Multi-camera monitoring method for grouping and polling automatic group flow and related device

Abstract

The invention discloses a multi-camera monitoring method for grouping and polling automatic group flow and a related device, and relates to the field of artificial intelligence or finance. The cameras deployed in a to-be-tested place are divided into a plurality of camera groups, and processes with the same number as the cameras contained in the camera groups are created. The plurality of camera groups are provided with a polling sequence, and if the number of workers in the video image collected by the camera group polled for the first time is greater than or equal to a preset compliance number, subsequent polling does not need to be carried out; if the number of the workers in the video images collected by the camera group subjected to first polling is smaller than the preset compliance number, in order to avoid wrong violation early warning, second polling is needed, namely, the number of the workers in the video images collected by the other camera group is obtained, and the number of the workers in the video images collected by the other camera group is obtained; and if the sum of the number of the workers in the video images acquired by the two groups of polled camera groups is greater than or equal to the preset compliance number, subsequent polling is not needed, and so on, so that the requirements on computing power and bandwidth are reduced.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a multi-camera monitoring method and related devices for group polling and automatic group flow. Background Art

[0002] Financial enterprises, such as banks, need to install cameras in some places to detect whether the number of people entering the place meets the requirements through the video captured by the cameras. For example, the number of people entering the bank's sorting room cannot be less than 2. If the number of people entering the bank's sorting room is less than 2, a violation warning will be issued.

[0003] If the area of ​​the venue is large, multiple cameras need to be deployed to monitor the entire venue. Multiple cameras can capture images of all areas of the venue. In the relevant technology, it is necessary to determine the number of people in real time from the images captured by multiple cameras to determine whether the number of people entering the venue meets the requirements.

[0004] The related technology requires real-time processing of images captured by multiple cameras, which requires high computing power and network link bandwidth. Summary of the invention

[0005] In view of the above problems, the present application provides a multi-camera monitoring method and related devices for group polling and automatic group flow, so as to achieve the purpose of reducing computing power requirements. The specific scheme is as follows:

[0006] The first aspect of the present application provides a multi-camera monitoring method for group polling automatic group flow, comprising:

[0007] Obtaining a polling order of multiple camera groups, wherein different camera groups contain the same number of cameras;

[0008] Create a number of processes equal to the number of cameras included in the camera group;

[0009] Based on the polling order and the camera groups that have been polled, determining a target camera group that currently needs to be polled;

[0010] For each of the processes, determining a first number of staff members located in a collection area of ​​the camera based on a video image collected in real time by one of the cameras in the target camera group through the process, wherein different processes correspond to different cameras in the target camera group;

[0011] Determine through the main process whether the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is greater than or equal to the preset compliance number, if so, determine that it meets the specification, if not, return to the step based on the polling order and the camera groups that have been polled, determine the target camera group that needs to be polled currently, until all the camera groups have been polled; the main process is any process among the multiple processes;

[0012] If all the camera groups have been polled and the sum of the first numbers corresponding to the cameras in all the camera groups that have been polled is less than the preset compliance number, it is determined through the main process that it does not meet the specifications.

[0013] In a possible implementation, the step of obtaining the polling order of multiple camera groups includes:

[0014] Obtaining the collection areas corresponding to the multiple cameras located at the place to be monitored;

[0015] Determine that the cameras corresponding to the acquisition areas meeting the set conditions belong to the same camera group;

[0016] Obtaining first probabilities that the staff members are located in the collection areas corresponding to the respective camera groups;

[0017] The polling order is determined as a result of sorting the camera groups from high to low according to the first probability.

[0018] In a possible implementation, the setting condition is:

[0019] The time taken for a worker located in the collection area of ​​one camera to move to the collection area of ​​another camera is less than or equal to a preset time threshold.

[0020] In a possible implementation, the step of determining that the cameras corresponding to the acquisition areas meeting the set conditions belong to the same camera group includes:

[0021] Obtaining second probabilities that the staff members are located in the collection areas corresponding to the cameras;

[0022] It is determined that the cameras corresponding to the difference of the second probability that falls within a preset range belong to the same camera group.

[0023] In a possible implementation, the second probability step of obtaining the workers' locations in the collection areas corresponding to the cameras includes:

[0024] The historical activity information of the staff in the collection areas respectively corresponding to all the cameras is input into a pre-built probability prediction model, and the second probabilities respectively corresponding to all the cameras are obtained through the probability prediction model.

[0025] In a possible implementation, the method further includes:

[0026] The process started first among the multiple processes is determined as the main process.

