Intelligent dispatching method and system for social governance events

Abstract

The invention discloses an intelligent dispatching method and system for social governance events, and the method comprises the steps: obtaining audio and video data at different monitoring points, and carrying out the analysis to obtain the number of persons, movement data and posture data; when the detected number of persons exceeds a preset value, the person movement data is further analyzed to calculate a person aggregation degree value. If the aggregation degree value is higher than the historical average value of the monitoring points and lasts for a certain time, the system can extract voice text keywords from the audio and video data, and the social governance event type is determined in combination with personnel movement and posture data. Meanwhile, the system can also obtain the position distribution information of the inspection personnel, and formulate an optimal inspection task assignment scheme according to the event type and the position of the inspection personnel, thereby achieving the quick and accurate assignment of tasks, and improving the event response speed and handling efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of social governance events, and in particular to an intelligent distribution method and system for social governance events. Background Art

[0002] In the current urban management and public security maintenance work, it is a common practice to rely on manual inspection tasks for public security incidents. This traditional manual dispatch method usually involves the monitoring center staff to observe the situation of each monitoring point in real time through the video monitoring system, and manually record the location, time and nature of the suspected incident based on the observation results. Once a possible public security incident is discovered, the staff must immediately determine the type of incident and decide how many inspectors to send to handle it based on experience. However, this method has many shortcomings, such as manual judgment is easily affected by subjective factors, resulting in misjudgment of the type of incident; manual recording and decision-making processes are time-consuming, affecting the timeliness of incident handling; in addition, due to the lack of systematic data analysis tools, it is difficult for staff to fully grasp the historical data and current status of each monitoring point, resulting in unreasonable resource allocation and low inspection efficiency. Summary of the invention

[0003] In order to solve at least one of the technical problems mentioned above, the present invention provides an intelligent dispatching method and system for social governance events.

[0004] In a first aspect, the present invention provides an intelligent dispatching method for social governance events, the method comprising:

[0005] Obtain audio and video data from different monitoring points, and analyze the audio and video data to obtain the number of people, personnel movement data, and personnel posture data;

[0006] When the number of people is greater than the preset number of people, the personnel concentration value is obtained based on the personnel movement data analysis; if the personnel concentration value is greater than the historical average concentration value of the current monitoring point and lasts for a preset period of time, the audio and video data is analyzed to obtain voice text keywords, and the type of social governance event at the current monitoring point is determined based on the voice text keywords, personnel movement data and personnel posture data;

[0007] Obtain the inspection location distribution information of inspection personnel, and determine the inspection task assignment plan based on the inspection location distribution information and the type of social governance event.

[0008] Preferably, the inspection task assignment scheme is determined according to the inspection location distribution information and the social governance event type, including:

[0009] Construct the system of equations that minimize the objective function and constraints:

[0010] min∑n∈N D(L(n),m)·X(n);

[0011]

[0012] Where N is the set of all patrol officers, M is the set of monitoring points, E is the set of social governance event types, D(L(n),m) is the time required for patrol officer n to travel from the current location L(n) to monitoring point m, and D(L(n),m) is calculated based on the distance from the current location L(n) to monitoring point m and the transportation mode of patrol officer n. X(n) is a binary decision variable, indicating whether patrol officer n is assigned to monitoring point m, and X(n) = 1 indicates assignment, and X(n) = 0 indicates no assignment. C(e) indicates the minimum number of patrol officers required for monitoring point m to correspond to social governance event type e.

[0013] The integer linear programming algorithm is used to solve the equations and obtain the list of inspectors going to the corresponding monitoring points.

[0014] Preferably, obtaining the personnel concentration value according to personnel movement data analysis includes:

[0015] Based on the personnel movement data, filter out the personnel who have been in the monitoring field of view for a preset time interval and whose travel speed is less than the preset speed value;

[0016] The Euclidean distance between two persons is calculated based on the personnel movement data corresponding to the screened persons. The average distance of all Euclidean distances is calculated and the inverse is taken to obtain the personnel concentration value.

[0017] Preferably, the method of analyzing the audio and video data to obtain voice text keywords, and determining the type of social governance event at the current monitoring point according to the voice text keywords, personnel movement data and personnel posture data, includes:

[0018] Collect historical patrol personnel dispatch data, including the voice text keywords of each patrol personnel dispatch data, personnel movement data, personnel posture data, and the corresponding social governance event types and number of patrol personnel; perform data cleaning, encoding and format conversion on the collected data to obtain a data set;

[0019] The data set is divided into a training set and a test set according to a preset ratio, the original random forest model is trained using the training set, and the recognition accuracy of the random forest model is verified using the test set; when it is determined that the recognition accuracy of the random forest model meets the preset conditions, a social governance event type prediction model is generated;

[0020] The voice text keywords, personnel movement data and personnel posture data of the current monitoring point are input into the social governance event type prediction model, and the social governance event type of the current monitoring point is output.

