Camera scheduling method and device, equipment and storage medium

Abstract

The invention discloses a camera scheduling method and apparatus, a device and a storage medium. The method comprises the steps of obtaining positioning data of a target person in real time; judging whether the target person is currently allocated with the camera or not; if yes, scheduling each camera configured in the target area according to the positioning data and the allocated camera; if not, performing global scheduling on each camera; for each camera after scheduling, pulling monitoring video data, identifying the number of monitoring personnel, and determining the number of real personnel corresponding to the camera; if the number of the monitored persons is smaller than the number of the real persons, determining all target persons currently monitored by the camera, and selecting an optimal camera for scheduling based on all the target persons currently monitored by the camera so as to complete camera scheduling; and if the number of the monitoring personnel is not smaller than the number of the real personnel, camera scheduling is completed. According to the invention, large-range and dynamic monitoring requirements can be met, and the safety of target personnel is ensured.

Description

Technical Field

[0001] The present application relates to the technical field of camera scheduling, and in particular to a camera scheduling method, device, equipment and storage medium. Background Art

[0002] Security is a key focus in many fields. Whether it is a public place or an office place, effective monitoring of personnel activities can ensure the safety of personnel at work and in life. At present, monitoring is usually achieved through cameras installed in the place. For example, multiple cameras are arranged at fixed positions and angles in a work area to cover key work areas, thereby achieving real-time monitoring of staff.

[0003] However, whether in work scenes or life scenes, the movement of people is full of randomness. If we only rely on cameras with fixed positions and fixed angles for monitoring, it will be difficult to meet large-scale and dynamic monitoring needs, and we cannot achieve accurate monitoring of people, which is not conducive to the safety of people. Summary of the invention

[0004] In view of this, the present application provides a camera scheduling method, device, equipment and storage medium to solve the problem that the movement of people is full of randomness. If only relying on cameras with fixed positions and angles for monitoring, it is difficult to meet large-scale, dynamic monitoring needs, and it is impossible to achieve accurate monitoring of people, which is not conducive to the safety of people.

[0005] To achieve the above objectives, the proposed solution is as follows:

[0006] In a first aspect, a camera scheduling method includes:

[0007] Acquire the location data of one or more target persons in the target area in real time;

[0008] For each of the target persons, determining whether the target person has been currently assigned a camera for monitoring the target person;

[0009] If the target person has been currently assigned a camera to monitor him, scheduling each camera configured in the target area according to the positioning data of the target person and the camera to monitor him;

[0010] If the target person is not currently assigned a camera to monitor him, then all cameras configured in the target area are globally scheduled;

[0011] For each camera after scheduling, pull the surveillance video data of the camera, identify the number of monitored persons in the surveillance video data, and determine the actual number of persons corresponding to the camera according to the positioning data;

[0012] If the number of monitored persons is less than the number of real persons, all target persons currently monitored by the camera are determined, and based on all target persons currently monitored by the camera, the best camera is selected from the cameras configured in the target area for scheduling, so as to complete the camera scheduling;

[0013] If the number of monitored personnel is not less than the number of actual personnel, the camera scheduling is completed.

[0014] Preferably, the scheduling of each camera configured in the target area according to the positioning data of the target person and the camera monitoring the target person includes:

[0015] Taking the camera that monitors the object as a first camera, determining a visual domain currently monitored by the first camera;

[0016] Determine whether the location data of the target person is within the currently monitored visual domain;

[0017] If the positioning data of the target person is within the currently monitored visual area, the first camera is dispatched to continue monitoring the target person.

[0018] Preferably, it also includes:

[0019] If the positioning data of the target person is not located within the currently monitored visual area, all target persons currently monitored by the first camera are determined as first persons;

[0020] Determining location data of each of the first persons;

[0021] Extracting each visual field of the first camera from a pre-established camera configuration data set;

[0022] Determining whether the positioning data of each of the first persons and the positioning data of the target person are both located within at least one visual field of the first camera;

[0023] If yes, dispatching the first camera to continue monitoring the target person;

[0024] If not, all cameras configured in the target area are globally scheduled.

[0025] Preferably, the global scheduling of the cameras configured in the target area includes:

[0026] Based on the positioning data, determining each camera to be selected corresponding to the target person from each camera;

[0027] Extracting each visual area of ​​each of the to-be-selected cameras from a pre-established camera configuration data set;

[0028] For each visual zone of each of the to-be-selected cameras, determining whether the positioning data of the target person is located within the visual zone;

[0029] If yes, the visual domain is used as the visual domain to be selected, thereby obtaining one or more visual domains to be selected corresponding to the target person;

[0030] Determining the area of ​​each of the visual areas to be selected;

[0031] The visual domain to be selected with the largest area is used as the target visual domain corresponding to the target person;

[0032] According to the target visual area, each camera configured in the target area is globally scheduled.

[0033] Preferably, the globally scheduling of the cameras configured in the target area according to the target visual area includes:

[0034] For each of the target persons, determine the cameras associated with the target person as associated cameras, and summarize them into a camera association set;

[0035] Determine other target persons associated with the target person, and aggregate the target person and the other associated target persons into a person association set;

[0036] According to the camera association set, the person association set and the target visual area corresponding to each target person in the person association set, establishing each camera allocation combination corresponding to the target person;

[0037] Screening from each of the camera allocation combinations to determine a target combination corresponding to the target person;

[0038] Based on the target combination corresponding to each target person, each camera configured in the target area is globally scheduled.

[0039] Preferably, screening from each of the camera allocation combinations to determine a target combination includes:

[0040] Matching each of the camera allocation combinations with a pre-established personnel scheduling strategy to obtain a first score set corresponding to each of the camera allocation combinations;

[0041] Screening each of the first score sets according to a preset optimization rule to obtain a screening result;

[0042] If the screening result includes only one first score set, the first score set is used as the target combination;

[0043] If the screening result includes at least two first score sets, the at least two first score sets are matched with the pre-established camera scheduling strategies respectively to obtain corresponding second score sets respectively, and the second score sets are screened according to the preferred rule to determine the target combination.

[0044] Preferably, the selecting the best camera from the cameras configured in the target area for scheduling based on all target persons currently monitored by the camera to complete the camera scheduling includes:

[0045] All target persons currently monitored by the camera are regarded as first persons;

[0046] Selecting cameras corresponding to all the first persons from cameras configured in the target area as corresponding cameras;

[0047] Determine the current north direction, randomly select a first person from each of the first persons, use the location of the positioning data of the first person as the origin, and establish a rectangular coordinate system based on the origin and the current north direction;

[0048] Selecting a corresponding camera in each of the four quadrants of the rectangular coordinate system as each second camera;

[0049] The best camera is selected from each of the second cameras.

