Aggregation alarm method based on personnel positioning scene

Through data acquisition, time period division and area division modules without the need for specific hardware equipment, real-time monitoring and alerting of the gathering of people is achieved, the problem of insufficient hardware dependence and flexibility in the existing technology is solved, and public safety and monitoring are improved efficiency.

CN119049240AActive Publication Date: 2024-11-29JIANGSU XINTA DIGITAL TECH RES INST CO LTD
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411163729.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-29
Estimated Expiration
2044-08-23

AI Technical Summary

Technical Problem

The existing personnel positioning system relies on specific hardware equipment, which increases costs and limits the flexibility of its application, making it difficult to accurately and timely judge the gathering of people, affecting public safety and preventing potential risks.

Method used

The data acquisition module collects the number of people in the target area in real time. The period division module divides the day into multiple periods and determines the high-frequency or low-frequency periods. The area division module subdivates the target area as a small area. The aggregation behavior alarm module uses the population density and time periods to determine the population. Type judges aggregation behavior and issues alarms to achieve real-time monitoring without the need for specific hardware devices.

Benefits of technology

Accurate prediction and real-time alerts of the gathering of people are realized, monitoring efficiency and response speed are improved, application costs are reduced, and system flexibility and penetration are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049240A_ABST
    Figure CN119049240A_ABST
Patent Text Reader

Abstract

The invention discloses a gathering alarm method based on a personnel positioning scene, relates to the technical field of gathering alarm, and solves the problems that a personnel positioning system usually needs to depend on specific hardware equipment, so that the cost of the system is increased, and the application flexibility and the popularization rate of the system are limited. According to the method, the gathering behavior of each area can be judged, an alarm can be given when the gathering behavior is detected, and related pictures are sent to the display module, the real-time alarm mechanism allows a worker to know the gathering situation at the initial stage and take measures in time to prevent potential safety risks, and in an emergency, the safety of the worker is improved. The rapid and accurate response plays a crucial role in avoiding situation expansion, and meanwhile, the current gathering situation is displayed in a visual mode, so that workers can be helped to understand the field situation more visually, and the decision making efficiency and accuracy are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of gathering alarm, and in particular to a gathering alarm method based on a personnel positioning scenario. Background Art

[0002] In the personnel positioning scenario, existing technologies mainly rely on various sensors and positioning systems, such as GPS, Wi-Fi, Bluetooth, RFID and other wireless communication technologies, to track and determine the location of individuals. These technologies can monitor the location changes of personnel in real time and are widely used in security monitoring, elderly and child care, employee management and other fields. For example, in large public events or construction sites, these positioning technologies can monitor the distribution of personnel in real time, effectively manage the flow of people and prevent potential safety risks.

[0003] The application with publication number CN116704723A discloses a method for detecting crowd gathering, which has the following advantages: by obtaining the number of surrounding personnel and the density of personnel through the spherical space generated by personnel positioning equipment, the degree of crowd gathering in any area can be effectively controlled; the detection radius of each detection point is adjusted according to the degree of danger of the work performed by the personnel, and because of the error of the positioning device, the alarm range is divided into a core alarm area and a non-core alarm area, thereby improving the safety of detection; the alarm situation is graded according to the intersection of different alarm areas caused by the flow of personnel, and corresponding emergency measures are taken according to the alarm level, thereby reducing casualties caused by safety accidents.

