is determined to be reactive avoidance, the system can play voice prompts to the target person through the public address system, guiding them to the nearest safe exit and displaying safe route guidance. If it is determined to be high-risk abnormal behavior, the system can immediately send alarm information to the security personnel's mobile terminals, along with the target person's real-time location and image, and simultaneously activate an area lockdown mechanism to restrict the person's further movement to prevent potential sabotage or dangerous behavior.

[0084] Through the above technical solution, this embodiment, by introducing analysis of the direction of personnel movement relative to the disaster source and combining it with a preset reference angle for judgment, achieves intelligent classification of personnel deviation behavior. It can accurately identify whether the deviation is an instinctive, reactive avoidance behavior or an abnormal movement with potential safety risks. This differentiated judgment capability makes subsequent prevention and control more precise and efficient, avoiding the need for blanket mandatory measures for all deviation behaviors, thereby reducing unnecessary resource waste, improving the targeting of emergency response, protecting personnel safety, and maintaining order in emergency situations.

[0085] In some embodiments, in step S105, determining the nature of the person's deviation behavior based on the first direction angle and the preset reference angle may include, but is not limited to, the following steps:

[0086] Step S201: Obtain the movement trajectory information of the target personnel after deviating from the preset evacuation route;

[0087] Step S202: Perform trajectory analysis on the motion trajectory information to obtain the trajectory analysis results;

[0088] Step S203: If the trajectory analysis result shows that there is a preset movement pattern, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance. The preset movement pattern includes a sudden drop in speed or a reversal of direction.

[0089] Step S204: If the trajectory analysis result indicates that there is no preset motion mode, then determine whether the included angle of the first direction is less than the preset reference angle;

[0090] Step S205: If the included angle in the first direction is less than the preset reference angle, then the nature of the personnel deviation behavior is determined to be a high-risk abnormal behavior.

[0091] Step S206: If the included angle of the first direction is greater than or equal to the preset reference angle, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance.

[0092] In some embodiments, relying solely on the included angle of a first direction and a preset reference angle may lead to misjudgment of the nature of a person's behavior. For example, when a person is in a state of panic, their movement trajectory may exhibit non-linear characteristics, and simply judging by angle is insufficient to accurately distinguish between stress-induced risk avoidance and high-risk abnormal behavior. Therefore, it is necessary to more accurately identify the true intentions behind a person's deviation from the prescribed evacuation route to avoid unnecessary panic or delays in rescue. Information on the movement trajectory of the target person after deviating from the preset evacuation path can be obtained first. This information includes more detailed behavioral characteristics such as the person's speed and direction changes. Then, trajectory analysis is performed on the movement trajectory information to obtain trajectory analysis results. These patterns are usually instinctive stress-induced risk avoidance responses when faced with sudden danger. If the trajectory analysis results indicate the existence of a preset movement pattern, the nature of the person's deviation behavior is determined to be stress-induced risk avoidance. The preset movement pattern includes a sudden drop in speed or a reversal of direction. This mechanism can capture non-linear movements caused by panic or sudden situations, avoiding misjudgment as high-risk abnormal behavior. If the trajectory analysis result indicates that there is no preset movement pattern, the analysis backtracks to the judgment of the first direction angle. It is determined whether the first direction angle is less than the preset reference angle. If the first direction angle is less than the preset reference angle, it indicates that the movement direction of the person deviates significantly from the direction away from the disaster source, and the nature of the person's deviation behavior is determined to be high-risk abnormal behavior. If the first direction angle is greater than or equal to the preset reference angle, even if the person deviates from the preset path, their movement direction is still generally tending towards the safe area, and the nature of the person's deviation behavior is determined to be stress-induced risk avoidance. This allows for a more comprehensive and accurate understanding of the true intention of the person's deviation behavior, effectively distinguishing between instinctive risk avoidance and potential risk behavior, and thus providing a more accurate decision-making basis for subsequent prevention and control, avoiding misjudgments or delays caused by insufficient information.

[0093] Understandably, motion trajectory information refers to a continuous set of location data of a target person within a specific time period. This data can include timestamps, coordinates, speed, acceleration, etc. It can be obtained through real-time tracking and recording of personnel movement using video analytics systems or through sensor fusion technology. Its purpose is to provide richer behavioral context than a single location point, enabling in-depth behavioral pattern recognition. A sudden drop in speed refers to a rapid decrease in the target person's movement speed from a high level to a low level within a short period. Specifically, this can manifest as a sudden deceleration from running to walking or even stopping. Its purpose is to indicate that the person may have encountered a sudden obstacle, been startled, or is taking evasive action. A reversal of direction refers to a significant change in the target person's direction of movement, approaching 180 degrees. Specifically, this can manifest as a sudden turn from forward movement to the rear or side. Its purpose is to indicate that the person may be avoiding a hazard, searching for a new escape route, or reacting quickly to changes in the environment.

[0094] To illustrate this technical solution more clearly, a specific example is provided below. When the video human capture system detects that a target person has deviated from the preset evacuation route, a behavior analysis module immediately activates to acquire the target person's movement trajectory information after deviating from the route. This information can consist of continuous GPS positioning data, Wi-Fi positioning data, or pixel coordinate sequences obtained from video image analysis, along with timestamps, forming a detailed location data list. For example, the system can record the target person's precise location coordinates once per second. Subsequently, a trajectory analysis algorithm module processes this movement trajectory information in real time. This module calculates the target person's changes in speed and direction of movement within a preset time period after the deviation occurred (e.g., the most recent 5 seconds). Specifically, it calculates instantaneous speed by comparing the distance and time difference between consecutive location points, and calculates the change in direction by comparing the angle between consecutive movement direction vectors. If the trajectory analysis algorithm identifies a preset movement pattern, such as a target person's speed suddenly dropping from rapid movement (e.g., 3 meters per second) to near-stopping (e.g., 0.5 meters per second) within a short period, while their direction of movement reverses by nearly 180 degrees, this typically indicates that the person may have encountered a sudden obstacle or been startled and has taken emergency avoidance action. In this case, the system will immediately determine the nature of the person's deviation behavior as stress-induced avoidance. Conversely, if the trajectory analysis results show that the aforementioned preset movement pattern does not exist, meaning the target person's movement trajectory does not exhibit obvious stress-induced avoidance characteristics, the system will further determine whether the first directional angle is less than a preset reference angle. For example, if the first directional angle (i.e., the angle between the target person's direction of movement and the direction relative to the disaster source) is less than a preset reference angle (e.g., 45 degrees), this may mean that the target person is moving towards the disaster source or that their direction of movement is significantly deviated from the evacuation direction. In this case, the system will determine the nature of the person's deviation behavior as high-risk abnormal behavior. If the angle in the first direction is greater than or equal to the preset reference angle, even if no obvious stress-induced avoidance pattern is detected, but the direction of movement is still generally away from the disaster source, the system will determine the nature of the person's deviation behavior as stress-induced avoidance. Through this hierarchical judgment mechanism, the system can more accurately identify the true intention of the person's behavior.

