A monitoring and management method for a gymnasium

Abstract

The present application discloses a monitoring and management method for a stadium. The method includes: performing connected component analysis on the audience seat distribution image to obtain all audience connected components; based on all the audience connected components in the audience seat distribution image, using a preset sub-region dynamic partitioning mechanism to divide the audience seat area into each sub-region; based on each sub-region, comparing the status image of each audience with a preset viewing status image library to determine the concerned audience; every preset time unit, according to the first dynamic influence value and the second static influence value in the sub-region, calculating the order influence degree value of the sub-region, determining the target sub-region, and controlling the broadcast port closest to the concerned audience in the target sub-region to give an order warning. Thereby, it is possible to perform sub-region monitoring according to the differentiated characteristics of the audience seat distribution, improve the viewing experience of the audience, and enhance the order management efficiency of the audience seat.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring and management, and in particular to a monitoring and management method for a gymnasium. Background Art

[0002] Stadiums typically host large crowds of spectators and athletes. Surveillance systems can be used to monitor crowd flow and prevent overcrowding, thereby improving safety. They can also monitor events within the stadium in real time and take timely action to maintain order. Therefore, to ensure both the spectator experience and stadium safety, it's necessary to monitor the audience seats and address any unhealthy behavior.

[0003] The current technology for monitoring the audience seats in sports stadiums mainly relies on security personnel patrolling the audience seat area. Obviously, this monitoring method has at least the following problems: patrolling the audience seat area by security personnel cannot effectively reduce the workload of security personnel, nor can it reduce the number of security personnel, thereby increasing security costs. On the other hand, security personnel cannot achieve all-round real-time monitoring of the audience seat area, nor can it reflect the intelligence of the audience monitoring method, and thus cannot promptly persuade and stop bad viewing behavior, reducing the viewing experience of other spectators, and thus cannot effectively ensure the safety of people in the gymnasium.

[0004] The Chinese invention patent application number 202410239395.4 discloses a monitoring and management system for gymnasiums. The system divides the auditorium into multiple sub-areas. By analyzing the viewing influence index corresponding to the target audience in each sub-area, security deployment for each sub-area is carried out, reducing security costs and effectively ensuring the stability of order in the auditorium.

[0005] However, for the stadium's auditorium, traditional monitoring methods often use pre-set fixed sub-areas for monitoring. This method cannot adapt to the randomness, fluidity and variability of audience seating in the auditorium, and does not take into account the dynamic distribution characteristics of the audience in the entire auditorium. When some audience members engage in bad viewing behavior, the standards for the degree of influence of the audience's dense distribution area and the audience's discrete distribution area on the surrounding audience are not the same. Therefore, the division of sub-areas based on the characteristics of differentiated audience distribution is extremely important for the accuracy and pertinence of the subsequent analysis of the viewing influence index of the target audience in each sub-area. Summary of the Invention

[0006] This application provides a monitoring and management method for a gymnasium, which can differentiate the distribution characteristics of the audience seats to perform sub-area monitoring, improve the audience's viewing experience, and enhance the efficiency of order management in the audience seats.

[0007] The present application provides a monitoring and management method for a stadium, including:

[0008] S101, obtaining a spectator seat distribution image, which is composed of a number of seat grids assigned with status values, performing connected component analysis on the spectator seat distribution image to obtain all spectator connected components;

[0009] S102, based on all the spectator connected components in the spectator seat distribution image, using a preset sub-region dynamic partitioning mechanism to divide the spectator seat area into each sub-region, and obtaining the status images of each spectator in each sub-region in real time;

[0010] S103, based on each sub-region, comparing the status images of each spectator with a preset viewing status image library to obtain the duration for which the status image of each spectator hits, and determining the spectators with a duration greater than a preset duration threshold as the concerned spectators;

[0011] S104, every preset time unit, calculating the order influence degree value of the sub-region according to the first dynamic influence value and the second static influence value in the sub-region;

[0012] S105, according to the order influence degree values of each sub-region, determining the target sub-region, and controlling the broadcast port closest to the concerned spectator in the target sub-region to give an order warning.

[0013] Preferably, the seat grid is assigned a unique coordinate (row, column), and the status value of the seat grid is set to 0 or 1. When the seat status is idle, the status value is assigned 0, and when the seat status is occupied, the status value is assigned 1; performing connected component analysis on the seat grids with a status value of 1, and all the connected seat grids form a connected component.