[0004] The existing technology still has some shortcomings in practical applications. The personnel positioning system usually needs to rely on specific hardware devices, which not only increases the cost of the system, but also limits its application flexibility and popularity. Therefore, developing a new gathering alarm method that can accurately and timely judge the gathering of people is of great significance for improving public safety and preventing potential risks. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a gathering alarm method based on a personnel positioning scenario, which solves the problems in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a gathering alarm method based on a personnel positioning scenario, comprising: The data collection module is used to collect the number of people in the target area in real time, obtain the specific area of ​​the target area, and send the number of people information to the subsequent time period division module, where the target area is the area that can be captured by multiple cameras; The time period division module is used to receive the number of people information in real time, divide a day into multiple time periods, determine the crowd density of each time period according to the number of people information, then select a time period as the target time period in turn, obtain the crowd density of the target time period n days before the current time, and determine whether the target time period is a high-frequency time period or a low-frequency time period through a preset formula, thereby dividing multiple time periods of a day into high-frequency time periods or low-frequency time periods, where n is a preset value; A region division module divides the target region into multiple small regions through a preset region division model, determines an analysis period, sequentially uses the multiple small regions as analysis regions, and determines whether the analysis region is a key region or a non-key region, wherein the analysis period is determined in a high-frequency period or a low-frequency period; The area determination module is used to divide the target area into four types of areas according to the time period division module and the area division module, namely, the key area of ​​the high-frequency period, the non-key area of ​​the high-frequency period, the key area of ​​the low-frequency period, and the non-key area of ​​the low-frequency period; The gathering behavior alarm module is used to determine whether gathering behavior occurs in each type of area according to the four types of areas. When it is determined that gathering behavior occurs, a gathering alarm is issued and the shooting picture is sent to the subsequent display module; Preferably, the method for determining the high-frequency time period and the low-frequency time period in the time period division module comprises the following steps: S1: Divide a day into y periods, where y is a preset value; S2: Randomly select a time period as the target time period, divide the target time period into x hours on average, and at the end of each hour period, obtain the specific number of people Rk in the target area based on the number of people information, and then obtain the specific area S of the target area. By calculating Gk=Rk / S, the crowd density Gk of the target time period at this moment is obtained, and then the formula is used The result U is used as the crowd density of the target period, where x is a preset value, 1≦k≦x; S3: Then obtain the crowd density Ui of the target period n days before the current time, where n>1, 1≦i≦n, and in this embodiment, n=90; S4: Then calculate the average value Up of the n groups of population density Ui, that is, S5: Then calculate the deviation value H of the n groups of crowd density Ui, that is S6: Determine the size of the deviation value H.

[0007] Preferably, in step S6, it also includes: S61: when H≤D1, the average value Up is used as the crowd density value Usp at the target time, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target time period is determined to be a high-frequency time period, and when Usp≦D2, the target time period is determined to be a low-frequency time period; S62: When H>D1, it is determined that the deviation value H is too large, and Ui is sorted in descending order according to the calculated |Ui-Up|, and the first-ranked Ui is deleted each time and the deviation value H is recalculated until H≦D1; Then count the number D3 of remaining Ui, calculate the ratio of D3 to n, and compare it with the preset ratio D4: When the ratio of D3 to n is greater than the preset ratio D4, the average value of the remaining Ui is calculated as the crowd density value Usp of the target period, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target period is determined to be a high-frequency period, and when Usp≦D2, the target period is determined to be a low-frequency period; When the ratio of D3 to n is less than or equal to the preset ratio D4, the average value of deleting Ui is calculated as the crowd density value Usp at the target time, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target time period is determined to be a high-frequency time period, and when Usp≦D2, the target time period is determined to be a low-frequency time period; Among them, D1 and D2 are preset values.

[0008] Preferably, after step S6, the method further includes: S7: Then repeat steps S2-S6 to divide the y time periods into high-frequency time periods or low-frequency time periods.

[0009] Preferably, in the area division module, the division of key areas and non-key areas includes the following steps: P1: Divide the target area into m small areas on average, each with the same area, denoted as Sj, 1≦j≦m, and P2: Select a high-frequency period or, if there is no high-frequency period, select a period with the largest crowd density value Usp as the analysis period, and divide the analysis period into x hours of time; P3: Select any one of the m small areas as the analysis area. At the end of each hour, obtain the specific number of people Rq in the analysis area, and then calculate the crowd density Gq=Rq / Sj in the analysis area at this moment. Then use the formula The specific crowd density Zj of the analysis area in the analysis period is obtained, wherein 1≦q≦x. Preferably, after step P3, the method further includes: P4: Then repeat P3 to obtain the specific crowd density Zj of all m small areas as the analysis area during the analysis period, and then compare the crowd density Zj with the preset value D5.

