Public area intelligent management method and device, equipment and storage medium

Abstract

The invention discloses a public area intelligent management method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a video stream through a camera monitoring network; performing character detection and behavior recognition on the video stream to obtain an individual coordinate of each character and an individual behavior sequence corresponding to each character; mapping the individual coordinates and the individual behavior sequence to a public area digital map to obtain a people flow density prediction value of each sub-area in the next time period; and based on the people flow density predicted value and a preset threshold value, generating a resource allocation instruction and a tourist path guide instruction to perform public area management, thereby remarkably reducing the instantaneous people flow gathering degree and shortening the congestion duration during the peak period of festivals and holidays.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a public area intelligent management method and device, equipment and a storage medium. BACKGROUND

[0002] With the development of society, the daily reception capacity of public areas is repeatedly breaking records, and tourists' demand for traffic efficiency, play comfort and safety is also increasing. The management of scenic spots urgently needs a means to perceive the flow situation in advance and intervene actively to maintain good order without increasing manpower.

[0003] However, the existing technology generally uses gate counting, Wi-Fi probes or simple video crowd counting methods to obtain instantaneous passenger flow, and then increases security or sets temporary isolation barriers according to experience; these methods can only provide lagging overall number of people, lack of insight into individual movement intentions, and lead to repeated problems of crowd gathering during holiday peak periods.

[0004] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0005] The main purpose of the present application is to provide a public area intelligent management method and device, equipment and a storage medium, which aims to solve the technical problem of how to reduce the degree of crowd gathering in public areas during holidays in advance.

[0006] To achieve the above purpose, the present application provides a public area intelligent management method, which comprises the following steps:

[0007] Obtain a video stream through a camera monitoring network;

[0008] Detect and recognize characters in the video stream to obtain individual coordinates of each character and individual behavior sequences corresponding to each character;

[0009] Map the individual coordinates and the individual behavior sequences to a public area digital map to obtain a crowd density prediction value of each sub-area in the next period;

[0010] Based on the crowd density prediction value and a preset threshold, generate resource allocation instructions and tourist path guidance instructions for public area management.

[0011] In an embodiment, the step of detecting and recognizing characters in the video stream to obtain individual coordinates of each character and individual behavior sequences corresponding to each character comprises:

[0012] Decode the video stream frame by frame to obtain continuous image frames;

[0013] processing the continuous image frames by a pre-trained target detection network to obtain individual coordinates of each person;

[0014] cutting a person image sequence of the continuous image frames centered on the individual coordinates;

[0015] processing the person image sequence by a spatio-temporal behavior recognition network to obtain an individual behavior sequence corresponding to each person.

[0016] In an embodiment, the step of mapping the individual coordinates and the individual behavior sequence to a public area digital map to obtain a people flow density prediction value of each sub-region in a next period includes:

[0017] reading a pre-stored public area digital map, wherein the public area digital map contains sub-region numbers and sub-region entrance coordinates;

[0018] distance matching the individual coordinates and the sub-region entrance coordinates to obtain a sub-region number to which each person belongs;

[0019] predicting a next position of each person according to the individual behavior sequence;

[0020] counting a number of all persons falling into each sub-region number in a next period to obtain a people flow density prediction value.

[0021] In an embodiment, the step of generating a resource allocation instruction and a tourist path guidance instruction based on the people flow density prediction value and a preset threshold to perform public area management includes:

[0022] comparing the people flow density prediction value with a preset threshold to obtain a congestion sub-region list;

[0023] querying a demand staff number of a corresponding sub-region from a resource scheduling strategy table according to the congestion sub-region list to obtain a resource allocation instruction;

[0024] calculating a target path for each person to avoid a congestion sub-region based on the congestion sub-region list to obtain the tourist path guidance instruction;

[0025] sending the resource allocation instruction to an on-site management terminal and pushing the tourist path guidance instruction to a mobile terminal held by a corresponding person.

[0026] In an embodiment, the step of calculating a shortest path for each person to avoid a congestion sub-region to obtain the tourist path guidance instruction includes:

[0027] obtaining individual coordinates of each person and a target scenic spot coordinate of the person;

[0028] construct a non-congestion topology path starting from the individual coordinate and ending at the target scenic spot coordinate;

[0029] perform travel time evaluation on the non-congestion topology path, select a path with the minimum travel time as a shortest path according to the evaluation result, and encapsulate the shortest path as the tourist path guidance instruction.

[0030] In an embodiment, the method further comprises:

[0031] obtaining environment data, wherein the environment data includes smoke concentration values, temperature values, and ground vibration values;

[0032] when any of the environment data exceeds a corresponding preset threshold value, marking an abnormal sub-region;

[0033] sending a sub-region number of the abnormal sub-region and an on-site verification instruction to an on-site management terminal.