[0027] In a possible implementation, the step of determining the first number of staff members located in the acquisition area of ​​the camera based on the video image acquired in real time by one of the cameras in the target camera group through the process includes:

[0028] The process inputs the video image captured by the camera in real time into a pre-built staff type differentiation model, and the first number of staff members corresponding to the target staff type in the capture area of ​​the camera is obtained through the staff type differentiation model.

[0029] A second aspect of the present application provides a multi-camera monitoring device for group polling and automatic group flow, comprising:

[0030] A first acquisition module is used to acquire the polling order of multiple camera groups, where different camera groups have the same number of cameras;

[0031] A creation module, used to create the same number of processes as the number of cameras included in the camera group;

[0032] A first determining module, used for determining a target camera group that needs to be polled currently based on the polling order and the camera groups that have been polled;

[0033] A second determination module is used to determine, for each of the processes, a first number of staff members located in a collection area of ​​the camera based on a video image collected in real time by one of the cameras in the target camera group through the process, wherein different processes correspond to different cameras in the target camera group;

[0034] A third determination module is used to determine whether the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is greater than or equal to a preset compliance number through a main process, and if so, determine that it meets the specification; if not, trigger the first determination module until all the camera groups have been polled; the main process is any process among the multiple processes;

[0035] The fourth determination module is used to determine that it does not meet the specifications through the main process if all the camera groups have been polled and the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is less than the preset compliance number.

[0036] A third aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an edge computing box, the edge computing box implements the multi-camera monitoring method of group polling and automatic group flow according to the first aspect or any implementation method of the first aspect.

[0037] A fourth aspect of the present application provides an edge computing box, comprising at least one processor and a memory connected to the processor, wherein:

[0038] The memory is used to store computer programs;

[0039] The processor is used to execute the computer program so that the edge computing box can implement the multi-camera monitoring method of group polling and automatic group flow according to the above-mentioned first aspect or any implementation method of the first aspect.

[0040] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an edge computing box, the edge computing box can implement the multi-camera monitoring method of group polling and automatic group flow according to the first aspect or any implementation method of the first aspect.

[0041] By means of the above technical scheme, the present application provides a multi-camera monitoring method for group polling automatic group flow, in which the cameras deployed at the place to be tested are divided into multiple camera groups, and the same number of processes as the number of cameras contained in the camera group are created, thereby greatly reducing the number of processes running simultaneously and reducing the demand for computing power. Multiple camera groups are provided with a polling order. If the number of staff members in the video image collected by the camera group of the first polling is greater than or equal to the preset compliance number, there is no need for subsequent polling, thereby saving the bandwidth resources of the camera and the edge computing box, and further reducing the demand for computing power; if the number of staff members in the video image collected by the camera group of the first polling is less than the preset compliance number, in order to avoid erroneous violation warnings, a second polling is required, that is, obtaining the number of staff members in the video image collected by another camera group, and if the sum of the number of staff members in the video image collected by the two camera groups polled is greater than or equal to the preset compliance number, there is no need for subsequent polling, thereby saving the bandwidth resources of the camera and the edge computing box, and further reducing the demand for computing power. The same is true for the follow-up, which will not be repeated here. In summary, compared with the related art, the embodiments of the present application greatly reduce the computing power requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0043] Figure 1 A schematic diagram of a system architecture provided for this application;

[0044] Figure 2 A schematic diagram of the structure of a server 200 provided in this application;

[0045] Figure 3 A schematic diagram of a flow chart of a multi-camera monitoring method for group polling and automatic group flow provided in an embodiment of the present application;

[0046] Figure 4 A schematic diagram of the structure of a multi-camera monitoring device for group polling and automatic group flow provided in an embodiment of the present application;

[0047] Figure 5 This is a schematic diagram of the structure of the edge computing box in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0050] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] It should be noted that the information of the staff involved in this application (including but not limited to user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the staff or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0052] See also Figure 1 , Figure 1 A schematic diagram of a system architecture is shown. The system may include an edge computing box 100, a server 200, and multiple cameras 300.

[0053] The server 200 may include one or more servers ( Figure 1 A server is included as an example for explanation).

[0054] The camera 300 is used to capture still images or videos. The subject (object or organism) generates an optical image through the lens and projects it onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the optical signal into an electrical signal, which is then transmitted to the ISP for conversion into a digital image signal.

[0055] Exemplarily, a plurality of cameras 300 are deployed at a location to be monitored. Exemplarily, the location to be monitored may be a bank sorting room, a bank vault, or a bank loading area.