[0010] Preferably, in step P4, it further includes: When the small area with Zj>D5 is determined as a key area, the small area with Zj≦D5 is determined as a non-key area, where D5 is a preset value, which is specifically set by the staff; Preferably, in the gathering behavior alarm module, determining whether gathering behavior occurs in each type of area includes the following steps: AS1: In the key areas of high-frequency time periods, the specific number of people is collected in real time every T1 time period. When the number of people collected for the first time exceeds L1, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module; AS2: In non-key areas during high-frequency periods, the specific number of people is collected in real time every T2 time period. When the number of people collected for the first time exceeds L2, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the captured images are sent to the subsequent display module; AS3: In the key area of ​​the low-frequency period, the specific number of people is collected in real time every T3 time period. When the number of people collected for the first time exceeds L3, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module; AS4: In non-key areas during low-frequency periods, the specific number of people is collected in real time every T4 time period. When the number of people collected for the first time exceeds L4, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the captured images are sent to the subsequent display module; Among them, T1, T2, T3 and T4 are preset values, and T4>T3>T2>T1, L1, L2, L3, L4 are preset values, and L1>L2>L3>L4.

[0011] Preferably, it also includes: The special alarm module for gathering behavior is used to determine whether people are gathering near the boundaries of different areas; The display module is used to display the gathering picture of the current gathering alarm.

[0012] Preferably, in the gathering behavior special alarm module, the specific judgment method is: At the boundary of the key area in the high-frequency period, the total number v of the adjacent areas of the boundary is obtained, and the specific type of area of ​​the four types of areas to which the v adjacent areas belong is determined. Then, the specific number of people at the boundary is collected in real time every Tr time period. When the number of people collected for the first time exceeds L1 / v, it is marked as "1". When it is marked as "1" three times in a row, the boundary close to the key area boundary of the high-frequency period in the v adjacent areas is determined, and the number of people collected at the corresponding adjacent area boundary exceeds Lr / v. When the corresponding number of people collected exceeds Lr / v, it is determined to be a gathering behavior, a gathering alarm is issued, and the shooting picture is sent to the subsequent display module. Otherwise, no processing is performed; Among them, r=1, 2, 3, 4, which is selected according to the specific type of area of ​​the four types of areas to which the adjacent area belongs: 1 represents the boundary of the key area during the high-frequency period, 2 represents the boundary of the non-key area during the high-frequency period, 3 represents the boundary of the key area during the low-frequency period, and 4 represents the boundary of the non-key area during the low-frequency period; Similarly, the boundaries of the other three regions are judged to see whether clustering behavior is formed, and all the regional boundaries are judged.

[0013] The present invention provides a gathering alarm method based on personnel positioning scenario. Compared with the prior art, it has the following beneficial effects: (1) This method can collect information on the number of people in the target area in real time and accurately calculate the crowd density. By dividing a day into multiple time periods and further dividing the time period into hourly periods, the method can analyze the changes in the number of people in each hourly period in more detail and determine high-frequency and low-frequency periods based on past data trends. This cumulative analysis of historical data helps predict possible crowd gatherings in the future and provides a scientific basis for event warning and resource allocation.

[0014] (2) This method divides the target area into multiple small areas through a regional division model, and dynamically determines key and non-key areas based on crowd density. This method can make monitoring resources more concentrated in areas with higher population density, thereby improving monitoring efficiency and response speed. In addition, the algorithm can adjust the regional division according to real-time data, which means that the system can adjust the monitoring focus in real time according to actual changes, rather than relying on preset divisions. Such flexibility is extremely beneficial for dealing with unpredictable gathering events.