[0050] In a second aspect, a camera scheduling device includes:

[0051] A positioning data acquisition module is used to acquire the positioning data of one or more target persons in the target area in real time;

[0052] A camera allocation judgment module is used to judge, for each target person, whether the target person has been currently assigned a camera for monitoring the target person;

[0053] A first scheduling module is used to schedule various cameras configured in the target area according to the positioning data of the target person and the camera monitoring the target person if the target person has been currently assigned a camera monitoring the target person;

[0054] A second scheduling module is used to globally schedule various cameras configured in the target area if the target person is not currently assigned a camera for monitoring him;

[0055] A personnel number determination module is used to pull the surveillance video data of each camera after scheduling, identify the number of monitored personnel in the surveillance video data, and determine the actual number of personnel corresponding to the camera according to the positioning data;

[0056] Returning to the execution module, if the number of monitored persons is less than the number of real persons, determining all target persons currently monitored by the camera, and selecting the best camera from the cameras configured in the target area for scheduling based on all target persons currently monitored by the camera, so as to complete the camera scheduling;

[0057] The scheduling completion module is used to complete the camera scheduling if the number of monitored personnel is not less than the number of real personnel.

[0058] In a third aspect, a camera scheduling device includes a memory and a processor;

[0059] The memory is used to store programs;

[0060] The processor is used to execute the program to implement each step of the camera scheduling method as described in any one of the first aspects.

[0061] In a fourth aspect, a storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the camera scheduling method as described in any one of the first aspects are implemented.

[0062] It can be seen from the above technical solution that the present application obtains the positioning data of one or more target persons located in the target area in real time; for each of the target persons, it is determined whether the target person has been currently assigned a camera to monitor him; if the target person has been currently assigned a camera to monitor him, then according to the positioning data of the target person and the camera to monitor him, the various cameras configured in the target area are scheduled; if the target person is not currently assigned a camera to monitor him, then the various cameras configured in the target area are globally scheduled; for each camera after scheduling, the monitoring video data of the camera is pulled, the number of monitored persons in the monitoring video data is identified, and the number of real persons corresponding to the camera is determined according to the positioning data; if the number of monitored persons is less than the number of real persons, all target persons currently monitored by the camera are determined, and based on all target persons currently monitored by the camera, the optimal camera is selected from the various cameras configured in the target area for scheduling to complete the camera scheduling; if the number of monitored persons is not less than the number of real persons, the camera scheduling is completed. On the one hand, the present application obtains the positioning data of the target personnel in real time, and judges whether to allocate the camera to monitor it, so as to perform different operations. If it has been allocated, the camera is scheduled according to the allocated camera and its positioning data, so as to reduce the impact on other cameras; if it has not been allocated, the cameras in the target area are globally scheduled, and the optimal scheduling and monitoring are achieved from a global perspective. In order to ensure the accuracy of monitoring the target personnel, a judgment mechanism is set after the scheduling is completed, and the monitoring video data is pulled, and the number of monitored personnel is identified, and the number of real personnel at the positioning data level is compared. According to the comparison, it is judged whether the scheduling is completed, so as to achieve real-time monitoring and real-time scheduling, and the accuracy of monitoring will not be affected by the random movement of the target personnel, and it can meet the large-scale and dynamic monitoring needs and ensure the safety of the target personnel. On the other hand, the present application efficiently integrates multi-source data such as monitoring data and positioning data, realizes excellent camera scheduling and allocation, can maximize the coverage of target personnel, avoids monitoring blind spots and resource waste, and improves resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0064] Figure 1An optional flow chart of a camera scheduling method provided in an embodiment of the present application;

[0065] Figure 2 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0066] Figure 3 A schematic diagram of a camera orientation provided in an embodiment of the present application;

[0067] Figure 4 A camera scheduling schematic diagram provided in an embodiment of the present application;

[0068] Figure 5 A schematic diagram of the actual situation of a target area provided in an embodiment of the present application;

[0069] Figure 6 A schematic diagram of an angle provided in an embodiment of the present application;

[0070] Figure 7 A schematic diagram of the structure of a camera scheduling device provided in an embodiment of the present application;

[0071] Figure 8 A schematic diagram of the structure of a camera scheduling device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0073] Security is a key concern in many fields. Whether it is a public place or an office place, effective monitoring of personnel activities can ensure the safety of personnel at work and in life. At present, monitoring is usually achieved through cameras installed in the place. For example, multiple cameras are arranged at fixed positions and angles in a work area to cover key work areas, thereby achieving real-time monitoring of staff. However, whether in work scenes or life scenes, the movement of personnel is full of randomness, especially in the vast outdoor environment. If only fixed-position and fixed-angle cameras are used for monitoring, it is difficult to meet large-scale and dynamic monitoring needs, and accurate monitoring and tracking of personnel cannot be achieved, which is not conducive to personnel safety.

[0074] For example, in the power grid scenario, the substation is one of the core components of the power system and plays an extremely important role in ensuring the stable operation of the power grid. However, due to the presence of a large number of high-voltage equipment, complex line layouts and diverse working environments (such as outdoor work) in the substation, workers face a variety of potential safety risks, including but not limited to protection against electric shock, falling, and object impact. These safety issues not only threaten the lives of personnel, but also pose a serious challenge to the normal operation of the power system. With the development of my country's power industry and the continuous expansion of the main grid investment scale, the number of newly built and expanded substations continues to increase, which has led to increasingly heavy maintenance and inspection tasks in substations. Therefore, in order to effectively cope with the management pressure brought about by this trend and ensure the safety of the operation process, it is particularly important to establish an efficient and reliable monitoring system.

[0075] In order to solve the defects of the above-mentioned prior art, an embodiment of the present invention provides a camera scheduling method that can be used in many general or special computing device environments or configurations, such as: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor devices, distributed computing environments including any of the above devices or equipment, etc.

[0076] The method can also be applied to various computer terminals or intelligent terminals, and its execution subject can be a processor or server of the computer terminal or intelligent terminal. The method flow chart of the method is as follows: Figure 1 As shown, specifically including:

[0077] S1: Acquire the positioning data of one or more target persons in the target area in real time.

[0078] The target area may refer to a work area, a public area, etc., and is not restricted to outdoors or indoors. A range may also be pre-set as the target area, and the target personnel within the target area may be monitored. This application does not impose any restrictions on this. The target personnel refers to personnel who enter the target area and / or are originally within the target area, and may be staff, passers-by, etc.

[0079] Optionally, with the permission of the target personnel, they can be asked to wear a work hat or safety helmet with an integrated positioning terminal in advance. The positioning terminal can upload real-time positioning data to the back-end server through the firewall and switch to ensure safe and reliable data transmission. At the same time, the face recognition function can be activated to bind the identity information of the target personnel; positioning data can also be obtained in real time through thermal imaging, GPS, Beidou RTK and other technologies.

[0080] Beidou is a service provided by the independently developed Beidou Satellite Navigation System (BDS). It uses a series of technologies and algorithms to improve positioning accuracy to centimeter level or even higher. This high-precision positioning capability is very important for application scenarios that require precise positioning information, such as personnel monitoring, autonomous driving, precision agriculture, deformation monitoring, etc. For example, in a substation, high-precision positioning can be used to accurately track the location and movement of personnel or equipment, thereby optimizing asset management and service response.