[0095] Through the above technical solution, this embodiment prioritizes trajectory analysis of the target person's movement trajectory information to identify whether there are preset movement patterns such as sudden drops in speed or reversals in direction. This allows for direct determination of whether the person is in a state of stress-induced risk avoidance, thus avoiding misjudgments that may occur based solely on directional angles. When these obvious stress-induced risk avoidance patterns are not present, the judgment is further refined by combining the first directional angle. This hierarchical judgment mechanism enables the system to effectively distinguish between instinctive risk avoidance behaviors caused by panic or emergencies and abnormal movements with potential safety risks, thereby providing security personnel with more accurate decision-making basis and avoiding unnecessary waste of resources or delays in rescue opportunities.

[0096] In some embodiments, step S202 involves performing trajectory analysis on the motion trajectory information to obtain trajectory analysis results, which may include, but is not limited to, the following steps:

[0097] Obtain a list of location data from the motion trajectory information. The list of location data includes multiple consecutive location coordinates of the target person.

[0098] Based on the location data list, calculate the changes in movement speed and direction within a preset time period after the deviation occurs;

[0099] If the change in movement speed satisfies the condition of a sudden drop in speed and the change in movement direction satisfies the condition of a reversal of direction, then the trajectory analysis result is determined to be that a preset motion pattern exists; otherwise, the trajectory analysis result is determined to be that a preset motion pattern does not exist.

[0100] The conditions for a sudden drop in speed are that the moving speed decreases from greater than a first speed threshold to less than a second speed threshold and the deceleration rate is greater than a preset deceleration rate threshold, and the first speed threshold is greater than the second speed threshold. The conditions for a reversal of direction are that the reversal angle is greater than a preset angle threshold and the angular velocity is greater than a preset angular velocity threshold. The reversal angle is the angle between the direction before the sudden drop in speed and the direction after the sudden drop in speed.

[0101] In some embodiments, qualitative trajectory analysis alone is insufficient to accurately determine a person's movement pattern. For example, speed is relative, and changes in direction may only involve minor adjustments. Therefore, it is necessary to quantify movement trajectory information more precisely and identify preset movement patterns more accurately. A list of location data, including multiple consecutive coordinate points of the target person, can be obtained from the movement trajectory information. Then, based on the location data list, the changes in movement speed and direction within a preset time period during which the deviation occurred are calculated, transforming the originally qualitative movement trajectory into quantifiable speed and direction parameters, making the judgment of the person's movement pattern more objective and accurate. If the change in movement speed meets the condition of a sudden drop in speed and the change in movement direction meets the condition of a reversal in direction, the trajectory analysis result is determined to indicate the existence of a preset movement pattern; otherwise, the trajectory analysis result is determined to indicate the absence of a preset movement pattern. The condition of a sudden drop in speed is that the movement speed decreases from greater than a first speed threshold to less than a second speed threshold, and the deceleration rate is greater than a preset deceleration rate threshold, effectively capturing rapid deceleration behavior caused by emergency situations. The first speed threshold is greater than the second speed threshold. The conditions for reversing direction are that the reversal angle is greater than a preset angle threshold and the angular velocity is greater than a preset angular velocity threshold. The reversal angle is the angle between the direction before the speed drops and the direction after the speed drops. It can identify the compound action of "sudden braking and turning" that people often take when trying to avoid danger.

[0102] Understandably, the location data list refers to a series of spatial coordinate points continuously recorded by the system within a specific time period for a target person. Specifically, this can be achieved through real-time tracking of personnel via a video analytics system, serializing and storing their geographical locations or pixel coordinates at different time points. The purpose is to provide foundational data for subsequent quantitative analysis of movement status. The preset deviation time period refers to a specific time window used by the system to analyze the movement trajectory of a person after detecting a deviation from a preset evacuation path. This can be a fixed duration, such as 5 or 10 seconds, or a dynamically adjusted time period. Its purpose is to limit the analysis scope and ensure the real-time and relevant nature of the deviation assessment. Changes in movement speed refer to the changes in the target person's speed within the preset deviation time period. Specifically, this can be achieved by calculating the time difference and distance between consecutive location coordinate points to obtain instantaneous speed, and further analyzing the trends of these instantaneous speeds. The purpose is to quantify the changes in the speed of the person's movement. Changes in movement direction refer to the changes in the target person's direction of movement within the preset deviation time period. Specifically, this can be achieved by calculating the displacement vectors formed between consecutive location coordinate points and analyzing the deflection of these vectors. The purpose is to quantify the adjustment of the person's direction of movement.

[0103] To illustrate this technical solution more clearly, a specific example is used below. Assume a video-based human capture and control system continuously acquires the real-time location information of a target person and stores it as a series of timestamps and coordinate point pairs, forming motion trajectory information. When the system detects that the target person deviates from the preset evacuation path, it triggers the trajectory analysis process. Specifically, the system extracts a list of location data from the motion trajectory information over a recent period; for example, it can acquire a location coordinate point recorded every 0.1 seconds within the past 5 seconds. Then, based on these continuous location coordinate points, the system calculates the changes in the target person's speed and direction of movement within the preset time period after the deviation occurred. For example, the instantaneous speed can be obtained by calculating the Euclidean distance between adjacent coordinate points divided by the time interval, and the change in direction can be obtained by calculating the angle between adjacent displacement vectors.

