Recording and broadcasting equipment-based face recognition method and device, storage medium and equipment

Abstract

The invention relates to a face recognition method and device based on recording and broadcasting equipment, a storage medium and equipment, and the method comprises the steps: determining the node information of a target recording and broadcasting equipment node for executing a face recognition task according to the residual calculation power information of each recording and broadcasting equipment node in response to a calling request of recording and broadcasting equipment nodes; the face recognition task is sent to the target recording and broadcasting equipment node to be executed and the face recognition information returned by the target recording and broadcasting equipment node is received, and idle recording and broadcasting equipment computing resources are utilized to assist the central node in performing face recognition calculation, so that the face recognition efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of face recognition, and in particular to a face recognition method, device, storage medium and equipment based on recording and broadcasting equipment. Background Art

[0002] A smart campus refers to an integrated environment for work, study and life on campus based on the Internet of Things. This integrated environment uses various application service systems as carriers to fully integrate teaching, scientific research, management and campus life. For example, facial recognition technology can be used to identify the identities of people entering and leaving the campus, thereby ensuring the safety of school property and students.

[0003] However, when there are large crowds of people, such as when students go to school or go home from school, the amount of face recognition tasks and data increases dramatically, face recognition takes a long time, and the efficiency of face recognition is low. Summary of the invention

[0004] The embodiments of the present application provide a face recognition method, apparatus, storage medium and device based on a recording and broadcasting device, which can utilize idle resources of a recording and broadcasting machine to assist in face recognition calculations and improve face recognition efficiency.

[0005] In a first aspect, an embodiment of the present application provides a face recognition method based on a recording and broadcasting device, which is applied to a central node, the central node is communicatively connected with at least one recording and broadcasting device node, and one recording and broadcasting device node corresponds to one recording and broadcasting device, and the method comprises the following steps:

[0006] In response to the call request of the recording and broadcasting device node, obtain the remaining computing power information and face recognition tasks of each recording and broadcasting device node;

[0007] Determine the target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information, and obtain the node information of the target recording and broadcasting device node;

[0008] According to the node information, the face recognition task is sent to the target recording and broadcasting device node, and the face recognition information obtained after the target recording and broadcasting device node executes the face recognition task is received.

[0009] In a second aspect, an embodiment of the present application provides a face recognition device based on a recording and broadcasting device, which is applied to a central node, the central node is communicatively connected with at least one recording and broadcasting device node, and one recording and broadcasting device node corresponds to one recording and broadcasting device, and the device includes:

[0010] A computing power information acquisition module is used to respond to a call request from a recording and broadcasting device node and obtain the remaining computing power information and face recognition tasks of each recording and broadcasting device node;

[0011] A node information acquisition module, used to determine the target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information, and acquire the node information of the target recording and broadcasting device node;

[0012] The face recognition information receiving module is used to send the face recognition task to the target recording and broadcasting device node according to the node information, and receive the face recognition information obtained after the target recording and broadcasting device node executes the face recognition task.

[0013] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the face recognition method based on a recording and broadcasting device as described in any one of the above.

[0014] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable by the processor; the processor is communicatively connected to at least one recording and broadcasting device node;

[0015] When the processor executes the computer program, the steps of the face recognition method based on the recording and broadcasting device as described in any one of the above items are implemented.

[0016] In an embodiment of the present application, in response to a call request from a recording and broadcasting device node, the node information of a target recording and broadcasting device node that executes a face recognition task is determined based on the remaining computing power information of each recording and broadcasting device node. By sending the face recognition task to the target recording and broadcasting device node for execution and receiving the face recognition information returned by the target recording and broadcasting device node, idle recording and broadcasting device computing resources are used to assist the central node in performing face recognition calculations, thereby improving face recognition efficiency.

[0017] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of an application scenario of a face recognition method based on a recording and broadcasting device in one embodiment of the present invention;

[0019] Figure 2 This is a flow chart of a face recognition method based on a recording and broadcasting device in one embodiment of the present invention;

[0020] Figure 3 This is a flowchart of a face recognition method based on a recording and broadcasting device in another embodiment of the present invention;

[0021] Figure 4 This is a flowchart of a face recognition method based on a recording and broadcasting device in another embodiment of the present invention;

[0022] Figure 5 is a flow chart of step S120 in one embodiment of the present invention;

[0023] Figure 6 This is a flowchart of a face recognition method based on a recording and broadcasting device in another embodiment of the present invention;

[0024] Figure 7 This is a flowchart of a face recognition method based on a recording and broadcasting device in another embodiment of the present invention;

[0025] Figure 8 This is a structural schematic diagram of a face recognition device based on a recording and broadcasting device in one embodiment of the present invention;

[0026] Fig. 9 The figure is a schematic diagram of the structure of an electronic device in one embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0028] It should be clear that 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 the embodiments of the present application.