[0081] RTK (Real-Time Kinematic) is the abbreviation of real-time dynamic differential positioning technology. It is a technology that uses carrier phase observations for real-time processing to obtain centimeter-level precision positions. It usually requires one or more fixed base stations and one or more mobile stations (i.e., receivers). For example, the base station sends information about the known position to the mobile station, and the mobile station calculates its own precise position based on the received signal. RTK technology can also be widely used in surveying, mapping, unmanned driving and other fields, and can provide real-time high-precision positioning data. RTK technology is used in some existing personnel monitoring and tracking solutions. Although it can accurately determine positioning data, it lacks integration with visual monitoring, and it is difficult to provide detailed information on the scene of the target area, and it is impossible to determine whether the behavior / status of the target personnel is standardized and safe. However, this application can use RTK technology to determine positioning data, and then integrate and connect with the camera to achieve excellent personnel monitoring and camera scheduling.

[0082] The number of target persons in the target area is uncertain, it may be one or more. As time goes by, more and more target persons may enter the target area, and some target persons may leave the target area. Therefore, the target persons exist in the target area in a dynamic form. Therefore, this application needs to obtain the positioning data of the target persons in real time. In addition, obtaining the positioning data can determine the position of the target persons, which is helpful for camera scheduling.

[0083] S2: For each of the target persons, determine whether the target person has been currently assigned a camera for monitoring the target person.

[0084] For the server that executes this method, before monitoring each target person, it is necessary to determine whether he or she is currently being monitored, that is, to determine whether a camera has been assigned to monitor him or her. This is because different scheduling strategies can be executed based on different judgment results, thereby achieving effective utilization of camera resources.

[0085] S3: If the target person has been currently assigned a camera for monitoring him / her, then the cameras configured in the target area are scheduled according to the positioning data of the target person and the camera for monitoring him / her.

[0086] In this step, if the person is currently being monitored, in order to rationally utilize resources, improve scheduling efficiency, avoid unnecessary scheduling, and improve the convenience of monitoring, the monitoring method or monitoring strategy may not be changed, and the assigned camera may continue to be used to monitor the target person under reasonable circumstances.

[0087] In addition, since the target person may have short-term rapid movements or be in the process of moving all the time, it is possible to determine whether the currently assigned camera can meet the monitoring needs based on the real-time positioning data. If so, continue to use the assigned camera to monitor the target person. If not, take into account the target persons and cameras around them from multiple aspects and schedule them together.

[0088] Therefore, the camera scheduling method of the present application can be changed at any time according to the real-time situation, which greatly increases the scheduling and processing efficiency for emergencies and improves the accuracy of personnel monitoring.

[0089] S4: If the target person is not currently assigned a camera to monitor him, a global scheduling is performed on each camera configured in the target area.

[0090] If the target person is not currently assigned a camera to monitor him / her, then it is necessary to consider the overall situation and reasonably dispatch the various cameras configured in the target area to monitor the target person reasonably. From the perspective of scheduling, an excellent monitoring method is implemented in the target area to cover the target person with the highest probability of monitoring.

[0091] It should be noted that the scheduling or global scheduling of cameras in this application includes which target person or persons the camera is assigned to, the orientation / angle of the camera itself, and the adjustment of the camera parameters itself to increase freedom.

[0092] S5: For each camera after scheduling, the surveillance video data of the camera is pulled, the number of monitored persons in the surveillance video data is identified, and the actual number of persons corresponding to the camera is determined according to the positioning data.

[0093] This step is to determine the accuracy of monitoring, so as to determine whether the camera scheduling method is excellent. Therefore, after scheduling, each camera is judged, its monitoring video data is pulled, and the number of people in it is identified as the number of monitored personnel. AI technology can be used to identify people in the monitoring video data, set the size of the human frame, and use the size of the human frame to filter the identified targets to determine the number of people.

[0094] At the same time, the actual number of people corresponding to the camera is determined based on the positioning data (the actual number of people is the number of all people currently within the monitoring range of the camera). This is because the target person may be in the monitoring range of the camera, but is in a stationary working state and is blocked by obstacles, or cannot be actually monitored due to backlight and other reasons, resulting in inaccurate camera scheduling, causing the number of monitored people to be different from the actual number of people. Therefore, this step judges whether the current scheduling is appropriate and accurate from the perspective of the number of people.

[0095] This can ensure effective coverage of the target personnel, making the application have strong environmental adaptability and able to meet the scheduling of cameras in complex areas and environments.

[0096] S6: If the number of monitored persons is less than the actual number of persons, all target persons currently monitored by the camera are determined, and based on all target persons currently monitored by the camera, the optimal camera is selected from the cameras configured in the target area for scheduling to complete the camera scheduling.

[0097] If the number of monitored personnel is less than the actual number of personnel, then the current scheduling of the camera is not accurate, so it needs to be adjusted. Since this step is based on the number to determine whether the scheduling is completed, the adjustment needs to focus on the target personnel currently monitored by the camera. The purpose of the adjustment is to ensure that the number of monitored personnel after adjustment is not less than the actual number of personnel. Therefore, after determining that the number of monitored personnel is less than the actual number of personnel, first determine all the target personnel currently monitored by the camera, and then select a camera as the optimal camera based on these target personnel, so that the optimal camera monitors these target personnel, and then schedules and completes it.

[0098] This can ensure full utilization of camera resources, enable more target personnel to be monitored, better understand the status of the target area, and improve the accuracy and robustness of scheduling.

[0099] S7: If the number of monitored personnel is not less than the number of real personnel, the camera scheduling is completed.

[0100] The number of monitored personnel is greater than or equal to the number of real personnel, which is the standard for completing camera scheduling set by this application, and then the scheduling is confirmed to be completed. This application can also determine the location information of the target person through the camera monitoring data, and compare the location information with the location data, so as to locate the target person more accurately.

[0101] It is understandable that the status of the target personnel is uncertain and dynamic, so after the scheduling is completed, real-time judgment, adjustment and scheduling are required to achieve accurate monitoring at all times and ensure the safety of personnel.

[0102] It can be seen from the above technical scheme that, on the one hand, the present application obtains the positioning data of the target personnel in real time, and judges whether to allocate the camera to monitor it, so as to perform different operations. If it has been allocated, the camera is scheduled according to the allocated camera and its positioning data, so as to reduce the impact on other cameras; if it has not been allocated, the cameras in the target area are globally scheduled, and the optimal scheduling and monitoring are achieved from a global perspective. In order to ensure the accuracy of monitoring the target personnel, a judgment mechanism is set after the scheduling is completed, and the monitoring video data is pulled, and the number of monitored personnel is identified, and the number of real personnel at the positioning data level is compared. According to the comparison, it is judged whether the scheduling is completed, so as to achieve real-time monitoring and real-time scheduling, and the accuracy of monitoring will not be affected by the random movement of the target personnel, and it can meet the large-scale and dynamic monitoring needs and ensure the safety of the target personnel. On the other hand, the present application efficiently integrates multi-source data such as monitoring data and positioning data, realizes excellent camera scheduling and allocation, can maximize the coverage of target personnel, avoids monitoring blind spots and waste of resources, and improves resource utilization.