[0104] Furthermore, the system determines whether these calculated changes in movement speed meet the conditions for a sudden drop in speed. For example, a first speed threshold of 2 m / s can be set, a second speed threshold of 0.5 m / s, and a preset deceleration rate threshold of 1 m / s². If the target person's speed decreases from greater than 2 m / s to less than 0.5 m / s, and its deceleration rate exceeds 1 m / s², then the conditions for a sudden drop in speed are considered met. Simultaneously, the system determines whether the change in movement direction meets the conditions for a reversal of direction. For example, a preset angle threshold of 90 degrees and a preset angular velocity threshold of 30 degrees / s can be set. If the angle between the direction before and after the sudden drop in speed is greater than 90 degrees, and the angular velocity is greater than 30 degrees / s, then the conditions for a reversal of direction are considered met. If both conditions—the conditions for a sudden drop in speed and the conditions for a reversal of direction—are simultaneously met, the system determines that the trajectory analysis results indicate the existence of a preset movement pattern, and thus classifies the nature of the person's deviation behavior as reactive risk avoidance. Conversely, if any condition is not met, the trajectory analysis result is determined to be that no preset movement pattern exists. In this case, the system will continue to determine the nature of the person's deviation behavior based on the first direction angle and the preset reference angle to distinguish between high-risk abnormal behavior and stress-induced risk avoidance. In this way, the system can accurately identify the instinctive risk avoidance actions taken by people in emergency situations, avoiding misjudgments.

[0105] Through the above technical solution, this embodiment can accurately quantify and analyze personnel movement trajectory information, overcoming the limitations of traditional qualitative analysis. By comprehensively judging sudden drops in movement speed and reversals in movement direction, and introducing specific thresholds and rate conditions, the system can more accurately identify personnel's stress-induced risk-avoidance patterns in emergency situations, such as sudden stops and turns caused by avoiding danger. This avoids misjudging normal, non-emergency movement adjustments as high-risk abnormal behaviors, thereby improving the accuracy of determining the nature of personnel deviation behaviors. This provides a more reliable basis for subsequent prevention and control decisions, helping to take timely and appropriate guidance or intervention measures to ensure personnel safety.

[0106] In some embodiments, after implementing prevention and control measures based on information about the nature of the individual's deviant behavior, the method further includes:

[0107] Step S301: Obtain the updated current position;

[0108] Step S302: Determine the prevention and control compliance status of the target personnel based on the updated current location;

[0109] Step S303: Determine the movement trend of the target personnel based on the updated current location and disaster source location;

[0110] Step S304: If the prevention and control compliance status is "not complying with prevention and control" or the movement trend changes, then update the personnel deviation behavior nature information to "high-risk abnormal behavior".

[0111] Step S305: Implement prevention and control measures based on the updated information on the nature of personnel deviation behaviors.

[0112] In some embodiments, since the prevention and control measures are only implemented once based on the initial assessment of the nature of the personnel's deviation behavior, there is a lack of real-time feedback on the effectiveness of the prevention and control measures and continuous monitoring of the subsequent behavior of the target personnel. This may result in changes in the current location of the target personnel, and the effectiveness of the prevention and control measures may not meet expectations. The target personnel may not have followed the prevention and control measures, or their movement trend may have changed. All of these situations may lead to changes in the nature of the personnel's deviation behavior, affecting the implementation of the prevention and control measures.

[0113] To improve the accuracy of identifying deviations in personnel behavior, the updated current location can be obtained first. Based on this updated location, the target personnel's compliance status with prevention and control measures can be determined, assessing whether they have acted in accordance with previous control instructions. Then, based on the updated current location and the location of the disaster source, the target personnel's movement trend can be determined to assess whether they are moving in a safe direction. If the compliance status is non-compliance or the movement trend changes to a level unfavorable to safety, it indicates that the original control measures may have failed or the target personnel's risk level has increased, and the deviation behavior information is updated to high-risk abnormal behavior. Control measures are then implemented again based on the updated deviation behavior information. This cyclical monitoring, assessment, updating, and re-control mechanism allows the prevention and control system to dynamically adjust its strategies based on the real-time behavior of target personnel and environmental changes.

[0114] Understandably, the "control compliance status" refers to the assessment result of whether the subsequent behavior of the target personnel is consistent with the requirements of the control instructions after receiving them. Specifically, this can be determined by analyzing whether the target personnel's movement direction is consistent with the safe direction, or whether their current location is within the designated safe area. Its purpose is to assess the actual effectiveness of the implemented control measures. "Movement trend" refers to the tendency of the target personnel's movement direction and speed to change within a specific time period. Specifically, this can be determined by analyzing the target personnel's continuous location data, combined with the location of the disaster source, to determine whether they are moving towards a safe area or a dangerous area. Its purpose is to predict the future behavioral trends of the target personnel and the potential risks.

[0115] To illustrate this technical solution more clearly, a specific example is used below. Suppose that in an emergency evacuation scenario, the system, based on the initial deviation behavior of a target person (e.g., determining it to be "stress-induced avoidance"), sends a voice broadcast instructing them to move towards the east exit. After issuing this instruction, the system continuously acquires the target person's updated current location using video capture devices and positioning sensors. For example, 30 seconds after the instruction is issued, the system obtains the target person's new coordinates. Then, based on this new location, the system analyzes the target person's compliance status with the control measures. If the system instruction is to move east, and the target person's direction of movement remains east, it is considered compliant; if the target person turns in another direction, it is considered non-compliant. Simultaneously, the system combines the target person's updated location with the known location of a disaster source, such as a fire source, to determine the target person's movement trend. If the target person is moving away from the fire source, the trend is safe; if they are approaching the fire source, the trend is dangerous. Suppose that in this assessment, the system detects that although the target person was initially identified as exhibiting "stress-induced risk aversion," after receiving instructions to move eastward, their movement direction reversed to the north, which is closer to the disaster source. This indicates a failure to comply with prevention and control measures and a change in movement trend. At this point, the system will immediately update the target person's deviation behavior information to "high-risk abnormal behavior." Subsequently, based on this updated "high-risk abnormal behavior" information, the system will immediately take corresponding prevention and control measures. For example, in addition to continuing to send voice warnings, it may activate physical barriers in the area or send an emergency interception command to the nearest security personnel to forcibly guide the target person away from the danger zone.