[0029] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0030] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0031] In addition, in the description of this application, unless otherwise specified, "several" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0032] Recording and broadcasting equipment can be used to integrate and synchronously record video, audio, and image signals of electronic equipment recorded on site to generate standardized streaming media files, which can then be used for external live broadcast, storage, post-editing, on-demand, etc., and has been widely used in smart campuses.

[0033] In the existing technology, recording and broadcasting equipment usually only has the function of processing recorded audio and video, and its computing resources are not well utilized. At the same time, when performing face recognition analysis on surveillance videos, the existing technology often requires the configuration of multiple high-performance GPU servers to meet the computing power requirements of face recognition analysis. However, high-performance GPU servers are expensive, resulting in high face recognition costs.

[0034] Therefore, in response to the above problems, the present application provides a face recognition method based on a recording and broadcasting device, which can be applied to a central node.

[0035] See also Figure 1 , which is a schematic diagram of an application scenario of a face recognition method based on a recording and broadcasting device in an embodiment; Figure 1 As shown, the scenario includes: a central node 101 and at least one recording and broadcasting device node 102; the recording and broadcasting device node 102 can be communicatively connected with the central node 101 through a common communication connection method such as a network.

[0036] The central node 101 may include a processor and a memory. The memory stores an operating system that enables the central node 101 to operate normally and a management program for managing the recording and broadcasting device node 102. The processor runs the operating system and the management program to manage the recording and broadcasting device node 102.

[0037] Each recording and broadcasting device node 102 corresponds to a recording and broadcasting device. The recording and broadcasting device here can be an existing recording and broadcasting device on campus. There is no need to configure a high-performance GPU server additionally, which can effectively reduce the cost of face recognition.

[0038] The face recognition method based on recording and broadcasting equipment of the present application can make full use of the computing resources of idle or free recording and broadcasting equipment on campus to assist in face recognition, improve face recognition efficiency, and reduce face recognition costs.

[0039] Specifically, see Figure 2The face recognition method based on the recording and broadcasting device of the present application comprises the following steps:

[0040] S110: In response to a call request from a recording and broadcasting device node, obtain remaining computing power information and face recognition tasks of each recording and broadcasting device node;

[0041] The call request is used to call the recording and broadcasting device node to perform the face recognition task. Specifically, the call request of the recording and broadcasting device node can be issued at times when there is a large flow of people, such as when students go to school or go home from school, so as to call the computing resources of the recording and broadcasting device node.

[0042] The remaining computing power information is used to determine the remaining computing power of each recording and broadcasting device node.

[0043] The face recognition task is a face recognition task that needs to be assigned to a recording and broadcasting device node, and the face recognition task may include a video or image to be recognized.

[0044] S120: determining a target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information, and acquiring node information of the target recording and broadcasting device node;

[0045] When determining the target recording and broadcasting device node, the target recording and broadcasting device node may be determined according to the screening conditions set by the user. For example, the recording and broadcasting device node with the largest remaining computing power may be used as the target recording and broadcasting device node.

[0046] The node information may include the name, address, and other information of the target recording and broadcasting device node. In an embodiment of the present application, each recording and broadcasting device node may be pre-registered in the central node, and the central node may include the node information of each pre-registered recording and broadcasting device node. After determining the target recording and broadcasting device node, the central node obtains the pre-registered node information of the target recording and broadcasting device node.

[0047] S130: Sending the face recognition task to the target recording and broadcasting device node according to the node information, and receiving face recognition information obtained after the target recording and broadcasting device node executes the face recognition task.

[0048] The target recording and broadcasting device node is a node for performing face recognition tasks determined by the user based on the remaining computing power information.

[0049] The target recording and broadcasting device node may include one recording and broadcasting device node or multiple recording and broadcasting device nodes. When the target recording and broadcasting device node includes multiple recording and broadcasting device nodes, the central node may divide the face recognition task into a plurality of face recognition subtasks according to the division conditions preset by the user, determine the recording and broadcasting device node corresponding to each face recognition subtask, and send the face recognition subtask to the corresponding recording and broadcasting device node. Specifically, when dividing the face recognition task, it may be divided into executable minimum computing task packages.