[0103] An application scenario of this method is Figure 2 As shown, it includes target personnel, backend servers (including camera scheduling system and AI target recognition system), and cameras. Among them, the camera scheduling system is used to execute the above steps S1 to S4 to perform scheduling and global scheduling, and the AI ​​target recognition system is used to execute the above steps S5 to S7 to perform quantity recognition and judgment.

[0104] The scheduling or global scheduling of cameras in this solution includes the allocation of cameras to one or several target persons, the adjustment of the direction / angle of the camera itself, the adjustment of the camera parameters itself, and many other aspects.

[0105] The adjustment of the camera's own parameters involves PTZ (Pan / Tilt / Zoom), which refers to the camera's horizontal rotation (Pan), vertical tilt (Tilt) and lens zoom (Zoom) functions. These functions enable the camera to be remotely controlled, covering a wider monitoring range, and can also zoom in on specific areas for monitoring. Using the camera's PTZ parameters for monitoring can enhance security and maintenance efficiency. It can also be combined with high-precision positioning technology (such as PTK technology) to adjust the camera's PTZ parameters so that it can automatically adjust the direction and focal length to capture the target person with a better perspective. For example, the camera can be remotely operated to check the status of the device or monitor the activities of a specific person, without the need for the monitoring personnel to be present in person, saving manpower and improving monitoring accuracy.

[0106] For the Z parameter (Zoom), this application determines the horizontal field of view (HFOV), vertical field of view (VFOV) and maximum visible radius of the camera under different Z parameters, as shown in Table 1 below:

[0107] Table 1

[0108]

[0109] Among them, the horizontal field of view angle refers to the angle range that the camera can capture in the horizontal direction, the vertical field of view angle refers to the angle range that the camera can capture in the vertical direction, and the visible radius depends on the actual content of the picture captured by the camera and its intelligent recognition capabilities.

[0110] According to Table 1 above, for each camera configured in the target area, a corresponding camera configuration data set can be set. The camera configuration data set includes the PTZ parameters, horizontal field of view angle, vertical field of view angle, visual radius, camera orientation, and number of visual areas of each camera. The setting process is as follows:

[0111] Collect and calibrate the longitude and latitude of each camera installation location, and determine the north angle. For the P parameter (Pan), the range of values ​​that can be taken is 0~3600. This setting can make the angles richer. The angle obtained by dividing 3600 by 10 is more than the angle obtained by directly defining 0°~360°. For example, an angle of 359.5° can be obtained. For any value of the P parameter, the camera orientation under the P parameter can be calculated. The calculation formula is:

[0112] ,

[0113] in, refers to the camera direction. It refers to the North Corner.

[0114] The camera orientation diagram can be shown as Figure 3 As shown, in Figure 3 In the figure, the area formed by the dotted line is the visible area of ​​the camera under the current orientation. It can be seen that the boundary line of the visible area contains an arc. The schematic diagram of scheduling (adjusting the orientation) of the camera can be shown as follows Figure 4 shown.

[0115] The latitude and longitude coordinates of the camera's visible area in this direction can be calculated:

[0116] ,

[0117] ,

[0118] ,

[0119] ,

[0120] , ,

[0121] .

[0122] in, is the arc of the visible area, Indicates converting the current angle value to radians. is the current angle value; for Horizontal offset in direction, for Horizontal offset in direction, is the visible radius, is the difference in latitude, is the radius of the Earth, is the circumference of a circle, is the latitude of the camera installation location, is the longitude of the camera installation location, is the difference in longitude, is the latitude of any point on the arc, is the longitude of any point on the arc, and the longitude and latitude coordinates of any point on the arc of the visible area are obtained, so that the range of the visible area can be determined. It can be understood that the direction and the visible area are complementary to each other. Once the direction is determined, the visible area corresponding to the direction can be determined, and once the visible area is determined, the direction corresponding to the visible area can be determined.

[0123] This application does not adjust the T parameter (Tilt), which can be set in advance.

[0124] Therefore, according to the above process, the range of the visual field of a camera under a set of PTZ parameters can be determined. At the same time, considering the different horizontal field of view angles under different Z parameters, the grid density of the camera's visual field can be planned. Taking Z=1 as an example, its HFOV=55.4553°. In order to make the target person as close as possible to the middle of the camera's field of view, the horizontal field of view angle is adjusted, and HFOV=55.4553°*0.5=27.72765° is taken. 27.72765° is taken as the true value, so the grid density is HFOV / 3≈10°. Since the range of the P parameter is set to 0~3600, , that is, the step size of the P parameter is 100, so when Z=1, calculate the number of all possible visual areas of the camera for:

[0125] ;

[0126] The calculated number of visible areas is 36, that is, when Z=1 and P=100, the camera has 36 visible areas. During the camera scheduling process, these 36 visible areas can be scheduled and selected.

[0127] In summary, the camera configuration data set can be shown in Table 2 below, taking one camera as an example:

[0128] Table 2

[0129]

[0130] In the method provided in the embodiment of the present invention, the process of scheduling each camera configured in the target area according to the positioning data of the target person and the camera monitoring the target person is specifically described as follows:

[0131] Taking the camera that monitors the object as a first camera, determining a visual domain currently monitored by the first camera;

[0132] Determine whether the location data of the target person is within the currently monitored visual domain;

[0133] If the positioning data of the target person is within the currently monitored visual area, the first camera is dispatched to continue monitoring the target person.

[0134] Specifically, if the target person has been currently assigned a camera to monitor him, then under normal circumstances, the first camera has been assigned to the target person and monitors the target person within one of his visual domains. At this time, it is necessary to determine whether this situation has changed, that is, to determine whether the positioning data of the target person is within the visual domain currently monitored by the first camera. If so, it means that it is accurate and reasonable to continue to use the first camera to monitor him. Therefore, the first camera is scheduled to continue to monitor the target person, and the PTZ parameters, orientation or visual domain of the first camera do not need to be adjusted. Among them, the ray intersection method can be used to determine whether the positioning data is within the currently monitored visual domain. For example, a ray in any direction is emitted at the position corresponding to the positioning data. If the intersection point of the ray and the visual domain is an odd number, it means that the positioning data is within the visual domain. If it is an even number, it is not within the visual domain.