[0116] Through the above technical solution, this embodiment can promptly obtain the latest location of target personnel and assess whether they have followed prevention and control instructions and whether their movement trend is safe. When it is detected that a target personnel has not followed instructions or their movement trend has changed adversely, the system can quickly identify and update the nature of their behavior as high-risk abnormal behavior, thereby triggering corresponding prevention and control measures. This avoids the accumulation of risks caused by the failure of prevention and control or changes in personnel behavior, ensures the continuous effectiveness of prevention and control measures, and improves the accuracy and reliability of personnel evacuation and risk management in emergency situations.

[0117] In some embodiments, in step S302, determining the prevention and control compliance status of the target personnel based on the updated current location may include, but is not limited to, the following steps:

[0118] Step S401: Obtain the safe direction indicated by the prevention and control response;

[0119] Step S402: Update the movement direction based on the updated current position;

[0120] Step S403: Perform a consistency analysis based on the safety direction and the updated movement direction to obtain the consistency analysis results;

[0121] Step S404: If the consistency analysis result is consistent, then the control and prevention compliance status is determined to be "controlled and prevention measures have been followed"; otherwise, the control and prevention compliance status is determined to be "not controlled and prevention measures have been followed".

[0122] In some embodiments, judging compliance with control measures solely based on the updated current location may not be accurate enough, as individuals may move in a safe direction but not fully comply with control measures, or they may have entered a safe zone that is not necessarily absolutely safe. Therefore, a more precise assessment of whether individuals are truly complying with control measures is needed. First, the safe direction indicated by the control response can be obtained; this is the expected movement direction planned by the system for the individuals based on the current disaster situation and evacuation strategy. Then, the movement direction is updated based on the updated current location, reflecting the actual movement trajectory of the individuals. A consistency analysis is then performed based on the safe direction and the updated movement direction. By comparing the expected direction with the actual direction, it can be determined whether the individual's movement is synchronized with the control instructions, distinguishing between movement that is merely moving away from the disaster source without a clear direction and movement that is truly in accordance with the safe guidance. If the consistency analysis result is consistent, it indicates that the individual's actions are as expected and they are moving towards a safe area, thus confirming compliance with control measures. Otherwise, it indicates a deviation between their movement direction and the safe guidance, confirming non-compliance with control measures, which may require further intervention.

[0123] Understandably, the safe direction indicated by the prevention and control response refers to the expected movement direction sent to the target personnel by the prevention and control system in an emergency, based on the location of the disaster source, evacuation route planning, or the setting of safe areas. It can be represented by a vector direction, angle range, or the direction of a specific path segment, with the purpose of providing clear action guidance to the target personnel.

[0124] To illustrate this technical solution more clearly, a specific example is used below. Suppose that in an emergency evacuation scenario, the control system issues an evacuation instruction to the target personnel via broadcast or smart terminal. This instruction explicitly instructs the target personnel to move due east to reach the nearest safe exit. At this time, the system parses and obtains the indicated safe direction from the control response, for example, representing it as a vector (1,0) or an angle of 0 degrees. Simultaneously, the system continuously acquires the real-time location data of the target personnel through a video analysis module or GPS positioning module. Based on these continuous location points, for example, acquiring location coordinates once per second, the system calculates the displacement vector of the target personnel over a recent period, thereby updating their current direction of movement. For example, if the target personnel moved from coordinates (X1,Y1) to (X2,Y2) in the past second, their direction of movement can be calculated as a vector direction from (X1,Y1) to (X2,Y2). Subsequently, the system performs a consistency analysis between the acquired safe direction (e.g., due east) and the updated direction of movement (e.g., the actual direction the target personnel moved). Specifically, the angle between these two directions can be calculated. If the calculated angle is less than a preset threshold, such as 15 degrees, the two directions are considered consistent, indicating that the person is moving in the safe direction expected by the system. In this case, the system will determine the person's compliance status as "compliant with control measures." Conversely, if the calculated angle is greater than or equal to the threshold, the directions are considered inconsistent, indicating that the person's movement direction deviates from the safety guidelines, and the system will determine their compliance status as "non-compliant with control measures." In this way, the system can accurately determine in real time whether the person is acting in accordance with evacuation instructions, providing a basis for subsequent prevention and control decisions.

[0125] Through the above technical solution, this embodiment, based on the judgment of directional consistency, can identify whether the target personnel are indeed moving in the expected safe direction, avoiding misjudging behavior that appears to be away from danger but is of unclear direction as compliance with prevention and control measures. This embodiment can provide more accurate compliance status information, providing a reliable basis for subsequent prevention and control decisions, thereby improving the accuracy of personnel behavior judgment and the effectiveness of prevention and control measures in emergency situations.

[0126] In some embodiments, in step S302, determining the prevention and control compliance status of the target personnel based on the updated current location may include, but is not limited to, the following steps:

[0127] Step S501: Obtain the evacuation area indicated by the prevention and control response;

[0128] Step S502: Based on the updated current location and evacuation area, perform a location relationship analysis to obtain the location relationship analysis results;

[0129] Step S503: If the location relationship analysis result is that the location is safe, then the control and prevention compliance status is determined to be "controlled and prevention measures have been followed"; otherwise, the control and prevention compliance status is determined to be "not controlled and prevention measures have been followed".

