A personnel data processing method, device and system based on hot zone

Abstract

The present application discloses a personnel data processing method, device and system based on thermal zones, the method comprising: determining whether a person appears within the shooting range of a camera, wherein the shooting range of the camera is divided into multiple areas, each area being a thermal zone; obtaining the movement trajectory of the person within the shooting range from the video captured by the camera; obtaining the cross-zone behavior of the person appearing in the movement trajectory, wherein the cross-zone behavior is the behavior of the person moving from one thermal zone to another. This application solves the problem in the prior art that the passenger flow situation cannot be better judged due to only focusing on a single thermal zone, thereby obtaining the flow of people between thermal zones, providing data support for more accurate judgment of passenger flow situation.

Description

Technical Field

[0001] The present application relates to the field of camera video analysis, and more specifically, to a personnel data processing method, device, and system based on hot zones. Background Art

[0002] Customer flow heatmaps have important applications in intelligent markets across various industries, including exhibition halls, city squares, corporate production lines, bank lobbies, bank workstations, construction sites, and chain supermarkets. In existing technologies, heatmaps are primarily used to identify thermal zones. For example, they can calculate the flow of people in a certain area and generate a heatmap distribution based on the flow of people.

[0003] The existing heatmap distribution focuses on the formation and changes of a single thermal area (referred to as a hot zone). That is, when the passenger flow in a thermal area changes, it can be reflected in the corresponding heatmap of the thermal area. The inventors found that in most cases, a region may include multiple thermal areas. This approach of focusing only on a single thermal area cannot effectively determine passenger flow. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, and system for processing personnel data based on thermal zones, so as to at least solve the problem in the prior art of being unable to better judge passenger flow conditions due to focusing only on a single thermal zone.

[0005] According to one aspect of the present application, a personnel data processing method based on hot zones is provided, including: determining whether a person appears within the shooting range of a camera, wherein the shooting range of the camera is divided into multiple areas, each area being a hot zone; obtaining a movement trajectory of the person within the shooting range from a video captured by the camera; and obtaining cross-zone behavior of the person appearing in the movement trajectory, wherein the cross-zone behavior is the behavior of the person moving from one hot zone to another.

[0006] Furthermore, after obtaining the cross-zone behavior of the personnel appearing in the movement trajectory, the method also includes: saving the cross-zone behavior; receiving statistical conditions, and searching for cross-zone behaviors corresponding to the statistical conditions in all saved cross-zone behaviors of the personnel; and performing statistics on the cross-zone behaviors corresponding to the statistical conditions to obtain crowd flow data between hot zones.

[0007] Furthermore, it also includes: displaying the shooting range of the camera through a user interface; receiving an operation of dividing the shooting range of the camera by a user through the user interface; and dividing the shooting range of the camera into multiple hot zones according to the received operation.

[0008] Furthermore, displaying the shooting range of the camera through the user interface includes: displaying a real-scene image within the shooting range of the camera in the first part of the user interface, and displaying a plan view of the scene within the shooting range corresponding to the real-scene image in the second part of the user interface; wherein the real-scene image is a real image obtained by the camera shooting the shooting range, and the plan view is a plane layout view corresponding to the real image; dividing the shooting range of the camera into the multiple hot zones according to the received operation includes: dividing the real-scene image and the plan view into the multiple hot zones according to the division operation, wherein the real-scene image and the hot zones in the plan view correspond one to one.

[0009] Furthermore, receiving the operation of the user to divide the shooting range of the camera through the user interface and dividing the multiple hot zones according to the operation includes: receiving the operation input by the user on the plan view, and dividing the plan view into multiple hot zones according to the operation; converting the coordinates of the multiple hot zones in the plan view into coordinates in the real view according to the coordinate relationship between the plan view and the real view, to obtain multiple hot zones in the real view; or, receiving the operation input by the user on the real view, and dividing the real view into multiple hot zones according to the operation; converting the coordinates of the multiple hot zones in the real view into coordinates in the plan view according to the coordinate relationship between the plan view and the real view, to obtain multiple hot zones in the plan view; or, receiving a first division operation input by the user on the plan view and a second division operation input by the user on the real view; dividing the plan view into multiple hot zones according to the first division operation, and dividing the real view into hot zones corresponding one to one to the multiple hot zones on the plan view according to the second division operation.

[0010] Furthermore, obtaining the cross-zone behavior of the person appearing in the movement trajectory includes: the movement trajectory obtained from the video of the camera is the movement trajectory of the person in the real-scene image, and the movement trajectory of the movement in the real-scene image in the plan view is obtained according to the coordinate correspondence between the real-scene image and the plan view; and obtaining the cross-zone behavior of the person appearing according to the movement trajectory of the person in the plan view.

[0011] Further, in the case where the thermal zone includes multiple sub-thermal zones, the cross-zone behavior includes at least one of the following: moving from a sub-thermal zone of one thermal zone to another thermal zone, moving from a sub-thermal zone of one thermal zone to a sub-thermal zone of the other thermal zone, and moving from a sub-thermal zone of one thermal zone to another sub-thermal zone of the same thermal zone.

[0012] According to another aspect of the present application, a personnel data processing device based on hot zones is also provided, including: a determination module for determining whether a person appears within the shooting range of a camera, wherein the shooting range of the camera is divided into multiple areas, each area being a hot zone; an acquisition module for acquiring the movement trajectory of the person within the shooting range from the video captured by the camera; and a second acquisition module for acquiring the cross-zone behavior of the person appearing in the movement trajectory, wherein the cross-zone behavior is the behavior of the person moving from one hot zone to another.