[0014] Preferably, the preset sub-region dynamic partitioning mechanism specifically includes:

[0015] A1. Based on the spectator seat distribution image, traversing the connected components in sequence, forming a divergence region by diverging the contour boundary of the currently traversed connected component outward by a preset distance, determining whether there is an intersection with other connected components in the divergence region, and taking the connected components with intersections as the adjacent connected components of the current connected component;

[0016] A2. Calculating the merging trend factors of the current connected component and the adjacent connected components respectively. If the merging trend factor is greater than the preset trend value, merging the corresponding adjacent connected component into the current connected component, and obtaining the maximum circumscribed circle of the merged region;

[0017] A3. If the area value of the largest circumscribed circle reaches the preset area threshold, stop further merging, and use the largest circumscribed rectangle of all connected regions within the largest circumscribed circle as a sub-region; if the area value of the largest circumscribed circle does not reach the preset area threshold, use the largest circumscribed circle as the current connected region and continue to execute steps A2 to A3;

[0018] A4. Continue to traverse the remaining connected regions, execute steps A1 to A3 until all connected regions are traversed, and count and generate all sub-regions.

[0019] Preferably, the merging trend factor is calculated according to the following formula:

[0020]

[0021] where, is the merging trend factor between connected region A and connected region B, d is the Euclidean distance between the centroids of the two connected regions, M is the number of seat grids on the contour boundary of connected region A, N is the number of seat grids on the contour boundary of connected region B, p is the number of grid pairs (coordinates of the grid in connected region A, coordinates of the grid in connected region B) with the distance between the seat grids on the contour boundary of connected region A and the seat grids on the contour boundary of connected region B less than the preset distance value, is the ratio of the number of seat grids on the contour boundary of connected region A to the total number of seat grids on the contour boundaries of the two connected regions, is the ratio of the number of seat grids on the contour boundary of connected region B to the total number of seat grids on the contour boundaries of the two connected regions, and are respectively the influence weight values of the centroid distance and contour boundary distance of the two connected regions on the merging trend factor.

[0022] Preferably, determining the target sub-region according to the order influence degree value of each sub-region specifically includes:

[0023] When the order influence degree value of the sub-region is greater than the preset degree threshold, determine this sub-region as the target sub-region.

[0024] Preferably, S104 specifically includes:

[0025]

[0026] where, T is the order influence degree value of the corresponding sub-region within the preset time unit, r1 is the first dynamic influence value of the concerned audience within this sub-region, r2 is the second static influence value of the concerned audience within this sub-region, and e1 and e2 are respectively the weight factors of the first dynamic influence value and the second static influence value on the order influence degree value.

[0027] Preferably, the obtaining method of the first dynamic influence value of the sub-region within a preset time unit includes:

[0028] S201, divide the preset time unit into k time nodes, denoted as t_i (i = 1, 2,..., k), and the contour area values of each concerned viewer collected at each time node t_i are denoted as (t_i), forming a data point sequence of time nodes and contour area values: {(t_1, (t_1)), (t_2, (t_2)),..., (t_k, (t_k))};

[0029] S202, according to the data point sequence, fit the change curve graph of the contour area value (x) - time node x, where the X-axis of the curve graph is the time node t_i and the Y-axis is the contour area value (t_i);

[0030] S203, determine the minimum contour area value in the change curve graph, denoted as _min, and calculate the first dynamic influence value of the sub-region according to the following formula:

[0031]

[0032] where r1 is the first dynamic influence value of the corresponding sub-region within the preset time unit, w is the total number of concerned viewers in the sub-region, a and b are the start and end time nodes of the preset time unit respectively, is the function value at x of the change curve corresponding to the jth concerned viewer, is the minimum contour area value of the jth concerned viewer, is the adjustment factor of the jth concerned viewer.

[0033] Preferably, the obtaining method of the second static influence value of the sub-region within a preset time unit includes:

[0034] S301, determine the connected domain where all concerned viewers in the sub-region are located as the target connected domain, denoted as (q = 1, 2,..., Q);

[0035] S302, obtain the average value of the distances between each concerned viewer in each target connected domain, and determine it as the influence propagation value ;

[0036] S303. Based on step A2, calculate the average value of all the merging trend values calculated pairwise for the target connected regions, and determine it as the cross-influence value H between the target connected regions;

[0037] S304. Calculate the second static influence value of the sub-region according to the following formula:

[0038]

[0039] where r2 is the second static influence value of the sub-region within a preset time unit, is the number of concerned audiences in the sub-region, W is the total number of audiences in the sub-region, is the cross-influence value between the target connected regions, is the influence propagation value of the q-th target connected region, Q is the total number of target connected regions in the sub-region, and are respectively the weight factors of the number of concerned audiences and the target connected region on the second static influence value.