[0130] In some embodiments, since consistency analysis based solely on movement direction may not fully and accurately reflect whether a target person is truly within a safe area or is moving towards one, especially in complex or dynamically changing disaster scenarios, relying solely on direction may lead to misjudgments. Therefore, a more accurate determination of the target person's compliance status with prevention and control measures is needed. This can be achieved by first obtaining the evacuation zone indicated by the prevention and control response. This evacuation zone is a safe area defined based on the current disaster situation and preset prevention and control strategies, providing a clear geographical safety target for the target person. Then, based on the updated current location and evacuation zone, a location relationship analysis is performed to obtain the results. This analysis aims to accurately determine whether the target person's current location is within or outside the evacuation zone. If the location relationship analysis result indicates a safe location, meaning they have successfully entered or are within the evacuation zone, the compliance status is determined to be "compliant with prevention and control measures"; otherwise, the compliance status is determined to be "non-compliant with prevention and control measures." This embodiment does not rely solely on the consistency of movement direction but directly assesses whether the person has reached or is in the designated safe space, thus avoiding the inaccuracies that may arise from relying solely on direction.

[0131] It is understandable that the evacuation area indicated by the prevention and control response refers to the geographical area that is pre-set or dynamically generated based on the disaster situation and prevention and control strategy, and is used to guide people to avoid danger and ensure safety. It can be one or more polygonal areas, circular areas, or complex-shaped areas defined by a series of coordinate points. Its purpose is to provide clear safe space guidance for target personnel.

[0132] Through the above technical solution, this embodiment can accurately identify whether target personnel have entered or remained within a safe area, thereby avoiding the uncertainty that may arise from judging solely based on the direction of movement. Therefore, this embodiment effectively solves the problem of accurately determining whether personnel are following prevention and control measures in complex or dynamic disaster scenarios, providing a solid foundation for subsequent risk assessment and intervention measures.

[0133] In some embodiments, in step S403, a consistency analysis is performed based on the safety direction and the updated movement direction to obtain the consistency analysis result, which may include, but is not limited to, the following steps:

[0134] Generate a preset safety angle sector based on the safety direction;

[0135] If the updated movement direction is within the preset safe angle sector, the consistency analysis result is determined to be consistent; otherwise, the consistency analysis result is determined to be inconsistent.

[0136] In some embodiments, simply comparing the safe direction and the direction of movement may not be accurate enough, as the direction of movement may fluctuate slightly around the safe direction. Judging a movement as "not following control measures" based solely on minor deviations could be overly sensitive, leading to frequent alarms and unnecessary interventions. Therefore, it is necessary to more accurately determine whether the movement direction substantially conforms to the safe direction to avoid misjudgments. A preset safe angle sector can be generated based on the safe direction. This sector defines an acceptable range of movement with a certain angular width centered on the safe direction. If the updated direction of movement falls within the preset safe angle sector, it indicates that the target person's movement direction is substantially consistent with the safe direction, and the consistency analysis result is determined to be consistent, thus determining the control control compliance status as "following control measures." Otherwise, it indicates that the target person's movement direction has deviated from the safe guidelines, and the consistency analysis result is determined to be inconsistent, thus determining the control control compliance status as "not following control measures." This embodiment effectively avoids misjudgments caused by minor fluctuations in the direction of movement, improving the accuracy and robustness of the control control compliance determination.

[0137] Understandably, a preset safety angle sector refers to an area formed by extending a certain angle range to both sides with the safety direction as the central axis. Specifically, it can be a sector-shaped area defined by a preset angle value on the central direction and the left and right sides. Its purpose is to provide an acceptable tolerance range for the direction of personnel movement in order to accommodate natural deviations in actual movement.

[0138] Through the above technical solution, this embodiment can more accurately determine whether the movement direction of a target person substantially conforms to the guidance of a safe direction. By introducing a preset safety angle sector, the system can tolerate reasonable directional deviations during movement, avoiding misjudgments caused by minor fluctuations. This effectively solves the problem that simple directional comparisons may be overly sensitive, reduces unnecessary alarms and interventions, thereby improving the accuracy and reliability of the control and prevention status determination, enabling the system to more effectively guide personnel to move in a safe direction and ensure personnel safety.

[0139] In some embodiments, in step S502, a positional relationship analysis is performed based on the updated current location and the avoidance area to obtain the positional relationship analysis result, which may include, but is not limited to, the following steps:

[0140] Obtain disaster source information, including type, intensity, diffusion characteristics, and environmental topography information;

[0141] Calculate the potential impact range of the disaster source based on the disaster source information;

[0142] Update the boundaries of the evacuation zone based on the dynamic evolution of the disaster source and its potential impact range;

[0143] The updated safe zone is divided into a core safe zone and a peripheral transition zone.

[0144] If the updated current location is within the core security area, then the location relationship analysis result is determined to be location safe;

[0145] If the updated current location is not within the core security area, then determine whether the updated current location is within the edge transition area;

[0146] If the updated current location is in the edge transition zone and the movement trend tends towards the core safe zone, then the location relationship analysis result is determined to be safe; otherwise, the location relationship analysis result is determined to be unsafe.

[0147] In some embodiments, considering only location relationships is insufficient because evacuation zones dynamically evolve with changes in the disaster source, and simple location assessments cannot distinguish whether personnel are truly safe or merely in transitional areas. Therefore, a more accurate assessment of location relationships and a more precise determination of the compliance status of prevention and control measures are needed. This can be achieved by first acquiring disaster source information, including its type, intensity, diffusion characteristics, and environmental topography. Different types of disasters have different impact patterns, and environmental topography significantly influences their diffusion paths. Then, based on the disaster source information, the potential impact range of the disaster source is calculated, and the boundaries of the evacuation zone are updated according to the dynamic evolution of the disaster source and its potential impact range. This ensures the effectiveness of the evacuation zone, keeping it synchronized with the actual disaster threat and avoiding misjudgments due to outdated evacuation zones. The updated evacuation zone is then further divided into core safe zones and peripheral transitional zones. The core safe zone represents an area that is absolutely safe under the current disaster conditions, while the peripheral transitional zone indicates areas that, although not in direct danger, still require caution or may require further guidance. This refined zoning makes the safety assessment of personnel locations more accurate and detailed.