[0135] In addition, if the location data of the target person is not within the currently monitored visual area, the following operations are performed:

[0136] Determine all target persons currently monitored by the first camera as first persons;

[0137] Determining location data of each of the first persons;

[0138] Extracting each visual field of the first camera from a pre-established camera configuration data set;

[0139] Determining whether the positioning data of each of the first persons and the positioning data of the target person are both located within at least one visual field of the first camera;

[0140] If yes, dispatching the first camera to continue monitoring the target person;

[0141] If not, all cameras configured in the target area are globally scheduled.

[0142] Specifically, in this case, the camera that monitors the target person should be replaced, or the PTZ parameters, orientation or field of view of the camera should be changed. At the same time, it is necessary to avoid monitoring of other target persons as much as possible. Therefore, all target persons currently monitored by the first camera are first determined, which refers to all target persons within the field of view currently monitored by the first camera, and their positioning data are determined respectively. The field of view of the first camera is determined from the camera configuration data set and screened. As long as there is at least one field of view that can cover all target persons within the field of view currently monitored by the first camera and the target person, it means that the first camera can meet the requirements of accurate monitoring of these target persons. Therefore, in this case, there is no need to replace the first camera. You only need to adjust the field of view of the first camera to the field of view that has just been screened out. This will not affect these target persons or other cameras, and efficiently realize camera scheduling.

[0143] If there is no camera that can cover all target persons within the visual area currently monitored by the first camera and the visual area of ​​the target person, other cameras need to be dispatched. However, other cameras also involve the corresponding monitored target persons. Therefore, in this case, it is necessary to globally dispatch all cameras configured in the target area to achieve optimal scheduling and monitoring from an overall and global perspective.

[0144] The following embodiment describes in detail the process of globally scheduling each camera configured in the target area in the present application.

[0145] Based on the positioning data, determining each camera to be selected corresponding to the target person from each camera;

[0146] Extracting each visual area of ​​each of the to-be-selected cameras from a pre-established camera configuration data set;

[0147] For each visual zone of each of the to-be-selected cameras, determining whether the positioning data of the target person is located within the visual zone;

[0148] If yes, the visual domain is used as the visual domain to be selected, thereby obtaining one or more visual domains to be selected corresponding to the target person;

[0149] Determining the area of ​​each of the visual areas to be selected;

[0150] The visual domain to be selected with the largest area is used as the target visual domain corresponding to the target person;

[0151] According to the target visual area, each camera configured in the target area is globally scheduled.

[0152] Specifically, the global scheduling of each camera configured in the target area may involve all cameras or some cameras, and the specific situation requires specific analysis. First, based on the positioning data, determine the camera to be selected corresponding to the target person. This can be based on whether the positioning data of the target person is located in at least one visual field of the camera as a standard. For example, if the positioning data of a target person is located in a visual field of a camera, then the camera can be used as the camera to be selected for the target person; the visual radius can also be used for calculation, such as calculating the distance between the positioning data and the longitude and latitude coordinates of the camera installation location, and determining the maximum visual radius of the camera at the same time, comparing the distance with the visual radius. If the distance is less than the visual radius, it is determined that the target person is located in the visual field of the camera, and the camera is used as the camera to be selected. If the distance is greater than the visual radius, the camera cannot monitor the target person, so the camera cannot be used as the camera to be selected for the target person. Subsequently, the camera to monitor the target person can be selected from these cameras to be selected, and then scheduled.

[0153] After determining the camera to be selected, it is also necessary to determine which visual domain the camera will monitor. All its visual domains are extracted from the camera configuration data set, and then the visual domains that can cover the target person's positioning data are determined one by one as the visual domain to be selected. In order to monitor the target person as much as possible and to the maximum extent, the visual domain with the largest area to be selected is used as the target visual domain, thereby obtaining the target visual domain corresponding to each target person, and then performing global scheduling.

[0154] In addition, when there is resource competition among cameras in the target area, or when a target person is monitored to have illegal behavior / is in an illegal state, global scheduling is also required. Resource competition situations include: the number of target persons in the target area is greater than the number of cameras, or a camera cannot monitor multiple target persons at the same time in the direction of one of its visual fields, but these multiple target persons cannot be monitored by any other camera. If a target person is monitored to have illegal behavior, it is necessary to prioritize monitoring of the illegal behavior / is in an illegal state to ensure the safety of the target person and ensure that security incidents are responded to in a timely manner. It also helps to promptly discover potential safety hazards, achieve refined management, and improve the level and efficiency of safety management of target areas and target persons.

[0155] In another case, if there are idle cameras, you can call the idle cameras after the scheduling is completed to achieve multi-camera multi-view monitoring.

[0156] Optionally, the globally scheduling each camera configured in the target area according to the target visual area includes:

[0157] For each of the target persons, determine the cameras associated with the target person as associated cameras, and summarize them into a camera association set;

[0158] Determine other target persons associated with the target person, and aggregate the target person and the other associated target persons into a person association set;

[0159] According to the camera association set, the person association set and the target visual area corresponding to each target person in the person association set, establishing each camera allocation combination corresponding to the target person;

[0160] Screening from each of the camera allocation combinations to determine a target combination corresponding to the target person;

[0161] Based on the target combination corresponding to each target person, each camera configured in the target area is globally scheduled.

[0162] Specifically, although global scheduling is performed, the operation must be carried out from the perspective of each target person. From a global perspective, a target person corresponds to one or more cameras, and these one or more cameras may correspond to other target persons. Therefore, the target person and the camera will affect each other. Therefore, when performing global scheduling, it is very important to clarify the target persons and cameras that can affect each other, so as to avoid a greater impact on other target persons and cameras. It may even be possible that global scheduling will not cause any impact on certain target persons and cameras. Therefore, the concept of link is introduced. For each target person, the process of determining the cameras associated with the target person and the other target persons associated with the target person is actually carried out in an interlaced manner. That is to say, after determining an associated camera, this associated camera will correspond to one or more other target persons, and these one or more other target persons may correspond to other cameras. Then, by repeating the cycle, you can get the associated cameras and associated persons on the entire link, and then summarize them into sets, and then combine the target visual area to establish a camera allocation combination.

[0163] After determining multiple camera allocation combinations, it is determined that each camera allocation combination can meet the monitoring of part / all of the target personnel. However, in order to achieve better scheduling, cover more target personnel, or make the coverage rate of special situations higher, it is necessary to select the best one as the target combination corresponding to the target personnel for scheduling.

[0164] It can be understood that the present application considers that the target combinations corresponding to the target persons belonging to the same link are the same.

[0165] The process of screening from each of the camera allocation combinations to determine the target combination may include:

[0166] Matching each of the camera allocation combinations with a pre-established personnel scheduling strategy to obtain a first score set corresponding to each of the camera allocation combinations;

[0167] Screening each of the first score sets according to a preset optimization rule to obtain a screening result;

[0168] If the screening result includes only one first score set, the first score set is used as the target combination;

[0169] If the screening result includes at least two first score sets, the at least two first score sets are matched with the pre-established camera scheduling strategies respectively to obtain corresponding second score sets respectively, and the second score sets are screened according to the preferred rule to determine the target combination.