[0056] Exemplarily, the camera 300 may be connected to an NVR (Network Video Recorder), which receives, stores and manages the video images captured by the camera.

[0057] For example, the edge computing box 100 can obtain corresponding real-time video images from the video streams transmitted to the NVR by multiple cameras 300. The multi-camera monitoring method of group polling automatic group stream provided in the embodiment of the application is used to determine whether the number of staff entering the monitored place is compliant.

[0058] Exemplarily, the edge computing box 100 may include a storage unit, a power distribution unit, a computing unit, and the like.

[0059] Exemplarily, the edge computing box 100 sends the result of whether the number of staff entering the monitored place is compliant to the server 200. If the number of staff entering the monitored place is not compliant, the edge computing box 100 will issue a violation alarm and simultaneously push the real-time monitoring screen at the time of violation to the server 200.

[0060] Exemplarily, the server 200 may send the violation alarm to the corresponding management personnel via SMS, phone call, corporate WeChat and other channels.

[0061] Next describe Figure 1 The product form of the server 200;

[0062] Figure 2 A schematic diagram of the structure of a server 200 is provided, such as Figure 2 As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.

[0063] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0064] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0065] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0066] It should be understood that the above-mentioned edge computing box 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned edge computing box 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with the function of executing instructions, such as CPU, DSP, etc., or a hardware system without the function of executing instructions, such as ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without the function of executing instructions and hardware systems with the function of executing instructions.

[0067] It is understandable that the space of the place to be monitored is large. If a small number of cameras are set up, there may be blind spots in monitoring. If the staff is located in the blind spots, the number of staff who actually enter the place to be monitored may be greater than or equal to the preset compliance number. However, since some staff are located in the blind spots, the number of staff in the video image collected by the camera is less than the preset compliance number, resulting in an error in the violation alarm. Based on this, it is necessary to deploy more cameras in the monitoring place to avoid blind spots in monitoring. The following is an example of the place to be monitored being a bank clearing room.

[0068] The bank clearing room is a very important security location, which is distributed all over the country and involves many high-standard compliance requirements, regulatory requirements and internal control requirements. However, due to the large area of ​​bank clearing rooms and the complex environment with many blind spots, the previous manual monitoring methods have the disadvantages of huge workload, high labor costs, difficulty in complete coverage and easy omissions. With the development of science and technology, it is also possible to use a combination of deep learning and edge computing for intelligent real-time monitoring, but bank clearing rooms are often large in area and complex in environment, and it is difficult for a few cameras to fully cover them. If multiple cameras are used directly for real-time monitoring and reasoning at the same time, the computing power and network link bandwidth requirements of the edge computing box will be relatively high.

[0069] It is understandable that for the financial industry, especially banks, once the cameras in the places to be monitored are deployed, it is very difficult to change the location of the cameras in the future, considering the safety and compliance process issues. In the embodiment of the present application, multiple cameras are deployed in the places to be monitored, and the ratio of the area of ​​the acquisition area of ​​each camera to the area of ​​the places to be monitored is not much different. For example, 8 cameras are deployed in the places to be monitored, and the ratio of the area of ​​the acquisition area of ​​each camera to the area of ​​the places to be monitored is about 12.5%, so it is impossible to use a few cameras as the main cameras, that is, there is no master-slave distinction between the cameras.

[0070] In the related art, it is necessary to create multiple processes at the same time, and the number of multiple processes is the same as the number of cameras deployed in the place to be monitored. Different processes correspond to different cameras, and each process determines the number of staff members located in the collection area of ​​a camera based on the video image collected in real time by a camera. If the sum of the numbers corresponding to all cameras is greater than or equal to the preset compliance number, it is determined to be in compliance with the specification, otherwise, it is not in compliance with the specification. Since multiple processes are started at the same time, the required computing power is very high.

[0071] Based on this, the present application groups the cameras deployed at the places to be monitored to obtain multiple camera groups, and different camera groups contain the same number of cameras; the number of processes created by the present application is the same as the number of cameras contained in the camera group, thereby greatly reducing the number of processes started at the same time and reducing the demand for computing power.

[0072] It can be understood that multiple camera groups all have a polling order, assuming that the polling order is called the first polling, the second polling, the third polling, ... For example, there are 8 cameras deployed in the monitored place, and the 8 cameras are divided into four groups. The first polling refers to polling one camera group in the four camera groups; the second polling refers to polling a group of cameras that have not been polled in the four camera groups; the third polling refers to polling a group of cameras that have not been polled in the four camera groups; the fourth polling refers to polling a group of cameras that have not been polled in the four camera groups; four pollings can poll all camera groups.