[0050] Facial recognition information may include information such as the features of the recognized face and the identity corresponding to the face.

[0051] In an embodiment of the present application, in response to a call request from a recording and broadcasting device node, the node information of a target recording and broadcasting device node that executes a face recognition task is determined based on the remaining computing power information of each recording and broadcasting device node. By sending the face recognition task to the target recording and broadcasting device node for execution and receiving the face recognition information returned by the target recording and broadcasting device node, idle recording and broadcasting device computing resources are used to assist the central node in performing face recognition calculations, thereby improving face recognition efficiency.

[0052] In step S110, the call request of the recording and broadcasting device node can be generated according to the control instruction issued by the user to the central node, or, in one embodiment, the call request of the recording and broadcasting device node can be generated by the central node when it is judged that its own computing power can hardly meet the requirements of the face recognition task. Specifically, the central node can judge whether its own computing power meets the requirements of the face recognition task according to the judgment conditions pre-set by the user. For example, the judgment condition can be that the current time period is in a time period with high traffic flow, or the current traffic flow at a certain location exceeds a threshold. At this time, it is determined that the central node's own computing power can hardly meet the requirements of the face recognition task, and a call request for the recording and broadcasting device node is generated.

[0053] like Figure 3 As shown, in an optional embodiment, before obtaining the remaining computing power information and face recognition tasks of each recording and broadcasting device node, the following steps are also included:

[0054] S210: Get current time;

[0055] S220: If the current time is within the target time period, generate a call request for the recording and broadcasting device node;

[0056] The target time period may be a time period in which the amount of face recognition tasks and data will increase dramatically, and the time period may be determined by collecting and counting the campus flow at historical time points. Specifically, it may be considered that there are fewer people active on campus during class time and at night, that is, the amount of face recognition tasks is small, and the target time period may be the first N minutes of the student's class time and / or the last M minutes of the student's get out of class time, and / or the break time, where N and M may be any natural numbers and may be determined based on the statistical results of the campus flow at each time period.

[0057] In an embodiment of the present application, if the current time is within the target time period when the amount of face recognition tasks increases dramatically, a call request for the recording and broadcasting device node is generated, thereby utilizing the surplus computing power of the recording and broadcasting device to share the face recognition tasks of the central node, thereby reducing the cost of face recognition and improving the efficiency of face recognition at times when the flow of people is large.

[0058] Or, if Figure 4 As shown, in another optional embodiment, before obtaining the remaining computing power information and face recognition tasks of each recording and broadcasting device node, the following steps are also included:

[0059] S310: Obtaining the flow of people at the target location;

[0060] S320: If the human flow value at the target location exceeds the target human flow threshold, a call request for a recording and broadcasting device node is generated.

[0061] The target location may be an entrance or exit of a campus or a teacher's entrance or exit, and the crowd flow value at the target location may be obtained by detecting a crowd flow detection device installed at the target location.

[0062] The target traffic threshold can be a traffic threshold pre-set by the user. When the traffic value at the target location exceeds the target traffic threshold, it means that there are a large number of people at the current location, and the face recognition task and data volume are large. At this time, by generating a call request for the recording and broadcasting device node, the recording and broadcasting device node is used to share the face recognition task of the central node, thereby improving the face recognition efficiency at times when the traffic is large.

[0063] When the face recognition task and data volume are small, the face recognition task can be directly performed by the central node without additionally calling the recording and broadcasting device node. Specifically, the face recognition method based on the recording and broadcasting device also includes the following steps:

[0064] If the current time is not within the target time period, or if the flow of people at the target location does not exceed the target flow threshold, the face recognition task is performed and face recognition information is obtained.

[0065] If the current time is not in the target time period, for example, it is class time for students, or it is in a time period with less human activity such as at night, or the flow of people at the target location does not exceed the target flow threshold, it means that the amount of face recognition tasks and data that need to be processed are relatively small. At this time, the central node directly executes the face recognition task and obtains the face recognition information, and the recording and broadcasting device of the recording and broadcasting device node can be in a closed state, thereby saving power.

[0066] In step S120, the remaining computing power information may include the number of physical cores, the average remaining utilization of each physical core per unit time, and the full load computing capacity value of each physical core.