[0021] In a second aspect, the present invention further provides an intelligent dispatching system for social governance events, the system comprising:

[0022] The data acquisition module is used to acquire audio and video data from different monitoring points, and analyze the audio and video data to obtain the number of personnel, personnel movement data, and personnel posture data;

[0023] The social governance event type prediction module is used to obtain the personnel concentration value based on the personnel movement data analysis when the number of personnel is greater than the preset number of personnel; if the personnel concentration value is greater than the historical average concentration value of the current monitoring point and lasts for a preset period of time, the audio and video data is analyzed to obtain voice text keywords, and the social governance event type of the current monitoring point is determined based on the voice text keywords, personnel movement data and personnel posture data;

[0024] The inspection task dispatch plan generation module is used to obtain the inspection location distribution information of the inspection personnel, and determine the inspection task dispatch plan based on the inspection location distribution information and the type of social governance event.

[0025] Preferably, the inspection task assignment scheme is determined according to the inspection location distribution information and the social governance event type, including:

[0026] Construct the system of equations that minimize the objective function and constraints:

[0027] min∑ n∈N D(L(n),m)·X(n);

[0028]

[0029] Where N is the set of all patrol officers, M is the set of monitoring points, E is the set of social governance event types, D(L(n),m) is the time required for patrol officer n to travel from the current location L(n) to monitoring point m, and D(L(n),m) is calculated based on the distance from the current location L(n) to monitoring point m and the transportation mode of patrol officer n. X(n) is a binary decision variable, indicating whether patrol officer n is assigned to monitoring point m, and X(n) = 1 indicates assignment, and X(n) = 0 indicates no assignment. C(e) indicates the minimum number of patrol officers required for monitoring point m to correspond to social governance event type e.

[0030] The integer linear programming algorithm is used to solve the equations and obtain the list of inspectors going to the corresponding monitoring points.

[0031] Preferably, obtaining the personnel concentration value according to personnel movement data analysis includes:

[0032] Based on the personnel movement data, filter out personnel who have been in the monitoring field of view for a preset time interval and whose travel speed is less than a preset speed value;

[0033] The Euclidean distance between two persons is calculated based on the personnel movement data corresponding to the screened persons. The average distance of all Euclidean distances is calculated and the inverse is taken to obtain the personnel concentration value.

[0034] Preferably, the method of analyzing the audio and video data to obtain voice text keywords, and determining the type of social governance event at the current monitoring point according to the voice text keywords, personnel movement data and personnel posture data, includes:

[0035] Collect historical patrol personnel dispatch data, including the voice text keywords of each patrol personnel dispatch data, personnel movement data, personnel posture data, and the corresponding social governance event types and number of patrol personnel; perform data cleaning, encoding and format conversion on the collected data to obtain a data set;

[0036] The data set is divided into a training set and a test set according to a preset ratio, the original random forest model is trained using the training set, and the recognition accuracy of the random forest model is verified using the test set; when it is determined that the recognition accuracy of the random forest model meets the preset conditions, a social governance event type prediction model is generated;

[0037] The voice text keywords, personnel movement data and personnel posture data of the current monitoring point are input into the social governance event type prediction model, and the social governance event type of the current monitoring point is output.

[0038] In a third aspect, the present invention further provides an electronic device comprising a processor and a memory, wherein the memory is used to store computer program code, and the computer program code comprises computer instructions. When the processor executes the computer instructions, the electronic device executes the method as described in the first aspect above and any possible implementation thereof.

[0039] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program includes program instructions, and when the program instructions are executed by a processor of an electronic device, the processor executes the method as described in the first aspect above and any possible implementation thereof.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1) The present invention monitors the number of people, movement trajectories and posture changes by acquiring and analyzing audio and video data from different monitoring points, and then, after discovering abnormal gatherings, accurately determines the type of event by combining voice text keywords, personnel movement and posture data; subsequently, based on the real-time location distribution of the inspection personnel, the selected means of transportation and the distance from the monitoring point, the navigation data interface is used to calculate the optimal arrival path and time, thereby ensuring that the inspection tasks can be quickly and accurately assigned to the most suitable personnel. This not only effectively solves the problem of inaccurate assignment of existing inspection personnel and reduces resource waste, but also significantly improves the speed and efficiency of responding to social governance incidents, providing strong technical support for maintaining social stability.