[0040] Preferably, after the S203, it further includes:

[0041] S204. Obtain the influenced objects of the j-th concerned audience. The influenced objects are determined as the audiences existing in the adjacent grids of the seat grid where the concerned audience is located. The adjacent grids are determined as the seat grids within a preset spacing in the left, right, and upper directions of the concerned audience. Count the number of the influenced objects of the concerned audience, denoted as G, and mark the influenced objects as g (g = 1, 2,..., G);

[0042] S205. Use the method of steps S201 to S202 to obtain the contour area value (x)-variation curve graph of the time node x;

[0043] S206. Determine the maximum contour area value in the variation curve graph of the influenced object, denoted as _max, and calculate the occlusion index of the j-th concerned audience according to the following formula:

[0044]

[0045] where, is the occlusion index of the j-th concerned audience in the corresponding sub-region, G is the total number of the influenced objects of the j-th concerned audience, a and b are respectively the start and end time nodes of the preset time unit, is the function value at x of the variation curve corresponding to the g-th influenced object of the j-th concerned audience, is the maximum contour area value of the g-th influenced object of the j-th concerned audience, is the adjustment factor for the g-th affected object of the j-th concerned viewer;

[0046] S207. Take the average of the occlusion indices of all concerned viewers in the sub-region to obtain the dynamic occlusion index of the sub-region. Replace the original first dynamic influence value with the result of multiplying the first dynamic influence value of the sub-region in step S203 by the dynamic occlusion index, and update the first dynamic influence value of the sub-region within the preset time unit.

[0047] Preferably, step S105 further includes:

[0048] B1. Start continuously monitoring the order influence degree value of the target sub-region within the preset time unit, and generate an order influence degree value sequence { , ,..., };

[0049] B2. When it is monitored that the fluctuation amplitude of the order influence degree value of the target sub-region in the z-th time unit is greater than the first preset amplitude value, deploy the security guard closest to the target sub-region to go to the scene to maintain order;

[0050] Among them, the fluctuation amplitude of the order influence degree value of the target sub-region in the z-th time unit is calculated according to the following formula:

[0051]

[0052] Among them, is the fluctuation amplitude of the order influence degree value of the target sub-region in the z-th time unit, is the order influence degree value of the target sub-region in the -th time unit, is the order influence degree value of the target sub-region in the -1-th time unit;

[0053] The method further includes:

[0054] Every preset period, repeat steps S101 to S102 to periodically update the sub-regions of the auditorium area division.

[0055] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0056] Through connected component analysis and a preset dynamic sub-region division mechanism, dynamic sub-region division according to the actual distribution of the audience is realized, which can more accurately reflect the dynamic distribution characteristics of the audience in the auditorium and provide a basis for subsequent analysis and management; by calculating the order influence degree value of each sub-region and determining the target sub-region for broadcast warning, the order problems in the auditorium can be quickly responded to and processed, improving the order maintenance efficiency; the dynamic sub-region division mechanism can make the monitoring more focused on the areas with dense audiences, reduce ineffective monitoring, and improve the monitoring efficiency.

[0057] By fitting the curve of the contour area value of the concerned audience changing with time and calculating the dynamic influence value based on the minimum contour area value and the adjustment factor, the calculation is made more accurate and practically meaningful, which helps to more accurately reflect the degree of influence of the audience's behavior on the order; by considering factors such as the average distance between the concerned audiences within the target connected component, the cross-influence value between the connected components, the number of concerned audiences, and the number of target connected components, the static influence value is calculated, which can more comprehensively reflect the influence of the audience distribution and behavior on the order; the order influence degree of the sub-region can be more accurately evaluated, so as to take measures for persuasion in a timely manner, which helps to improve the efficiency and accuracy of the monitoring and management of the stadium and reduce the situations of misjudgment and missed judgment.

[0058] By calculating the occlusion index of the influenced objects of each concerned audience in the sub-region and introducing the change situation of the contour area value of the influenced object side of the concerned audience, and judging the occlusion situation based on the maximum contour area value of the influenced object, the occlusion situation of the concerned audience to the surrounding audiences is considered, making the calculation of the dynamic influence value more comprehensive and accurate.

[0059] After the target sub-region is detected, continuously monitor the order influence degree value of the target sub-region. When it is detected that the fluctuation range of the order influence degree value exceeds the preset threshold, it indicates that there are relatively serious order problems in the target sub-region and the broadcast effect is not reflected. Immediately deploy security guards to the scene to maintain order; this immediate response mechanism helps to intervene when the order problem first appears and prevent the problem from deteriorating, improving the safety management level of the stadium.

[0060] By repeating the execution of the dynamic sub-region division mechanism at preset intervals to periodically update the sub-regions of the auditorium area division, this mechanism takes into account the mobility of the people in the stadium auditorium changing with time, ensures that the sub-region division is always in line with the actual distribution of the audience, and provides accurate basic data for subsequent analysis and management. Brief Description of the Drawings

[0061] Figure 1 It is a schematic flowchart of the monitoring and management method for the stadium in the embodiment of the present invention. Detailed Embodiment

[0062] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings. The preferred embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0063] It should be noted that the terms "vertical", "horizontal", "upper", "lower", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only embodiments.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0065] Embodiment 1: Figure 1 It is a schematic flow chart of the monitoring and management method for a stadium according to an embodiment of the present invention.

[0066] As Figure 1 shown, a monitoring and management method for a stadium includes the following steps:

[0067] S101. Obtain the audience seat distribution image. The audience seat distribution image is composed of a number of seat grids each assigned a status value, and each seat grid corresponds to a unique coordinate. Perform connected component analysis on the audience seat distribution image to obtain all the connected components of the audience.