[0073] In the process of polling and processing video images collected by multiple camera groups, if the number of staff members in the video images collected by the camera group polled for the first time is greater than or equal to the preset compliance number, no subsequent polling is required, thereby saving bandwidth resources of the camera and edge computing box, and further reducing the demand for computing power; if the number of staff members in the video images collected by the camera group polled for the first time is less than the preset compliance number, in order to avoid erroneous violation warnings, a second polling is required, that is, to obtain the number of staff members in the video images collected by another camera group. If the sum of the number of staff members in the video images collected by the two camera groups is greater than or equal to the preset compliance number, no subsequent polling is required, thereby saving bandwidth resources of the camera and edge computing box, and further reducing the demand for computing power. The same is true for the future, and I will not repeat it here.

[0074] Reference Figure 3 , Figure 3 A flowchart of a multi-camera monitoring method for group polling and automatic group flow provided in an embodiment of the present application is shown as follows: Figure 3 As shown, a multi-camera monitoring method for group polling and automatic group flow provided in an embodiment of the present application may include steps S301 to S306, and these steps are described in detail below.

[0075] Step S301: obtaining a polling order of multiple camera groups, wherein different camera groups include the same number of cameras.

[0076] Exemplarily, the polling order of the multiple camera groups may be random.

[0077] Step S302: creating a number of processes equal to the number of cameras included in the camera group.

[0078] The following example is used to illustrate that if the number of cameras deployed in the place to be monitored is 8, and they are divided into 4 groups, each group has two cameras, then 2 processes can be created. Compared with creating 8 processes in the related art, the computing power requirement is greatly reduced.

[0079] Step S303: Based on the polling sequence and the camera groups that have been polled, determine the target camera group that currently needs to be polled.

[0080] The following example illustrates this.

[0081] Assume that there are 4 camera groups in total, namely: the first camera group, the second camera group, the third camera group and the fourth camera group; if the polling order is: the first camera group → the second camera group → the third camera group → the fourth camera group, if the camera group that has been polled is the second camera group, then the target camera group that needs to be polled currently is the third camera group.

[0082] Step S304: For each of the processes, the first number of staff members located in the acquisition area of ​​the camera is determined based on the video image acquired in real time by one of the cameras in the target camera group through the process, and different processes correspond to different cameras in the target camera group.

[0083] The following description takes the case where the target camera group includes two cameras as an example, assuming that the two cameras in the target camera group are the first camera and the second camera. Assuming that the two processes are the first process and the second process, the first process determines the first number of staff members located in the collection area of ​​the first camera based on the video image collected in real time by the first camera; the second process determines the first number of staff members located in the collection area of ​​the second camera based on the video image collected in real time by the second camera.

[0084] It can be understood that a process may include multiple threads. For example, a process includes three threads, namely the first thread, the second thread and the third thread. The first thread is used to pull the video image from the camera, parse it, and store the parsed video image in the first queue; the second thread is used to obtain the parsed video image from the first queue and analyze the first number of staff members in the video image; the third thread is used to store the first number in the shared memory.

[0085] Step S305: The main process determines whether the sum of the first numbers corresponding to each camera in all camera groups that have been polled is greater than or equal to the preset compliance number. If so, it is determined to be in compliance with the specification. If not, return to step S303 until all the camera groups have been polled.

[0086] The main process is any process among the multiple processes.

[0087] Exemplarily, different processes may communicate with each other through a shared memory. After different processes obtain the first number, the first number may be stored in the shared memory.

[0088] For example, each process processes one frame of video image to obtain the first number. After the process completes processing one frame of video image to obtain the first number and stores the first number in the shared memory, if other processes have not completed processing the video image, the process needs to wait for other processes to complete processing the video image and obtain the first number.

[0089] Exemplarily, any process is randomly selected from the multiple processes as the main process. Exemplarily, the process started first from the multiple processes is determined as the main process.

[0090] In order for those skilled in the art to better understand the processing flow between the various processes in the embodiment of the present application, the following example is used for illustration. Assume that the number of the multiple processes is 2, and they are the first process and the second process respectively; each camera group contains two cameras, namely the first camera and the second camera respectively. The processing flow of the two processes includes the following steps A1 to A3.

[0091] Step A1: The first process determines a first number of staff members located in a collection area of ​​the first camera based on a video image collected in real time by the first camera, and stores the first number in a shared memory.

[0092] Step A2: The second process determines the first number of staff members located in the collection area of ​​the second camera based on the video image collected in real time by the second camera, and stores the first number in the shared memory.