[0013] Furthermore, it also includes: a saving module, which is used to save the cross-zone behavior; a search module, which is used to receive statistical conditions and search for cross-zone behaviors corresponding to the statistical conditions in the saved cross-zone behaviors of all personnel; a statistical module, which is used to count the cross-zone behaviors corresponding to the statistical conditions to obtain the flow of people data between hot zones.

[0014] Furthermore, it also includes: a division module, which is used to display the shooting range of the camera through a user interface; receive the user's operation of dividing the shooting range of the camera through the user interface; and divide the shooting range of the camera into multiple hot zones according to the received operation.

[0015] Furthermore, the division module is used to: display a real scene image within the shooting range of the camera in the first part of the user interface, and display a plan view of the scene within the shooting range corresponding to the real scene image in the second part of the user interface; wherein the real scene image is a real image obtained by the camera shooting the shooting range, and the plan view is a plane layout view corresponding to the real image; dividing the shooting range of the camera into the multiple hot zones according to the received operation includes: dividing the real scene image and the plan view into the multiple hot zones according to the division operation, wherein the real scene image and the hot zones in the plan view correspond one to one.

[0016] Further, the division module is used to: receive the operation input by the user on the plan view, and divide the plan view into multiple hot zones according to the operation; convert the coordinates of the multiple hot zones in the plan view into coordinates in the real view according to the coordinate relationship between the plan view and the real view, to obtain multiple hot zones in the real view; or, receive the operation input by the user on the real view, and divide the real view into multiple hot zones according to the operation; convert the coordinates of the multiple hot zones in the real view into coordinates in the plan view according to the coordinate relationship between the plan view and the real view, to obtain multiple hot zones in the plan view; or, receive a first division operation input by the user on the plan view and a second division operation input by the user on the real view; divide the plan view into multiple hot zones according to the first division operation, and divide the real view into hot zones corresponding one to one to the multiple hot zones on the plan view according to the second division operation.

[0017] Furthermore, the second acquisition module is used to: obtain the movement trajectory obtained from the video of the camera as the movement trajectory of the person in the real-scene image, and obtain the movement trajectory of the movement trajectory in the real-scene image in the plan view according to the coordinate correspondence between the real-scene image and the plan view; obtain the cross-area behavior of the person according to the movement trajectory of the person in the plan view.

[0018] Further, in the case where the thermal zone includes multiple sub-thermal zones, the cross-zone behavior includes at least one of the following: moving from a sub-thermal zone of one thermal zone to another thermal zone, moving from a sub-thermal zone of one thermal zone to a sub-thermal zone of the other thermal zone, and moving from a sub-thermal zone of one thermal zone to another sub-thermal zone of the same thermal zone.

[0019] According to another aspect of the present application, an electronic device is provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory is used to store a computer program; and the processor is used to implement the above-mentioned method steps when executing the computer program stored in the memory.

[0020] According to another aspect of the present application, a personnel data processing system based on a hot zone is also provided, comprising: a camera and the above-mentioned electronic device.

[0021] In an embodiment of the present application, a method is adopted to determine whether a person appears within the shooting range of a camera, wherein the shooting range of the camera is divided into multiple areas, each area being a hot zone; the movement trajectory of the person within the shooting range is obtained from the video captured by the camera; and the cross-area behavior of the person appearing in the movement trajectory is obtained, wherein the cross-area behavior is the behavior of the person moving from one hot zone to another. This application solves the problem in the prior art that only focuses on a single thermal area and is unable to better judge the passenger flow situation, thereby being able to obtain the flow of people between hot zones, providing data support for more accurate judgment of passenger flow situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0023] Figure 1 is a flow chart of a personnel data processing method based on hot zones according to an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of the hot zone division according to an embodiment of the present application;

[0025] Figure 3 is a flow chart of trend analysis according to an embodiment of the present application;

[0026] Figure 4 2 is a schematic diagram of the effect of trend analysis according to an embodiment of the present application;

[0027] Figure 5 It is a schematic diagram of the process of analyzing and reporting thermal data of sub-thermal zones according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0029] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] In the following embodiments, hotspots and pedestrian flows are processed using content captured by a camera. The camera can be a webcam, which can send captured content to a computing device for processing. In one scenario, if the camera has sufficient computing power, some of the steps in the following embodiments can be performed directly by the camera. Considering that the camera's primary function is to capture images, all related processing or calculations can be offloaded to the computing device. The computing device can be an edge computing device, physically deployed at the same location as the camera, connected to it via a local area network. Alternatively, the computing device can be located at a different location from the camera, connected to the computing device via a network. Alternatively, the computing device can be understood as a system comprising a terminal device and a server, with the terminal device connected to the camera and the server connected to the terminal device. If the terminal device has sufficient computing power, the steps in the following embodiments can be performed by the terminal device. If the terminal device has limited computing power, some steps in the following embodiments can be performed on the terminal device, while other steps can be offloaded to a service on the server. As long as the steps in the following embodiments can be executed, they will achieve the corresponding technical effect, regardless of the hardware combination used for execution.