[0068] Specifically, each seat in the audience seat distribution image is regarded as a grid, and the seat of each seat grid is defined as (row, column); its status value is set to 0 or 1. When the seat status is idle, the status value is assigned 0, and when the seat status is occupied, the status value is assigned 1. Perform connected component analysis on the seat grids with a status value of 1, and all the connected seat grids form a connected component.

[0069] S102. Based on all the connected components of the audience in the audience seat distribution image, use the preset sub-region dynamic partitioning mechanism to divide the audience seat area into sub-regions, and obtain the status images of each audience in each sub-region in real time.

[0070] In some embodiments, the preset sub-region dynamic partitioning mechanism is specifically:

[0071] A1. Based on the image of the auditorium distribution, traverse the connected components in sequence. Expand the contour boundary (the set of seat grids on the outer edge of the connected component) of the currently traversed connected component outward by a preset distance to form a divergence area. Determine whether there is an intersection with other connected components in the divergence area, and regard the connected components with intersections as the adjacent connected components of the current connected component.

[0072] A2. Calculate the merging trend factors of the current connected component and the adjacent connected components respectively. If the merging trend factor is greater than the preset trend value, merge the corresponding adjacent connected component into the current connected component, and obtain the maximum circumscribed circle of the merged area.

[0073] Specifically, the merging trend factor is calculated according to the following formula:

[0074]

[0075] Among them, is the merging trend factor between connected component A and connected component B, d is the Euclidean distance between the centroids of the two connected components, M is the number of seat grids on the contour boundary of connected component A, N is the number of seat grids on the contour boundary of connected component B, p is the number of grid pairs (the grid coordinates of connected component A, the grid coordinates of connected component B) where the distance between the seat grids on the contour boundary of connected component A and the seat grids on the contour boundary of connected component B is less than the preset distance value. It should be noted that the same seat grid coordinates are not repeated among different grid pairs; is the ratio of the number of seat grids on the contour boundary of connected component A to the total number of seat grids on the contour boundaries of the two connected components, is the ratio of the number of seat grids on the contour boundary of connected component B to the total number of seat grids on the contour boundaries of the two connected components, and are the influence weight values of the centroid distance and contour boundary distance of the two connected components on the merging trend factor respectively, which are set according to the actual situation. That is, it is determined according to the actual application scenario and professional experience. By adjusting the magnitudes of the two influence weight values, the decision-making importance of the two factors for merging can be dynamically adjusted. In the context of stadium monitoring and management, it can be set through methods such as historical data analysis or evaluation by professional researchers. For example, if historical data or professional evaluation shows that the behavior of the audience is more affected by the centroid distance of the connected components and less sensitive to the proximity of the boundaries, then the two corresponding influence weight values can be reasonably set so that and have a greater and smaller impact on the merging trend factor respectively.

[0076] A3. If the area value of the largest circumscribed circle reaches the preset area threshold, stop further merging, and use the largest circumscribed rectangle of all connected regions within the largest circumscribed circle as a sub-region; if the area value of the largest circumscribed circle does not reach the preset area threshold, use the largest circumscribed circle as the current connected region and continue to execute steps A2 to A3.

[0077] A4. Continue to traverse the remaining connected regions, execute steps A1 to A3 until all connected regions are traversed, and count and generate all sub-regions.

[0078] S103. Based on each sub-region, compare the status images of each audience with the preset viewing status image library to obtain the duration for which the status images of each audience are hit, and determine the audiences with a duration greater than the preset duration threshold as the concerned audiences.

[0079] Specifically, the preset viewing status image library stores status images corresponding to the bad behaviors (disruptive behaviors) of different audiences, which are used to represent the behavioral characteristics of the audiences during viewing, and different status images correspond to different behavioral characteristics.

[0080] S104. Every preset time unit (which can be set to 2 minutes), calculate the order influence degree value of the sub-region according to the first dynamic influence value and the second static influence value within the sub-region.

[0081] S105. According to the order influence degree values of each sub-region, determine the target sub-region, and control the broadcast port closest to the concerned audience in the target sub-region to give an order warning.

[0082] Specifically, when the order influence degree value of the sub-region is greater than the preset degree threshold, determine the sub-region as the target sub-region. Among them, the preset degree threshold is set according to the actual situation and expert experience.

[0083] Among them, in the auditorium area, a number of broadcast ports are evenly distributed, which are used to broadcast preset voice content to prompt the audience with relevant information.

[0084] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0085] Through connected component analysis and the preset dynamic sub-region division mechanism, the dynamic sub-region division according to the actual distribution of the audience is realized, which can more accurately reflect the dynamic distribution characteristics of the audience in the auditorium and provide a basis for subsequent analysis and management; by calculating the order influence degree values of each sub-region and determining the target sub-region for broadcast warning, the order problems in the auditorium can be quickly responded to and processed, improving the order maintenance efficiency.