[0148] If the updated current location is within the core safety zone, it indicates that the individual is in a fully safe state, and the location relationship analysis result is determined to be safe. If the updated current location is not within the core safety zone, it is determined whether the updated current location is in the edge transition zone. For personnel in the edge transition zone, simply determining their location is insufficient, as they may only be temporarily passing through or have not yet fully entered the safe zone. If the updated current location is in the edge transition zone and the movement trend is towards the core safety zone, it indicates that the personnel are actively moving in a safer direction, and the location relationship analysis result is determined to be safe; otherwise, the location relationship analysis result is determined to be unsafe. This judgment logic, which comprehensively considers both location and movement trend, avoids misjudging personnel who are merely temporarily located in the transition zone but have not taken appropriate safety measures as safe, thereby improving the accuracy and effectiveness of prevention and control.

[0149] Understandably, the potential impact range refers to the maximum geographical area that a disaster might affect, predicted through physical models or simulations based on the characteristics of the disaster source. Specifically, this could involve simulating smoke diffusion or chemical spills using fluid dynamics models, or calculating the area of ​​tremor sensation using seismic wave propagation models. The purpose is to provide a scientific basis for delineating evacuation zones. Dynamic evolution refers to the changes in key parameters of the disaster source, such as its nature, intensity, and diffusion path, over time. Specifically, this could include the expansion or contraction of a fire, changes in wind direction and speed leading to alterations in smoke diffusion direction, and increases or decreases in the rate of vaporization of leaked materials. The purpose is to ensure that the boundaries of evacuation zones can respond in real-time to the latest developments of the disaster and maintain their effectiveness.

[0150] Through the above technical solutions, this embodiment can achieve a more accurate assessment of the location relationships of target personnel. By acquiring comprehensive disaster source information and calculating its potential impact range, a scientific basis can be provided for the delineation of evacuation zones. By updating the boundaries of evacuation zones according to the dynamic evolution of disaster sources and their potential impact range, it can be ensured that evacuation zones always remain synchronized with actual disaster threats, improving the accuracy of evacuation instructions. By dividing evacuation zones into core safety zones and peripheral transition zones, a more refined safety assessment can be provided. By comprehensively considering the relationship between personnel's current location and the divided evacuation zones, and combining this with their movement trends, the true safety status of personnel can be assessed more accurately, avoiding misjudging personnel who are only temporarily located in transition zones but have not taken correct evacuation measures as safe, thereby improving the effectiveness of prevention and control.

[0151] In some embodiments, after determining the nature of the person's deviation behavior based on the first directional angle and a preset reference angle, the method further includes:

[0152] Get the topological properties of the off-path segments;

[0153] If the topology property is a single restricted channel, then obtain the direction of motion before the deviation occurs;

[0154] Calculate the angle of the second direction based on the direction of movement and the direction of motion before the deviation occurred;

[0155] If the angle in the second direction is greater than the preset reverse angle, the nature of the personnel's deviation behavior is updated to stress-induced risk avoidance; otherwise, the nature of the personnel's deviation behavior is updated to high-risk abnormal behavior.

[0156] In some embodiments, because movement choices are limited in paths with specific topological properties such as single-channel confinement, the angle of the first direction alone may not accurately determine the nature of the behavior. For example, in a narrow passage, even if a person's deviation direction is close to the direction of the disaster source, it may be because the physical limitations of the passage prevent them from choosing a better escape route. In this case, simply classifying it as stress-induced avoidance may not be accurate. To improve the accuracy of the judgment, the topological properties of the deviation path segment can be obtained first, enabling the identification of whether the environment in which the person is located is a single-channel confinement. If the topological property is a single-channel confinement, the direction of movement before the deviation occurs can be obtained. Then, based on the direction of movement before the deviation and the direction of movement, the angle of the second direction can be calculated, which can more accurately reflect whether the person has made an active, significantly different reverse movement from the original direction. If the angle of the second direction is greater than the preset reverse angle, it indicates that even if the direction of the deviation is not completely consistent with the direction of the disaster source, the person's behavior is an active and significant change in the original direction of travel. This is usually a reaction of the instinct to avoid danger, and the nature of the person's deviation behavior is updated to stress-induced avoidance. Otherwise, it indicates that the person's deviation behavior is not obvious reverse avoidance and may have other non-avoidance purposes, and the nature of the person's deviation behavior is updated to high-risk abnormal behavior.

[0157] Understandably, the topological properties of a deviation path segment refer to the environmental geometry and connectivity characteristics of the actual physical path segment traversed by personnel after deviating from the pre-set evacuation route. This can be obtained by analyzing Geographic Information System (GIS) data, building floor plans, or real-time environmental sensor data. Its purpose is to identify whether the path segment has specific spatial constraints or structural characteristics. A single restricted passage refers to a physical space with significant width limitations, strong directionality, and typically allowing only one-way or limited two-way passage, such as narrow corridors, stairwells, doorways, or narrow gaps that only allow one person to pass at a time. Its purpose is to identify specific environments where personnel's freedom of movement is restricted. A preset reversal angle refers to a pre-defined angle threshold used to determine whether a change in personnel direction constitutes significant reverse movement. This can be empirically set based on the safety requirements of the actual scenario and personnel behavior patterns. Its purpose is to distinguish between normal turning and drastic directional adjustments during emergency avoidance.