[0067] like Figure 5 As shown, according to the remaining computing power information, determining the target recording and broadcasting device node for executing the face recognition task, and obtaining the node information of the target recording and broadcasting device node specifically include:

[0068] S121: acquiring a state variable of a recording and broadcasting device node according to the number of physical cores, the average remaining utilization rate of each physical core in the unit time, and the full load computing capacity value of each physical core;

[0069] Specifically, the state variables of the recording and broadcasting device node are obtained in the following manner:

[0070] s t，i =m i u i p i

[0071] Among them, s t，i represents the state variable of recording and broadcasting device node i at time t, m i Indicates the number of physical cores of recording and broadcasting device node i, u i represents the average remaining utilization of each physical core of the recording and broadcasting device node i per unit time, p i Indicates the full load computing capacity value of each physical core of recording and broadcasting device node i.

[0072] For n recording and broadcasting device nodes, the system status is:

[0073] S t ={s t，1 ,s t，2 , ..., s t，n}

[0074] Among them, S t Indicates the system status of n recording and broadcasting device nodes at time t, where n indicates the number of recording and broadcasting device nodes.

[0075] S122: constructing a reward function for a recording and broadcasting device node according to the number of physical cores and the full load computing capacity value of each physical core;

[0076] The reward function can be determined based on the benefits that the user needs to obtain. In an embodiment of the present application, the reward function can be constructed for the computing resources of the recording and broadcasting device node. Specifically, the reward function can be constructed based on the product of the number of physical cores and the full-load computing capacity value of each physical core.

[0077] Construct the reward function of the recording and broadcasting device node in the following way:

[0078] R t =m t p t

[0079] Among them, R t represents the reward function of the recording and broadcasting device node at time t, m tIndicates the number of physical cores of the recording and broadcasting device node at time t, p t Indicates the full load computing capacity value of each physical core of the recording and broadcasting device node at time t.

[0080] S123: constructing an action variable based on the recording and broadcasting device node; wherein the action variable indicates that the face recognition task is assigned to the current recording and broadcasting device node;

[0081] Specifically, the action variable a i It means that the face recognition task is assigned to the i-th recording and broadcasting machine node to run, then the action space corresponding to the action variable can be:

[0082] A t ={a1, a2, ..., a n}

[0083] Wherein, n represents the number of recording and broadcasting device nodes.

[0084] S124: Obtain an expected reward value according to the state variable, the reward function and the action variable, obtain an action variable corresponding to the maximum expected reward value based on the Q-Learning reinforcement learning algorithm, and determine a target recording and broadcasting device node according to the action variable.

[0085] Q-Learning reinforcement learning refers to the use of the state-action value function Q(S, A) to find the optimal action selection measurement. The state-action value function Q(S, A) is used to calculate the expected reward value obtained by taking the action in the current state.

[0086] Specifically, when obtaining the action variable corresponding to the maximum expected reward value based on the Q-Learning reinforcement learning algorithm, first, the Q table and action table A are initialized; the action variables in the action space are selected based on the greedy strategy, and the expected reward value is obtained through the state variables, reward function and action variables to update the value of the Q table, and the action variable that can obtain the maximum benefit is determined according to the value of the Q table, thereby determining the target recording and broadcasting device node.

[0087] The Q table may include expected reward values ​​obtained by taking different actions in each state, the value of the initialized Q table may be 0, and the action table A includes various actions.

[0088] The greedy strategy refers to taking the best or optimal (i.e. most favorable) choice in each step when solving a problem, hoping to lead to the best or optimal algorithm.

[0089] Specifically, in an embodiment of the present application, when selecting action variables in the action space based on a greedy strategy, a probability value ranging from 0 to 1 is first randomly generated. If the probability value is greater than or equal to the target threshold ε, the action with the largest expected reward value is selected. If the probability value is less than the target threshold ε, an action in the action space is randomly selected. The present application can avoid the Q-Learning reinforcement learning algorithm from falling into local search and improve the accuracy of the calculation by randomly generating probability values. The probability value can be generated using an existing random generation algorithm.

[0090] Specifically, update the expected reward value in the Q table as follows:

[0091] Q(S t , A t )′←Q(S t , A t )+α[R t+1 +γmaxQ(S t+1 , A t+1 )-Q(S t , A t )]

[0092] Among them, Q(S t , A t )′ represents the expected reward value at time t after update, Q(S t , A t ) represents the expected reward value at time t, α represents the learning rate, which represents the magnitude of each Q value update and determines the convergence speed of the algorithm, γ represents the discount factor, R t+1 represents the expected reward value at time t+1, maxQ(S t+1 , A t+1 ) represents the maximum expected reward value at time t+1.

