Tracking and monitoring method for target crowd in limited range

Abstract

The invention discloses a target crowd tracking and monitoring method in a limited range, relates to the technical field, and is used for solving the problems that a traditional target crowd tracking system usually starts all monitoring equipment in a whole area in real time, so that a large amount of unnecessary calculation burden and resource waste are caused, and especially in some time periods, the monitoring time is short. In the prior art, only a small number of targets need to be tracked, monitoring pictures in other areas do not need to be continuously processed, data storage and transmission pressure can be caused, and the efficiency of the whole system is reduced; according to the method, the'important area 'in the area can be accurately recognized by recording the moving paths and the staying time of the vehicles and the persons in unit time, and different from traditional full-time-period or full-area monitoring, the method only starts monitoring in a specific area or time point through intelligent analysis, so that the burden of the system is greatly reduced, and the monitoring efficiency is improved. The operation efficiency and stability of the system are improved, and storage and bandwidth resources are saved.

Description

Technical Field

[0001] The present invention relates to the field of monitoring and tracking technology, and more particularly, to a method for tracking and monitoring a target population within a limited range. Background Art

[0002] Limited-scope target crowd tracking and surveillance technology involves monitoring groups of people or individuals within a designated area, such as a shopping mall, airport, or city block. Tracking systems must be able to identify and track specific targets (such as people or vehicles) and accurately locate them in space and time. This technology integrates advanced technologies such as computer vision, deep learning, and sensor fusion, and is widely used in fields such as public safety, traffic management, and smart retail. With continuous technological advancements, particularly the application of deep learning algorithms, tracking accuracy and application scenarios in this field are expanding.

[0003] The existing technology has the following deficiencies:

[0004] Traditional target crowd tracking systems typically activate all surveillance devices (such as video surveillance) in the entire area in real time, resulting in a significant amount of unnecessary computing overhead and resource waste. This is especially true during certain periods of time, when only a few targets may need to be tracked, while surveillance footage from other areas may not require continuous processing. This not only increases computing resource consumption but also places pressure on data storage and transmission, reducing overall system efficiency.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for tracking and monitoring a target population within a limited range to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for tracking and monitoring a target population within a limited range includes the following steps:

[0009] Step S1: Enable all monitoring of a traffic section, record the movement paths and corresponding stay time of vehicles and people on the traffic section, and record the maximum viewing angle area of each monitoring section;

[0010] Step S2: Map the action path and the corresponding dwell time into a grid coordinate system, quantify them, and perform weighting to obtain important areas. The optimal monitoring angle is determined based on the maximum viewing angle of each monitoring area.

[0011] Step S3: record the important areas after the optimal monitoring angle is implemented and compare them with the important areas when all monitoring is enabled to obtain a matching degree, and obtain historical important times based on the historical CPU usage data;

[0012] Step S4: Determine whether to enable all monitoring for auxiliary monitoring based on the matching degree and historical important time.

[0013] In a preferred embodiment, step S1 includes the following contents:

[0014] Enable full monitoring of traffic sections, transmit each monitoring information to the central processing unit in real time through video stream, identify and track the position of vehicles and people in real time, generate path data for each vehicle and person, and record the entry timestamp when vehicles and people enter an area, and the exit timestamp when vehicles and people leave the area. The difference between the two timestamps is the time the vehicle and person stayed in the area;

[0015] Calculate the maximum coverage area based on the maximum viewing angle of each monitor and the installation height of the monitor. The maximum coverage area formula is: A=πR 2 -B, where A is the maximum coverage area, R is the radius of the maximum coverage area, and B is the area of the obstruction under the maximum area. The formula for calculating the radius of the maximum coverage area is: Where: h is the installation height of the monitoring system, and θ is the viewing angle when the monitoring system has the maximum coverage area.

[0016] In a preferred embodiment, step S2 includes the following:

[0017] The path of each vehicle and person is represented by coordinate points. By recording the position of the vehicle and person in each frame, a complete path trajectory is generated. The recorded path data of the vehicle and person is converted from the monitoring perspective into grid coordinates. The real-time position of the vehicle and person is mapped to the grid coordinate system based on the monitoring perspective parameters. The monitoring area is grid-divided into several small grids. The size of the monitoring area is L×W (length and width), which is divided into m×n small grids, where the size of each small grid is: Each grid corresponds to a two-dimensional coordinate point (x i ,y j ), where x i ∈[1,m],y j ∈[1,n];

[0018] Vehicles and characters in grid (x i ,y j ) is recorded as T ij , the accumulation of residence time is expressed by the following formula: T ij =∑ΔT ij, where: ΔT ij The number of vehicles and people in each frame of the image in the grid (x i ,y j ) in the dwell time increment.