[0015] (3) This method can not only determine the gathering behavior in each area, but also issue an alarm when the gathering behavior is detected and send the relevant images to the display module. This real-time alarm mechanism allows staff to understand the gathering situation at the initial stage and take timely measures to prevent potential safety risks. In an emergency, a quick and accurate response is crucial to avoiding the escalation of the situation. At the same time, by displaying the current gathering situation in a visual way, it can help staff understand the on-site situation more intuitively and improve the efficiency and accuracy of decision-making. In addition, this method does not need to rely on specific hardware equipment, which greatly reduces the application cost and improves flexibility and popularity. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 It is a framework diagram of a gathering alarm method based on a personnel positioning scenario of the present invention; DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0019] Embodiment 1

[0020] See also Figure 1 ,The present invention provides a gathering alarm method based on personnel positioning scenario, including; The data collection module is used to collect the number of people in the target area in real time, obtain the specific area of ​​the target area, and send the number of people information to the subsequent time period division module, where the target area is the area that can be captured by multiple cameras; The time period division module is used to receive the number of people information in real time, divide a day into multiple time periods, determine the crowd density of each time period according to the number of people information, then select a time period as the target time period in turn, obtain the crowd density of the target time period n days before the current time, and determine whether the target time period is a high-frequency time period or a low-frequency time period through a preset formula, thereby dividing multiple time periods of a day into high-frequency time periods or low-frequency time periods; The method for determining the high-frequency period and the low-frequency period includes the following steps: S1: Divide a day into y time periods, where y is a preset value, which is specifically set by the staff. In this embodiment, y=24; S2: Randomly select a time period as the target time period, divide the target time period into x hours on average, and at the end of each hour period, obtain the specific number of people Rk in the target area based on the number of people information, and then obtain the specific area S of the target area. By calculating Gk=Rk / s, the crowd density Gk of the target time period at this moment is obtained, and then the formula is used to calculate the crowd density Gk of the target time period at this moment. The result U is used as the crowd density of the target period, where x is a preset value, which is set by the staff, 1≦k≦x; S3: Then obtain the crowd density Ui of the target period n days before the current time, where n>1, 1≦i≦n, and in this embodiment, n=90; S4: Then calculate the average value Up of the n groups of population density Ui, that is, S5: Then calculate the deviation value H of the n groups of crowd density Ui, that is S6: Determine the size of the deviation value H: S61: when H≤D1, the average value Up is used as the crowd density value Usp at the target time, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target time period is determined to be a high-frequency time period, and when Usp≦D2, the target time period is determined to be a low-frequency time period; S62: When H>D1, it is determined that the deviation value H is too large, and Ui is sorted in descending order according to the calculated |Ui-Up|, and the first-ranked Ui is deleted each time and the deviation value H is recalculated until H≦D1; Then count the number D3 of remaining Ui, calculate the ratio of D3 to n, and compare it with the preset ratio D4: When the ratio of D3 to n is greater than the preset ratio D4, the average value of the remaining Ui is calculated as the crowd density value Usp of the target period, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target period is determined to be a high-frequency period, and when Usp≦D2, the target period is determined to be a low-frequency period; When the ratio of D3 to n is less than or equal to the preset ratio D4, the average value of deleting Ui is calculated as the crowd density value Usp at the target time, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target time period is determined to be a high-frequency time period, and when Usp≦D2, the target time period is determined to be a low-frequency time period; Wherein, D1 and D2 are preset values, and the specific parameters of D1, D2 and D4 are set by the staff. In this embodiment, D4 is set to 0.55; S7: Then repeat steps S2-S6 to divide the y time periods into high-frequency time periods or low-frequency time periods; It should be noted that a high-frequency period indicates that the number of people coming and going is relatively frequent during this period, while a low-frequency period indicates that the number of people coming and going is relatively sparse during this period.