[0093] In the present application, when the value of the Q table is updated, the iteration may be stopped when all recording and broadcasting machine nodes reach full load calculation, or the iteration may be stopped when the target number of iterations is reached, wherein the target number of iterations may be set according to actual needs. Alternatively, in another embodiment, the iteration may be stopped when other conditions for stopping iteration set by the user are met, and after the iteration is stopped, the action variable that can obtain the maximum benefit is determined according to the value of the Q table, thereby determining the target recording and broadcasting device node.

[0094] The face recognition method based on recording and broadcasting equipment of the present application can be applied to track the trajectory of key personnel on campus, such as Figure 6 As shown, in one embodiment, the face recognition task includes a video, and the method further includes the following steps:

[0095] S410: Determine whether the video contains a target person according to the face recognition information, and if the video contains the target person, store the video in a storage path corresponding to the target person;

[0096] S420: In response to the target person tracking instruction, a target video is obtained from a storage path corresponding to the target person, and activity track information of the target person on campus is obtained according to the target video.

[0097] The target person in the embodiment of the present application may be a student, teacher or other key personnel on campus.

[0098] The video may be a surveillance video collected by a surveillance camera on campus.

[0099] To determine whether the target person is included in the video, the face recognition information may be compared with the face recognition information of the target person to determine whether the target person is included in the video.

[0100] The face recognition information of the target person may be information pre-input by the user, and the face recognition information of the target person may include the image, face features, identity and other information of the target person.

[0101] In the present application, facial recognition information may include facial features. When the facial features in the video are consistent with the facial features of the target person, it is determined that the video contains the target person. When the facial features in the video are inconsistent with the facial features of the target person, it is determined that the video does not contain the target person.

[0102] When the video contains a target person, the video containing the target person is stored in the storage path corresponding to the target person. The storage path here can correspond to a target person, so that when the user's target person tracking instruction is received, the video containing the target person can be obtained from the storage path, and the target person's activity trajectory on campus can be tracked according to the video content, so as to facilitate users to monitor the target person's activity trajectory on campus.

[0103] like Figure 7 As shown, in another embodiment, the face recognition method based on the recording and broadcasting device of the present application can be applied to track the trajectory of strangers on campus. Specifically, the face recognition task includes a video, and the method further includes the following steps:

[0104] S510: Determine whether the video contains non-campus personnel according to the face recognition information, and if the video contains non-campus personnel, store the video in a storage path corresponding to the non-campus personnel;

[0105] S520: In response to the non-campus personnel tracking instruction, a target video is obtained from a storage path corresponding to the non-campus personnel, and activity track information of the non-campus personnel on campus is obtained according to the target video.

[0106] To determine whether the video contains non-campus personnel, the facial recognition information can be compared with the facial recognition information of campus personnel to determine whether the video contains non-campus personnel.

[0107] The facial recognition information of campus personnel may be information of one or more campus personnel pre-entered by the user, and the facial recognition information of campus personnel may include images, facial features, identities, and other information of the campus personnel.

[0108] In the present application, non-campus personnel may refer to the current person who is not a campus person, that is, a person who is not registered in the preset database, such as a stranger, etc. Campus personnel may be a person pre-registered in the database, such as a student, teacher or other faculty member. Face recognition information may include facial features. When the facial features in the video are consistent with the facial features of campus personnel, it is determined that the video does not contain non-campus personnel. When the facial features in the video are inconsistent with the facial features of campus personnel, it is determined that the video contains non-campus personnel.

[0109] When the video contains non-campus persons, the video containing the non-campus persons will be stored in the storage path corresponding to the non-campus persons. The storage path here can be a separate path, so that when a non-campus person tracking instruction is received, the video containing the non-campus persons can be obtained from the storage path, and the activity trajectory of the non-campus persons on campus can be tracked according to the video content, so as to facilitate users to monitor the activity trajectory of non-campus persons on campus and ensure campus safety.