[0019] According to the residence time of each grid and the number of times vehicles and people appear, a heat map is generated. The heat of each grid is expressed as: H ij =α·T ij +β·N ij , where: H ij is the grid (x i ,y j )'s heat, T ij is the total stay time of vehicles and people in the grid, N ij is the number of times vehicles and people pass through the grid, α and β are weight coefficients that adjust the contribution of residence time and number of appearances to heat;

[0020] By normalizing the heat of all grids, the range of heat values is [0,1], and the formula is expressed as: Where: H' ij is the normalized heat value, max(H ij ) and min(H ij ) are the maximum and minimum heat values in the heat map respectively;

[0021] The calculation formula for setting the heat weighted overlap is: overlap = ∑H' ij ×C ij , where: H' ij is the normalized heat value, C ij It is whether the monitoring covers the location (i.e. 1 or 0). The decision is made by comparing the maximum coverage area of each monitoring and the overlap of the high-heat area in the heat map, and the monitoring with the highest overlap is selected as the optimal monitoring perspective.

[0022] In a preferred embodiment, step S3 includes the following contents:

[0023] The monitoring with the optimal monitoring angle is the same as the above steps. The path of each vehicle and person is represented by a coordinate point. By recording the position of the vehicle and person in each frame, a complete path trajectory is generated. The recorded path data of the vehicle and person is converted from the monitoring angle into grid coordinates. The real-time position of the vehicle and person is mapped to the grid coordinate system based on the monitoring angle parameters. The monitoring area is grid-divided into several small grids. The size of the monitoring area is L×W (length and width), which is divided into m×n small grids, where the size of each small grid is: Each grid corresponds to a two-dimensional coordinate point (x i ,y j ), where xi ∈[1,m],y j ∈[1,n];

[0024] Vehicles and characters in grid (x i ,y j ) is recorded as T ij2 , the accumulation of residence time is expressed by the following formula: T ij2 =∑ΔT ij2 , where: ΔT ij2 The number of vehicles and people in each frame of the image in the grid (x i ,y j ) in the dwell time increment;

[0025] According to the residence time of each grid and the number of times vehicles and characters appear, the heat of each grid is expressed as: H ij2 =α·T ij2 +β·N ij2 , where: H ij2 is the grid (x i ,y j )'s heat, T ij2 is the total stay time of vehicles and people in the grid, N ij2 is the number of times vehicles and people pass through the grid, α and β are weight coefficients that adjust the contribution of residence time and number of appearances to heat;

[0026] By normalizing the heat of all grids, the formula is expressed as: , where: H' ij2 is the normalized heat value, max(H ij2 ) and min(H ij2 ) are the maximum and minimum heat values in the heat map respectively;

[0027] Record and store historical data of system CPU usage and corresponding timestamps. Based on the historical data of system CPU usage, identify activities in certain time periods that are of high importance and record the corresponding timestamps, which are historically important times.

[0028] In a preferred embodiment, step S4 includes the following contents:

[0029] All monitoring will be turned on to obtain the normalized heat value H' of each grid in the area ij The normalized new heat value H' of each grid obtained by opening the optimal monitoring angle ij2 The matching degree of the optimal monitoring perspective is obtained by comparison, and the formula is expressed as: Where: H' ij It is to open all monitoring to obtain the normalized heat value of each grid in the area, H' ij2It is the normalized heat value of each grid obtained by opening the optimal monitoring angle;

[0030] After receiving the matching degree and the historical important time, the matching degree and the historical important time are defined as input variables and divided into different fuzzy sets respectively;

[0031] All monitoring start-up conditions are defined as output variables and divided into fuzzy sets;

[0032] Formulate fuzzy rules to describe the impact of matching degree and historically significant time definitions on all monitoring activation situations;

[0033] Perform fuzzy reasoning based on fuzzy rules to determine all monitoring activation options.