[0170] Specifically, setting the personnel scheduling strategy and camera scheduling strategy can minimize the number and range of camera scheduling, while maximizing the coverage of target personnel. Taking the power work area as an example, the personnel scheduling strategy is shown in Table 3 below:

[0171] Table 3

[0172]

[0173] The personnel scheduling strategy contains 7 behaviors / states, each of which corresponds to a priority. The priority level is based on the serial number in Table 3, that is, priority 1 is the highest, and so on. Except for the 7th priority, the other behaviors / states indicate that the target personnel are in illegal or dangerous behaviors / states, so the first 6 priorities are relatively high, and their corresponding behaviors / states need to be monitored intensively.

[0174] The camera scheduling strategy is shown in Table 4 below:

[0175] Table 4

[0176]

[0177] In order to achieve maximum monitoring, the cameras are also prioritized, specifically explained as follows:

[0178] Priority 1: For example, there are two situations now, 1) Camera a monitors target person 1 and target person 2 at the same time, and camera b monitors target person 3; 2) Camera a monitors target person 4, and camera b monitors target person 3. Now you need to choose from these two situations. It can be seen that situation 1) can monitor three target persons, while situation 2) can only monitor two target persons. Therefore, at this time, you need to choose situation 1).

[0179] Priority 2: For example, the existing cameras a and b can both monitor target person 1, where camera a faces the front of target person 1 and camera b faces the back of target person 1, then camera a is selected to monitor target person 1.

[0180] Priority 3: During global scheduling, for a target person and a camera, if the target person can be monitored without any scheduling of the camera, the camera will be assigned to the target person for monitoring.

[0181] Priority 4: During global scheduling, for a target person and a camera, if there is no need to replace other cameras to monitor the target person, and only the direction (visual field) of the camera needs to be changed, then the camera is assigned to the target person and monitored according to the changed visual field.

[0182] It should be noted that although the personnel scheduling strategy and the camera scheduling strategy are two different strategies, the overall priority of the personnel scheduling strategy is higher than that of the camera scheduling strategy. Therefore, the priorities in the camera scheduling strategy can also be postponed according to the priorities in the personnel scheduling strategy, that is, they are postponed to priority 8, priority 9, priority 10, and priority 11 respectively. Therefore, in the above process, the personnel scheduling strategy is matched first, and then the camera scheduling strategy is matched. In this way, in some cases, only the personnel scheduling strategy can be matched to determine the target combination.

[0183] Take the personnel scheduling strategy as an example to illustrate the process of matching and obtaining the first score set:

[0184] First, an empty set [(),(),…,()] is constructed according to the priority number of the personnel scheduling strategy, where the number of elements is the priority number, and the position order of the elements in the empty set is the same as the level order of the priority. Then, according to the priority level set in the personnel scheduling strategy, from priority 1 to priority 7, it is determined whether the camera scheduling strategy contains the behavior / state involved in the priority, and the number of personnel involved in the behavior / state involved in the priority is determined, and the number of personnel is filled in the corresponding position of the empty set, such as "0", "1", "2", etc. For example, in an example, a first score set is: [(0), (1),…].

[0185] The preferred rule is: taking the process of screening the first score set as an example, each first score set is compared, and the sizes of the elements corresponding to priority 1 in the first score set are first compared, and the first score set where the largest element is located is screened out. At the same time, the number of screened first score sets needs to be determined: if the number is 1, the camera allocation combination corresponding to the first score set is directly used as the target combination; if the number is not 1, the sizes of the elements corresponding to priority 2 in the screened first score sets are compared, and so on.

[0186] Then in this process, before comparing priority 7, only one first score set can be screened out, or only after comparing priority 7 can only the only first score set be finally determined, and even after comparing priority 7, at least two first score sets remain. Among them, the situation of at least two remaining means that simply screening with the personnel scheduling strategy cannot obtain an optimal target combination, then it is necessary to use the camera scheduling strategy to continue screening the remaining at least two first score sets. The screening process is the same as the screening method with the personnel scheduling strategy, which will not be repeated here.

[0187] In an extreme case, if at least two second score sets remain after the screening according to the camera scheduling strategy is completed, one is randomly selected and the corresponding camera allocation combination is used as the target combination.

[0188] In one example, a substation scenario is simulated, such as Figure 5 As shown in the figure, there are a total of 12 cameras configured in the target area. The total range of the visual field of each camera (the range surrounded by the dotted box) is quantized into a 6×6 grid. Each grid corresponds to a visual field and a set of PTZ parameters. There are a total of five target persons in the target area, among which target person 5 does not wear a safety helmet, and the other four target persons behave normally. The arrow indicates the moving direction of target person 5. It can be seen that target person 5 moves from the visual field of camera a to the visual field where camera a and camera b intersect. In addition, target persons 1, 3 and 4 are currently only in the visual field of camera a, and target person 2 is only in the visual field of camera f.

[0189] Below, a, b, c, d, e, f, g, h, i, g, k, and l represent the 12 cameras respectively, and 1, 2, 3, 4, and 5 represent the five target persons respectively. The target persons on the same link are 1, 3, 4, and 5. Starting from any one of these four target persons, we can get the corresponding three camera allocation combinations, which are [a-1, b-5], [a-3, a-4, b-5], and [a-5]. According to the personnel scheduling strategy, the first score set of the camera allocation combination [a-1, b-5] is [(), (), (), (), (), (1, (1)], and the camera allocation combination [a-3, a-4, b-5] is [a-2, b-3, and b-4]. The first score set of camera allocation combination [a-5] is [(), (), (), (), (), (), (1), (2)], and the first score set of camera allocation combination [a-5] is [(), (), (), (), (), (), (1, (0)]. According to the optimization rule, it can be seen that the screening cannot be completed by comparing the sizes of the elements corresponding to priority 1~6. Only by comparing the sizes of the elements corresponding to priority 7 can the target combination be finally determined. The sizes of the elements corresponding to priority 7 in these three first score sets are 2>1>0. Therefore, the camera allocation combination [a-3,a-4,b-5] corresponding to the first score set with the element corresponding to priority 7 as "2" is taken as the target combination of target persons 1, 3, 4, and 5.

[0190] For target person 2, there is no other associated target person, nor any other associated camera except camera f, so its target combination is [f-2].

[0191] The following embodiment describes in detail the process of selecting the optimal camera from the cameras configured in the target area for scheduling based on all target persons currently monitored by the camera in the present application to complete the camera scheduling.

[0192] All target persons currently monitored by the camera are regarded as first persons;

[0193] Selecting cameras corresponding to all the first persons from cameras configured in the target area as corresponding cameras;

[0194] Determine the current north direction, randomly select a first person from each of the first persons, use the location of the positioning data of the first person as the origin, and establish a rectangular coordinate system based on the origin and the current north direction;

[0195] Selecting a corresponding camera in each of the four quadrants of the rectangular coordinate system as each second camera;

[0196] The best camera is selected from each of the second cameras.