[0110] See also Figure 8 The present application also provides a face recognition device based on a recording and broadcasting device, which is applied to a central node, the central node is communicatively connected with at least one recording and broadcasting device node, and one recording and broadcasting device node corresponds to one recording and broadcasting device, and the device includes:

[0111] The computing power information acquisition module 201 is used to respond to the call request of the recording and broadcasting device node and obtain the remaining computing power information and face recognition tasks of each recording and broadcasting device node;

[0112] The node information acquisition module 202 is used to determine the target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information, and acquire the node information of the target recording and broadcasting device node;

[0113] The face recognition information receiving module 203 is used to send the face recognition task to the target recording and broadcasting device node according to the node information, and receive the face recognition information obtained after the target recording and broadcasting device node executes the face recognition task.

[0114] In one embodiment, the node information acquisition module 202 includes:

[0115] A state variable acquisition unit, used to acquire the state variable of the recording and broadcasting device node according to the number of physical cores, the average remaining utilization rate of each physical core in the unit time, and the full load computing capacity value of each physical core;

[0116] A reward function construction unit, used to construct a reward function of a recording and broadcasting device node according to the number of physical cores and the full load computing capacity value of each physical core;

[0117] An action variable construction unit, used to construct an action variable based on a recording and broadcasting device node; wherein the action variable indicates that a face recognition task is assigned to a current recording and broadcasting device node;

[0118] The target recording and broadcasting device node determination unit is used to obtain the expected reward value according to the state variable, the reward function and the action variable, obtain the action variable corresponding to the maximum expected reward value based on the Q-Learning reinforcement learning algorithm, and determine the target recording and broadcasting device node according to the action variable.

[0119] In one embodiment, the reward function is constructed based on the product of the number of physical cores and the full load computing capacity value of each physical core.

[0120] In one embodiment, the face recognition task includes a video, and the face recognition apparatus based on the recording and broadcasting device further includes:

[0121] A target person detection unit, used to determine whether the video contains a target person according to the face recognition information, and if the video contains the target person, store the video in a storage path corresponding to the target person;

[0122] The target person tracking unit is used to respond to the target person tracking instruction, obtain the target video from the storage path corresponding to the target person, and obtain the activity track information of the target person in the campus according to the target video.

[0123] In one embodiment, the face recognition task includes a video, and the face recognition apparatus based on the recording and broadcasting device further includes:

[0124] A non-campus person detection unit, used to determine whether the video contains non-campus people according to the face recognition information, and if the video contains non-campus people, store the video in a storage path corresponding to the non-campus people;

[0125] The non-campus person tracking unit is used to respond to a non-campus person tracking instruction, obtain a target video from a storage path corresponding to the non-campus person, and obtain activity track information of the non-campus person in the campus according to the target video.

[0126] In one embodiment, the face recognition device based on the recording and broadcasting device further includes:

[0127] Time acquisition module, used to obtain the current time;

[0128] A first call request generating module, used to generate a call request for a recording and broadcasting device node if the current time is within a target time period;

[0129] Alternatively, the face recognition device based on the recording and broadcasting device further includes:

[0130] A crowd flow value acquisition module is used to obtain the crowd flow value of the target location;

[0131] The second call request generating module is used to generate a call request for the recording and broadcasting device node if the pedestrian flow value at the target location exceeds the target pedestrian flow threshold.

[0132] In one embodiment, the face recognition device based on the recording and broadcasting device further includes:

[0133] The face recognition task execution module is used to execute the face recognition task and obtain face recognition information if the current time is not within the target time period, or if the flow of people at the target location does not exceed the target flow threshold.

[0134] It should be noted that the face recognition device based on the recording and broadcasting device provided in the above embodiment only uses the division of the above functional modules as an example when executing the face recognition method based on the recording and broadcasting device. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the face recognition device based on the recording and broadcasting device provided in the above embodiment and the face recognition method based on the recording and broadcasting device in the above embodiment belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0135] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the face recognition method based on the recording and broadcasting device as described in any one of the above are implemented.

[0136] The present application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0137] See also Fig. 9 , the present application also provides an electronic device 300, including a memory 301, a processor 302, and a computer program stored in the memory 301 and executable by the processor 302; the processor 302 is communicatively connected with at least one recording and broadcasting device node;

[0138] When the processor 302 executes the computer program, the steps of the face recognition method based on the recording and broadcasting device as described in any one of the above are implemented.

[0139] The present invention is not limited to the above-mentioned embodiments. If various changes or modifications to the present invention do not depart from the spirit and scope of the present invention, and if these changes and modifications fall within the scope of the claims and equivalent technologies of the present invention, the present invention is also intended to include these changes and modifications.