[0158] To illustrate this technical solution more clearly, a specific example is used below. After the system initially determines the nature of a person's deviation behavior based on the first directional angle and a preset reference angle—for example, initially identifying it as reactive risk avoidance—the system immediately obtains the topological attributes of the deviation path segment the person is currently on to further verify its accuracy. This can be achieved by querying pre-stored community map data or Building Information Modeling (BIM) data, which includes information such as the width, length, connectivity, and presence of obstacles in the passageway. If the query results show that the topological attribute of the deviation path segment is a single restricted passage, such as a narrow staircase less than 1.5 meters wide, allowing only upward or downward movement, the system further analyzes the person's historical movement trajectory data to obtain their movement direction before the deviation occurred, i.e., before entering the narrow staircase. For example, by analyzing their position point sequence over the past 5 seconds, the system calculates their average movement direction vector. Subsequently, the system calculates the second directional angle between the movement direction before the deviation and the current movement direction. Assuming the person initially moved upwards but encountered a blockage at the top of the stairs and immediately turned downwards, the second directional angle would be close to 180 degrees. If the calculated second directional angle is greater than the preset reverse angle, for example, greater than 135 degrees, this indicates that the person made a significant reverse movement within the narrow stairwell. Even if their current direction may not be directly away from the disaster source, such a drastic change in direction is usually due to an instinctive avoidance mechanism. Therefore, the system will update the person's deviation behavior information as reactive avoidance. Conversely, if the second directional angle is small, for example, less than 45 degrees, it indicates that the person's deviation behavior is not a significant reverse avoidance and may involve other risks. In this case, the system will update the person's deviation behavior information as high-risk abnormal behavior. In this way, even in physically confined spaces, the system can more accurately identify the person's true intentions and avoid misjudgments.

[0159] Through the above technical solution, this embodiment effectively solves the problem that, in paths with specific topological attributes such as a single restricted passage, the nature of personnel deviation behavior cannot be accurately determined solely based on the included angle of the first direction. By acquiring the topological attributes of the deviation path segment, and, for the case of a single restricted passage, further acquiring the direction of movement before the deviation occurred and calculating the included angle of the second direction, the system can more accurately determine whether the personnel's directional adjustment is due to stress-induced risk avoidance, thereby avoiding misjudging suboptimal path selection caused by physical constraints as high-risk abnormal behavior. This makes the judgment of the nature of personnel deviation behavior more refined and accurate, providing a more reliable basis for subsequent differentiated prevention and control, and avoiding unnecessary waste of resources or delays in rescue opportunities.

[0160] The beneficial effects of implementing the embodiments of the present invention include: the embodiments of this application first obtain the current location of the target personnel, the starting location of the deviation from the preset evacuation route, and the location of the disaster source; then, based on the current location and the starting location of the deviation from the preset evacuation route, the movement direction of the target personnel is calculated, and based on the current location and the location of the disaster source, the relative direction of the disaster source is calculated; then, based on the movement direction and the relative direction of the disaster source, the first direction angle is calculated, and based on the first direction angle and the preset reference angle, the nature information of the personnel deviation behavior is determined; finally, based on the nature information of the personnel deviation behavior, prevention and control are carried out, thereby enabling the determination of deviation behavior by combining positional relationship and movement direction to achieve prevention and control, thereby improving the accuracy and safety of identification.

[0161] like Figure 2 As shown, this embodiment of the invention also provides a video portrait capture and control system, including:

[0162] The location acquisition module 601 is used to acquire the current location of the target personnel, the starting location of the deviation from the preset evacuation route, and the location of the disaster source;

[0163] The personnel movement direction determination module 602 is used to calculate the movement direction of the target personnel based on the current position and the starting position deviating from the preset evacuation route;

[0164] The disaster source relative direction determination module 603 is used to calculate the relative direction of the disaster source based on the current location and the location of the disaster source;

[0165] Angle calculation module 604 is used to calculate the angle in the first direction based on the direction of movement and the relative direction of the disaster source;

[0166] The nature determination module 605 is used to determine the nature information of the personnel deviation behavior based on the first direction angle and the preset reference angle.

[0167] Control module 606 is used for prevention and control based on information about the nature of personnel deviation behavior.

[0168] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0169] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

Claims

1. A video portrait capture and control method, characterized in that, Includes the following steps: Obtain the current location of the target personnel, their starting point of deviation from the preset evacuation route, and the location of the disaster source; Calculate the movement direction of the target personnel based on the current location and the starting position of the deviation from the preset evacuation route; Calculate the relative direction of the disaster source based on the current location and the location of the disaster source; Calculate the angle between the first direction and the relative direction of the disaster source based on the direction of movement; Based on the included angle of the first direction and the preset reference angle, determine the nature of the person's deviation behavior; Based on the nature of the personnel's deviant behavior, prevention and control measures will be implemented; The step of determining the nature of the person's deviation behavior based on the first directional angle and the preset reference angle includes: Obtain the movement trajectory information of the target personnel after they deviate from the preset evacuation route; The motion trajectory information is analyzed to obtain the trajectory analysis results; If the trajectory analysis result indicates the existence of a preset movement pattern, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance, and the preset movement pattern includes a sudden drop in speed or a reversal of direction; If the trajectory analysis result indicates that there is no preset motion pattern, then determine whether the included angle of the first direction is less than the preset reference angle; If the included angle of the first direction is less than the preset reference angle, then the nature of the personnel deviation behavior is determined to be a high-risk abnormal behavior; If the angle of the first direction is greater than or equal to the preset reference angle, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance. The process of performing trajectory analysis on the motion trajectory information to obtain trajectory analysis results includes: Obtain a list of location data from the motion trajectory information, the list of location data including multiple consecutive location coordinates of the target person; Based on the location data list, calculate the changes in movement speed and direction of movement within a preset time period during which the deviation occurs; If the change in movement speed satisfies the condition of a sudden drop in speed and the change in movement direction satisfies the condition of a reversal of direction, then the trajectory analysis result is determined to indicate the existence of a preset movement pattern; otherwise, the trajectory analysis result is determined to indicate the absence of a preset movement pattern. The conditions for the sudden drop in speed are that the moving speed decreases from greater than a first speed threshold to less than a second speed threshold and the deceleration rate is greater than a preset deceleration rate threshold, wherein the first speed threshold is greater than the second speed threshold. The conditions for the reversal of direction are that the reversal angle is greater than a preset angle threshold and the angular velocity is greater than a preset angular velocity threshold. The reversal angle is the angle between the direction before the sudden drop in speed and the direction after the sudden drop in speed.