[0031] Considering that the existing technology can obtain the changes in the flow of people data in a single hot zone, this processing method based on the flow of people data in a single hot zone cannot reflect the connection between hot zones. Based on this, a method for processing personnel data based on hot zones is provided in this embodiment. Figure 1 is a flowchart of a personnel data processing method based on hot zones according to an embodiment of the present application. Figure 1 The steps shown in the figure can obtain the flow data between hot zones, which reflects the mutual connection between the changes in the flow data between hot zones. Figure 1 The steps involved are described.

[0032] Step S102: determining whether a human body appears within the shooting range of the camera, wherein the shooting range of the camera is divided into multiple areas, each area being a hot zone;

[0033] Step S104, obtaining the movement trajectory of the human body within the shooting range from the video captured by the camera;

[0034] Step S106 , obtaining the cross-zone behavior of the human body appearing in the movement trajectory, wherein the cross-zone behavior is the behavior of the human body moving from one hot zone to another hot zone.

[0035] In the above steps, as long as a person appears, the person's movement trajectory can be obtained, and the cross-zone behavior of the person moving across hot zones can be obtained from the movement trajectory. In the above steps, the shooting range of a single camera is divided into multiple hot zones. In this way, through the above steps, the cross-zone behavior of people within the shooting range of a single camera can be obtained. This solves the problem of the existing technology that only focuses on a single thermal area and cannot better judge the passenger flow situation, thereby obtaining the flow of people between hot zones.

[0036] In order to facilitate users to view cross-zone behaviors, in an optional implementation, the cross-zone behaviors can also be saved; then, statistical conditions are received, and cross-zone behaviors corresponding to the statistical conditions are searched in the saved cross-zone behaviors of all personnel; statistics are collected on the cross-zone behaviors corresponding to the statistical conditions to obtain the flow of people data between hot zones.

[0037] This optional implementation method can record the person's movement trajectory, obtain the person's cross-zone behavior from the movement trajectory, and then obtain cross-zone crowd flow data based on statistical conditions. In this optional method, when the crowd flow data of a hot zone changes, not only can the change in the crowd flow data of the hot zone be known, but also which hot zone the person in the hot zone has moved to, and which hot zone the person in the hot zone came from. In this way, each independent hot zone is linked, and the flow of people between hot zones can be obtained, providing data support for more accurate judgment of passenger flow.

[0038] In the above steps, the movement trajectory of the person is identified. Considering that the above hot zones are hot zones obtained by dividing a camera, the movement trajectory of the person can be obtained by identifying the body of the person. Alternatively, the movement trajectory of the person can be identified by recognizing the face, but considering that when performing face recognition, the same camera can only recognize faces in a certain direction, and faces in other directions require the deployment of new cameras; in order to solve this problem, multiple cameras can be deployed at different angles of a scene, and the field of view of each of these cameras is divided into multiple hot zones that are the same. That is, the field of view of each camera is divided into multiple hot zones, and the hot zones of each camera are corresponding. Face recognition is performed by multiple cameras to obtain the movement trajectory of the person. Alternatively, it is also possible to consider combining the face with the body to identify the trajectory of the person, which will not be repeated here.

[0039] In practical applications, the recognition of personnel trajectories based on the human body is more efficient. Therefore, the human body is used as an example for explanation below.

[0040] This embodiment can be applied in places where people flow data needs to be obtained (for example, stores, etc.). Stores usually install a camera for monitoring. In the above steps, the shooting range of the camera can be divided into multiple hot zones. When the camera captures a person entering a hot zone (for example, entering a hot zone from outside the shooting range), the movement trajectory of the person can be obtained through the video captured by the camera. In order to facilitate subsequent data processing, as long as the movement trajectory of the person crosses zones, the cross-zone situation is recorded. These data are recorded and saved, providing data support for subsequent processing.

[0041] To facilitate users in viewing cross-zone pedestrian flow data within a particular hot zone, a user interface can be provided. This user interface can be accessed through a browser or provided by a program installed on a computer. For user convenience, an application installed on a mobile terminal can also be provided to provide the user interface. This user interface can be used to input statistical conditions. For example, multiple pre-defined hot zones can be displayed within the user interface. A user-selected hot zone can be received through the user interface. For example, if a user selects the first hot zone, all cross-zone behaviors occurring within the first hot zone are counted and the statistical results are displayed. For another example, a time period can be input within the user interface. If a user selects the first hot zone and the past day as the time period, all cross-zone behaviors occurring within the first hot zone within that day are counted and the statistical results are displayed. Cross-zone behaviors occurring within the first hot zone can be of two types: movement from the first hot zone to another hot zone, and movement from another hot zone to the first hot zone. When querying, if a user is interested in the flow of people within a particular hot zone (e.g., the first hot zone), they can select that hot zone to view the data.

[0042] The division of hot zones can be performed automatically. For example, a division rule can be pre-configured, and the division rule is based on the identified objects placed within the shooting range. After the camera captures the area within the shooting range, it divides the shooting range into multiple hot zones according to the pre-configured division rule.

[0043] While this automatic hot zone division method is convenient, it may result in the divided hot zones not meeting user needs. To address this issue, in an optional embodiment, multiple hot zones divided by the user can be received through a user interface. In this optional embodiment, the camera's shooting range can be displayed through the user interface; an operation of dividing the camera's shooting range through the user interface is received; and the camera's shooting range is divided into multiple hot zones based on the received operation. This division method provides users with maximum flexibility, allowing users to divide hot zones according to their needs.