[0034] In an embodiment, the method further comprises:

[0035] statistically obtaining a historical passenger flow density prediction value of a historical period in a fixed time window to obtain a historical passenger flow sequence;

[0036] processing the historical passenger flow sequence through an attention mechanism network to obtain a predicted peak period and a predicted peak area in a future period;

[0037] adjusting the preset threshold value based on the predicted peak period;

[0038] pushing a staggered peak prompting information including the predicted peak period and the predicted peak area to a corresponding mobile terminal held by a tourist when the tourist enters the park.

[0039] In addition, to achieve the above-mentioned purpose, the application further provides a public area intelligent management device, characterized in that the device comprises:

[0040] an acquisition module configured to acquire a video stream through a camera monitoring network;

[0041] an identification analysis module configured to perform character detection and behavior recognition on the video stream to obtain individual coordinates of each character and an individual behavior sequence corresponding to each character;

[0042] a prediction module configured to map the individual coordinates and the individual behavior sequence to a public area digital map to obtain a passenger flow density prediction value of each sub-region in a next period;

[0043] a scheduling generation module configured to generate a resource allocation instruction and a tourist path guidance instruction based on the passenger flow density prediction value and a preset threshold value to manage the public area.

[0044] In addition, to achieve the above object, the application further provides a public area intelligent management device, which comprises a memory, a processor and a public area intelligent management program stored in the memory and executable on the processor, and the public area intelligent management program is configured to implement the steps of the public area intelligent management method.

[0045] In addition, to achieve the above object, the application further provides a storage medium, which stores a public area intelligent management program, and the public area intelligent management program implements the steps of the public area intelligent management method when executed by a processor.

[0046] In addition, to achieve the above object, the application further provides a computer program product, which comprises a computer program, and the computer program implements the steps of the public area intelligent management method when executed by a processor.

[0047] The one or more technical solutions provided by the application have at least the following technical effects:

[0048] The video stream is obtained through the camera monitoring network, and the person detection and behavior recognition are implemented for each frame, so as to obtain the individual coordinates and individual behavior sequence; after the sequence is mapped into the public area digital map, the next position of the individual is extrapolated by using the continuous space-time probability model, so as to construct the crowd flow density prediction value of each sub-area in the next period. When the prediction value of a certain sub-area is higher than a preset threshold value, the system immediately triggers the resource allocation instruction and the tourist path guiding instruction: the former calculates the required on-site staff number according to the sub-area area and congestion level and sends it to the on-site management terminal, and the latter generates a detour scheme for each tourist based on the shortest path algorithm and pushes it to the mobile terminal. Since all the calculations are completed before the actual arrival of the crowd flow, the on-site only needs to execute the instructions, and no additional manpower is needed to reduce the crowd flow density of the sub-area to below the threshold value before the predicted time, so that the degree of crowd flow aggregation in the public area during the holiday is reduced in advance without increasing the on-site personnel. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application together with the specification.

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced in the following. Obviously, for those skilled in the field, the other drawings can also be obtained without creative labor based on these drawings.

[0051] Figure 1 The flowchart provided by the first embodiment of the public area intelligent management method of the application;

[0052] Figure 2 The overall framework diagram provided by the first embodiment of the public area intelligent management method of the present application is shown in the following figure:

[0053] Figure 3 The terminal device workflow diagram provided by the first embodiment of the public area intelligent management method of the present application is shown in the following figure:

[0054] Figure 4 The flowchart provided by the second embodiment of the public area intelligent management method of the present application is shown in the following figure:

[0055] Figure 5 The module structure diagram of the public area intelligent management device of the embodiment of the present application is shown in the following figure:

[0056] Figure 6 The device structure diagram of the hardware running environment involved in the public area intelligent management method of the embodiment of the present application is shown in the following figure.

[0057] The purpose implementation, functional features and advantages of the present application will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0058] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0059] In order to better understand the technical solutions of the present application, the following will be described in detail in conjunction with the drawings and specific embodiments of the specification.

[0060] It should be noted that the execution subject of the present embodiment can be a computing service device with data processing, network communication and program running functions, such as tablet computers, personal computers, mobile phones and the like, or an electronic device capable of realizing the above functions, a public area intelligent management device and the like. The present embodiment and the following embodiments will be described taking the public area intelligent management device as an example.

[0061] Based on this, the present embodiment provides a public area intelligent management method, which is described in detail as follows with reference to the accompanying drawings: Figure 1 , Figure 1 The flowchart of the first embodiment of the public area intelligent management method of the present application is shown in the following figure:

[0062] In the present embodiment, the public area intelligent management method comprises steps S10-S40:

[0063] Step S10, video stream is acquired through the camera monitoring network;

[0064] It should be noted that the camera monitoring network refers to a network composed of multiple high-definition cameras deployed in public areas. These cameras can capture dynamic images in real time and transmit video data in the form of digital signals to the backend server. Video stream refers to the dynamic video data continuously output by the camera, containing information of each frame of picture in time sequence, usually transmitted and stored in a specific codec format (such as H.264, H.265).