[0042] 2) The present invention can accurately calculate the optimal response path and minimum response time of the inspectors according to their actual location, mode of transportation and specific needs of the monitoring points, thereby achieving rapid response and efficient handling of social governance events; by combining the objective function of minimizing the total time cost with the constraints, it ensures that each monitoring point can obtain sufficient inspection force support, avoids redundant configuration of human resources, effectively improves the accuracy and response speed of inspection task assignment, and ultimately achieves the maximization of resource utilization and a significant improvement in social governance efficiency.

[0043] 3) The present invention screens low-speed moving personnel who are continuously within the monitoring field of view within a preset time interval, calculates the average value of the Euclidean distance between these personnel, and then takes the inverse to determine the personnel concentration value, thereby realizing effective monitoring and analysis of the crowd dynamics in a specific area, accurately identifying potential safety hazards, avoiding false alarms for ordinary pedestrian flows, and clarifying the crowd density through specific mathematical calculations, providing a scientific basis for public safety management, thereby significantly enhancing the level of intelligence in urban governance and emergency response capabilities.

[0044] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.

[0046] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0047] Figure 1 A flowchart of an intelligent dispatching method for social governance events provided by an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of the structure of an intelligent dispatching system for social governance events provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0050] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0051] The existing method of manually dispatching patrol tasks for public security incidents has defects such as low efficiency, slow response and unreasonable resource allocation.

[0052] See also Figure 1 , Figure 1 The following is a flow chart of an intelligent dispatching method for social governance events provided by an embodiment of the present invention. Figure 1 As shown, the method includes:

[0053] S100, obtaining audio and video data from different monitoring points, and analyzing the audio and video data to obtain the number of personnel, personnel movement data, and personnel posture data;

[0054] Audio and video data are collected from various monitoring points, and processed through image recognition algorithms to extract information on the number of people. At the same time, the movement trajectory and speed of each person are recorded, and the changes in the size of the detection frame are used to reflect the changes in the individual's posture. It should be noted that "audio and video data" refers to the sound and video signals captured by the camera, and the "number of people" is obtained by the number of faces or human contours detected in the video frame. The "personnel movement data" not only includes personal speed information, but also includes their position changes within a certain period of time. The "personnel posture data" infers the individual's movement status by analyzing the size and shape changes of the human body detection frame, such as whether they are running, squatting or falling. Specifically, for the statistics of the "number of people", a target detection algorithm is usually used, such as the YOLO algorithm, which can mark and count each face or body in the video frame, and estimate the speed and position changes of the individual through the tracking algorithm Kalman filter to obtain "personnel movement data", while the extraction of "personnel posture data" can use the OpenPose framework to identify multiple key points of the human body (such as head, shoulders, elbows, wrists, etc.), so as to analyze the posture of the person according to the changes of these points. In a possible embodiment, the size of the person detection frame can also be used to approximate the person posture data to reduce the amount of data calculation.

[0055] S200: When the number of people is greater than the preset number of people, the concentration value of people is obtained based on the analysis of the personnel movement data; if the concentration value of people is greater than the historical average concentration value of the current monitoring point and lasts for a preset period of time, the audio and video data are analyzed to obtain voice text keywords, and the type of social governance event at the current monitoring point is determined based on the voice text keywords, personnel movement data and personnel posture data;

[0056] When it is detected that the number of people at a monitoring point exceeds the preset threshold, the movement patterns of these people are further analyzed to calculate the "people concentration value". The people concentration value reflects the degree of concentration of the crowd. If the people concentration value is not only higher than the historical average concentration value of the monitoring point but also lasts for a certain period of time, the next step is triggered: further processing of the audio and video data to extract the voice text keywords, which include but are not limited to "help", "beating", "murder", "robbery", "fire" and other words closely related to public safety. Through statistical analysis of the frequency of occurrence of these keywords, the nature of the social governance events that may occur at the monitoring point can be preliminarily determined. At the same time, combined with the aforementioned personnel movement data and posture data, the specific type of social governance event can be more accurately determined, and then the number of patrol personnel to be dispatched can be determined. It should be noted that the "preset time period" refers to the shortest length of time that the system determines that the abnormal situation continues, and the "historical average concentration value" is a benchmark value obtained based on the gathering of people at the location in the same time period in the past.