[0093] Step A3: the first process searches the shared memory for the first number stored in the second process, and if not, determines that the first process is the main process; or the second process searches the shared memory for the first number stored in the first process, and if not, determines that the second process is the main process.

[0094] In an optional implementation, after the main process is determined during the first polling of the camera groups, the main process remains unchanged during subsequent polling of other camera groups.

[0095] In an optional implementation, during each polling of the camera group, the main process determined may be the same or different.

[0096] Exemplarily, in order to prevent other processes except the main process from executing the step of "determining whether the sum of the first numbers corresponding to each camera in all camera groups that have been polled is greater than or equal to the preset compliance number", after the main process obtains the first numbers corresponding to each camera in all camera groups that have been polled from the shared memory, the shared memory can be locked.

[0097] Step S306: If all the camera groups have been polled and the sum of the first numbers corresponding to the cameras in all the camera groups that have been polled is less than the preset compliance number, the main process determines that it does not meet the specifications.

[0098] The following is an example of 4 camera groups, each camera group including 2 cameras. In the process of the first polling of the camera group, the video images respectively collected by the two cameras in the camera group at the first collection time are processed to obtain the first number; in the process of the second polling of the camera group, the video images respectively collected by the two cameras in the camera group at the second collection time are processed to obtain the first number; in the process of the third polling of the camera group, the video images respectively collected by the two cameras in the camera group at the third collection time are processed to obtain the first number; in the process of the fourth polling of the camera group, the video images respectively collected by the two cameras in the camera group at the fourth collection time are processed to obtain the first number; it is obvious that the first collection time is earlier than the second collection time, the second collection time is earlier than the third collection time, and the third collection time is earlier than the fourth collection time.

[0099] It can be understood that the processing speed of the process is very fast, and the difference between the fourth collection time and the first collection time is in milliseconds. For example, the difference between the fourth collection time and the first collection time is 300 milliseconds. In a short period of time, the staff cannot teleport from one collection area to another, that is, the same staff member will not be counted twice. Therefore, even if asynchronous polling is used, the number of staff in the monitored place obtained by this application is accurate.

[0100] The embodiment of the present application provides a multi-camera monitoring method for group polling automatic group flow. The cameras deployed at the place to be tested are divided into multiple camera groups, and the same number of processes as the number of cameras contained in the camera group are created, thereby greatly reducing the number of processes running at the same time and reducing the demand for computing power. Multiple camera groups have a polling order. If the number of staff members in the video image collected by the camera group of the first polling is greater than or equal to the preset compliance number, there is no need to perform subsequent polling, thereby saving the bandwidth resources of the camera and the edge computing box, and further reducing the demand for computing power; if the number of staff members in the video image collected by the camera group of the first polling is less than the preset compliance number, in order to avoid erroneous violation warnings, a second polling is required, that is, obtaining the number of staff members in the video image collected by another camera group. If the sum of the number of staff members in the video image collected by the two camera groups that have been polled is greater than or equal to the preset compliance number, there is no need to perform subsequent polling, thereby saving the bandwidth resources of the camera and the edge computing box, and further reducing the demand for computing power. The same is true for the follow-up, and it will not be repeated here. In summary, compared with the related art, the embodiments of the present application greatly reduce the computing power requirements.

[0101] It is understandable that there are multiple ways to implement step S301. The embodiment of the present application provides but is not limited to the following method, which includes but is not limited to the following steps B1 to B4.

[0102] Step B1: Obtain the collection areas corresponding to the multiple cameras located at the place to be monitored.

[0103] Exemplarily, the acquisition area of ​​a camera is determined by the field of view of the camera.

[0104] Step B2: Determine that the cameras corresponding to the acquisition areas meeting the set conditions belong to the same camera group.

[0105] Exemplarily, the set conditions include, but are not limited to: the time taken for a worker located in a collection area of ​​one camera to move to a collection area of ​​another camera is less than or equal to a preset time threshold.

[0106] Exemplarily, the preset time threshold may be determined based on actual conditions, such as 500 milliseconds. If the time to move from one collection area to another is less than 500 milliseconds, it is possible that the same staff member appears in different collection areas within 500 milliseconds (although the probability of this is extremely small, and the judgment is continuous, and an occasional error will not affect the overall judgment result), resulting in an erroneous situation of counting the same staff member twice.

[0107] Exemplarily, step B2 may include the following steps B21 to B22.

[0108] Step B21: Obtain second probabilities that the staff members are located in the collection areas corresponding to the cameras.