[0065] As shown in Figure 2 , a camera is installed in the public area to obtain real-time monitoring video data. After obtaining the video, AI recognition is performed on the people and environment in the video, and analysis is performed on the people to generate data and record the situation. The location of the person in the public area is recorded in real time, and the next action of the current person is analyzed. Through the above information, the dynamic of each person in the public area can be obtained, and a public area personnel map is generated. AI recognition obtains the real-time location of each person, generates a real-time map of the public area, and shares it with tourists. The system provides the best play route based on the current situation through algorithm, improving the tourist experience. After obtaining the above AI information, the public area manager understands the real-time situation of the public area, and allocates resources scientifically using algorithms, such as increasing staff in areas with high flow of people, to achieve efficient use of human resources. Using the AI recognition function of the camera, the environment of the public area is monitored, and real-time alarm is given when there is a fire or police situation, and relevant personnel are automatically notified to handle it. AI recognition is used to monitor uncivilized behavior in the public area, such as smoking in areas where smoking is prohibited. Information is sent to staff for processing, and information is sent to tourists.

[0066] As shown in Figure 3 , the multiple cameras in the public area form a monitoring area and the terminal is successfully connected. Using the AI recognition of the local cameras in the park, the monitoring area detects that someone is present or someone is active, or an abnormal situation occurs. The local camera transmits the video of the monitoring area to the terminal in real time. The terminal performs AI recognition processing on the obtained video. After AI recognition of the people and environment in the video stream and AI behavior analysis of the people's behavior, the terminal records the current data and generates a personnel map.

[0067] The terminal realizes personnel mobilization, environment monitoring, and public area flow distribution based on AI recognition data, generates the nearest play route for tourists according to the algorithm, and pushes information.

[0068] Steps 3-6 are repeated continuously.

[0069] Step S20, character detection and behavior recognition are performed on the video stream to obtain the individual coordinates of each character and the individual behavior sequence corresponding to each character;

[0070] It should be noted that the two processes of person detection and behavior recognition are integrated, and the purpose is to extract the position information and behavior information of each person from the video to provide data support for subsequent prediction and management. Person detection is responsible for locating the position of the person, and behavior recognition is responsible for analyzing the behavior of the person, and the combination of the two can comprehensively understand the dynamic situation of the personnel in the public area.

[0071] In a feasible implementation, step S20 includes steps A11-A14:

[0072] A11: Decoding the video stream frame by frame to obtain continuous image frames;

[0073] It should be noted that frame-by-frame decoding means that the video stream is decomposed into a series of static image frames according to the time sequence, and each image frame represents the picture of the video at a certain time point. This step is usually completed by video processing software or hardware decoder, and the decoded image frames are saved in common image formats (such as JPEG, PNG) for subsequent image processing algorithms to analyze.

[0074] A12: Process the continuous image frames through the pre-trained target detection network to obtain the individual coordinates of each person;

[0075] It should be noted that the pre-trained target detection network is a computer vision model based on deep learning, which has been trained on a large amount of labeled data and can recognize people in images and determine their positions. Individual coordinates refer to the specific position of a person in the image plane coordinate system, usually represented by (x, y) coordinate points representing the center position of the person or the diagonal coordinates of the bounding box. The bounding box is a rectangular area that tightly encloses the detected person, and its coordinates can be represented by the coordinates of the upper left corner and the lower right corner.

[0076] A13: Extract the person image sequence from the continuous image frames centered on the individual coordinates;

[0077] It should be noted that the person image sequence containing the person is extracted from the continuous image frames centered on the person's coordinates. Specifically, the area where the person is located is cropped from each image frame according to the coordinates of the bounding box to form an image sequence containing only the person. This helps the subsequent analysis of the behavior of the person, because only the part related to the person is retained, and irrelevant information such as the background is removed, improving the processing efficiency and accuracy.

[0078] A14: Process the person image sequence through the spatio-temporal behavior recognition network to obtain the individual behavior sequence corresponding to each person.

[0079] It should be noted that the spatio-temporal behavior recognition network is a deep learning model capable of analyzing the motion changes of a person in time and space, thereby identifying the behavior patterns of the person. The individual behavior sequence refers to the sequence of behavior labels of each person in continuous time, such as continuous records of behaviors such as "walking", "standing", "waving", and "running". The model extracts key features and classifies through analysis of the sequence of person images, and finally outputs the behavior label sequence.

[0080] Step S30, mapping the individual coordinates and individual behavior sequence to the public area digital map to obtain the crowd density prediction value of each sub-area in the next period;

[0081] It should be noted that the position and behavior information of the person are mapped to the digital map, and the future crowd distribution is predicted through a spatio-temporal prediction model (such as a time series analysis or a deep learning prediction model), so as to predict the congested area in advance. The prediction result is intuitively displayed on the digital map in the form of a heat map, helping the management personnel to quickly identify high-risk areas.