[0057] S300, obtaining the inspection location distribution information of the inspection personnel, and determining the inspection task assignment plan according to the inspection location distribution information and the type of social governance event.

[0058] Obtain the location distribution information of all available patrol personnel, and formulate a patrol task dispatch plan based on this information and the type of social governance event. The formulation of the dispatch plan can be optimized according to a variety of strategies, such as the principle of the shortest distance or the least time for patrol personnel to reach the scene, and ensure that the number of dispatched personnel meets the standard patrol personnel requirements for handling this type of social governance. "Patrol location distribution information" refers to the real-time location of all currently available patrol personnel on the map, and "Patrol task dispatch plan" refers to the personnel scheduling plan designed by the system to respond to specific types of social governance events based on the location distribution data of all currently available patrol personnel, combined with other The type of transportation they each choose, using the real-time traffic information provided by the navigation data interface, comprehensively evaluates the estimated time required for each inspector to reach the monitoring point from the current location. The "navigation data interface" refers to the data source connected to a third-party or proprietary map service API, which can obtain real-time traffic conditions. The latest road conditions, congestion levels and other information can be obtained through the navigation data interface, and a more accurate estimated arrival time can be calculated. At the same time, the choice of "transportation mode" will also affect the calculation result of the arrival time. For example, riding a bicycle or walking may be affected by traffic lights, while driving a motorcycle or car needs to consider factors such as traffic flow.

[0059] In this embodiment, the number of people, movement trajectories and posture changes are monitored by acquiring and analyzing audio and video data from different monitoring points. After abnormal gatherings are found, the type of event is accurately determined by combining voice text keywords, personnel movement and posture data; subsequently, based on the real-time location distribution of the inspection personnel, the selected means of transportation and the distance from the monitoring point, the navigation data interface is used to calculate the optimal arrival path and time, thereby ensuring that the inspection tasks can be quickly and accurately assigned to the most suitable personnel. This not only effectively solves the problem of inaccurate assignment of existing inspection personnel and reduces resource waste, but also significantly improves the speed of response and handling efficiency of social governance incidents, providing strong technical support for maintaining social stability.

[0060] Preferably, the inspection task assignment scheme is determined according to the inspection location distribution information and the social governance event type, including:

[0061] Construct the system of equations that minimize the objective function and constraints:

[0062] min∑ n∈N D(L(n),m)·X(n);

[0063]

[0064] Where N is the set of all patrol officers, M is the set of monitoring points, E is the set of social governance event types, D(L(n),m) is the time required for patrol officer n to travel from the current location L(n) to monitoring point m, and D(L(n),m) is calculated based on the distance from the current location L(n) to monitoring point m and the transportation mode of patrol officer n. X(n) is a binary decision variable, indicating whether patrol officer n is assigned to monitoring point m, and X(n) = 1 indicates assignment, and X(n) = 0 indicates no assignment. C(e) indicates the minimum number of patrol officers required for monitoring point m to correspond to social governance event type e.

[0065] The integer linear programming algorithm is used to solve the equations and obtain the list of inspectors going to the corresponding monitoring points.

[0066] The above equations are an integer linear programming model used to solve the problem of how to determine the inspection task allocation scheme based on the location distribution of inspection personnel and the type of social governance events. Specifically, min means that the objective function is to minimize the total time cost, ∑ n∈N D(L(n),m)·X(n) is the total time cost. D(L(n),m) is the time required for inspector n to travel from the current location L(n) to the monitoring point m. D(L(n),m) is calculated based on the distance from the current location L(n) to the monitoring point m and the transportation mode of inspector n. The constraints are: It means that the social governance event type e corresponding to each monitoring point must at least meet the required number of patrol personnel C(e), and the constraints are: Ensure that each inspector can only be assigned to one monitoring point, with the following constraints: X(n) is a binary decision variable, indicating whether patrol inspector n is assigned to monitoring point m. X(n)=1 indicates assignment, and X(n)=0 indicates no assignment.

[0067] By solving the aforementioned integer linear programming model, we can get the list of inspectors who go to the corresponding monitoring points, that is, which monitoring point each inspector should go to and which inspectors should go to which monitoring point. The role of the integer linear programming model is to optimize the inspection task dispatching plan to minimize the total time cost while meeting the number of inspectors required for all monitoring points. By using the integer linear programming algorithm to solve, we can get an optimal inspection task dispatching plan so that the inspectors can reach the monitoring points as quickly as possible, improving the response speed and handling efficiency.