[0086] The dynamic sub - region division mechanism enables monitoring to focus more on areas with a dense audience, reducing ineffective monitoring and improving monitoring efficiency.

[0087] Example Two:

[0088] Example Two further defines the order influence degree value of the sub - regions in Example One to more accurately comprehensively evaluate the order influence degree value according to the dynamic and static influences within the sub - regions.

[0089] In some embodiments, step S104 specifically includes:

[0090] Calculate the order influence degree value of the sub - region within a preset time unit according to the following formula:

[0091]

[0092] Where T is the order influence degree value of the corresponding sub - region within a preset time unit, r1 is the first dynamic influence value of the audience of concern within the sub - region, r2 is the second static influence value of the audience of concern within the sub - region, and e1 and e2 are the weight factors of the first dynamic influence value and the second static influence value on the order influence degree value, which are set according to the actual situation.

[0093] In some embodiments, the method for obtaining the first dynamic influence value of the sub - region within a preset time unit includes:

[0094] S201, divide the preset time unit into k time nodes, denoted as t_i (i = 1, 2,..., k), and denote the contour area value of each audience of concern collected at each time node t_i as (t_i), forming a data point sequence of time nodes and contour area values: {(t_1, (t_1)), (t_2, (t_2)),..., (t_k, (t_k))}.

[0095] Where (t_i) represents the contour area value of the j - th audience of concern collected at the time node t_i, and denote the total number of audiences of concern in the sub - region as w.

[0096] S202, according to the data point sequence, fit the change curve graph of the contour area value (x) - time node x, where the X - axis of the curve graph is the time node t_i and the Y - axis is the contour area value (t_i).

[0097] S203, determine the minimum contour area value in the change curve graph, denoted as _min, the first dynamic influence value of the sub-region is calculated according to the following formula:

[0098]

[0099] where r1 is the first dynamic influence value of the corresponding sub-region within the preset time unit, w is the total number of concerned audiences within the sub-region, a and b are the start and end time nodes of the preset time unit respectively, is the function value at x of the change curve corresponding to the j-th concerned audience, is the minimum contour area value of the j-th concerned audience, is the adjustment factor of the j-th concerned audience.

[0100] Among them, the adjustment factor of the j-th concerned audience is specifically calculated according to the following formula:

[0101]

[0102] where k is the number of time nodes within the preset time unit, (t_i) represents the contour area value collected by the j-th concerned audience at the time node t_i, is the average value of the contour area values of the j-th concerned audience within the preset time unit.

[0103] In some embodiments, the method for obtaining the second static influence value of the sub-region within the preset time unit includes:

[0104] S301, determining the connected domain where all the concerned audiences within the sub-region are located as the target connected domain, marked as (q = 1, 2,..., Q);

[0105] S302, obtaining the average value of the distances between the concerned audiences in each target connected domain, and determining it as the influence propagation value of this target connected domain ;

[0106] S303, based on step A2, calculating the average value of all the combined trend values calculated pairwise between the target connected domains, and determining it as the cross-influence value H between the target connected domains;

[0107] S304, calculating the second static influence value of the sub-region according to the following formula:

[0108]

[0109] where r2 is the second static influence value of the sub-region within the preset time unit, is the number of concerned audiences within the sub-region, and W is the total number of audiences within the sub-region, is the cross - influence value between target connected regions, is the influence propagation value of the q - th target connected region, and Q is the total number of target connected regions in the sub - area, and are the number of concerned audiences and the weight factor of the target connected region on the second static influence value respectively, which are set according to the actual situation. That is, it is determined according to the actual application scenario and professional experience. By adjusting the magnitudes of the two weight values, the importance degrees of the two factors for the second static influence value of the sub - area are dynamically adjusted. The number of concerned audiences and the number of target connected regions are both important factors affecting order, but their relative importance for the second static influence value may vary due to factors such as venue scale, audience characteristics, and activity type. By setting specific weight factors, the balance of different influencing factors can be achieved, making the calculation of the second static influence value more in line with the actual situation. Suppose in a large stadium, the auditorium is divided into multiple sub - areas. Through observation, it is found that in some cases, the influence of the number of concerned audiences on order is greater than that of the number of target connected regions. Based on this observation, the weight factor of the number of concerned audiences can be set to 0.7, and the weight factor of the number of target connected regions can be set to 0.3, as long as the sum of and is 1.

[0110] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0111] By fitting the change curve of the contour area value of the concerned audiences over time and calculating the dynamic influence value based on the minimum contour area value and the adjustment factor, the calculation is more accurate and meaningful, which helps to more accurately reflect the degree of influence of the audience behavior on order; by considering factors such as the average distance between the concerned audiences within the target connected region, the cross - influence value between the connected regions, the number of concerned audiences, and the number of target connected regions, the static influence value is calculated, which can more comprehensively reflect the influence of the audience distribution and behavior on order; it can more accurately evaluate the degree of order influence of the sub - area, so as to take measures for persuasion in a timely manner, which helps to improve the efficiency and accuracy of the stadium monitoring management and reduce the situations of misjudgment and missed judgment.