[0082] It should be understood that the terminal realizes AI recognition on each video, obtains the real-time position of each public area visitor, records the information, predicts the next public area to be visited by the visitor, and generates a real-time map and a trend map of the public area visitor. The terminal pushes the public area map to each visitor to improve the visitor's playing experience, and provides the visitor with the best playing route through an algorithm, thereby realizing the management and control of the crowd in the public area. After the public area manager obtains the real-time map of the public area generated by AI recognition, the manager reasonably dispatches resources, automatically dispatches staff to places with large crowds to help, and realizes the scientific use of human resources in the public area. The terminal realizes AI recognition on the people in the video, obtains whether the current person has dangerous behavior or the visitor needs help, and the like, the system automatically notifies the public area staff, and monitors and processes the results. The terminal realizes AI recognition on the environment in the video, monitors whether the environment appears dangerous, such as fire, damaged road, and the like, finds problems, and reports in real time and notifies relevant personnel to handle the problems.

[0083] In a feasible implementation, step S30 includes steps A21-A24:

[0084] A21: reading a pre-stored public area digital map, wherein the public area digital map contains sub-area numbers and sub-area entrance coordinates;

[0085] It should be noted that the pre-stored public area digital map is a pre-made electronic map that divides the public area into multiple sub-areas, each sub-area has a unique number, and the specific coordinate position of each sub-area entrance is recorded. The sub-area number is a unique code used to identify different areas within the public area, and the entrance coordinate refers to the position of each sub-area entrance in the map coordinate system. The digital map is usually stored in geographic information system (GIS) format and contains rich spatial information.

[0086] A22: Match the individual coordinates with the sub-area entrance coordinates to obtain the sub-area number to which each person belongs;

[0087] It should be noted that distance matching refers to calculating the Euclidean distance between the coordinates of the person and the coordinates of each sub-area entrance, determining which sub-area entrance the person is closest to, and thus determining the sub-area where the person is currently located. This step realizes the positioning of the person on the map through coordinate calculation, and usually uses the nearest neighbor algorithm or spatial index structure (such as R-tree) to improve matching efficiency.

[0088] A23: Predict the next position of each person based on the individual behavior sequence;

[0089] It should be noted that by analyzing the movement trend and behavior pattern of the person, the position that the person can appear at a future time is predicted. For example, if the person is moving in a certain direction for several consecutive frames, the system will predict the next position of the person based on this movement trend. The prediction algorithm can be based on simple linear regression, or a complex deep learning model (such as LSTM network) can be used to capture the spatio-temporal dependence.

[0090] A24: Count the number of people falling into each sub-area number in the next time period, and obtain the people flow density prediction value.

[0091] It should be noted that the number of people in each sub-area in the predicted next time period is counted, and the people flow density of each sub-area is calculated. Specifically, the predicted next position of all people is mapped to the sub-area number, and then the number of people in each sub-area is counted. The people flow density prediction value is usually expressed in the form of the number of people per square meter, which provides a basis for subsequent resource allocation and people flow guidance.

[0092] Step S40, based on the people flow density prediction value and the preset threshold, generating resource allocation instructions and tourist path guidance instructions for public area management.

[0093] It should be noted that the predicted crowd density is compared with the set threshold value, and when the predicted density exceeds the threshold value, the corresponding instruction is generated. The resource allocation instruction is an instruction sent to the management personnel, instructing them how to adjust the resource allocation, such as assigning more staff to the congested area, increasing temporary facilities, etc. The tourist path guidance instruction is a suggestion sent to tourists, guiding them to choose a better tour route to achieve effective management of the public area. The instructions can be issued through management terminals (such as handheld devices, control center screens) and tourist mobile terminals (such as mobile phone applications, SMS services).

[0094] Further, the present scheme also includes:

[0095] Obtain environmental data, wherein the environmental data includes smoke concentration value, temperature value and ground vibration value;

[0096] When any of the environmental data exceeds the corresponding preset threshold value, mark the abnormal sub-area;

[0097] Send the sub-area number of the abnormal sub-area and the on-site verification instruction to the on-site management terminal.

[0098] It should be noted that the environmental data refers to the information about the environmental state of the public area collected by various sensors (such as smoke sensors, temperature sensors, vibration sensors). The smoke concentration value refers to the concentration of smoke particles in the air, usually expressed in the number of smoke particles per cubic meter; the temperature value is the thermodynamic temperature of the environment, usually expressed in Celsius or Fahrenheit; the ground vibration value is the intensity of ground vibration, usually expressed in the form of acceleration or speed. These data are transmitted in real time to the backend system through Internet of Things (IoT) devices for monitoring the environmental safety of the public area.

[0099] When any of the environmental data exceeds the corresponding preset threshold value, mark the abnormal sub-area. The preset threshold value refers to the upper or lower limit of the data set according to safety standards or management requirements. For example, the smoke concentration threshold can be set to 0.1 milligrams per cubic meter, the temperature threshold can be set to 40 degrees Celsius, and the ground vibration threshold can be set to 0.5 meters per second 2 .