[0068] In this embodiment, the optimal response path and minimum response time of the inspectors can be accurately calculated according to the actual location of the inspectors, the mode of transportation and the specific needs of the monitoring points, so as to achieve rapid response and efficient handling of social governance events; by combining the objective function of minimizing the total time cost with the constraint conditions, it not only ensures that each monitoring point can obtain sufficient inspection force support, but also avoids redundant configuration of human resources, effectively improves the accuracy and response speed of inspection task assignment, and ultimately achieves the maximization of resource utilization and a significant improvement in social governance efficiency.

[0069] Preferably, obtaining the personnel concentration value according to personnel movement data analysis includes:

[0070] Based on the personnel movement data, filter out the personnel who have been in the monitoring field of view for a preset time interval and whose travel speed is less than the preset speed value;

[0071] The Euclidean distance between two persons is calculated based on the personnel movement data corresponding to the screened persons. The average distance of all Euclidean distances is calculated and the inverse is taken to obtain the personnel concentration value.

[0072] Based on the personnel movement data, the personnel who have remained in the monitoring field of view within the preset time interval and whose travel speed is lower than a preset speed value are screened out. This screening process is achieved by continuously tracking the human detection frame in the video frame and recording their position changes. Only those individuals who have not left the monitoring area within a given time and whose movement speed does not exceed the preset threshold (such as walking speed) will be included in the analysis scope. Then, all qualified personnel screened out are further processed, and the Euclidean distance between any two individuals is calculated. The average of all calculated Euclidean distances is calculated, and the inverse of the average distance is taken as an indicator to measure the density of personnel, that is, the personnel concentration value.

[0073] "Preset time interval" refers to the time period used to determine whether an individual stays in the same area for a long time. It is usually set according to the specific situation, such as one minute or five minutes; "Preset speed value" is the threshold used to distinguish between static or slowly moving people and quickly passing people, and is generally set to the lower limit of walking speed; "Monitoring angle of view" refers to the visible area covered by the camera. Any human activity occurring in this area is within the monitoring range; "Euclidean distance" is a concept in mathematics used to describe the straight-line distance between two points. It is used to quantify the relative proximity between the center points of two human detection frames.

[0074] In this embodiment, by screening low-speed moving personnel who are continuously within the monitoring field of view within a preset time interval, and calculating the average Euclidean distance between these personnel and then taking the inverse to determine the personnel concentration value, effective monitoring and analysis of crowd dynamics in a specific area is achieved, which can accurately identify potential safety hazards and avoid false alarms for ordinary pedestrian flows. In addition, the crowd density is clarified through specific mathematical calculations, which provides a scientific basis for public safety management, thereby significantly enhancing the level of intelligence in urban governance and emergency response capabilities.

[0075] Preferably, the method of analyzing the audio and video data to obtain voice text keywords, and determining the type of social governance event at the current monitoring point according to the voice text keywords, personnel movement data and personnel posture data, includes:

[0076] Collect historical patrol personnel dispatch data, including the voice text keywords of each patrol personnel dispatch data, personnel movement data, personnel posture data, and the corresponding social governance event types and number of patrol personnel; perform data cleaning, encoding and format conversion on the collected data to obtain a data set;

[0077] The data set is divided into a training set and a test set according to a preset ratio, the original random forest model is trained using the training set, and the recognition accuracy of the random forest model is verified using the test set; when it is determined that the recognition accuracy of the random forest model meets the preset conditions, a social governance event type prediction model is generated;

[0078] The voice text keywords, personnel movement data and personnel posture data of the current monitoring point are input into the social governance event type prediction model, and the social governance event type of the current monitoring point is output.

[0079] The inspection personnel assignment data are collected from the historical database. These data cover the voice text keywords in each event processing, that is, the key phrases or words extracted from the on-site recordings through speech recognition technology, personnel movement data, specifically the personnel trajectory information recorded by the video surveillance system, and personnel posture data, the personnel behavior pattern characteristics derived from the posture recognition algorithm. In addition, it also includes the corresponding social governance event type labels and the number of inspection personnel actually involved in the processing process; then, these raw data are preprocessed, including data cleaning to remove invalid or erroneous information, encoding conversion to facilitate machine learning algorithm processing, and format standardization to make the data set consistent, so as to obtain a data set that can be used for model training. Furthermore, the prepared data set is divided into a training set and a test set according to a preset ratio. The former is used to train the random forest model in the initial state, and the latter is used to verify the accuracy of the trained model in identifying the type of social governance events. The model performance is optimized by adjusting the parameters in the random forest model, such as the number of trees, the maximum depth, etc., and the recognition accuracy of the model is repeatedly verified by using the test set data until the preset conditions are met, that is, the classification accuracy of the model for new data reaches the predetermined threshold. At this time, it is confirmed that the model has good generalization ability and can accurately predict the type of social governance events of unknown data, and save it as a social governance event type prediction model. In practical applications, the voice text keywords, personnel movement data, and personnel posture data collected in real time at the current monitoring point are used as input and passed to the trained social governance event type prediction model. The type of social governance event that may occur at the current monitoring point is output through the operation of the model, thereby assisting decision makers to respond quickly and improve the efficiency and pertinence of event handling.