[0021] A region division module divides the target region into multiple small regions through a preset region division model, determines an analysis period, sequentially uses the multiple small regions as analysis regions, and determines whether the analysis region is a key region or a non-key region, wherein the analysis period is determined in a high-frequency period or a low-frequency period; The division of key areas and non-key areas includes the following steps: P1: Divide the target area into m small areas on average, each with the same area, denoted by and P2: Select a high-frequency period or, if there is no high-frequency period, select a period with the largest crowd density value Usp as the analysis period, and divide the analysis period into x hours of time; P3: Select any one of the m small areas as the analysis area. At the end of each hour, obtain the specific number of people Rq in the analysis area, and then calculate the crowd density Gq=Rq / Sj in the analysis area at this moment. Then use the formula Get the specific crowd density Zj of the analysis area in the analysis period, where 1≦q≦x; P4: Then repeat P3 to obtain the specific crowd density Zj of all m small areas as the analysis area in the analysis period, and then compare the crowd density Zj with the preset value D5. When Zj>D5, the small area is determined as the key area, and when Zj≦D5, the small area is determined as the non-key area, where D5 is a preset value, which is specifically set by the staff; The area determination module is used to divide the target area into four types of areas according to the time period division module and the area division module, namely, the key area of ​​the high-frequency period, the non-key area of ​​the high-frequency period, the key area of ​​the low-frequency period, and the non-key area of ​​the low-frequency period; The gathering behavior alarm module is used to determine whether gathering behavior occurs in each type of area according to the four types of areas. When it is determined that gathering behavior occurs, a gathering alarm is issued and the captured image is sent to the subsequent display module.

[0022] Embodiment 2

[0023] In the specific implementation process, this embodiment is based on the first embodiment and is different from the first embodiment in that this embodiment mainly determines whether agglomeration occurs in each type of area and includes the following steps: AS1: In the key areas of high-frequency time periods, the specific number of people is collected in real time every T1 time period. When the number of people collected for the first time exceeds L1, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module; AS2: In non-key areas during high-frequency periods, the specific number of people is collected in real time every T2 time period. When the number of people collected for the first time exceeds L2, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the captured images are sent to the subsequent display module; AS3: In the key area of ​​the low-frequency period, the specific number of people is collected in real time every T3 time period. When the number of people collected for the first time exceeds L3, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module; AS4: In non-key areas during low-frequency periods, the specific number of people is collected in real time every T4 time period. When the number of people collected for the first time exceeds L4, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the captured images are sent to the subsequent display module; Among them, T1, T2, T3 and T4 are preset values, and T4>T3>T2>T1, and the specific parameters are set by the staff. L1, L2, L3, L4 are preset values, and L1>L2>L3>L4; the specific parameters are set by the staff.

[0024] Embodiment 3

[0025] In the specific implementation process, this embodiment is based on the first and second embodiments, and the difference from the first and second embodiments is that this embodiment further includes a display module.

[0026] Embodiment 4

[0027] In the specific implementation process, this embodiment is based on the first and second embodiments, and is different from the first and second embodiments in that this embodiment further includes a gathering behavior special alarm module and a display module; The gathering behavior special alarm module is used to determine whether a crowd has formed a gathering behavior near the boundaries of different areas. The specific judgment method is: All regional boundaries are judged, taking the key regional boundaries of high-frequency periods as an example: At the boundary of the key area in the high-frequency period, the total number v of the adjacent areas of the boundary is obtained, and the specific type of area of ​​the four types of areas to which the v adjacent areas belong is determined. Then, the specific number of people at the boundary is collected in real time every Tr time period. When the number of people collected for the first time exceeds L1 / v, it is marked as "1". When it is marked as "1" three times in a row, the boundary close to the key area boundary of the high-frequency period in the v adjacent areas is determined, and the number of people collected at the corresponding adjacent area boundary exceeds Lr / v. When the corresponding number of people collected exceeds Lr / v, it is determined to be a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module. Otherwise, no processing is performed. Among them, r=1, 2, 3, 4, which is selected according to the specific type of area of ​​the four types of areas to which the adjacent areas belong: 1 represents the boundary of the key area in the high-frequency period, 2 represents the boundary of the non-key area in the high-frequency period, 3 represents the boundary of the key area in the low-frequency period, and 4 represents the boundary of the non-key area in the low-frequency period; Similarly, the other three regions are examined to determine whether their boundaries form clustering behavior; The display module is used to display the gathering picture of the current gathering alarm, so that the staff can understand the gathering situation on the scene at the first time and judge whether a dangerous situation occurs.

[0028] Embodiment 5

[0029] The specific implementation process of this embodiment includes all the implementation processes of the above three groups of embodiments.