[0100] When the actual monitoring data exceeds these thresholds, the system will mark the corresponding sub-area as an abnormal sub-area, indicating that the area may have safety hazards and needs further verification and handling. Send the sub-area number of the abnormal sub-area and the on-site verification instruction to the on-site management terminal. This step is to send the information of the abnormal sub-area and the on-site verification instruction to the on-site management personnel, so that they can timely go to verify and handle the abnormal situation. The on-site management terminal can be a handheld device (such as a smartphone, a tablet computer) or a fixed workstation of the management personnel, and the instruction is pushed to the management personnel in the form of a notification or a task list, ensuring that they can respond quickly.

[0101] Still further, the present solution also includes:

[0102] Statistically obtaining a historical flow density prediction value of a historical period in a fixed time window, to obtain a historical flow sequence;

[0103] Processing the historical flow sequence through an attention mechanism network to obtain a predicted peak period and a predicted peak area of a future period;

[0104] Adjusting a preset threshold based on the predicted peak period;

[0105] Pushing a staggered peak prompt information containing the predicted peak period and the predicted peak area to a mobile terminal held by a corresponding person when the person enters a park.

[0106] It should be noted that the historical flow density prediction value of the historical period in the fixed time window is statistically obtained to obtain the historical flow sequence. The fixed time window refers to a fixed time period, such as every 15 minutes or every hour, for statistical historical data. The historical flow sequence refers to a sequence of flow density data recorded in these fixed time windows, reflecting the past flow trend. The selection of the time window can be adjusted according to actual needs to balance the granularity of the data and the accuracy of the prediction.

[0107] The historical flow sequence is processed through an attention mechanism network to obtain a predicted peak period and a predicted peak area of a future period. The attention mechanism network is a deep learning model that can automatically learn important features and temporal dependencies in data. Through analysis of the historical flow sequence, the model can predict possible peak periods and peak areas in the future. The attention mechanism can highlight the most important time points or features in the sequence for prediction, improving the accuracy of the prediction.

[0108] The preset threshold is adjusted based on the predicted peak period. This step is to adjust the preset threshold in advance according to the predicted peak period, so that the system can more flexibly cope with the flow pressure in the peak period. For example, when it is predicted that the flow will increase significantly in a certain period, the congestion warning threshold can be appropriately increased to avoid frequent false alarms; at the same time, more resources can be allocated to the predicted peak area in advance to ensure that the peak flow can be effectively coped with.

[0109] The staggered peak prompt information containing the predicted peak period and the predicted peak area is pushed to a mobile terminal held by a corresponding person when the person enters a park. This step is to push the prompt information containing the predicted peak period and area to the mobile terminal or other mobile device of the visitor when the visitor enters the public area, suggesting them to stagger their visit to improve the visit experience and reduce congestion. The push information can be sent through a mobile application, SMS or email, etc., and the content includes the specific time of the peak period, the possible congestion area and the recommended staggered visit route.

[0110] The embodiment provides a public area intelligent management method. Video streams are acquired through a camera monitoring network, human detection and behavior recognition are performed on each frame, individual coordinates and individual behavior sequences are obtained, the sequences are mapped into a public area digital map, a continuous space-time probability model is used to extrapolate a next position of the individual, and thus a flow density prediction value of each sub-area in a next period is constructed. When the prediction value of a certain sub-area is higher than a preset threshold value, a resource allocation instruction and a tourist path guiding instruction are triggered immediately: the former calculates a required number of on-site staff according to a sub-area area and a congestion level and is sent to an on-site management terminal, and the latter generates a detour scheme for each tourist based on a shortest path algorithm and is pushed to a mobile terminal. Since all calculations are completed before actual arrival of the flow, on-site staff only need to execute the instructions, and the flow density of the sub-area can be reduced to below the threshold value before the prediction time without additional increase of manpower, so that the flow gathering degree of the public area during holidays is reduced in advance without increase of on-site staff.

[0111] Based on the first embodiment, the same or similar contents in the second embodiment can be referred to the above description, and will not be described in detail. On this basis, please refer to Figure 4 , step S40, including steps S401-S404:

[0112] Step S401, comparing the flow density prediction value with a preset threshold value to obtain a congestion sub-area list;

[0113] It should be noted that the system compares the predicted flow density of each sub-area with a preset threshold value. The preset threshold value is a standard value set according to the bearing capacity and management requirements of the public area, and is used to judge whether the flow density of a certain area is too high. When the predicted density of a certain sub-area exceeds this threshold value, the area will be marked as a congestion sub-area and added to the congestion sub-area list. This list will be used as the basis for subsequent resource allocation and path guidance.

[0114] Step S402, querying the required number of staff of the corresponding sub-area from a resource scheduling strategy table according to the congestion sub-area list to obtain a resource allocation instruction;

[0115] It should be noted that the resource scheduling strategy table is a pre-defined data table, which specifies the required number of staff to be dispatched for each sub-area under different congestion levels. The system finds the required number of staff corresponding to each congestion sub-area in the strategy table according to the congestion sub-area list, and then integrates these information to generate a resource allocation instruction. This instruction specifies the required number of staff to be dispatched and the specific sub-area they should go to for on-site management and guidance.