[0109] Exemplarily, there are multiple methods for implementing step B21. The embodiment of the present application provides but is not limited to the following method: inputting the historical activity information of the staff in the collection areas corresponding to all the cameras into a pre-built probability prediction model, and obtaining the second probability corresponding to all the cameras through the probability prediction model.

[0110] It is understandable that if the current time is October 25, 2024, then staff have entered the monitored place before October 25, 2024. The historical activity information of the preset time period closest to the current time can be input into the pre-built probability prediction model, and the probabilities corresponding to all the cameras can be obtained through the probability prediction model.

[0111] Assuming that the preset time period is 25 days, the preset time period is between October 1, 2024 and October 25, 2024.

[0112] Exemplarily, the historical activity information includes, but is not limited to: the length of time the staff stays in the collection area, the number of times the staff enters and exits the collection area. When a staff leaves a collection area and then enters the collection area again, it is considered to be one entry and exit of the collection area.

[0113] Exemplarily, the probability prediction model is obtained by training by using the historical activity information of the sample staff as the input of the probability prediction model and using the probability of the sample staff being located in each collection area as the training target.

[0114] Step B22: Determine that the cameras whose corresponding second probability difference values ​​fall within a preset range belong to the same camera group.

[0115] It is understandable that since the probability of a worker moving from one acquisition area to another acquisition area within a short time (such as 500 milliseconds) is extremely low or even nonexistent, the camera groups can be obtained by dividing in the manner of step B22.

[0116] For example, the preset range may be determined based on actual conditions.

[0117] Exemplarily, the preset ranges corresponding to different camera groups are the same.

[0118] Exemplarily, the preset ranges corresponding to different camera groups may be different, for example, if two cameras with the closest corresponding second probabilities are determined to be grouped into the same camera group, then the preset ranges corresponding to different camera groups are different. This is described below with an example.

[0119] Assume that 4 cameras are deployed in the place to be monitored, and the 4 cameras are: the first camera, the second camera, the third camera, and the fourth camera. The second probabilities corresponding to the first camera, the second camera, the third camera, and the fourth camera are: 20.8%, 30.2%, 10%, and 30%, respectively. Assuming that each camera group includes 2 cameras, the second camera and the fourth camera can be divided into a camera group, and the preset range corresponding to the camera group is [0, 0.2%], and the first camera and the third camera are divided into a camera group, and the preset range corresponding to the camera group is (0.2%, 10.8%].

[0120] Step B3: Obtain first probabilities that the staff members are located in the acquisition areas corresponding to the respective camera groups.

[0121] Exemplarily, the acquisition area corresponding to the camera group is obtained by stitching the acquisition areas of each camera in the camera group; the first probability of the acquisition area corresponding to the camera group is the sum of the second probabilities of the acquisition areas corresponding to each camera in the camera group.

[0122] The second probability of the collection area corresponding to the camera refers to the probability that the staff is located in the collection area of ​​the camera.

[0123] Step B4: Determine the polling order as a result of sorting the camera groups from high to low according to the first probability.

[0124] It is understandable that different equipment is deployed in different collection areas, so the first probability of the staff in different collection areas is different. Polling each camera group in order from high to low according to the first probability can quickly determine the situation that meets the specifications when polling a few camera groups, greatly reducing the number of times the camera groups are polled.

[0125] It is understandable that there are many types of staff entering the place to be monitored, and the number of staff of the same type needs to be greater than or equal to the preset compliance number. Assuming that the number of total staff belonging to the target staff type needs to be monitored, the implementation method of step S304 includes the following steps: through the process, the video image collected by the camera in real time is input into a pre-built staff type distinction model, and the first number of staff corresponding to the target staff type located in the collection area of ​​the camera is obtained through the staff type distinction model.

[0126] For example, different types of workers have different work clothes, so the type of worker can be determined based on the worker's attire.

[0127] Exemplarily, the training process of the staff type distinction model includes: taking a sample video image as an input of the staff type distinction model, taking the type of staff in the sample video image as a training target, and training the obtained staff type distinction model.

[0128] In the embodiment of the present application, combined with the idea of ​​time-sharing computing power scheduling, multiple cameras are grouped, and each camera group is polled at the same time. For each camera group, a frame of video image captured by each camera in the camera group is read separately to infer the number of staff members in the monitored place.

[0129] The above introduces a multi-camera monitoring method for group polling and automatic group flow provided by an embodiment of the present application. The following will introduce a device for executing the above-mentioned multi-camera monitoring method for group polling and automatic group flow.