[0030] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0031] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A gathering alarm method based on personnel positioning scenario, characterized in that: include: The data collection module is used to collect the number of people in the target area in real time, obtain the specific area of ​​the target area, and send the number of people information to the subsequent time period division module, where the target area is the area that can be captured by multiple cameras; The time period division module is used to receive the number of people information in real time, divide a day into multiple time periods, determine the crowd density of each time period according to the number of people information, then select a time period as the target time period in turn, obtain the crowd density of the target time period n days before the current time, and determine whether the target time period is a high-frequency time period or a low-frequency time period through a preset formula, thereby dividing multiple time periods of a day into high-frequency time periods or low-frequency time periods, where n is a preset value; A region division module divides the target region into multiple small regions through a preset region division model, determines an analysis period, sequentially uses the multiple small regions as analysis regions, and determines whether the analysis region is a key region or a non-key region, wherein the analysis period is determined in a high-frequency period or a low-frequency period; The area determination module is used to divide the target area into four types of areas according to the time period division module and the area division module, namely, the key area of ​​the high-frequency period, the non-key area of ​​the high-frequency period, the key area of ​​the low-frequency period, and the non-key area of ​​the low-frequency period; The gathering behavior alarm module is used to determine whether gathering behavior occurs in each type of area according to the four types of areas. When it is determined that gathering behavior occurs, a gathering alarm is issued and the captured image is sent to the subsequent display module.

2. The method for gathering alarm based on personnel positioning scenario according to claim 1, characterized in that: The method for determining the high-frequency time period and the low-frequency time period in the time period division module includes the following steps: S1: Divide a day into y periods, where y is a preset value; S2: Randomly select a time period as the target time period, divide the target time period into x hours on average, and at the end of each hour period, obtain the specific number of people Rk in the target area based on the number of people information, and then obtain the specific area S of the target area. By calculating Gk=Rk / S, the crowd density Gk of the target time period at this moment is obtained, and then by the formula, The result U is used as the crowd density of the target period, where x is a preset value, 1≦k≦x; S3: Then obtain the crowd density Ui of the target period n days before the current time, where n>1, 1≦i≦n, and in this embodiment, n=90; S4: Then calculate the average value Up of the n groups of population density Ui, that is, ; S5: Then calculate the deviation value H of the n groups of crowd density Ui, that is ; S6: Determine the size of the deviation value H.

3. The method for gathering alarm based on personnel positioning scenario according to claim 2, characterized in that: In step S6, it also includes: S61: when H≤D1, the average value Up is used as the crowd density value Usp at the target time, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target time period is determined to be a high-frequency time period, and when Usp≦D2, the target time period is determined to be a low-frequency time period; S62: When H>D1, it is determined that the deviation value H is too large, and Ui is sorted in descending order according to the calculated |Ui-Up|, and the first-ranked Ui is deleted each time and the deviation value H is recalculated until H≦D1; Then count the number D3 of remaining Ui, calculate the ratio of D3 to n, and compare it with the preset ratio D4: When the ratio of D3 to n is greater than the preset ratio D4, the average value of the remaining Ui is calculated as the crowd density value Usp of the target period, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target period is determined to be a high-frequency period, and when Usp≦D2, the target period is determined to be a low-frequency period; When the ratio of D3 to n is less than or equal to the preset ratio D4, the average value of deleting Ui is calculated as the crowd density value Usp at the target time, and the crowd density value Usp is compared with the preset value D2. When Usp>D2, the target time period is determined to be a high-frequency time period, and when Usp≦D2, the target time period is determined to be a low-frequency time period; Among them, D1 and D2 are preset values.

4. A gathering alarm method based on personnel positioning scenario according to claim 2 or 3, characterized in that: After step S6, the method further includes: S7: Then repeat steps S2-S6 to divide the y time periods into high-frequency time periods or low-frequency time periods.

5. The method for gathering alarm based on personnel positioning scenario according to claim 1, characterized in that: In the area division module, the division of key areas and non-key areas includes the following steps: P1: Divide the target area into m small areas on average, each with the same area, denoted by ; P2: Select a high-frequency period or, if there is no high-frequency period, select a period with the largest crowd density value Usp as the analysis period, and divide the analysis period into x hours of time; P3: Select any one of the m small areas as the analysis area. At the end of each hour, obtain the specific number of people Rq in the analysis area, and then calculate the crowd density Gq=Rq / Sj in the analysis area at this moment. Then use the formula The specific population density Zj of the analysis area in the analysis period is obtained, where 1≦q≦x.