[0112] Embodiment Three:

[0113] In Embodiment One and Embodiment Two, although the dynamic sub - area division and the calculation of the order influence degree value have been realized, the occlusion situation of the concerned audiences on the surrounding audiences has not been fully considered. Only the dynamic influence value is calculated unilaterally according to the change situation of the contour area of the concerned audiences themselves, but it does not consider whether the behaviors of the concerned audiences will have sufficient or actual influence on the surrounding. The evaluation object is limited to the concerned audiences, while ignoring the possible changes of the objects that the concerned audiences may affect, which may lead to the inaccuracy or one - sidedness of the evaluation of the first dynamic influence value in the sub - area, thus affecting the accuracy of the monitoring management.

[0114] Therefore, the embodiments of the present application are optimized to a certain extent based on the above embodiments.

[0115] In some embodiments, after the step S203, the method further includes:

[0116] S204, obtaining the influencing objects of the j-th concerned viewer, where the influencing objects are determined as the viewers existing in the adjacent grids of the seat grid where the concerned viewer is located, and the adjacent grids are determined as the seat grids within a preset distance in the left, right, and upper directions of the concerned viewer, counting the number of the influencing objects of the concerned viewer, denoted as G, and marking the influencing objects as g (g = 1, 2,..., G).

[0117] Among them, the preset distance can be set according to the actual situation, which is used to represent the range that the bad behavior of the concerned viewer may affect.

[0118] S205, obtaining the contour area value of each influencing object by using the method of steps S201 to S202 (x) - the change curve graph with the time node x.

[0119] S206, determining the maximum contour area value in the change curve graph of the influencing object, denoted as _max, and calculating the occlusion index of the j-th concerned viewer according to the following formula:

[0120]

[0121] Among them, is the occlusion index of the j-th concerned viewer in the corresponding sub-region, G is the total number of the influencing objects of the j-th concerned viewer, a and b are the start and end time nodes of the preset time unit respectively, is the function value at x of the change curve corresponding to the g-th influencing object of the j-th concerned viewer, is the maximum contour area value of the g-th influencing object of the j-th concerned viewer, is the adjustment factor of the g-th influencing object of the j-th concerned viewer.

[0122] It can be understood that the adjustment factor of the g-th influencing object is specifically calculated according to the following formula:

[0123]

[0124] Among them, k is the number of time nodes within the preset time unit, (t_i) represents the contour area value collected by the g-th influencing object at the time node t_i, is the average value of the contour area values of the g-th influencing object within the preset time unit.

[0125] S207. Take the average of the occlusion indices of all concerned viewers within the sub-region to obtain the dynamic occlusion index of the sub-region. Replace the original first dynamic influence value with the result of multiplying the first dynamic influence value of the sub-region in step S203 by the dynamic occlusion index, and update the first dynamic influence value of the sub-region within the preset time unit.

[0126] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0127] By calculating the occlusion indices of the influencing objects of each concerned viewer in the sub-region, introducing the change situation of the contour area value of the influencing object of the concerned viewer, and judging the occlusion situation based on the maximum contour area value of the influencing object, the occlusion situation of the concerned viewer to the surrounding viewers is considered, making the calculation of the dynamic influence value more comprehensive and accurate.

[0128] Embodiment 4:

[0129] As a large public place, the personnel flow in the auditorium of the stadium is large, and order problems may occur at any time and change rapidly. Therefore, a management method that can continuously monitor the order influence degree value and respond immediately as needed is required. At the same time, since the personnel distribution in the auditorium changes with time, the division of sub-regions also needs to be updated regularly to maintain accuracy. In Embodiments 1 to 3, there is a lack of continuous monitoring of the continuously changing order influence degree value and an immediate response mechanism. At the same time, the dynamic influence of the personnel flow in the auditorium of the stadium on the division of sub-regions is not considered.

[0130] Therefore, the embodiments of the present application are optimized to a certain extent on the basis of the above embodiments.

[0131] In some embodiments, S105 further includes:

[0132] B1. Start continuously monitoring the order influence degree value of the target sub-region within the preset time unit, and generate an order influence degree value sequence { , ,..., }.

[0133] B2. When it is monitored that the fluctuation amplitude of the order influence degree value of the target sub-region in the z-th time unit is greater than the first preset amplitude value, deploy the security guard closest to the target sub-region to go to the scene to maintain order.

[0134] Among them, the first preset amplitude value is set according to the actual situation and expert experience.

[0135] Among them, the fluctuation amplitude of the order influence degree value of the target sub-region in the z-th time unit is calculated according to the following formula:

[0136]

[0137] Among them, is the fluctuation range of the order influence degree value of the target sub-region within the z-th time unit, is the order influence degree value of the target sub-region within the -th time unit, is the order influence degree value of the target sub-region within the -(z - 1)-th time unit.