[0034] The technical effects and advantages of the method for tracking and monitoring a target population within a limited range of the present invention are as follows:

[0035] By recording the movement paths and dwell times of vehicles and people per unit time, "critical areas" within an area can be accurately identified. This means that monitoring resources can be optimized based on actual needs, rather than indiscriminately monitoring the entire area, thus avoiding unnecessary waste of resources. Unlike traditional full-time or full-area monitoring, this method uses intelligent analysis to enable monitoring only in specific areas or at specific times, greatly reducing the burden on the system, improving its operational efficiency and stability, and saving storage and bandwidth resources. A comprehensive analysis of whether all monitoring needs to be enabled, rather than relying solely on manual judgment or set rules, reduces the possibility of false alarms or missed alarms during the monitoring process, ensuring that key areas receive timely attention, especially when important events occur. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a structural diagram of a method for tracking and monitoring a target population within a limited range according to the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] Example 1

[0039] Figure 1 The present invention provides a method for tracking and monitoring a target population within a limited range, comprising:

[0040] Step S1, start all monitoring of the traffic section and record the movement paths and corresponding stay time of vehicles, vehicles and people in the traffic section, and record the maximum viewing angle area of each monitoring;

[0041] Step S2: Map the action path and the corresponding dwell time into a grid coordinate system, quantify them, and perform weighting to obtain important areas. The optimal monitoring angle is determined based on the maximum viewing angle of each monitoring area.

[0042] Step S3: record the important areas after the optimal monitoring angle is implemented and compare them with the important areas when all monitoring is enabled to obtain a matching degree, and obtain historical important times based on the historical CPU usage data;

[0043] Step S4: Determine whether to enable all monitoring for auxiliary monitoring based on the matching degree and historical important time.

[0044] The specific implementation is as follows:

[0045] In step S1, all monitoring of the traffic section is turned on to record the movement paths and corresponding stay time of vehicles, vehicles and people in the traffic section, and the maximum viewing angle area of each monitoring, including the following details:

[0046] To enable full monitoring of traffic sections, a video surveillance system is used. This system requires a surveillance network consisting of multiple cameras covering the target area. Each camera transmits a real-time video stream to a central processing unit for subsequent image analysis and vehicle and person tracking. Vehicles and people are identified and tracked in real time. Every frame of the surveillance image can be used to identify their location and generate path data for each vehicle or person. When a vehicle or person enters an area, an entry timestamp is recorded. When a vehicle or person leaves the area, an exit timestamp is recorded. The difference between the two timestamps represents the duration of the vehicle or person's stay in the area.

[0047] Based on the maximum viewing angle of each monitor and the installation height of the monitor, the maximum coverage area can be calculated. The maximum coverage area formula is: A=πR 2 -B, where A is the maximum coverage area, R is the radius of the maximum coverage area, and B is the area of the obstruction under the maximum area. The formula for calculating the radius of the maximum coverage area is: Where: h is the installation height of the monitoring system, θ is the viewing angle when the monitoring system has the maximum coverage area;

[0048] For example, if the monitoring obstruction area is 15 square meters and the installation height is 3 meters and the viewing angle is 120 degrees, the coverage area can be obtained through geometric calculation: tan60°=3×1.732=5.196 meters, A=πR 2 -B≈π(5.196) 2-B≈84.7-15=69.7 square meters, so the maximum coverage area of the monitoring is 69.7.

[0049] In step S2, the action path and the corresponding dwell time are mapped to the grid coordinate system, quantified and weighted to obtain the important areas, and the optimal monitoring angle is determined in combination with the maximum viewing angle area of each monitoring area. The specific contents include:

[0050] The path of each vehicle and person can be represented by a series of coordinate points. By recording the position of the vehicle and person in each frame, a complete path trajectory is generated. The recorded path data of the vehicle and person is converted from the monitoring perspective into grid coordinates. The real-time position of the vehicle and person is mapped to the grid coordinate system using the monitoring perspective parameters (such as focal length, angle, etc.). The monitoring area is gridded into several small grids. This process can be accomplished by spatially discretizing the monitoring area. The size of the monitoring area is L×W (length and width), which is divided into m×n small grids, where the size of each small grid is: Each grid corresponds to a two-dimensional coordinate point (x i ,y j ), where x i ∈[1,m],y j ∈[1,n].

[0051] Vehicles and characters in grid (x i ,y j ) is recorded as T ij , then for each frame of the image, record the increment of the time that the vehicle and the person stay in each grid in the frame. The accumulation of the stay time can be expressed by the following formula: T ij =∑ΔT ij , where: ΔT ij The number of vehicles and people in each frame of the image in the grid (x i ,y j ) in the dwell time increment.