[0080] In this embodiment, multi-source heterogeneous data is integrated and applied to random forest model training to achieve automatic recognition and prediction of social governance event types. On the one hand, by learning from historical data, the model can automatically extract feature combinations related to different event types, thereby improving the accuracy and reliability of event classification; on the other hand, the prediction of real-time input data enables the system to make accurate judgments at the early stages of an event, thereby providing decision support for timely and effective intervention, enhancing foresight and initiative in urban management, and ultimately achieving the goal of improving the level of public safety management.

[0081] In summary, the method provided in this embodiment can at least achieve the following effects:

[0082] 1) The present invention monitors the number of people, movement trajectories and posture changes by acquiring and analyzing audio and video data from different monitoring points, and then, after discovering abnormal gatherings, accurately determines the type of event by combining voice text keywords, personnel movement and posture data; subsequently, based on the real-time location distribution of the inspection personnel, the selected means of transportation and the distance from the monitoring point, the navigation data interface is used to calculate the optimal arrival path and time, thereby ensuring that the inspection tasks can be quickly and accurately assigned to the most suitable personnel. This not only effectively solves the problem of inaccurate assignment of existing inspection personnel and reduces resource waste, but also significantly improves the speed and efficiency of responding to social governance incidents, providing strong technical support for maintaining social stability.

[0083] 2) The present invention can accurately calculate the optimal response path and minimum response time of the inspectors according to their actual location, mode of transportation and specific needs of the monitoring points, thereby achieving rapid response and efficient handling of social governance events; by combining the objective function of minimizing the total time cost with the constraints, it ensures that each monitoring point can obtain sufficient inspection force support, avoids redundant configuration of human resources, effectively improves the accuracy and response speed of inspection task assignment, and ultimately achieves the maximization of resource utilization and a significant improvement in social governance efficiency.

[0084] 3) The present invention screens low-speed moving personnel who are continuously within the monitoring field of view within a preset time interval, calculates the average value of the Euclidean distance between these personnel, and then takes the inverse to determine the personnel concentration value, thereby realizing effective monitoring and analysis of the crowd dynamics in a specific area, accurately identifying potential safety hazards, avoiding false alarms for ordinary pedestrian flows, and clarifying the crowd density through specific mathematical calculations, providing a scientific basis for public safety management, thereby significantly enhancing the level of intelligence in urban governance and emergency response capabilities.

[0085] See also Figure 2 In one embodiment, a system for intelligently dispatching social governance events is also provided, the system comprising:

[0086] The data acquisition module 100 is used to acquire audio and video data of different monitoring points, and analyze the audio and video data to obtain the number of personnel, personnel movement data and personnel posture data;

[0087] The social governance event type prediction module 200 is used to obtain the personnel concentration value based on the personnel movement data analysis when the number of personnel is greater than the preset number of personnel; if the personnel concentration value is greater than the historical average concentration value of the current monitoring point and lasts for a preset time period, the audio and video data are analyzed to obtain voice text keywords, and the social governance event type of the current monitoring point is determined based on the voice text keywords, personnel movement data and personnel posture data;

[0088] The inspection task assignment plan generation module 300 is used to obtain the inspection location distribution information of the inspection personnel and determine the inspection task assignment plan based on the inspection location distribution information and the type of social governance event.

[0089] Preferably, the inspection task assignment scheme is determined according to the inspection location distribution information and the social governance event type, including:

[0090] Construct the system of equations that minimize the objective function and constraints:

[0091] min∑ n∈N D(L(n),m)·X(n);

[0092]

[0093] Where N is the set of all patrol officers, M is the set of monitoring points, E is the set of social governance event types, D(L(n),m) is the time required for patrol officer n to travel from the current location L(n) to monitoring point m, and D(L(n),m) is calculated based on the distance from the current location L(n) to monitoring point m and the transportation mode of patrol officer n. X(n) is a binary decision variable, indicating whether patrol officer n is assigned to monitoring point m, and X(n) = 1 indicates assignment, and X(n) = 0 indicates no assignment. C(e) indicates the minimum number of patrol officers required for monitoring point m to correspond to social governance event type e.