6. The method for gathering alarm based on personnel positioning scenario according to claim 5, characterized in that: After step P3, the method further includes: P4: Then repeat P3 to obtain the specific crowd density Zj of all m small areas as the analysis area during the analysis period, and then compare the crowd density Zj with the preset value D5.

7. The method for gathering alarm based on personnel positioning scenario according to claim 5, characterized in that: In step P4, it also includes: When the small area Zj>D5 is determined as the key area, and when the small area Zj≦D5 is determined as the non-key area, D5 is a preset value.

8. The method for gathering alarm based on personnel positioning scenario according to claim 5, characterized in that: In the gathering behavior alarm module, judging whether gathering behavior occurs in each type of area includes the following steps: AS1: In the key areas of high-frequency time periods, the specific number of people is collected in real time every T1 time period. When the number of people collected for the first time exceeds L1, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module; AS2: In non-key areas during high-frequency periods, the specific number of people is collected in real time every T2 time period. When the number of people collected for the first time exceeds L2, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the captured images are sent to the subsequent display module; AS3: In the key area of ​​the low-frequency period, the specific number of people is collected in real time every T3 time period. When the number of people collected for the first time exceeds L3, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, and a gathering alarm is issued and the shooting picture is sent to the subsequent display module; AS4: In non-key areas during low-frequency periods, the specific number of people is collected in real time every T4 time period. When the number of people collected for the first time exceeds L4, it is marked as "1". When it is marked as "1" three times in a row, it is determined as a gathering behavior, a gathering alarm is issued, and the captured images are sent to the subsequent display module; Among them, T1, T2, T3 and T4 are preset values, and T4>T3>T2>T1, L1, L2, L3, L4 are preset values, and L1>L2>L3>L4.

9. The method for gathering alarm based on personnel positioning scenario according to claim 1, characterized in that: Also includes: The special alarm module for gathering behavior is used to determine whether people are gathering near the boundaries of different areas; The display module is used to display the gathering picture of the current gathering alarm.

10. The method for gathering alarm based on personnel positioning scenario according to claim 9, characterized in that: In the special alarm module for aggregated behavior, the specific judgment method is as follows: At the boundary of the key area in the high-frequency period, the total number v of the adjacent areas of the boundary is obtained, and the specific type of area of ​​the four types of areas to which the v adjacent areas belong is determined. Then, the specific number of people at the boundary is collected in real time every Tr time period. When the number of people collected for the first time exceeds L1 / v, it is marked as "1". When it is marked as "1" three times in a row, the boundary close to the key area boundary of the high-frequency period in the v adjacent areas is determined, and the number of people collected at the corresponding adjacent area boundary exceeds Lr / v. When the corresponding number of people collected exceeds Lr / v, it is determined to be a gathering behavior, a gathering alarm is issued, and the shooting picture is sent to the subsequent display module. Otherwise, no processing is performed; Among them, r=1, 2, 3, 4, which is selected according to the specific type of area of ​​the four types of areas to which the adjacent area belongs: 1 represents the boundary of the key area during the high-frequency period, 2 represents the boundary of the non-key area during the high-frequency period, 3 represents the boundary of the key area during the low-frequency period, and 4 represents the boundary of the non-key area during the low-frequency period; Similarly, the boundaries of the other three regions are judged to see whether clustering behavior is formed, and all the regional boundaries are judged.

Citation Information

Patent Citations

  • Large scale crowd video analysis system and method thereof

    CN105447458A

  • Congestion degree detection and early-warning system

    CN105791022A

  • GIS-based automatic identification method and device for abnormal gathering of personnel

    CN111400424A

  • Big data-based analysis adaptive dense scene people flow real-time monitoring system

    CN114283386A

  • Early warning system for detecting crowd gathering in real time and application method

    CN115984777A