[0052] Based on the dwell time of each grid and the number of times vehicles and people appear, a heat map can be generated to represent the heat of each grid in the monitoring area. The calculation formula can be: the number of times vehicles and people appear: represents the frequency of vehicles and people passing through a grid, and the dwell time of vehicles and people: represents the total time vehicles and people stay in a grid. Therefore, the heat of each grid can be expressed as: H ij =α·T ij +β·N ij , where: H ij is the grid (x i ,y j )'s heat, T ij is the total stay time of vehicles and people in the grid, Nij is the number of times vehicles and people pass through the grid, and α and β are the weight coefficients that adjust the contribution of residence time and number of appearances to heat.

[0053] To facilitate analysis, a heat map can be obtained by normalizing the heat of all grids to show the relative importance of each location in the monitoring area, ensuring that the range of heat values is uniform and the range of heat values is [0,1]. The formula can be expressed as: Where: H' ij is the normalized heat value, max(H ij ) and min(H ij ) are the maximum and minimum heat values in the heat map respectively.

[0054] For each monitoring, by comparing its maximum viewing angle area and the overlap of important areas in the heat map, the monitoring angle that best covers the important area is selected. The greater the overlap, the better the monitoring effect on the area. Extract the areas with higher heat from the normalized heat map. You can set a threshold and select the heat areas above the threshold, or directly consider the heat map as a continuous variable and set the heat weighted overlap calculation formula as: Overlap = ∑H' ij ×C ij , where: H' ij is the normalized heat value, C ij Is to monitor whether the position is covered (ie 1 or 0).

[0055] By comparing the maximum coverage area of each monitoring and the overlap of the high-heat areas in the heat map, decisions are made to ensure that the monitoring can cover the most important areas, and the monitoring with the highest overlap is selected as the optimal monitoring perspective.

[0056] In step S3, the important areas after the optimal monitoring angle is implemented are recorded and compared with the important areas when all monitoring is enabled to obtain a matching degree, and the historical important time is obtained based on the historical CPU usage data. The specific content includes:

[0057] The monitoring process for the optimal monitoring perspective is the same as the above steps. The positions of vehicles and people are identified, and the path data for each vehicle and person is generated. When a vehicle or person enters an area, the entry timestamp is recorded. When the vehicle or person leaves the area, the exit timestamp is recorded. The difference between the two timestamps is the time the vehicle or person stayed in the area. The path of each vehicle or person can be represented by a series of coordinate points. By recording the position of the vehicle and person in each frame, a complete path trajectory is generated. The recorded path data of the vehicle and person is converted from the monitoring perspective into grid coordinates. The real-time position of the vehicle and person is mapped to the grid coordinate system using the monitoring perspective parameters (such as focal length, angle, etc.). The monitoring area is gridded into several small grids. This process can be accomplished by spatially discretizing the monitoring area. The size of the monitoring area is L×W (length and width), which is divided into m×n small grids, where the size of each small grid is: Each grid corresponds to a two-dimensional coordinate point (x i ,y j ), where x i ∈[1,m],y j ∈[1,n].

[0058] Vehicles and characters in grid (x i ,y j ) is recorded as T ij2 , then for each frame of the image, record the increment of the time that the vehicle and the person stay in each grid in the frame. The accumulation of the stay time can be expressed by the following formula: T ij2 =∑ΔT ij2 , where: ΔT ij2 The number of vehicles and people in each frame of the image in the grid (x i ,y j ) in the dwell time increment.

[0059] Based on the dwell time of each grid and the number of times vehicles and people appear, a heat map can be generated to represent the heat of each grid in the monitoring area. The calculation formula can be: the number of times vehicles and people appear: represents the frequency of vehicles and people passing through a grid, and the dwell time of vehicles and people: represents the total time vehicles and people stay in a grid. Therefore, the heat of each grid can be expressed as: H ij2 =α·T ij2 +β·N ij2 , where: H ij2 is the grid (x i ,y j )'s heat, T ij2 is the total stay time of vehicles and people in the grid, N ij2is the number of times vehicles and people pass through the grid, and α and β are the weight coefficients that adjust the contribution of residence time and number of appearances to heat.

[0060] For ease of analysis, by normalizing the heat of all grids, the formula can be expressed as: Where: H' ij2 is the normalized heat value, max(H ij2 ) and min(H ij2 ) are the maximum and minimum heat values in the heat map respectively.