[0044] Alternatively, the two aforementioned division methods can be combined. First, the shooting range can be divided into multiple hot zones according to pre-configured division rules. Then, the divided hot zones can be displayed to the user through the aforementioned user interface, and the user can adjust any unsatisfactory hot zones through the user interface. This method can improve the efficiency of hot zone division while giving the user room for self-division, thereby improving the user experience.

[0045] The camera's shooting range can be displayed in two ways: one is to display a real image, which is called a real scene map, and the other is to display the camera's shooting range as a plan view. Specifically, the real scene map is the real image obtained by the camera capturing the shooting range, and the plan view is the floor plan corresponding to the real image. The user interface can display a real scene map for users to divide hot zones, or a plan view can be displayed for users to divide hot zones. Operating on the real scene map is closer to the real environment, while operating on the plan view is more conducive to drawing hot zones.

[0046] In order to take into account the advantages of both the plan view and the real-scene view, in an optional embodiment, the plan view and the real-scene view can be displayed simultaneously on the user interface, that is, displaying the shooting range of the camera through the user interface includes: displaying the real-scene view within the shooting range of the camera in the first part of the user interface, and displaying the plan view of the scene within the shooting range corresponding to the real-scene view in the second part of the user interface; the hot zones in the plan view and the real-scene view are the same, that is, the real-scene view and the plan view are divided into the multiple hot zones according to the division operation, wherein the hot zones in the real-scene view and the plan view correspond one to one.

[0047] There is a correspondence between the floor plan and the real-life map, and this correspondence can be established through modeling. As explained above, if the camera has a certain computing power, this modeling can be performed through the camera; if the camera's computing power is insufficient, a computing device can be provided to perform the functions of acquiring the movement trajectory in the above steps, acquiring and recording cross-zone behaviors, and performing statistics. In addition, the computing device is also used to provide various user interfaces. In an optional embodiment, the computing device may include two parts, one part is used to divide the hot zones and analyze the video of the camera, etc., which is also called an AI thinking box; the other part is a service provided by the server, which can be used to save cross-zone behaviors, perform statistics and return statistical results, etc., that is, the service can provide various statistical services, which is called passenger flow thermal service. The AI ​​thinking box can be understood as a device that provides computing services. Taking into account the limitations of the local storage data and computing power of the device, a passenger flow thermal service is provided to cooperate with it. The passenger flow mentioned here can be understood as human flow.

[0048] The AI ​​thinking box is used to analyze and report passenger flow data (especially cross-zone human behavior). The passenger flow heat map service summarizes and classifies the reported data and displays it intuitively in a graphical form by hot zone.

[0049] After using the AI ​​Thinking Box, the AI ​​Thinking Box establishes a mapping relationship between hot zones by modeling the base map (i.e., the floor plan) and the real scene map (here, the process of establishing a coordinate system between the base map and the real scene map is called modeling, and then the mapping relationship between hot zones is established through the coordinate system of the base map and the real scene map). Each base map hot zone corresponds to a real scene map hot zone. That is, the hot zone in this embodiment is a virtual hot zone, which includes the base map hot zone and the real scene map hot zone.

[0050] There are many ways to establish a coordinate correspondence between a plan view and a real-life view. For example, a predetermined number of points can be selected in the plan view, and then each of the predetermined number of points can be marked in the real-life view to obtain the coordinates of each point in the plan view and the coordinates of each point in the real-life view. The coordinate relationship between the plan view and the real-life view can be established based on the correspondence between each point in the plan view and the real-life view.

[0051] After establishing the coordinate correspondence between the plan view and the real view, the user can choose to divide the hot zones on the real view or on the plan view through the user interface, which is more convenient for the user. That is, the operation input by the user on the plan view can be received, and the plan view can be divided into multiple hot zones according to the operation; according to the coordinate relationship between the plan view and the real view, the coordinates of the multiple hot zones in the plan view are converted into coordinates in the real view to obtain multiple hot zones in the real view; or, the operation input by the user on the real view can be received, and the real view can be divided into multiple hot zones according to the operation; according to the coordinate relationship between the plan view and the real view, the coordinates of the multiple hot zones in the real view are converted into coordinates in the plan view to obtain multiple hot zones in the plan view.

[0052] As another way to establish the correspondence between the coordinates of the plan view and the coordinates of the real scene, multiple hot spots can be drawn on the plan view, and then the multiple hot spots can be drawn on the real scene. The hot spots on the real scene and the plan view are one-to-one corresponding, and then a coordinate correspondence is established based on the coordinates of each hot spot on the plan view and the coordinates of each hot spot on the real scene.

[0053] The difference between the above two methods of establishing coordinate correspondence is that the first method draws the hot zone after establishing the coordinate correspondence; the second method performs coordinate correspondence based on the hot zone drawn on the real scene map and the plan map.

[0054] After the hot zone division, an analysis task can also be issued in the user interface. The analysis task is used to start the AI ​​thinking box to work, that is, after receiving the analysis task, the AI ​​thinking box starts to obtain the movement trajectory of the human body within the shooting range from the video captured by the camera, and then obtains the cross-zone behavior of the human body appearing in the movement trajectory, and saves the cross-zone behavior (the saving here can be local saving or uploaded to the server for saving, such as uploading to the passenger flow heat map service for saving). In the user interface, you can also choose to stop the analysis task, so that the AI ​​thinking box stops analyzing the cross-zone behavior. This analysis task can also be called a trend analysis task. The trend analysis here refers to analyzing the direction of passenger flow and the number of people to reflect the correlation between each hot zone.