2. The method according to claim 1, characterized in that, After implementing prevention and control measures based on the information regarding the nature of the individual's deviant behavior, the method further includes: Get the updated current position; Based on the updated current location, determine the prevention and control compliance status of the target personnel; Based on the updated current location and the location of the disaster source, determine the movement trend of the target personnel; If the control and prevention status is not followed or the movement trend changes, then the information on the nature of the person's deviation behavior is updated to high-risk abnormal behavior. Based on the updated information regarding the nature of the personnel's deviant behavior, prevention and control measures will be implemented.

3. The method according to claim 2, characterized in that, The step of determining the prevention and control compliance status of the target personnel based on the updated current location includes: Obtain the safe direction indicated by the prevention and control response; Update the movement direction based on the updated current position; Based on the safety direction and the updated movement direction, a consistency analysis is performed to obtain the consistency analysis results; If the consistency analysis result is consistent, then the control and prevention compliance status is determined to be "followed control and prevention"; otherwise, the control and prevention compliance status is determined to be "not followed control and prevention".

4. The method according to claim 2, characterized in that, The step of determining the prevention and control compliance status of the target personnel based on the updated current location includes: Obtain the safe zones indicated by the prevention and control response; Based on the updated current location and the aforementioned safe zone, a location relationship analysis is performed to obtain the location relationship analysis results; If the location relationship analysis result indicates that the location is safe, then the control and prevention compliance status is determined to be "controlled and prevention measures have been followed"; otherwise, the control and prevention compliance status is determined to be "not controlled and prevention measures have been followed".

5. The method according to claim 3, characterized in that, The step of performing a consistency analysis based on the security direction and the updated movement direction to obtain the consistency analysis result includes: Generate a preset safety angle sector based on the stated safety direction; If the updated direction of movement is within the preset safe angle sector, the consistency analysis result is determined to be consistent; otherwise, the consistency analysis result is determined to be inconsistent.

6. The method according to claim 4, characterized in that, The step involves performing a positional relationship analysis based on the updated current location and the safe zone to obtain the positional relationship analysis results, including: Obtain disaster source information, which includes type, intensity, diffusion characteristics, and environmental topography information; Based on the disaster source information, calculate the potential impact range of the disaster source; The boundaries of the evacuation area are updated based on the dynamic evolution of the disaster source and the potential impact range. The updated evacuation zone is divided into a core safety zone and an edge transition zone. If the updated current location is within the core security area, then the location relationship analysis result is determined to be location safe; If the updated current position is not within the core security area, then determine whether the updated current position is within the edge transition area; If the updated current position is in the edge transition region and the movement trend tends towards the core security region, then the position relationship analysis result is determined to be a safe position; otherwise, the position relationship analysis result is determined to be an unsafe position.

7. The method according to claim 1, characterized in that, After determining the nature of the person's deviation behavior based on the first directional angle and the preset reference angle, the method further includes: Get the topological properties of the off-path segments; If the topological attribute is a single restricted channel, then obtain the direction of motion before the deviation occurs; Calculate the included angle of the second direction based on the direction of movement before the deviation occurred and the direction of movement; If the angle between the second direction and the second direction is greater than the preset reverse angle, then the nature information of the personnel deviation behavior is updated to stress-induced risk avoidance; otherwise, the nature information of the personnel deviation behavior is updated to high-risk abnormal behavior.

8. A video facial capture and control system, characterized in that, include: The location acquisition module is used to acquire the current location of the target personnel, the starting location of the deviation from the preset evacuation route, and the location of the disaster source; The personnel movement direction determination module is used to calculate the movement direction of the target personnel based on the current position and the starting position of the deviation from the preset evacuation path; The disaster source relative direction determination module is used to calculate the relative direction of the disaster source based on the current position and the disaster source position; Angle calculation module, used to calculate the angle in the first direction based on the direction of movement and the relative direction of the disaster source; The nature determination module is used to determine the nature information of the person's deviation behavior based on the first directional angle and the preset reference angle. The control module is used to perform prevention and control measures based on the information regarding the nature of the personnel's deviant behavior. The property determination module is also used for: Obtain the movement trajectory information of the target personnel after they deviate from the preset evacuation route; The motion trajectory information is analyzed to obtain the trajectory analysis results; If the trajectory analysis result indicates the existence of a preset movement pattern, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance, and the preset movement pattern includes a sudden drop in speed or a reversal of direction; If the trajectory analysis result indicates that there is no preset motion pattern, then determine whether the included angle of the first direction is less than the preset reference angle; If the included angle of the first direction is less than the preset reference angle, then the nature of the personnel deviation behavior is determined to be a high-risk abnormal behavior; If the angle of the first direction is greater than or equal to the preset reference angle, then the nature of the person's deviation behavior is determined to be stress-induced risk avoidance. The process of performing trajectory analysis on the motion trajectory information to obtain trajectory analysis results includes: Obtain a list of location data from the motion trajectory information, the list of location data including multiple consecutive location coordinates of the target person; Based on the location data list, calculate the changes in movement speed and direction of movement within a preset time period during which the deviation occurs; If the change in movement speed satisfies the condition of a sudden drop in speed and the change in movement direction satisfies the condition of a reversal of direction, then the trajectory analysis result is determined to indicate the existence of a preset movement pattern; otherwise, the trajectory analysis result is determined to indicate the absence of a preset movement pattern. The conditions for the sudden drop in speed are that the moving speed decreases from greater than a first speed threshold to less than a second speed threshold and the deceleration rate is greater than a preset deceleration rate threshold, wherein the first speed threshold is greater than the second speed threshold. The conditions for the reversal of direction are that the reversal angle is greater than a preset angle threshold and the angular velocity is greater than a preset angular velocity threshold. The reversal angle is the angle between the direction before the sudden drop in speed and the direction after the sudden drop in speed.

Images