[0094] The integer linear programming algorithm is used to solve the equations and obtain the list of inspectors going to the corresponding monitoring points.

[0095] Preferably, obtaining the personnel concentration value according to personnel movement data analysis includes:

[0096] Based on the personnel movement data, filter out the personnel who have been in the monitoring field of view for a preset time interval and whose travel speed is less than the preset speed value;

[0097] The Euclidean distance between two persons is calculated based on the personnel movement data corresponding to the screened persons. The average distance of all Euclidean distances is calculated and the inverse is taken to obtain the personnel concentration value.

[0098] Preferably, the method of analyzing the audio and video data to obtain voice text keywords, and determining the type of social governance event at the current monitoring point according to the voice text keywords, personnel movement data and personnel posture data, includes:

[0099] Collect historical patrol personnel dispatch data, including the voice text keywords of each patrol personnel dispatch data, personnel movement data, personnel posture data, and the corresponding social governance event types and number of patrol personnel; perform data cleaning, encoding and format conversion on the collected data to obtain a data set;

[0100] The data set is divided into a training set and a test set according to a preset ratio, the original random forest model is trained using the training set, and the recognition accuracy of the random forest model is verified using the test set; when it is determined that the recognition accuracy of the random forest model meets the preset conditions, a social governance event type prediction model is generated;

[0101] The voice text keywords, personnel movement data and personnel posture data of the current monitoring point are input into the social governance event type prediction model, and the social governance event type of the current monitoring point is output.

[0102] It can be understood that the functions or modules included in the system provided in this embodiment can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

[0103] The present invention also provides an electronic device, including a processor and a memory, wherein the memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any possible implementation manner.

[0104] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method as described in any possible implementation manner.

[0105] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0106] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. Those skilled in the art can also clearly understand that the descriptions of the various embodiments of the present invention have different focuses. For the convenience and brevity of description, the same or similar parts may not be repeated in different embodiments. Therefore, for parts not described or not described in detail in a certain embodiment, refer to the records of other embodiments.

Claims

1. An intelligent method for allocating social governance events, characterized in that: The method comprises: Obtain audio and video data from different monitoring points, and analyze the audio and video data to obtain the number of people, personnel movement data, and personnel posture data; When the number of people is greater than the preset number of people, the personnel concentration value is obtained based on the personnel movement data analysis; if the personnel concentration value is greater than the historical average concentration value of the current monitoring point and lasts for a preset period of time, the audio and video data is analyzed to obtain voice text keywords, and the type of social governance event at the current monitoring point is determined based on the voice text keywords, personnel movement data and personnel posture data; Obtain the inspection location distribution information of inspection personnel, and determine the inspection task assignment plan based on the inspection location distribution information and the type of social governance event.

2. The intelligent dispatching method for social governance events according to claim 1 is characterized in that: The inspection task allocation scheme is determined according to the inspection location distribution information and the social governance event type, including: Construct the system of equations that minimize the objective function and constraints: min∑ n∈N D(L(n),m)·X(n); Where N is the set of all patrol officers, M is the set of monitoring points, E is the set of social governance event types, D(L(n),m) is the time required for patrol officer n to travel from the current location L(n) to monitoring point m, and D(L(n),m) is calculated based on the distance from the current location L(n) to monitoring point m and the transportation mode of patrol officer n. X(n) is a binary decision variable, indicating whether patrol officer n is assigned to monitoring point m, and X(n) = 1 indicates assignment, and X(n) = 0 indicates no assignment. C(e) indicates the minimum number of patrol officers required for monitoring point m to correspond to social governance event type e. The integer linear programming algorithm is used to solve the equations and obtain the list of inspectors going to the corresponding monitoring points.

3. The intelligent distribution method of social governance events according to claim 1 is characterized in that: The personnel concentration value obtained by analyzing the personnel movement data includes: Based on the personnel movement data, filter out personnel who have been in the monitoring field of view for a preset time interval and whose travel speed is less than a preset speed value; The Euclidean distance between two persons is calculated based on the personnel movement data corresponding to the screened persons. The average distance of all Euclidean distances is calculated and the inverse is taken to obtain the personnel concentration value.