[0061] During system operation, a large amount of historical CPU usage data will be accumulated, and a large amount of system CPU usage historical data and corresponding timestamps will be recorded and stored. Based on the system CPU usage historical data, the system can identify that activities in certain time periods are of higher importance. The system will identify important timestamps where anomalies are most likely to occur and record them as historical important times.

[0062] In step S4, a decision is made based on the matching degree and historical important time whether all monitoring functions need to be enabled for auxiliary monitoring. The specific contents include:

[0063] All monitoring will be turned on to obtain the normalized heat value H' of each grid in the area ij The normalized new heat value H' of each grid obtained by opening the optimal monitoring angle ij2 The matching degree of the optimal monitoring perspective is obtained by comparison. The formula can be expressed as: Where: H' ij It is to open all monitoring to obtain the normalized heat value of each grid in the area, H' ij2 It is the normalized heat value of each grid obtained by opening the optimal monitoring perspective.

[0064] After receiving the matching degree and the historical important time, the matching degree and the historical important time are defined as input variables, and are divided into different fuzzy sets respectively.

[0065] For example, "Low", "Medium", and "High" are for matching degrees, and "In" and "Out" are for historically important times.

[0066] All monitoring opening conditions are treated as output variables and divided into fuzzy sets, for example, "Open", "Close" for the growth model.

[0067] Formulate a set of fuzzy rules to describe the impact of different input variables on the output variable. The definition of rules can be based on professional knowledge or obtained through data analysis and experiments. For example:

[0068] Mark the matching degree as P, the important historical time as L, and all monitoring start-up conditions as Monitor.

[0069] Rule 1:IF(P is Low)AND(L is In)THEN(Monitor is Open)

[0070] Rule 2:IF(P is High)AND(L is Out)THEN(Monitor is Close) ...

[0072] Perform fuzzy reasoning based on fuzzy rules to determine the growth model solution.

[0073] It should be noted that the division of fuzzy sets can be adjusted according to actual conditions. For example, although this embodiment takes three fuzzy sets as an example, in fact, the matching degree, historical important time and all monitoring activation conditions can be divided into more than three sets to facilitate better and more precise adjustment.

[0074] Furthermore, for the judgment of matching degree and historical important time, thresholds can be set according to actual conditions. For example, when the matching degree exceeds 80%, it is marked as "High", etc., which will not be elaborated here.

[0075] Based on historical important time, it is not necessary to enable all monitoring conditions: the matching degree reaches the threshold and the historical important time has not arrived. All monitoring conditions need to be enabled: the matching degree does not reach the threshold or the historical important time has arrived. Enable all monitoring when one or both of them are met.

[0076] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0077] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0078] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0079] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0080] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0081] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0082] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for tracking and monitoring a target population within a limited range, characterized in that: Including steps: Step S1: Enable all monitoring of a traffic section, record the movement paths and corresponding stay time of vehicles and people on the traffic section, and record the maximum viewing angle area of each monitoring section; Step S2: Map the action path and the corresponding dwell time into a grid coordinate system, quantify them, and perform weighting to obtain important areas. The optimal monitoring angle is determined based on the maximum viewing angle of each monitoring area. Step S3: record the important areas after the optimal monitoring angle is implemented and compare them with the important areas when all monitoring is enabled to obtain a matching degree, and obtain historical important times based on the historical CPU usage data; Step S4: Determine whether to enable all monitoring for auxiliary monitoring based on the matching degree and historical important time.

2. The method for tracking and monitoring a target population within a limited range according to claim 1, wherein: Enable full monitoring of traffic sections, transmit each monitoring information to the central processing unit in real time through video stream, identify and track the position of vehicles and people in real time, and generate path data for each vehicle and person. When a vehicle or person enters an area, the entry timestamp is recorded, and when the vehicle or person leaves the area, the exit timestamp is recorded. The difference between the two timestamps is the time the vehicle or person stayed in the area. Calculate the maximum coverage area based on the maximum viewing angle of each monitor and the installation height of the monitor. The maximum coverage area formula is: A=πR 2 -B, where A is the maximum coverage area, R is the radius of the maximum coverage area, and B is the area of the obstruction under the maximum area. The formula for calculating the radius of the maximum coverage area is: Where: h is the installation height of the monitoring system, and θ is the viewing angle when the monitoring system has the maximum coverage area.