[0116] Step S403, calculate the target path for each person to avoid the congestion sub-area based on the congestion sub-area list, and obtain the tourist path guidance instruction;

[0117] It should be noted that the optimal path for the tourist to avoid the congestion area is planned. The system will analyze the current position of each tourist and their target scenic spot, combined with the congestion sub-area list, to calculate a path that will not pass through the congestion area. The tourist path guidance instruction will contain information about this path, which will be used to guide the movement of the tourist.

[0118] In a possible implementation, step S403 includes steps A31-A33:

[0119] A31: Obtain the individual coordinates of each person and the target scenic spot coordinates of the person;

[0120] It should be noted that the individual coordinates are the current positions of the person in the public area, which are obtained through the previous detection steps. The target scenic spot coordinates are the positions of the scenic spot that the tourist plans to visit on the digital map, which can be obtained through the input of the tourist or historical behavior prediction. The system needs to obtain both sets of coordinates at the same time to prepare for path planning.

[0121] A32: Construct a congestion-free topological path with the individual coordinates as the starting point and the target scenic spot coordinates as the end point;

[0122] It should be noted that based on the digital map of the public area and the current congestion situation, a path network is constructed to connect the starting point and the end point. This path network will exclude all congestion sub-areas to ensure that the generated path is unobstructed. Topological path refers to the path structure constructed on the map according to certain rules, which considers the connection relationship between sub-areas, avoiding crossing obstacles or prohibited areas.

[0123] A33: Perform travel time evaluation on the congestion-free topological path, select the path with the shortest travel time as the shortest path according to the evaluation result, and encapsulate the shortest path as the tourist path guidance instruction.

[0124] It should be noted that the system will evaluate the estimated travel time of each candidate path based on factors such as the length of the path, the average moving speed of the tourist, and possible stopping points on the path. Then select the path with the shortest travel time from among them, and determine it as the shortest path recommended to the tourist. The information of this path will be arranged into the tourist path guidance instruction, which will be pushed to the tourist through their mobile device, helping the tourist to efficiently reach the destination.

[0125] Step S404, send the resource allocation instruction to the on-site management terminal, and push the tourist path guidance instruction to the mobile terminal held by the corresponding person.

[0126] It should be noted that the instructions generated in the foregoing are delivered to the corresponding execution parties. The field management terminal refers to the equipment used by the field management personnel, such as a handheld terminal or a computer in the control center, and they can adjust the resource allocation in the field in a timely manner according to the resource allocation instructions. At the same time, the tourist path guidance instructions are pushed to the tourists through their mobile devices (such as the park application on the smart phone), providing real-time navigation information for the tourists and helping them avoid congestion and improve the touring experience.

[0127] The embodiment provides a public area intelligent management method, which can predict in advance which areas are likely to be congested through real-time monitoring and analysis of video streams, and automatically trigger corresponding management measures. The resource allocation instructions enable the management personnel to timely deploy staff to the required places for effective on-site relief. The tourist path guidance instructions help tourists avoid congested areas, reasonably plan the touring route, reduce waiting time, and improve the comfort and satisfaction of touring. This intelligent management method not only improves the operation efficiency of the public area, but also significantly improves the experience quality of the tourists, achieving the dual improvement of management efficiency and tourist satisfaction.

[0128] It should be noted that the above examples are only for understanding the present application and do not constitute a limitation on the public area intelligent management method of the present application. More forms of simple changes based on this technical concept are within the protection scope of the present application.

[0129] The present application also provides a public area intelligent management device, which comprises Figure 5 The acquisition module 10 is configured to acquire video streams through the camera monitoring network.

[0130] The identification analysis module 20 is configured to perform person detection and behavior recognition on the video streams to obtain individual coordinates of each person and individual behavior sequences corresponding to each person.

[0131] The prediction module 30 is configured to map the individual coordinates and the individual behavior sequences to a public area digital map to obtain a people flow density prediction value of each sub-area in the next period.

[0132] The scheduling generation module 40 is configured to generate resource allocation instructions and tourist path guidance instructions based on the people flow density prediction value and a preset threshold to manage the public area.

[0133]

[0134] ​The public area intelligent management device provided by the application adopts the public area intelligent management method in the above embodiment, and can solve the technical problem of how to reduce the degree of human flow aggregation in public areas during holidays in advance. Compared with the prior art, the beneficial effects of the public area intelligent management device provided by the application are the same as those of the public area intelligent management method provided by the above embodiment, and the other technical features of the public area intelligent management device are the same as those disclosed in the above embodiment method, which will not be repeated here.

[0135] In an embodiment, the recognition analysis module 20 is further configured to decode the video stream frame by frame to obtain continuous image frames;

[0136] The pre-trained target detection network is used to process the continuous image frames to obtain individual coordinates of each person;

[0137] The individual coordinates are used as the center to cut the person image sequence of the continuous image frames;

[0138] The spatio-temporal behavior recognition network is used to process the person image sequence to obtain the individual behavior sequence corresponding to each person.