[0197] Specifically, before the above process, in order to prevent misjudgment, a time threshold and a distance threshold can be set, such as 5 seconds and 0.5 meters. After 5 seconds, determine whether the target person under the camera has moved more than 0.5 meters within these 5 seconds. If not, return to the step of pulling the surveillance video data of the camera and re-compare the number of people, so as to improve efficiency; if so, execute the above step of treating all target persons currently monitored by the camera as the first person. The corresponding camera refers to a camera that can monitor all first persons at the same time.

[0198] It can be understood that, with the location of the first person's positioning data as the origin, the direction formed by it and the north direction is the direction of the y-axis, thereby forming the y-axis to establish a rectangular coordinate system, please refer to Figure 6 , Figure 6 The image is divided into four quadrants, namely the first quadrant, the second quadrant, the third quadrant and the fourth quadrant. Then, in each quadrant, a corresponding camera is determined as the second camera. For each quadrant, the camera closest to the origin can be used as the second camera, thereby obtaining four second cameras.

[0199] Then, the process of selecting the best camera from each of the second cameras may include:

[0200] For each of the second cameras, pull the surveillance video data of the second camera, identify the number of monitored persons in the surveillance video data of the second camera, and determine the number of first persons;

[0201] Determine the relationship between the number of second cameras whose number of monitored persons is not less than the number of first persons and a preset value, and set the preset value to 1;

[0202] If the number is equal to 1, the second camera is taken as the optimal camera;

[0203] If the number is greater than 1, it means that more than one second camera is qualified as the optimal camera. Then, the second camera with the closest distance to the origin can be selected as the optimal camera.

[0204] If the number is less than 1, it means that there is no second camera that can fully observe all real people. Then the number of monitored personnel of these four second cameras should be compared, and the second camera with the largest number of monitored personnel should be selected as the optimal camera.

[0205] Furthermore, the present application can also utilize positioning data to fully record the movement trajectory of the target personnel in the target area, and use cameras to record behavior, to build a monitoring file of the target personnel throughout the entire process, to facilitate subsequent work analysis, risk assessment, and operation process playback, and to provide strong support for safety management.

[0206] and Figure 1 Corresponding to the method described above, the embodiment of the present invention also provides a camera scheduling device for Figure 1 In the specific implementation of the method, the camera scheduling device provided in the embodiment of the present invention can be used in a computer terminal or various mobile devices, combined with Figure 7 , the camera scheduling device is introduced, such as Figure 7 As shown, the device may include:

[0207] A positioning data acquisition module 10 is used to acquire the positioning data of one or more target persons in the target area in real time;

[0208] The camera allocation judgment module 20 is used to judge, for each target person, whether the target person has been currently assigned a camera for monitoring the target person;

[0209] A first scheduling module 30 is used to schedule various cameras configured in the target area according to the positioning data of the target person and the camera monitoring the target person if the target person has been currently assigned a camera monitoring the target person;

[0210] The second scheduling module 40 is used to globally schedule various cameras configured in the target area if the target person is not currently assigned a camera to monitor him;

[0211] A personnel number determination module 50 is used to pull the surveillance video data of each camera after scheduling, identify the number of monitored personnel in the surveillance video data, and determine the actual number of personnel corresponding to the camera according to the positioning data;

[0212] Return to the execution module 60, which is used to determine all target persons currently monitored by the camera if the number of monitored persons is less than the number of real persons, and select the best camera from the cameras configured in the target area for scheduling based on all target persons currently monitored by the camera, so as to complete the camera scheduling;

[0213] The scheduling completion module 70 is used to complete the camera scheduling if the number of monitored personnel is not less than the number of actual personnel.

[0214] It can be seen from the above technical scheme that, on the one hand, the present application obtains the positioning data of the target personnel in real time, and judges whether to allocate the camera to monitor it, so as to perform different operations. If it has been allocated, the camera is scheduled according to the allocated camera and its positioning data, so as to reduce the impact on other cameras; if it has not been allocated, the cameras in the target area are globally scheduled, and the optimal scheduling and monitoring are achieved from a global perspective. In order to ensure the accuracy of monitoring the target personnel, a judgment mechanism is set after the scheduling is completed, and the monitoring video data is pulled, and the number of monitored personnel is identified, and the number of real personnel at the positioning data level is compared. According to the comparison, it is judged whether the scheduling is completed, so as to achieve real-time monitoring and real-time scheduling, and the accuracy of monitoring will not be affected by the random movement of the target personnel, and it can meet the large-scale and dynamic monitoring needs and ensure the safety of the target personnel. On the other hand, the present application efficiently integrates multi-source data such as monitoring data and positioning data, realizes excellent camera scheduling and allocation, can maximize the coverage of target personnel, avoids monitoring blind spots and waste of resources, and improves resource utilization.

[0215] Furthermore, the embodiment of the present application provides a camera scheduling device. Optionally, Figure 8 The hardware structure diagram of the camera scheduling device is shown in FIG. Figure 8 The hardware structure of the camera scheduling device may include: at least one processor 01, at least one communication interface 02, at least one memory 03 and at least one communication bus 04.

[0216] In the embodiment of the present application, the number of the processor 01 , the communication interface 02 , the memory 03 , and the communication bus 04 is at least one, and the processor 01 , the communication interface 02 , and the memory 03 communicate with each other through the communication bus 04 .

[0217] The processor 01 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0218] The memory 03 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), etc., such as at least one disk memory.

[0219] The memory stores a program, and the processor can call the program stored in the memory, and the program is used to execute the following camera scheduling method, including:

[0220] Acquire the location data of one or more target persons in the target area in real time;

[0221] For each of the target persons, determining whether the target person has been currently assigned a camera for monitoring the target person;

[0222] If the target person has been currently assigned a camera to monitor him, scheduling each camera configured in the target area according to the positioning data of the target person and the camera to monitor him;

[0223] If the target person is not currently assigned a camera to monitor him, then all cameras configured in the target area are globally scheduled;

[0224] For each camera after scheduling, pull the surveillance video data of the camera, identify the number of monitored persons in the surveillance video data, and determine the actual number of persons corresponding to the camera according to the positioning data;

[0225] If the number of monitored persons is less than the number of real persons, all target persons currently monitored by the camera are determined, and based on all target persons currently monitored by the camera, the best camera is selected from the cameras configured in the target area for scheduling, so as to complete the camera scheduling;

[0226] If the number of monitored personnel is not less than the number of actual personnel, the camera scheduling is completed.

[0227] Optionally, the detailed functions and extended functions of the program may refer to the description of the camera scheduling method in the method embodiment.