4. The intelligent dispatching method for social governance events according to claim 1 is characterized in that: The method of analyzing the audio and video data to obtain voice text keywords and determining the type of social governance event at the current monitoring point based on the voice text keywords, personnel movement data, and personnel posture data includes: Collect historical patrol personnel dispatch data, including the voice text keywords of each patrol personnel dispatch data, personnel movement data, personnel posture data, and the corresponding social governance event types and number of patrol personnel; perform data cleaning, encoding and format conversion on the collected data to obtain a data set; The data set is divided into a training set and a test set according to a preset ratio, the original random forest model is trained using the training set, and the recognition accuracy of the random forest model is verified using the test set; when it is determined that the recognition accuracy of the random forest model meets the preset conditions, a social governance event type prediction model is generated; The voice text keywords, personnel movement data and personnel posture data of the current monitoring point are input into the social governance event type prediction model, and the social governance event type of the current monitoring point is output.

5. An intelligent dispatching system for social governance events, characterized in that: The system comprises: The data acquisition module is used to acquire audio and video data from different monitoring points, and analyze the audio and video data to obtain the number of personnel, personnel movement data, and personnel posture data; The social governance event type prediction module is used to obtain the personnel concentration value based on the personnel movement data analysis when the number of personnel is greater than the preset number of personnel; if the personnel concentration value is greater than the historical average concentration value of the current monitoring point and lasts for a preset period of time, the audio and video data is analyzed to obtain voice text keywords, and the social governance event type of the current monitoring point is determined based on the voice text keywords, personnel movement data and personnel posture data; The inspection task dispatch plan generation module is used to obtain the inspection location distribution information of the inspection personnel, and determine the inspection task dispatch plan based on the inspection location distribution information and the type of social governance event.

6. The intelligent dispatching system for social governance events according to claim 5 is characterized in that: The inspection task allocation scheme is determined according to the inspection location distribution information and the social governance event type, including: Construct the system of equations that minimize the objective function and constraints: min∑ n∈N D(L(n),m)·X(n); Where N is the set of all patrol officers, M is the set of monitoring points, E is the set of social governance event types, D(L(n),m) is the time required for patrol officer n to travel from the current location L(n) to monitoring point m, and D(L(n),m) is calculated based on the distance from the current location L(n) to monitoring point m and the transportation mode of patrol officer n. X(n) is a binary decision variable, indicating whether patrol officer n is assigned to monitoring point m, and X(n) = 1 indicates assignment, and X(n) = 0 indicates no assignment. C(e) indicates the minimum number of patrol officers required for monitoring point m to correspond to social governance event type e. The integer linear programming algorithm is used to solve the equations and obtain the list of inspectors going to the corresponding monitoring points.

7. The intelligent dispatching system for social governance events according to claim 5 is characterized in that: The personnel concentration value obtained by analyzing the personnel movement data includes: Based on the personnel movement data, filter out personnel who have been in the monitoring field of view for a preset time interval and whose travel speed is less than a preset speed value; The Euclidean distance between two persons is calculated based on the personnel movement data corresponding to the screened persons. The average distance of all Euclidean distances is calculated and the inverse is taken to obtain the personnel concentration value.

8. The intelligent dispatching system for social governance events according to claim 5 is characterized in that: The method of analyzing the audio and video data to obtain voice text keywords and determining the type of social governance event at the current monitoring point based on the voice text keywords, personnel movement data, and personnel posture data includes: Collect historical patrol personnel dispatch data, including the voice text keywords of each patrol personnel dispatch data, personnel movement data, personnel posture data, and the corresponding social governance event types and number of patrol personnel; perform data cleaning, encoding and format conversion on the collected data to obtain a data set; The data set is divided into a training set and a test set according to a preset ratio, the original random forest model is trained using the training set, and the recognition accuracy of the random forest model is verified using the test set; when it is determined that the recognition accuracy of the random forest model meets the preset conditions, a social governance event type prediction model is generated; The voice text keywords, personnel movement data and personnel posture data of the current monitoring point are input into the social governance event type prediction model, and the social governance event type of the current monitoring point is output.

9. An electronic device, characterized in that: include: A processor and a memory, the memory being used to store computer program code, the computer program code comprising computer instructions, and when the processor executes the computer instructions, the electronic device executes the intelligent dispatching method for social governance events as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes the intelligent dispatching method for social governance events as described in any one of claims 1 to 4.

Images