[0139] In an embodiment, the prediction module 30 is further configured to read a pre-stored public area digital map, wherein the public area digital map includes sub-area numbers and sub-area entrance coordinates;

[0140] The individual coordinates and the sub-area entrance coordinates are matched in distance to obtain the sub-area number to which each person belongs;

[0141] The individual behavior sequence is used to predict the next position of each person;

[0142] The number of all persons falling into each sub-area number in the next time period is counted to obtain the human flow density prediction value.

[0143] In an embodiment, the scheduling generation module 40 is further configured to compare the human flow density prediction value with a preset threshold to obtain a congestion sub-area list;

[0144] According to the congestion sub-area list, the number of required staff for the corresponding sub-area is queried from a resource scheduling strategy table to obtain a resource allocation instruction;

[0145] Based on the congestion sub-area list, a target path for each person to avoid the congestion sub-area is calculated to obtain a tourist path guidance instruction;

[0146] The resource allocation instruction is sent to a field management terminal, and the tourist path guidance instruction is pushed to a mobile terminal held by the corresponding person.

[0147] In an embodiment, the scheduling generation module 40 is further configured to obtain the individual coordinates of each person and the target scenic spot coordinates of the person;

[0148] constructing a non-congestion topology path starting from the individual coordinate and ending at the target scenic spot coordinate;

[0149] conducting a travel time evaluation on the non-congestion topology path, selecting a path with the minimum travel time as the shortest path according to the evaluation result, and encapsulating the shortest path as a tourist path guiding instruction.

[0150] In an embodiment, the dispatch generation module 40 is further configured to obtain environmental data, wherein the environmental data includes smoke concentration values, temperature values, and ground vibration values.

[0151] When any of the environmental data exceeds a corresponding preset threshold value, an abnormal sub-region is marked;

[0152] The sub-region number of the abnormal sub-region and the on-site verification instruction are sent to the on-site management terminal.

[0153] In an embodiment, the dispatch generation module 40 is further configured to obtain a historical human flow density prediction value of a historical period by fixed time window statistics, to obtain a historical human flow sequence.

[0154] The historical human flow sequence is processed by an attention mechanism network to obtain a predicted peak period and a predicted peak area in a future period.

[0155] The preset threshold value is adjusted based on the predicted peak period.

[0156] The off-peak prompting information including the predicted peak period and the predicted peak area is pushed to a corresponding mobile terminal held by a tourist when the tourist enters the park.

[0157] The present application provides a public area intelligent management device, which comprises at least one processor and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the public area intelligent management method in the above-mentioned embodiment one.

[0158] Reference will be made to the following description of the embodiments Figure 6The diagram illustrates a structural schematic suitable for implementing the intelligent public area management device of the embodiments of this application. The intelligent public area management device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6 The smart management device for public areas shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0159] like Figure 6 As shown, the public area intelligent management device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the public area intelligent management device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touch screens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. The communication device 1009 allows the public area intelligent management device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows public area intelligent management devices with various systems, it should be understood that it is not required to implement or possess all of the systems shown. More or fewer systems may be implemented alternatively.

[0160] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program codes for executing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiments disclosed in the present application are executed.

[0161] The public area intelligent management device provided by the present application adopts the public area intelligent management method in the above-mentioned embodiments, and can solve the technical problem of how to reduce the degree of flow aggregation in public areas during holidays in advance. Compared with the prior art, the public area intelligent management device provided by the present application has the same beneficial effects as the public area intelligent management method provided by the above-mentioned embodiments, and other technical features in the public area intelligent management device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0162] It should be understood that various parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the description of the above-mentioned embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0163] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0164] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e. computer program) for executing the public area intelligent management method in the above-mentioned embodiments.

[0165] The computer readable storage medium provided in the application may be, for example, a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium may include, but are not limited to, an electric connection with one or more conductive wires, a portable computer disk, a hard disk, a RAM (Random Access Memory), a ROM (Read Only Memory), an EPROM (Erasable Programmable Read Only Memory or flash memory), an optical fiber, a CD-ROM (CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the embodiment, the computer readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including but not limited to an electric wire, an optical cable, an RF (Radio Frequency), and the like, or any suitable combination of the above.

[0166] The computer readable storage medium described above may be contained in the public area intelligent management device, or may exist separately without being assembled into the public area intelligent management device.

[0167] The computer readable storage medium described above carries one or more programs, which, when executed by the public area intelligent management device, cause the public area intelligent management device to: obtain a video stream through a camera monitoring network; perform person detection and behavior recognition on the video stream to obtain individual coordinates of each person and an individual behavior sequence corresponding to each person; map the individual coordinates and the individual behavior sequence to a public area digital map to obtain a people flow density prediction value of each sub-area in a next time period; and generate resource allocation instructions and tourist path guidance instructions based on the people flow density prediction value and a preset threshold to perform public area management.

[0168] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0169] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functions, and operations that can be implemented in systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code that comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0170] The modules involved in the embodiments of the present application can be implemented in software or hardware. In some cases, the names of the modules do not constitute a limitation on the modules themselves.