[0228] The embodiment of the present application further provides a storage medium, which may store a program suitable for execution by a processor, and when the program is running, controls the device where the storage medium is located to execute the following camera scheduling method, including:

[0229] Acquire the location data of one or more target persons in the target area in real time;

[0230] For each of the target persons, determining whether the target person has been currently assigned a camera for monitoring the target person;

[0231] If the target person has been currently assigned a camera to monitor him, scheduling each camera configured in the target area according to the positioning data of the target person and the camera to monitor him;

[0232] If the target person is not currently assigned a camera to monitor him, then all cameras configured in the target area are globally scheduled;

[0233] For each camera after scheduling, pull the surveillance video data of the camera, identify the number of monitored persons in the surveillance video data, and determine the actual number of persons corresponding to the camera according to the positioning data;

[0234] If the number of monitored persons is less than the number of real persons, all target persons currently monitored by the camera are determined, and based on all target persons currently monitored by the camera, the best camera is selected from the cameras configured in the target area for scheduling, so as to complete the camera scheduling;

[0235] If the number of monitored personnel is not less than the number of actual personnel, the camera scheduling is completed.

[0236] Specifically, the storage medium may be a computer-readable storage medium, and the computer-readable storage medium may be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk or a ROM.

[0237] Optionally, the detailed functions and extended functions of the program may refer to the description of the camera scheduling method in the method embodiment.

[0238] In addition, the functional modules in the various embodiments of the present disclosure can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part. If the function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a live broadcast device, or a network device, etc.) to perform all or part of the steps of the methods of the various embodiments of the present disclosure.

[0239] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0240] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0241] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A camera scheduling method, characterized in that: include: Acquire the location data of one or more target persons in the target area in real time; For each of the target persons, determining whether the target person has been currently assigned a camera for monitoring the target person; If the target person has been currently assigned a camera to monitor him, scheduling the cameras configured in the target area according to the positioning data of the target person and the camera to monitor him; If the target person is not currently assigned a camera to monitor him, then all cameras configured in the target area are globally scheduled; For each camera after scheduling, pull the surveillance video data of the camera, identify the number of monitored persons in the surveillance video data, and determine the actual number of persons corresponding to the camera according to the positioning data; If the number of monitored persons is less than the number of real persons, all target persons currently monitored by the camera are determined, and based on all target persons currently monitored by the camera, the best camera is selected from the cameras configured in the target area for scheduling, so as to complete the camera scheduling; If the number of monitored personnel is not less than the number of actual personnel, the camera scheduling is completed.

2. The method according to claim 1, characterized in that The step of scheduling each camera configured in the target area according to the positioning data of the target person and the camera monitoring the target person includes: Taking the camera that monitors the object as a first camera, determining a visual domain currently monitored by the first camera; Determine whether the location data of the target person is within the currently monitored visual domain; If the positioning data of the target person is within the currently monitored visual area, the first camera is dispatched to continue monitoring the target person.

3. The method according to claim 2, characterized in that Also includes: If the positioning data of the target person is not located within the currently monitored visual area, all target persons currently monitored by the first camera are determined as first persons; Determining location data of each of the first persons; Extracting each visual field of the first camera from a pre-established camera configuration data set; Determining whether the positioning data of each of the first persons and the positioning data of the target person are both located within at least one visual field of the first camera; If yes, dispatching the first camera to continue monitoring the target person; If not, all cameras configured in the target area are globally scheduled.

4. The method according to any one of claims 1 to 3, characterized in that The global scheduling of the cameras configured in the target area includes: Based on the positioning data, determining each camera to be selected corresponding to the target person from each camera; Extracting each visual area of ​​each of the to-be-selected cameras from a pre-established camera configuration data set; For each visual zone of each of the to-be-selected cameras, determining whether the positioning data of the target person is located within the visual zone; If yes, the visual domain is used as the visual domain to be selected, thereby obtaining one or more visual domains to be selected corresponding to the target person; Determining the area of ​​each of the visual areas to be selected; The visual domain to be selected with the largest area is used as the target visual domain corresponding to the target person; According to the target visual area, each camera configured in the target area is globally scheduled.

5. The method according to claim 4, characterized in that The globally scheduling each camera configured in the target area according to the target visual area includes: For each of the target persons, determine the cameras associated with the target person as associated cameras, and summarize them into a camera association set; Determine other target persons associated with the target person, and aggregate the target person and the other associated target persons into a person association set; According to the camera association set, the person association set and the target visual area corresponding to each target person in the person association set, establishing each camera allocation combination corresponding to the target person; Screening from each of the camera allocation combinations to determine a target combination corresponding to the target person; Based on the target combination corresponding to each target person, each camera configured in the target area is globally scheduled.

6. The method according to claim 5, characterized in that The screening from each of the camera allocation combinations to determine the target combination includes: Matching each of the camera allocation combinations with a pre-established personnel scheduling strategy to obtain a first score set corresponding to each of the camera allocation combinations; Screening each of the first score sets according to a preset optimization rule to obtain a screening result; If the screening result includes only one first score set, the first score set is used as the target combination; If the screening result includes at least two first score sets, the at least two first score sets are matched with the pre-established camera scheduling strategies respectively to obtain corresponding second score sets respectively, and the second score sets are screened according to the preferred rule to determine the target combination.

7. The method according to claim 1, characterized in that The method of selecting the best camera from the cameras configured in the target area for scheduling based on all target persons currently monitored by the camera to complete the camera scheduling includes: All target persons currently monitored by the camera are regarded as first persons; Selecting cameras corresponding to all the first persons from cameras configured in the target area as corresponding cameras; Determine the current north direction, randomly select a first person from each of the first persons, use the location of the positioning data of the first person as the origin, and establish a rectangular coordinate system based on the origin and the current north direction; Selecting a corresponding camera in each of the four quadrants of the rectangular coordinate system as each second camera; The best camera is selected from each of the second cameras.

8. A camera scheduling device, characterized in that: include: A positioning data acquisition module is used to acquire the positioning data of one or more target persons in the target area in real time; A camera allocation judgment module is used to judge, for each target person, whether the target person has been currently assigned a camera for monitoring the target person; A first scheduling module is used to schedule various cameras configured in the target area according to the positioning data of the target person and the camera monitoring the target person if the target person has been currently assigned a camera monitoring the target person; A second scheduling module is used to globally schedule various cameras configured in the target area if the target person is not currently assigned a camera for monitoring him; A personnel number determination module is used to pull the surveillance video data of each camera after scheduling, identify the number of monitored personnel in the surveillance video data, and determine the actual number of personnel corresponding to the camera according to the positioning data; A first scheduling completion module is used for determining all target persons currently monitored by the camera if the number of monitored persons is less than the number of real persons, and selecting the best camera from the cameras configured in the target area for scheduling based on all target persons currently monitored by the camera, so as to complete the camera scheduling; The second scheduling completion module is used to complete the camera scheduling if the number of monitored personnel is not less than the number of real personnel.

9. A camera scheduling device, characterized in that: including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement the various steps of the camera scheduling method as described in any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the camera scheduling method as described in any one of claims 1 to 7 is implemented.

Images