[0138] In some embodiments, the method further includes:

[0139] Repeating steps S101 to S102 every preset period to periodically update the sub-regions of the auditorium area division. Wherein, the preset period can be set to 10 minutes.

[0140] Thereby, considering the mobility of the people in the stadium auditorium changing with time, the division of the sub-regions is periodically updated, providing a reliable and accurate image of the auditorium distribution for subsequent monitoring of the auditorium.

[0141] The technical solutions in the above embodiments of the present application at least have the following technical effects or advantages:

[0142] After detecting the target sub-region, continuously monitor the order influence degree value of the target sub-region. When it is detected that the fluctuation range of the order influence degree value exceeds the preset threshold, it indicates that there are relatively serious order problems in the target sub-region and the broadcast effect is not reflected, and security guards are quickly dispatched to the scene to maintain order; this immediate response mechanism helps to intervene when the order problem begins to appear, prevent the problem from deteriorating, and improve the safety management level of the stadium.

[0143] In addition, by repeating steps S101 to S102 every preset period to periodically update the sub-regions of the auditorium area division, this mechanism considers the mobility of the people in the stadium auditorium changing with time, ensuring that the division of the sub-regions is always consistent with the actual distribution of the audience, providing accurate basic data for subsequent analysis and management.

[0144] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A monitoring and management method for a gymnasium, characterized in that, Including: S101, obtaining the auditorium distribution image, which consists of seat grids assigned with status values, performing connected component analysis on the auditorium distribution image to obtain all audience connected components; S102, based on all the audience connected components in the auditorium distribution image, using a preset sub-region dynamic partitioning mechanism to divide the auditorium area into sub-regions, and obtaining the status images corresponding to each audience in each sub-region in real time; S103, based on each sub-region, comparing the status images of each audience with a preset viewing status image library to obtain the duration for which the status image of each audience is hit, and determining the audience with a duration greater than the preset duration threshold as the concerned audience; S104, every preset time unit, calculating the order influence degree value of the sub-region according to the first dynamic influence value and the second static influence value in the sub-region; S105. Determine the target sub-region based on the order influence degree values of each sub-region, and control the broadcast port closest to the concerned audience in the target sub-region to give order warnings; start continuously monitoring the order influence degree values of the target sub-region within a preset time unit, and generate an order influence degree value sequence { , ,... }, being the order influence degree value within the z-th time unit; when it is monitored that the fluctuation range of the order influence degree value of the target sub-region within the z-th time unit is greater than the first preset amplitude value, deploy the security guard closest to the target sub-region to go to the scene to maintain order; among them, the fluctuation range of the order influence degree value of the target sub-region within the z-th time unit is calculated according to the following formula: , being the fluctuation range of the order influence degree value of the target sub-region within the z-th time unit, being the order influence degree value of the target sub-region within the -th time unit, being the order influence degree value of the target sub-region within the -1-th time unit; Every preset period, repeat steps S101 to S102 to periodically update the sub-regions of the auditorium area division.

2. The monitoring and management method for a gymnasium according to claim 1, wherein The seat grid is assigned a unique coordinate, and the status value of the seat grid is set to 0 or 1. When the seat status is idle, the status value is assigned 0, and when the seat status is occupied, the status value is assigned 1; perform connected component analysis on the seat grids with a status value of 1, and all connected seat grids form a connected component.

3. The monitoring and management method for a gymnasium according to claim 2, wherein The preset sub-region dynamic partitioning mechanism specifically includes: A1. Based on the auditorium distribution image, sequentially traverse the connected components, expand the contour boundary of the currently traversed connected component outward by a preset distance to form a divergence region, and determine whether there is an intersection with other connected components in the divergence region. The connected components with intersections are used as the adjacent connected components of the current connected component; A2. Calculate the merging trend factor between the current connected component and the adjacent connected component respectively. If the merging trend factor is greater than the preset trend value, merge the corresponding adjacent connected component into the current connected component and obtain the minimum bounding circle of the merged region; A3. If the area value of the minimum bounding circle reaches the preset area threshold, stop further merging, and use the minimum bounding rectangle of all connected components within the minimum bounding circle as a sub-region; if the area value of the minimum bounding circle does not reach the preset area threshold, use the minimum bounding circle as the current connected component and continue to execute steps A2 to A3; A4. Continue to traverse the remaining connected components and execute steps A1 to A3 until all connected components are traversed, and count and generate all sub-regions.