[0171] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e. computer programs) for executing the above-mentioned public area intelligent management method, and can solve the technical problem of how to reduce the degree of public holiday crowd gathering in advance. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the public area intelligent management method provided by the above-mentioned embodiments, which will not be repeated here.

[0172] The application further provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the public area intelligent management method as described above.

[0173] The computer program product provided by the application can solve the technical problem of how to reduce the degree of public holiday crowd gathering in advance. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the public area intelligent management method provided by the above-mentioned embodiments, and are not described here.

[0174] The above only describes some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation, direct / indirect application in other related technical fields, or direct / indirect application in other related technical fields based on the technical concept of the application and the content of the specification and drawings are included in the patent protection scope of the application.

Claims

1. A method for intelligent management of public areas, characterized in that, The method includes: Video streams are acquired through a network of surveillance cameras; The video stream is subjected to person detection and behavior recognition to obtain the individual coordinates of each person and the corresponding individual behavior sequence of each person; Mapping the individual coordinates and the individual behavior sequence to a digital map of the public area yields the predicted population density for each sub-area in the next time period. Based on the predicted population density and preset threshold, resource allocation instructions and tourist route guidance instructions are generated for public area management.

2. The method as described in claim 1, characterized in that, The steps of performing person detection and behavior recognition on the video stream to obtain the individual coordinates of each person and the corresponding individual behavior sequence for each person include: The video stream is decoded frame by frame to obtain continuous image frames; The continuous image frames are processed by a pre-trained object detection network to obtain the individual coordinates of each character; A sequence of human images is extracted from consecutive image frames centered on the individual coordinates. The spatiotemporal behavior recognition network is used to process the image sequence of the person to obtain the individual behavior sequence corresponding to each person.

3. The method as described in claim 1, characterized in that, The step of mapping the individual coordinates and the individual behavior sequence to a digital map of a public area to obtain the predicted pedestrian density values ​​for each sub-area in the next time period includes: Read the pre-stored digital map of the public area, wherein the digital map of the public area includes sub-area numbers and sub-area entrance coordinates; The individual coordinates and the sub-region entrance coordinates are matched by distance to obtain the sub-region number to which each character belongs; Predict the next position of each character based on the individual behavior sequence; The number of people entering each sub-area number in the next time period is counted to obtain the predicted value of the crowd density.

4. The method as described in claim 1, characterized in that, The step of generating resource allocation instructions and visitor route guidance instructions based on the predicted population density and a preset threshold for public area management includes: The predicted pedestrian density is compared with a preset threshold to obtain a list of congested sub-regions. Based on the list of congested sub-regions, query the resource scheduling strategy table to obtain the number of staff required for the corresponding sub-region and obtain resource allocation instructions; Based on the list of congested sub-regions, a target path to avoid the congested sub-regions is calculated for each person, and the tourist path guidance instruction is obtained. The resource allocation instruction is sent to the on-site management terminal, and the tourist route guidance instruction is pushed to the mobile terminal held by the corresponding person.

5. The method as described in claim 4, characterized in that, The steps of calculating the shortest path to avoid congested sub-areas for each person and obtaining the tourist path guidance instructions include: Obtain the individual coordinates of each character and the coordinates of the target location for each character; Construct a congestion-free topology path that starts at the coordinates of the individual and ends at the coordinates of the target scenic spot; The travel time of the congestion-free topology path is evaluated, and the path with the shortest travel time is selected as the shortest path based on the evaluation results. The shortest path is then encapsulated as the tourist path guidance instruction.

6. The method as described in claim 1, characterized in that, The method further includes: Acquire environmental data, including smoke concentration, temperature, and ground vibration values; When any item in the environmental data exceeds the corresponding preset threshold, an abnormal sub-region is marked. The sub-region number of the abnormal sub-region and the on-site verification instruction are sent to the on-site management terminal.

7. The method as described in claim 1, characterized in that, The method further includes: By statistically analyzing the predicted values ​​of pedestrian density for historical periods within a fixed time window, a historical pedestrian flow sequence is obtained. The historical pedestrian flow sequence is processed by an attention mechanism network to obtain the predicted peak time period and predicted peak area for future periods. Adjust the preset threshold based on the predicted peak period; When visitors enter the park, a staggered peak time reminder message containing the predicted peak time period and the predicted peak area will be pushed to the mobile terminal held by the corresponding person.

8. A smart management device for public areas, characterized in that, The device includes: The acquisition module is used to acquire video streams through a video surveillance network; The identification and analysis module is used to perform person detection and behavior recognition on the video stream to obtain the individual coordinates of each person and the corresponding individual behavior sequence of each person; The prediction module is used to map the individual coordinates and the individual behavior sequence to a digital map of the public area to obtain the predicted value of the crowd density of each sub-area in the next time period; The scheduling generation module is used to generate resource allocation instructions and tourist path guidance instructions based on the predicted value of pedestrian flow density and preset threshold for public area management.

9. A public area intelligent management device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the public area intelligent management method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the public area intelligent management method as described in any one of claims 1 to 7.

Images