[0130] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of a multi-camera monitoring device for group polling and automatic group flow provided by an embodiment of the present application. Figure 4 As shown, the multi-camera monitoring device for group polling automatic group flow includes:

[0131] A first acquisition module 401 is used to acquire a polling order of multiple camera groups, where different camera groups have the same number of cameras;

[0132] A creation module 402, used to create a number of processes equal to the number of cameras included in the camera group;

[0133] A first determining module 403 is used to determine a target camera group that needs to be polled currently based on the polling order and the camera groups that have been polled;

[0134] A second determining module 404 is used to determine, for each of the processes, a first number of staff members located in a collection area of ​​the camera based on a video image collected in real time by one of the cameras in the target camera group through the process, wherein different processes correspond to different cameras in the target camera group;

[0135] The third determination module 405 is used to determine whether the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is greater than or equal to a preset compliance number through a main process, and if so, determine that it meets the specification; if not, trigger the first determination module until all the camera groups have been polled; the main process is any process among the multiple processes;

[0136] The fourth determination module 406 is used to determine that it does not meet the specifications through the main process if all the camera groups have been polled and the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is less than the preset compliance number.

[0137] In an optional implementation, the first acquisition module includes:

[0138] A first acquisition unit is used to acquire collection areas corresponding to a plurality of cameras located at a place to be monitored;

[0139] A first determining unit, used to determine that the cameras corresponding to the acquisition areas meeting the set conditions belong to the same camera group;

[0140] A second acquisition unit is used to acquire first probabilities that the staff members are located in the acquisition areas corresponding to the respective camera groups;

[0141] The second determining unit is used to determine the polling order as a result of sorting the camera groups from high to low according to the first probability.

[0142] In an optional implementation, the setting condition is:

[0143] The time taken for a worker located in the collection area of ​​one camera to move to the collection area of ​​another camera is less than or equal to a preset time threshold.

[0144] In an optional implementation, the first determining unit includes:

[0145] An acquisition subunit, used for acquiring second probabilities that the staff members are located in the collection areas corresponding to the cameras;

[0146] A determination subunit is used to determine that the cameras corresponding to the difference of the second probability that falls within a preset range belong to the same camera group.

[0147] In an optional implementation, the acquisition subunit is specifically used to: input the historical activity information of the staff in the collection areas corresponding to all the cameras into a pre-built probability prediction model, and obtain the second probability corresponding to all the cameras through the probability prediction model.

[0148] In an optional implementation, the method further includes:

[0149] The fifth determining module is used to determine the process started first from the multiple processes as the main process.

[0150] In an optional implementation, the second determination module is specifically used to: input the video image captured by the camera in real time into a pre-built staff type distinction model through the process, and obtain the first number of staff members corresponding to the target staff type in the capture area of ​​the camera through the staff type distinction model.

[0151] The present application also provides an edge computing box. Figure 5 As shown, it shows a schematic diagram of the structure of an edge computing box suitable for implementing the embodiment of the present application. The edge computing box in the embodiment of the present application may include but is not limited to fixed terminals such as laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 5 The edge computing box shown is merely an example and should not bring any limitations to the functionality and scope of use of the embodiments of the present application.

[0152] like Figure 5 As shown, the edge computing box may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 to a random access memory (RAM) 503. When the edge computing box is powered on, various programs and data required for the operation of the edge computing box are also stored in the RAM 503. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0153] Typically, the following devices may be connected to the I / O interface 505: input devices 506 including, for example, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 508 including, for example, memory cards, hard disks, etc.; and communication devices 509. The communication device 509 may allow the edge computing box to communicate with other devices wirelessly or by wire to exchange data. Figure 5 An edge computing box with various devices is shown, but it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have instead.

[0154] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an edge computing box, the edge computing box implements any of the multi-camera monitoring methods for group polling and automatic group flow provided in an embodiment of the present application.

[0155] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When one or more computer programs are executed by an edge computing box, the edge computing box can implement any of the multi-camera monitoring methods for group polling and automatic group flow provided in an embodiment of the present application.

[0156] It should be noted that the multi-camera monitoring method and related devices provided by the present invention can be used in the field of artificial intelligence or finance. The above is only an example and does not limit the application field of the multi-camera monitoring method and related devices provided by the present invention.

[0157] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0158] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including application-specific integrated circuits, special CPUs, GPUs, TPU special memories, special components, etc. In general, all functions performed by computer programs can be easily implemented with corresponding hardware, and the specific hardware structures used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for this application, software program implementation is a better implementation method in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, U disk, mobile hard disk, ROM, RAM, disk or optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, training equipment, or network equipment, etc.) to execute the methods described in each embodiment of the present application.