[0055] Figure 2 This is a schematic diagram of the hot zone division according to the embodiment of the present application, and the trend analysis task configuration process, Figure 2 A user interface for hot zone configuration (also called hot zone configuration interface dialog box) is provided in the text. The user interface can receive user operations and divide different hot zones according to the user operations. Figure 2 Explain this.

[0056] (1) Open the hot zone configuration interface dialog box. The left side of the dialog box displays the base map, and the right side displays the real scene. Figure 2 The interface shown in the figure displays the base map (i.e., the plan view) and the real scene (i.e., the real scene image) on the left and right. The base map and the real scene can also be displayed on the top and bottom of the user interface as needed.

[0057] (2) The function of drawing wall lines can be provided in the hot zone configuration interface dialog box. This function can draw the wall part in the map screen, so as to provide assistance for drawing hot zones. One of the more important functions of the hot zone configuration interface dialog box is drawing hot zones. For drawing hot zones, tools for drawing shapes (for example, rectangle drawing tools) can be provided to draw hot zones, and tools for drawing lines can also be provided to draw hot zones. When using the line drawing tools, the drawn lines are connected into a predetermined shape, and the area surrounded by the shape is the hot zone. Figure 2 In the example, a rectangle drawing tool is used to draw hot zone A in the left base map. In the right real scene, the corresponding hot zone A is drawn. The method of drawing other hot zones is similar. Figure 2 In the hot zone configuration interface dialog box shown, five hot zones are drawn, namely hot zone A, hot zone B, hot zone C, hot zone D and hot zone E.

[0058] (3) After the hot zone is drawn, an analysis task is issued, and hot zone A and other hot zones in the base map and hot zone A and other hot zones in the real scene are sent to the AI ​​thinking box device. The AI ​​thinking box models the image and establishes a coordinate mapping relationship between the base map hot zone and the real scene hot zone.

[0059] (4) After the AI ​​thinking box detects a person in the hot zone of the real-life image, it converts the coordinates of the person in the real-life image into coordinates in the base map, and reports the thermal event (the thermal event may carry the movement trajectory of the human body) to the passenger flow thermal map service program. It should be noted that in order to facilitate the statistics of thermal events, what is reported in this step are the thermal events that appear in the base map. Therefore, it is necessary to convert the coordinates of the movement trajectory of the person in the real-life image into coordinates in the base map. Of course, as another optional method, it is also possible to report thermal events in the real-life image. In this optional method, the AI ​​thinking box can directly report the thermal events in the real-life image, and there is no need to convert the coordinates in the real-life image into coordinates in the base map.

[0060] In another embodiment, a coordinate correspondence between the base map image and the real-life image can be pre-established, and then the hotspots in the real-life image can be generated based on the hotspots drawn by the user on the base map image. For example, hotspot A can be drawn on the base map image, and then the corresponding hotspot A can be automatically drawn on the real-life image based on the pre-established coordinate correspondence. This embodiment is described below.

[0061] (1) Open the hot zone configuration interface dialog box, which displays the base map.

[0062] (2) Draw hot zone A, hot zone B and other hot zones on the base map.

[0063] (3) After the hot zone is drawn, the analysis task is issued, and the hot zone A and other hot zones in the base map are sent to the AI ​​thinking box. The AI ​​thinking box calculates the coordinates of the hot zones in the corresponding real scene and establishes a mapping relationship.

[0064] (4) After the AI ​​thinking box detects a person in the hot zone of the real-scene image, it converts the coordinates of the person in the real-scene image into the coordinates in the base map and reports the thermal event to the passenger flow thermal map service program.

[0065] In the above two methods, the user can draw the hot zone in the real scene image and the plan view respectively, and then establish the coordinate correspondence between the real scene image and the plan view based on the correspondence of the drawn hot zones. Alternatively, the coordinate correspondence between the plan view and the real scene image can be established in advance, and then the hot zone in the real scene image can be calculated based on the hot zone drawn by the user in the plan view. In another embodiment, the user can also draw the hot zone in the real scene image, and then calculate the hot zone in the plan view based on the coordinate correspondence between the plan view and the real scene image.

[0066] After obtaining the hot zone in the above manner, the cross-zone behavior can be recorded, and after accumulating a certain amount of data, the data can be counted. The statistical conditions for data statistics can be input by the user. For example, the user inputs a time period and a hot zone identifier, and the statistical conditions are generated based on the time period and hot zone identifier input by the user. The statistical conditions are used to query the cross-zone behavior that occurs in the hot zone corresponding to the hot zone identifier within the time period. At this time, the hot zone identifier and time period carried in the statistical conditions can be obtained; among all the cross-zone behaviors that occur in the movement trajectory of the human body, the cross-zone behavior that occurs in the hot zone corresponding to the hot zone identifier within the time period is searched. After the relevant cross-zone behavior is found, the cross-zone behavior found is counted to obtain the crowd flow data of the hot zone.