4. The monitoring and management method for a gymnasium according to claim 3, wherein, The merging trend factor is calculated according to the following formula: Among them, is the merging trend factor between connected region A and connected region B, d is the Euclidean distance between the centroids of the two connected regions, M is the number of seat grids on the contour boundary of connected region A, N is the number of seat grids on the contour boundary of connected region B, and p is the number of grid pairs (the grid coordinates of connected region A, the grid coordinates of connected region B) with the distance between the seat grids on the contour boundary of connected region A and the seat grids on the contour boundary of connected region B less than the preset distance value. is the ratio of the number of seat grids on the contour boundary of connected region A to the total number of seat grids on the contour boundaries of the two connected regions. is the ratio of the number of seat grids on the contour boundary of connected region B to the total number of seat grids on the contour boundaries of the two connected regions. and are the influence weight values of the centroid distance and the contour boundary distance of the two connected regions on the merging trend factor respectively.

5. The monitoring and management method for a gymnasium according to claim 1, characterized in that, Determining the target sub-region according to the order influence degree value of each sub-region specifically includes: When the order influence degree value of the sub-region is greater than the preset degree threshold, determine the sub-region as the target sub-region.

6. The monitoring and management method for a gymnasium according to claim 4, wherein, The specific content of S104 includes: Among them, T is the order influence degree value of the corresponding sub-region within a preset time unit, r1 is the first dynamic influence value of the concerned audience in the sub-region, r2 is the second static influence value of the concerned audience in the sub-region, and e1 and e2 are the weight factors of the first dynamic influence value and the second static influence value on the order influence degree value respectively.

7. The monitoring and management method for a gymnasium according to claim 6, characterized in that, The obtaining method of the first dynamic influence value of the sub-region within a preset time unit includes: S201, divide the preset time unit into k time nodes, denoted as t_i (i = 1, 2, ..., k), and the profile area values of each concerned audience collected at each time node t_i are denoted as (t_i), forming a data point sequence of time nodes and profile area values: {(t_1, (t_1)), (t_2, (t_2)), ..., (t_k, (t_k))}; S202. Fit the contour area value of each concerned audience according to the data point sequence. (x) - Variation curve graph of time node x. The X-axis of the curve graph is time node t_i, and the Y-axis is the contour area value. (t_i); S203. Determine the minimum contour area value in the variation curve graph, denoted as _min, and calculate the first dynamic influence value of the sub-region according to the following formula: Among them, r1 is the first dynamic influence value of the corresponding sub-region within a preset time unit, w is the total number of concerned viewers within the sub-region, a and b are the start and end time nodes of the preset time unit, respectively. is the function value at x of the change curve corresponding to the j-th concerned viewer. is the minimum contour area value of the j-th concerned viewer. is the adjustment factor of the j-th concerned viewer.

8. The monitoring and management method for a stadium according to claim 6, wherein, The method for obtaining the second static influence value of the sub-region within a preset time unit includes: S301, determine the connected region where all the concerned viewers in the sub-region are located as the target connected region, and label it as (q = 1, 2,..., Q); S302. Obtain the average value of the distances between the concerned viewers in each target connected domain, and determine it as the influence propagation value of the target connected domain ; S303. Based on step A2, calculate the average value of all the merging trend values calculated pairwise for the target connected regions, and determine it as the cross-influence value H between the target connected regions; S304. Calculate the second static influence value of the sub-region according to the following formula: Among them, $r_2$ is the second static influence value of the sub-region within a preset time unit, $N$ is the number of concerned audiences within the sub-region, $W$ is the total number of audiences within the sub-region, $I$ is the cross-influence value between target connected regions, $P_q$ is the influence propagation value of the $q$-th target connected region, $Q$ is the total number of target connected regions within the sub-region, and $w_1$ and $w_2$ are the weight factors of the number of concerned audiences and the target connected region on the second static influence value respectively.

9. The monitoring and management method for a gymnasium according to claim 7, characterized in that, After the S203, it further includes: S204. Obtain the influencing objects of the j-th concerned viewer. The influencing objects are determined as the viewers existing in the adjacent grids of the seat grid where the concerned viewer is located. The adjacent grids are determined as the seat grids within a preset distance in the left, right, and up directions of the concerned viewer. Count the number of the influencing objects of the concerned viewer, denoted as G, and mark the influencing objects as g (g = 1, 2,..., G); S205, obtain the contour area value of each influencing object by using the method of steps S201 to S202 (x) - Variation curve graph of time node x; S206. Determine the maximum contour area value in the change curve graph of the affected object, denoted as _max, and calculate the occlusion index of the j-th concerned viewer according to the following formula: Among them, is the occlusion index for the j-th concerned viewer in the corresponding sub-region, G is the total number of influence objects of the j-th concerned viewer, a and b are the start and end time nodes of the preset time unit, is the function value at x of the change curve corresponding to the g-th influence object of the j-th concerned viewer, is the maximum contour area value of the g-th influence object of the j-th concerned viewer, is the adjustment factor of the g-th influence object of the j-th concerned viewer; S207. Take the average of the occlusion indices of all the concerned viewers in the sub-region to obtain the dynamic occlusion index of the sub-region, and replace the original first dynamic influence value with the result of multiplying the first dynamic influence value of the sub-region in step S203 by the dynamic occlusion index, and update the first dynamic influence value of the sub-region within a preset time unit.

Images