Claims

1. A face recognition method based on recording and broadcasting equipment, characterized in that: The method is applied on a central node, the central node is communicatively connected with at least one recording and broadcasting device node, and one recording and broadcasting device node corresponds to one recording and broadcasting device. The method comprises the following steps: In response to the call request of the recording and broadcasting device node, obtain the remaining computing power information and face recognition tasks of each recording and broadcasting device node; Determine the target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information, and obtain the node information of the target recording and broadcasting device node; According to the node information, the face recognition task is sent to the target recording and broadcasting device node, and the face recognition information obtained after the target recording and broadcasting device node executes the face recognition task is received.

2. The face recognition method based on recording and broadcasting equipment according to claim 1, characterized in that: The remaining computing power information includes the number of physical cores, the average remaining utilization rate of each physical core per unit time, and the full load computing capacity value of each physical core. The step of determining the target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information specifically includes: Acquire a state variable of a recording and broadcasting device node according to the number of physical cores, the average remaining utilization rate of each physical core in the unit time, and the full load computing capacity value of each physical core; Constructing a reward function for a recording and broadcasting device node according to the number of physical cores and the full load computing capacity value of each physical core; Constructing an action variable based on the recording and broadcasting device node; wherein the action variable indicates that the face recognition task is assigned to the current recording and broadcasting device node; The expected reward value is obtained according to the state variable, the reward function and the action variable, the action variable corresponding to the maximum expected reward value is obtained based on the Q-Learning reinforcement learning algorithm, and the target recording and broadcasting device node is determined according to the action variable.

3. The face recognition method based on recording and broadcasting equipment according to claim 2 is characterized in that: The reward function is constructed based on the product of the number of physical cores and the full load computing capacity value of each physical core.

4. The face recognition method based on recording and broadcasting equipment according to claim 1, characterized in that: The face recognition task includes a video, and the method further includes the following steps: Determine whether the video contains a target person according to the face recognition information, and if the video contains the target person, store the video in a storage path corresponding to the target person; In response to a target person tracking instruction, a target video is obtained from a storage path corresponding to the target person, and activity track information of the target person on campus is obtained according to the target video.

5. The face recognition method based on recording and broadcasting equipment according to claim 1, characterized in that: The face recognition task includes a video, and the method further includes the following steps: Determine whether the video contains non-campus characters according to the face recognition information, and if the video contains non-campus characters, store the video in a storage path corresponding to the non-campus characters; In response to a non-campus personnel tracking instruction, a target video is obtained from a storage path corresponding to the non-campus personnel, and activity track information of the non-campus personnel on campus is obtained based on the target video.

6. The face recognition method based on recording and broadcasting equipment according to claim 1, characterized in that: Before obtaining the remaining computing power information and face recognition tasks of each recording and broadcasting device node, the following steps are also included: Get the current time; If the current time is within the target time period, a call request for the recording and broadcasting device node is generated; Alternatively, obtain the flow of people at the target location; If the human flow value at the target location exceeds the target human flow threshold, a call request for the recording and broadcasting device node is generated.

7. The face recognition method based on recording and broadcasting equipment according to claim 6 is characterized in that: The following steps are also included: If the current time is not within the target time period, or if the flow of people at the target location does not exceed the target flow threshold, the face recognition task is performed and face recognition information is obtained.

8. A face recognition device based on a recording and broadcasting device, characterized in that: The device is applied on a central node, the central node is communicatively connected with at least one recording and broadcasting device node, one recording and broadcasting device node corresponds to one recording and broadcasting device, and the device includes: A computing power information acquisition module is used to respond to a call request from a recording and broadcasting device node and obtain the remaining computing power information and face recognition tasks of each recording and broadcasting device node; A node information acquisition module, used to determine the target recording and broadcasting device node for executing the face recognition task according to the remaining computing power information, and acquire the node information of the target recording and broadcasting device node; The face recognition information receiving module is used to send the face recognition task to the target recording and broadcasting device node according to the node information, and receive the face recognition information obtained after the target recording and broadcasting device node executes the face recognition task.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the face recognition method based on the recording and broadcasting device as described in any one of claims 1 to 7 are implemented.

10. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable by the processor; the processor is communicatively connected with at least one recording and broadcasting device node; When the processor executes the computer program, the steps of the face recognition method based on the recording and broadcasting device as described in any one of claims 1 to 7 are implemented.

Images