[0067] Alternatively, templates can be provided that store preconfigured statistical conditions. For example, Template 1 displays the traffic data for all hot zones within the past week by default, while Template 2 displays the traffic data for all traffic flows from hot zone A to hot zone B within the past month by default. When using templates, users don't need to enter statistical conditions each time; they can simply select the corresponding template to obtain the required statistics. After entering statistical conditions, users can also save them as templates for easy access next time.

[0068] The above analysis of passenger flow data between hot zones can also be called trend analysis. The so-called trend analysis can be understood as analyzing the direction and number of passenger flow, thereby reflecting the correlation between various hot zones. Figure 3 This is a flow chart of the trend analysis according to the embodiment of the present application. Figure 3 This section describes an example of trend analysis, in which the AI ​​Thinking Box and customer flow heat map services are used.

[0069] like Figure 3 As shown, the trend analysis process is as follows:

[0070] (1) After the user configures the regional heat analysis task of a single camera, he submits the analysis task to the AI ​​thinking box.

[0071] (2) The AI ​​thinking box takes the code stream from the camera for analysis, models the human body of the people appearing in the hot zone, and records the trajectory of the human body.

[0072] (3) The AI ​​thinking box analyzes the trajectory data of the human body in the passenger flow and reports the human body trajectory data across hot zones in the form of events.

[0073] (4) The passenger flow heat map service receives the cross-hot zone data reported by the AI ​​thinking box and stores it in the database. In this step, the passenger flow heat map service will determine whether the reported data is cross-hot zone data, and then store the cross-hot zone data and other data separately.

[0074] (5) The passenger flow heat map service collects statistics on data across hot zones based on time periods.

[0075] (6) Calculate and display the trend data between each hot zone. Figure 4 This is a schematic diagram of the effect of trend analysis according to an embodiment of the present application. Figure 4 You can select the time period for which you want to count the trend data. After selecting the time period, you can also select the hot zone for which you want to count. Figure 4 Select hot zone C and select the time period 00:00:00-23:59:59. After selection, the statistical data can be displayed. Figure 4 In the data shown, the number of people flowing from hot zone C to hot zone E is 1, the number of people flowing from hot zone C to hot zone B is 13, and the number of people flowing from hot zone C to hot zone D is 36. Figure 4 What is displayed is the number of people flowing out of the hot zone C. Optionally, the number of people flowing into the hot zone C during the time period can also be displayed. The display method is similar and will not be described in detail here.

[0076] In an optional embodiment, a hot zone may be further divided into sub-hot zones. For example, if a hot zone is relatively large, the hot zone may be divided into multiple sub-hot zones in order to more accurately calculate passenger flow within the hot zone. When a hot zone includes multiple sub-hot zones, the cross-zone behavior includes at least one of the following: moving from a sub-hot zone of a hot zone to another hot zone, moving from a sub-hot zone of a hot zone to a sub-hot zone of another hot zone, or moving from a sub-hot zone of a hot zone to another sub-hot zone of the same hot zone.

[0077] Figure 5 This is a flow chart of the analysis and reporting of sub-thermal zone thermal data according to the embodiment of the present application. Figure 5 Explain the process of trend analysis between sub-hot zones. Figure 5 The process shown in includes the following steps:

[0078] (1) After the user configures the regional heat analysis task of a single camera, he submits the analysis task to the AI ​​thinking box.

[0079] (2) The AI ​​thinking box takes the code stream from the camera for analysis, models the human body of the people appearing in the hot zone, and records the trajectory of the human body.

[0080] (3) The AI ​​thinking box analyzes the trajectory of the human body and reports the human body data appearing in the sub-thermal zone, that is, thermal data, in the form of events.

[0081] (4) The passenger flow heat map service receives the sub-hot zone thermal data reported by the AI ​​thinking box and stores it in the database. In this step, the passenger flow heat map service will store the sub-hot zone thermal data and other data separately.

[0082] (5) The passenger flow heat map service collects statistics on the thermal data of sub-hot zones according to time periods.

[0083] (6) Calculate the thermal data of the neutron hot zone in each hot zone and display them.

[0084] The division of sub-hot zones allows users to understand the passenger flow in smaller areas. For example, if there is a row of seats in a certain hot zone, this row of seats can be further subdivided to form sub-hot zones. Through thermal analysis, the seat thermal data can be analyzed, allowing merchants to understand which seats have more traffic and which have less traffic, so that merchants can further analyze the reasons for the high and low traffic, providing a reference for merchants to improve their operating efficiency.

[0085] Through the above implementation, the hot zone trend analysis of thermal data within the visual field of a single camera is supported, that is, the thermal trends between specific hot zones can be analyzed, so that users can understand the actual trends between hot zones. For example, in a certain time period, how many people go from hot zone A to hot zone B, and how many go from hot zone B to hot zone C. These flow data actually reflect the connection between hot zones. Understanding these trend analysis data is of great guiding significance for businesses to deeply understand customer needs.

[0086] In this embodiment, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and the processor is configured to run the computer program to execute the method in the above embodiment.

[0087] Alternatively, an electronic device may be provided, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory is used to store computer programs; and the processor is used to implement the above-mentioned method steps when executing the computer program stored in the memory.

[0088] In this embodiment, a personnel data processing system based on hot zones may also be provided, including: a camera and the above-mentioned electronic device.

[0089] The above program can be executed in a processor or stored in a memory (or computer-readable medium), which includes permanent and non-permanent, removable and non-removable media that can implement information storage by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape, disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0090] These computer programs can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps of the functions specified in one or more blocks can be implemented by different modules corresponding to different steps.