3. The method for tracking and monitoring a target population within a limited range according to claim 2, wherein: The path of each vehicle and person is represented by coordinate points. By recording the position of the vehicle and person in each frame, a complete path trajectory is generated. The recorded path data of the vehicle and person is converted from the monitoring perspective into grid coordinates. The real-time position of the vehicle and person is mapped to the grid coordinate system based on the monitoring perspective parameters. The monitoring area is grid-divided into several small grids. The size of the monitoring area is L×W (length and width), which is divided into m×n small grids, where the size of each small grid is: Each grid corresponds to a two-dimensional coordinate point (x i ,y j ), where x i ∈[1,m],y j ∈[1,n]; Vehicles and characters in grid (x i ,y j ) is recorded as T ij , the accumulation of residence time is expressed by the following formula: T ij =∑ΔT ij , where: ΔT ij The number of vehicles and people in each frame of the image in the grid (x i ,y j ) in the dwell time increment. According to the residence time of each grid and the number of times vehicles and people appear, a heat map is generated. The heat of each grid is expressed as: H ij =α·T ij +β·N ij , where: H ij is the grid (x i ,y j )'s heat, T ij is the total stay time of vehicles and people in the grid, N ij is the number of times vehicles and people pass through the grid, α and β are weight coefficients that adjust the contribution of residence time and number of appearances to heat; By normalizing the heat of all grids, the range of heat values is [0,1], and the formula is expressed as: Where: H' ij is the normalized heat value, max(H ij ) and min(H ij ) are the maximum and minimum heat values in the heat map respectively; The calculation formula for setting the heat weighted overlap is: overlap = ∑H' ij ×C ij , where: H' ij is the normalized heat value, C ij It is whether the monitoring covers the location (i.e. 1 or 0). The decision is made by comparing the maximum coverage area of each monitoring and the overlap of the high-heat area in the heat map, and the monitoring with the highest overlap is selected as the optimal monitoring perspective.

4. The method for tracking and monitoring a target population within a limited range according to claim 3, wherein: The monitoring with the optimal monitoring angle is the same as the above steps. The path of each vehicle and person is represented by a coordinate point. By recording the position of the vehicle and person in each frame, a complete path trajectory is generated. The recorded path data of the vehicle and person is converted from the monitoring angle into grid coordinates. The real-time position of the vehicle and person is mapped to the grid coordinate system based on the monitoring angle parameters. The monitoring area is grid-divided into several small grids. The size of the monitoring area is L×W (length and width), which is divided into m×n small grids, where the size of each small grid is: Each grid corresponds to a two-dimensional coordinate point (x i ,y j ), where x i ∈[1,m],y j ∈[1,n]; Vehicles and characters in grid (x i ,y j ) is recorded as T ij2 , the accumulation of residence time is expressed by the following formula: T ij2 =∑ΔT ij2 , where: ΔT ij2 The number of vehicles and people in each frame of the image in the grid (x i ,y j ) in the dwell time increment; According to the residence time of each grid and the number of times vehicles and characters appear, the heat of each grid is expressed as: H ij2 =α·T ij2 +β·N ij2 , where: H ij2 is the grid (x i ,y j )'s heat, T ij2 is the total stay time of vehicles and people in the grid, N ij2 is the number of times vehicles and people pass through the grid, α and β are weight coefficients that adjust the contribution of residence time and number of appearances to heat; By normalizing the heat of all grids, the formula is expressed as: , where: H' ij2 is the normalized heat value, max(H ij2 ) and min(H ij2 ) are the maximum and minimum heat values in the heat map respectively; Record and store historical data of system CPU usage and corresponding timestamps. Based on the historical data of system CPU usage, identify activities in certain time periods that are of high importance and record the corresponding timestamps, which are historically important times.

5. The method for tracking and monitoring a target population within a limited range according to claim 4, characterized in that: All monitoring will be turned on to obtain the normalized heat value H' of each grid in the area ij The normalized new heat value H' of each grid obtained by opening the optimal monitoring angle ij2 The matching degree of the optimal monitoring perspective is obtained by comparison, and the formula is expressed as: Where: H' ij It is to open all monitoring to obtain the normalized heat value of each grid in the area, H' ij2 It is the normalized heat value of each grid obtained by opening the optimal monitoring angle; After receiving the matching degree and the historical important time, the matching degree and the historical important time are defined as input variables and divided into different fuzzy sets respectively; All monitoring start-up conditions are defined as output variables and divided into fuzzy sets; Formulate fuzzy rules to describe the impact of matching degree and historically significant time definitions on all monitoring activation situations; Perform fuzzy reasoning based on fuzzy rules to determine all monitoring activation options.

Images