[0159] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0160] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A multi-camera monitoring method for group polling automatic group flow, characterized in that: include: Obtaining a polling order of multiple camera groups, wherein different camera groups contain the same number of cameras; Create a number of processes equal to the number of cameras included in the camera group; Based on the polling order and the camera groups that have been polled, determining a target camera group that currently needs to be polled; For each of the processes, determining a first number of staff members located in a collection area of ​​the camera based on a video image collected in real time by one of the cameras in the target camera group through the process, wherein different processes correspond to different cameras in the target camera group; Determine through the main process whether the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is greater than or equal to the preset compliance number, if so, determine that it meets the specification, if not, return to the step based on the polling order and the camera groups that have been polled, determine the target camera group that needs to be polled currently, until all the camera groups have been polled; the main process is any process among the multiple processes; If all the camera groups have been polled and the sum of the first numbers corresponding to the cameras in all the camera groups that have been polled is less than the preset compliance number, it is determined through the main process that it does not meet the specifications.

2. According to the multi-camera monitoring method of group polling automatic group flow according to claim 1, it is characterized in that: The step of obtaining the polling order of multiple camera groups includes: Obtaining the collection areas corresponding to the multiple cameras located at the place to be monitored; Determine that the cameras corresponding to the acquisition areas meeting the set conditions belong to the same camera group; Obtaining first probabilities that the staff members are located in the collection areas corresponding to the respective camera groups; The polling order is determined as a result of sorting the camera groups from high to low according to the first probability.

3. According to claim 2, the multi-camera monitoring method of group polling automatic group flow is characterized in that: The setting conditions are: The time taken for a worker located in the collection area of ​​one camera to move to the collection area of ​​another camera is less than or equal to a preset time threshold.

4. The multi-camera monitoring method of group polling automatic group flow according to claim 2 is characterized in that: The step of determining that the cameras corresponding to the acquisition areas meeting the set conditions belong to the same camera group includes: Obtaining second probabilities that the staff members are located in the collection areas corresponding to the cameras; It is determined that the cameras corresponding to the difference of the second probability that falls within a preset range belong to the same camera group.

5. The multi-camera monitoring method of group polling automatic group flow according to claim 4 is characterized in that: The second probability step of obtaining the location of the staff in the collection area corresponding to each camera includes: The historical activity information of the staff in the collection areas respectively corresponding to all the cameras is input into a pre-built probability prediction model, and the second probabilities respectively corresponding to all the cameras are obtained through the probability prediction model.

6. The multi-camera monitoring method of group polling automatic group flow according to any one of claims 1 to 5, characterized in that: Also includes: The process started first among the multiple processes is determined as the main process.

7. The multi-camera monitoring method according to any one of claims 1 to 5, characterized in that: The step of determining the first number of staff members located in the acquisition area of ​​the camera based on the video image acquired in real time by one of the cameras in the target camera group through the process comprises: The process inputs the video image captured by the camera in real time into a pre-built staff type differentiation model, and the first number of staff members corresponding to the target staff type in the capture area of ​​the camera is obtained through the staff type differentiation model.

8. A multi-camera monitoring device with group polling and automatic group flow, characterized in that: include: A first acquisition module is used to acquire the polling order of multiple camera groups, where different camera groups have the same number of cameras; A creation module, used to create the same number of processes as the number of cameras included in the camera group; A first determining module, used for determining a target camera group that needs to be polled currently based on the polling order and the camera groups that have been polled; A second determination module is used to determine, for each of the processes, a first number of staff members located in a collection area of ​​the camera based on a video image collected in real time by one of the cameras in the target camera group through the process, wherein different processes correspond to different cameras in the target camera group; A third determination module is used to determine whether the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is greater than or equal to a preset compliance number through a main process, and if so, determine that it meets the specification; if not, trigger the first determination module until all the camera groups have been polled; the main process is any process among the multiple processes; The fourth determination module is used to determine that it does not meet the specifications through the main process if all the camera groups have been polled and the sum of the first numbers corresponding to each camera in all the camera groups that have been polled is less than the preset compliance number.

9. A computer program product, characterized in that It includes computer-readable instructions. When the computer-readable instructions are executed on an edge computing box, the edge computing box implements the multi-camera monitoring method of group polling and automatic group flow as described in any one of claims 1 to 7.

10. An edge computing box, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the edge computing box can implement the multi-camera monitoring method of group polling automatic group flow as described in any one of claims 1 to 7.

Images