[0091] This embodiment provides such a device. This device is called a personnel data processing device based on hot zones, and includes: a determination module for determining whether a person appears within the shooting range of a camera, wherein the shooting range of the camera is divided into multiple areas, each area being a hot zone; a first acquisition module for acquiring the movement trajectory of the person within the shooting range from the video captured by the camera; and a second acquisition module for acquiring the cross-zone behavior of the person appearing in the movement trajectory, wherein the cross-zone behavior is the behavior of the person moving from one hot zone to another.

[0092] The system or device is used to implement the functions of the method in the above-mentioned embodiment. Each module in the system or device corresponds to each step in the method, which has been explained in the method and will not be repeated here.

[0093] For example, it also includes: a saving module for saving the cross-zone behavior; a search module for receiving statistical conditions and searching for cross-zone behaviors corresponding to the statistical conditions in the saved cross-zone behaviors of all personnel; a statistical module for performing statistics on the cross-zone behaviors corresponding to the statistical conditions to obtain the flow of people data between hot zones.

[0094] For example, it also includes: a division module, which is used to display the shooting range of the camera through a user interface; receive the user's operation of dividing the shooting range of the camera through the user interface; and divide the shooting range of the camera into multiple hot zones according to the received operation.

[0095] Optionally, the division module is used to: display a real scene image within the shooting range of the camera in the first part of the user interface, and display a plan view of the scene within the shooting range corresponding to the real scene image in the second part of the user interface; wherein the real scene image is a real image obtained by the camera shooting the shooting range, and the plan view is a plane layout view corresponding to the real image; dividing the shooting range of the camera into the multiple hot zones according to the received operation includes: dividing the real scene image and the plan view into the multiple hot zones according to the division operation, wherein the real scene image and the hot zones in the plan view correspond one to one.

[0096] Optionally, the division module is used to: receive the operation input by the user on the plan view, and divide the plan view into multiple hot zones according to the operation; convert the coordinates of the multiple hot zones in the plan view into coordinates in the real view according to the coordinate relationship between the plan view and the real view, to obtain multiple hot zones in the real view; or, receive the operation input by the user on the real view, and divide the real view into multiple hot zones according to the operation; convert the coordinates of the multiple hot zones in the real view into coordinates in the plan view according to the coordinate relationship between the plan view and the real view, to obtain multiple hot zones in the plan view; or, receive a first division operation input by the user on the plan view and a second division operation input by the user on the real view; divide the plan view into multiple hot zones according to the first division operation, and divide the real view into hot zones corresponding one to one to the multiple hot zones on the plan view according to the second division operation.

[0097] For another example, the second acquisition module is used to: obtain the movement trajectory obtained from the video of the camera as the movement trajectory of the person in the real-scene image, obtain the movement trajectory of the movement trajectory in the real-scene image in the plan view according to the coordinate correspondence between the real-scene image and the plan view; obtain the cross-area behavior of the person according to the movement trajectory of the person in the plan view.

[0098] Optionally, it also includes: an establishment module for obtaining the coordinates of each hot zone in the real scene image and the coordinates of each hot zone in the plan view, and establishing a coordinate correspondence between the real scene image and the plan view according to the correspondence between the hot zones in the real scene image and the plan view; or, marking multiple points in the real scene image and the plan view, wherein the points marked in the real scene image and the plan view correspond one to one in position, and establishing a coordinate correspondence between the real scene image and the plan view according to the coordinates of the multiple points marked in the real scene image and the plan view.

[0099] For another example, the search module is used to: obtain the hot zone identifier and time period carried in the statistical condition; and search for the cross-zone behavior that occurs in the hot zone corresponding to the hot zone identifier within the time period among all cross-zone behaviors of all personnel in their movement trajectories.

[0100] Optionally, in the case where the thermal zone includes multiple sub-thermal zones, the cross-zone behavior includes at least one of the following: moving from a sub-thermal zone of one thermal zone to another thermal zone, moving from a sub-thermal zone of one thermal zone to a sub-thermal zone of the other thermal zone, and moving from a sub-thermal zone of one thermal zone to another sub-thermal zone of the same thermal zone.

[0101] The above embodiment solves the problem in the prior art that passenger flow cannot be better judged due to focusing only on a single thermal zone, thereby obtaining the flow of people between thermal zones, providing data support for more accurate judgment of passenger flow.

[0102] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A personnel data processing method based on hot zones, characterized in that: include: Determining whether a person appears within a shooting range of a camera, wherein the shooting range of the camera is divided into a plurality of areas, each area being a hot zone; Obtaining a movement trajectory of the person within the shooting range from the video captured by the camera; Acquire cross-zone behavior of the person appearing in the movement trajectory, wherein the cross-zone behavior is an action of the person moving from one hot zone to another hot zone; The method further comprises: Displaying a real scene image within the shooting range of the camera in a first portion of the user interface, and displaying a plan view of the scene within the shooting range corresponding to the real scene image in a second portion of the user interface; wherein the real scene image is a real image obtained by the camera shooting the shooting range, and the plan view is a plan layout view corresponding to the real image; receiving an operation input by a user on the plan view to divide the shooting range of the camera, and dividing the plan view into a plurality of hot zones according to the operation; converting the coordinates of the plurality of hot zones in the plan view into the coordinates of the real view according to the coordinate relationship between the plan view and the real view image, to obtain a plurality of hot zones in the real view image; the real view image and the hot zones in the plan view have a one-to-one correspondence; or, receiving an operation input by the user on the real scene image to divide the shooting range of the camera, and dividing the real scene image into a plurality of hot zones according to the operation; converting the coordinates of the plurality of hot zones in the real scene image into the coordinates in the plan view according to the coordinate relationship between the plan view and the real scene image, to obtain a plurality of hot zones in the plan view; the real scene image and the hot zones in the plan view have a one-to-one correspondence; or, Receive a first division operation input by the user on the plan view and a second division operation input by the user on the real scene view; divide the plan view into a plurality of hot zones according to the first division operation, and divide the real scene view into hot zones corresponding one-to-one to the plurality of hot zones on the plan view according to the second division operation.

2. The method according to claim 1, characterized in that After obtaining the cross-zone behavior of the person appearing in the movement trajectory, the method further includes: Saving the cross-region behavior; Receive statistical conditions, and search for cross-region behaviors corresponding to the statistical conditions in all saved cross-region behaviors of personnel; Statistics are collected on cross-zone behaviors corresponding to the statistical conditions to obtain crowd flow data within hot zones.

3. The method according to claim 1, characterized in that Obtaining the cross-zone behavior of the person appearing in the movement trajectory includes: The movement trajectory obtained from the video of the camera is the movement trajectory of the person in the real scene image, and the movement trajectory of the movement trajectory in the real scene image in the plan view is obtained according to the coordinate correspondence between the real scene image and the plan view; The cross-area behavior of the person is obtained according to the movement trajectory of the person on the plan.

4. The method according to any one of claims 1 to 3, characterized in that In the case where a hot zone includes multiple sub-hot zones, the cross-zone behavior includes at least one of the following: Move from a sub-hot zone of one hot zone to another hot zone, move from a sub-hot zone of one hot zone to a sub-hot zone of said another hot zone, move from a sub-hot zone of one hot zone to another sub-hot zone of the same hot zone.

5. A personnel data processing device based on hot zones, characterized in that: include: a determination module, configured to determine whether a person is present within a shooting range of a camera, wherein the shooting range of the camera is divided into a plurality of areas, each area being a hot zone; A first acquisition module is used to obtain the movement trajectory of the person within the shooting range from the video captured by the camera; A second acquisition module is configured to acquire cross-zone behavior of the person appearing in the movement trajectory, wherein the cross-zone behavior is the behavior of the person moving from one hot zone to another hot zone; a partitioning module, configured to: display a real scene image within the shooting range of the camera in a first portion of the user interface, and display a plan view of the scene within the shooting range corresponding to the real scene image in a second portion of the user interface; wherein the real scene image is a real image obtained by the camera shooting the shooting range, and the plan view is a plan layout view corresponding to the real image; receiving an operation input by a user on the plan view to divide the shooting range of the camera, and dividing the plan view into a plurality of hot zones according to the operation; converting the coordinates of the plurality of hot zones in the plan view into the coordinates of the real view according to the coordinate relationship between the plan view and the real view image, to obtain a plurality of hot zones in the real view image; the real view image and the hot zones in the plan view have a one-to-one correspondence; or, receiving an operation input by the user on the real scene image to divide the shooting range of the camera, and dividing the real scene image into a plurality of hot zones according to the operation; converting the coordinates of the plurality of hot zones in the real scene image into the coordinates in the plan view according to the coordinate relationship between the plan view and the real scene image, to obtain a plurality of hot zones in the plan view; the real scene image and the hot zones in the plan view have a one-to-one correspondence; or, Receive a first division operation input by the user on the plan view and a second division operation input by the user on the real scene view; divide the plan view into a plurality of hot zones according to the first division operation, and divide the real scene view into hot zones corresponding one-to-one to the plurality of hot zones on the plan view according to the second division operation.

6. The device according to claim 5, characterized in that The system further includes: a saving module for saving the cross-zone behavior; a search module for receiving a statistical condition and searching for a cross-zone behavior corresponding to the statistical condition in all the saved cross-zone behaviors of people; a statistics module for collecting statistics on the cross-zone behaviors corresponding to the statistical condition to obtain the flow of people data between hot zones; The second acquisition module is configured to: obtain the movement trajectory of the person in the real scene image from the video of the camera as the movement trajectory of the person in the real scene image, obtain the movement trajectory of the movement trajectory in the real scene image in the plan view according to the coordinate correspondence between the real scene image and the plan view; and obtain the cross-zone behavior of the person according to the movement trajectory of the person in the plan view; In the case where the thermal zone includes multiple sub-thermal zones, the cross-zone behavior includes at least one of the following: moving from a sub-thermal zone of one thermal zone to another thermal zone, moving from a sub-thermal zone of one thermal zone to a sub-thermal zone of the other thermal zone, and moving from a sub-thermal zone of one thermal zone to another sub-thermal zone of the same thermal zone.

7. An electronic device, characterized in that: comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; The memory is used to store computer programs; A processor, configured to implement the method steps described in any one of claims 1 to 4 when executing the computer program stored in the memory.

8. A personnel data processing system based on hot zones, characterized in that: include: A camera and an electronic device as claimed in claim 7.

Images